PowerMac 1.6GHz Benchmarks [Archive]

Longey Nowze

Aug 20, 2003, 10:17 PM

for those who think benchmarks are useless: why do car manufactures tell you how fast their car goes from 0-100km/h? and then tell you how fast it goes from 100Km/h-0? why do they give you fuel consumption numbers?

I know some benchmarks are useless for most people sometimes as useless as telling you how fast the windows go up in a car! SPEC are useless to me but some benchmarks show where the computer is faster and where it is slower this SP G5 is faster in memory due to the faster bus which is a good thing to know! real world tests are ok and when you give us PS tests even though important to a lot of people will make wintel users will never shut up, but do we really care?! they will say stuff like it's written for mac! it's the only fast up on mac! which I think is stupid! a all apps should be optimised! most software being slow today isn't the chips fault! it's bad programming! look at how fast X is getting with every update?! I'm a firm believer that we don't need faster CPUs as much as we need better programming! CPUs have evolved so much but programming seems to be well... it's not catching up!

to the person who asked about browsers that support DP I believe OmniWeb and Netscape maybe even mozilla support DP, I'm not sure if safari does I haven't read anything that says it does...

THANK YOU
MaT

mvc

Aug 20, 2003, 10:20 PM

Originally posted by Powerbook G5
Benchmarks are just a nice, modern-day replacement for pissing contests :) Much cleaner, too.

LOL - look, ma, see how far along the bench my mark is!

No, when using a computer 12 hours a day 6 days a week (and having no life), trust me, its all about the interface being Snappy™

Thats the only benchmark that matters to me. Bring on Panther and a G5, and then my daily work experience might finally be as snappy as XP on my 1.8ghz P4! :(

BrandonRP0123

Aug 20, 2003, 10:22 PM

Just by reading the G5 performance primer, it's news to me that the G5 setup doesn't have an L3 cache. Interesting. Although, I think the L3 cache was put into the G4 mix to make it faster than it really was (as even the 900Mhz G3 lacks an L3?)

Longey Nowze

Aug 20, 2003, 10:22 PM

Originally posted by Fender2112
hmmm...I may have to rethink this whole G5 thing. I really don't want to get a Dell. I'm sure I can figure out something. Do you thing I'll void the warranty if I cut the off the handles with a hacksaw?

why don't you make it support the desk?! :p

or a better idea why don't you extend the legs of the desk?

Powerbook G5

Aug 20, 2003, 10:26 PM

Originally posted by Fender2112
hmmm...I may have to rethink this whole G5 thing. I really don't want to get a Dell. I'm sure I can figure out something. Do you thing I'll void the warranty if I cut the off the handles with a hacksaw?

I know this might sound a bit stupid...but how about getting a new desk? :)

BrandonRP0123

Aug 20, 2003, 10:28 PM

Originally posted by Fender2112

Anyone know if the handles are removable?. The G5 measures 20.1" and I only have 18.75" under my desk. I didn't realize it was so big. I could set it off to the side but the cabels will most likely be about 2" too short. :(

Sounds like you need some cinder blocks. I know a guy that made cutouts in his CPU cabinet part of his desk for room for the handles on the G4 (instead of taking them off). The handles stuck out a bit through the top of his desk and he used them for cable running.

The fact that the G5 is 20.1" tall with non removable handles dashes my hopes of spending 1999 on a 1.6 and using it as a rack-mouted server versus 3000 on an xserve.

Stella

Aug 20, 2003, 10:29 PM

The G5 doesn't support L3 cache IIRC because it doesn't need it.

Originally posted by BrandonRP0123
Just by reading the G5 performance primer, it's news to me that the G5 setup doesn't have an L3 cache. Interesting. Although, I think the L3 cache was put into the G4 mix to make it faster than it really was (as even the 900Mhz G3 lacks an L3?)

PieMac

Aug 20, 2003, 10:34 PM

Originally posted by Fender2112
After seeing the G5 next to that G4, I checked the dimentions and the G5 is not going to fit under my desk, or on top, or under. I may have to put it in the garage. :)

Anyone know if the handles are removable?. The G5 measures 20.1" and I only have 18.75" under my desk. I didn't realize it was so big. I could set it off to the side but the cabels will most likely be about 2" too short. :(

I'm assuming that the 20.1" measurement includes the handles?

daRAT

Aug 20, 2003, 10:34 PM

Originally posted by Longey Nowze
for those who think benchmarks are useless: why do car manufactures tell you how fast their car goes from 0-100km/h? and then tell you how fast it goes from 100Km/h-0? why do they give you fuel consumption numbers?

I know some benchmarks are useless for most people sometimes as useless as telling you how fast the windows go up in a car! SPEC are useless to me but some benchmarks show where the computer is faster and where it is slower this SP G5 is faster in memory due to the faster bus which is a good thing to know! real world tests are ok and when you give us PS tests even though important to a lot of people will make wintel users will never shut up, but do we really care?! they will say stuff like it's written for mac! it's the only fast up on mac! which I think is stupid! a all apps should be optimised! most software being slow today isn't the chips fault! it's bad programming! look at how fast X is getting with every update?! I'm a firm believer that we don't need faster CPUs as much as we need better programming! CPUs have evolved so much but programming seems to be well... it's not catching up!

to the person who asked about browsers that support DP I believe OmniWeb and Netscape maybe even mozilla support DP, I'm not sure if safari does I haven't read anything that says it does...

THANK YOU
MaT

Yes, and those car "benchmarks" are done by professional drivers, not your average person. Thats why the gas mileage is an estimate.

I also would like to hear from an average user that has a G5, and how it does on PS, Bryce, Painter, downloading photos from a camera, copying from the harddrive to an external drive...

You know, normal day to day chores ;]

mim

Aug 20, 2003, 10:36 PM

Originally posted by mvc
LOL - look, ma, see how far along the bench my mark is!

No, when using a computer 12 hours a day 6 days a week (and having no life), trust me, its all about the interface being Snappy™

Thats the only benchmark that matters to me. Bring on Panther and a G5, and then my daily work experience might finally be as snappy as XP on my 1.8ghz P4! :(

Yeh, we all keep talking about this, and I think we all agree that it's often the little things that save you time and frustration...but seriously, your XP box is snappy and your mac is not? That is really the opposite of my (and many others here experience). I have a pretty new machine at work, and some fairly old macs that we use for dtp/etc. Now those macs aren't speed demons when it comes to processing huge photoshop files, but for zooming in/out resizing the work window you can't beat them.

It may be a memory thing. Probably registry thing too (for windows...) my xp box had a LOT of crud on it while the macs are faily clean. But they are easier to keep clean.

Anyway, it's good to see the G5's arriving. I hope when they start putting the 970 into iMacs that the Powermacs all go to Dp's though. Not only to offset the cost of developing the DP boards, but it should give Apple less of a reason to restrict the consumer machine speeds.

rice_web

Aug 20, 2003, 10:39 PM

I'm still torn three ways: single 1.2, dual-867, or single 1.8 G5. Though, I'm now leaning towards the single 1.2.

I was amazed at the low scores of the G5, however, I'm still waiting on a few more benchmarks.

Powerbook G5

Aug 20, 2003, 10:42 PM

Well, using both OS 9 and XP, I've found that nothing is as snappy as a well maintained Mac OS 9 running on my G3...I can only imagine how fast it would be on a newer G4 system. It makes XP seem slow, for sure, though. If Panther is as snappy as my PowerBook G3 400 MHz running at 400 MHz, I'd be really happy. If a G5 cannot produce those results, then there is something wrong...

mim

Aug 20, 2003, 10:45 PM

Originally posted by BrandonRP0123
Just by reading the G5 performance primer, it's news to me that the G5 setup doesn't have an L3 cache. Interesting. Although, I think the L3 cache was put into the G4 mix to make it faster than it really was (as even the 900Mhz G3 lacks an L3?)

Stella's correct. The memory is running at full bus speed (or vise-versa), making L3 irrelevent. Those 8Gb's ARE your L3 :D

BrandonRP0123

Aug 20, 2003, 10:50 PM

Originally posted by mim
Stella's correct. The memory is running at full bus speed (or vise-versa), making L3 irrelevent. Those 8Gb's ARE your L3 :D

Forgive my ignorance, but the DDR RAM Stevie presented as the G5 using at WWDC is DDR 400Mhz. At best (assuming that DDR means you get double the bandwidth out of it), that only gives you 800Mhz of bandwidth to play with. While this is fine for the 1.6 its not as fast as the system bus in the 1.8 (900Mhz) and the 2.0 (1000Mhz)

Anyone know how this works?

Powerbook G5

Aug 20, 2003, 10:51 PM

Wait until faster RAM comes out, I suppose

rice_web

Aug 20, 2003, 10:53 PM

The Ancients

Aug 20, 2003, 10:53 PM

Reliability, that's why!

At least the Fiesta would get you there, AND back...

Originally posted by Powerbook G5
Why drive to the grocery store in a Ford Festiva when you can drive to the grocery store in a twin turbo Lotus? :D

Longey Nowze

Aug 20, 2003, 10:53 PM

Originally posted by daRAT
Yes, and those car "benchmarks" are done by professional drivers, not your average person. Thats why the gas mileage is an estimate.

I also would like to hear from an average user that has a G5, and how it does on PS, Bryce, Painter, downloading photos from a camera, copying from the harddrive to an external drive...

You know, normal day to day chores ;]

that's kinda my point, the G5 is a Pro machine, I think it will be used by professionals more than consumers, and some car programmes have lap times rather than from the 0-100 like BBC's Top Gear which gives you a better way to compare cars, so PS FCP and Maya and the like should be better tests to run compiling software code is too, but isn't that what some benchmarks do? I don't know about downloading photos from a camera isn't that dependant on the camera? and weather it uses USB or FW? and the type of storage it has? I don't think the computer has much to do with its speed.

Thank You
MaT

adamfilip

Aug 20, 2003, 10:55 PM

Personally i think the benchmark results SUCK!

i want more then this!

MacAztec

Aug 20, 2003, 10:55 PM

Originally posted by rice_web
I'm still torn three ways: single 1.2, dual-867, or single 1.8 G5. Though, I'm now leaning towards the single 1.2.

I was amazed at the low scores of the G5, however, I'm still waiting on a few more benchmarks.

What? Low scores?!?

That thing is as fast as a Dual 1.42 in most tests, and faster in some more! And remember, this is a SINGLE processor.

It would SMOKE a Dual 867 or a Single 1.2GHz G4.

Has xBench fully supported the G5 yet? I think once all the apps that were used are optimized (like photoshop was), there will be better numbers.

NavyIntel007

Aug 20, 2003, 10:57 PM

Originally posted by Powerbook G5
Why drive to the grocery store in a Ford Festiva when you can drive to the grocery store in a twin turbo Lotus? :D

Speaking of cars... I was driving through miami today and saw a Delorean...

Unfortunately someone painted it red... :mad:

adamfilip

Aug 20, 2003, 11:00 PM

a painted Delorean.. thats criminal!

Mac Kiwi

Aug 20, 2003, 11:06 PM

Thats interesting Cinebench is C4D.Until recently G5s have been crashing on the Cinebench OGL test while running 8.1,so something must have been fixed just recently.

When Cinebench was introduced the PCs were on average getting double the Macs scores and we the Mac C4D users took a lot of flak for it was well.

http://www.imashination.com/bench.html

BrandonRP0123

Aug 20, 2003, 11:07 PM

Originally posted by adamfilip
a painted Delorean.. thats criminal!

Just as bad as the dummies that have old corvettes (you know, with popup headlights) and have installed after market `kits' to rid them of it.

Its weird seeing an old vette with bolt on fog lights and headlamps.

NavyIntel007

Aug 20, 2003, 11:08 PM

The benchmarks suck...

This could be because 10.2.7 isn't close to being the optimal OS for the G5 and/or xBench is not yet set up to test the G5. I'm willing to bet it's both. when 10.3 comes out I imagine the G5 will start to pull away from the pack when it comes to speed.

If I had a Dual G4 >= 1 Ghz, I wouldn't upgrade. The dualies will make up some for the lack of speed. Single processor G4's... probably consider the 1.8 G5 for the PCI-X.

But sadly, I need a small laptop... so I got the 12" powerbook. Damn Military service, I should have got a real job! :D

Powerbook G5

Aug 20, 2003, 11:08 PM

I've seen a red DeLorean, too...maybe it's the same car...I'm only 35-40 minutes from Miami. It is criminal to paint one, it'd be like someone painting a G5 or PowerBook lime green... *cough Alienware cough*

NavyIntel007

Aug 20, 2003, 11:09 PM

Originally posted by BrandonRP0123
Just as bad as the dummies that have old corvettes (you know, with popup headlights) and have installed after market `kits' to rid them of it.

Its weird seeing an old vette with bolt on fog lights and headlamps.

Yeah but painting a delorean... that's just wrong... I almost pulled over to buy it from him. On the way home, stop by HomeDepot and get paint thinner and start working.

BrandonRP0123

Aug 20, 2003, 11:12 PM

Originally posted by Powerbook G5
*cough Alienware cough*

Ah yes. Here we go :D

Wonder Boy

Aug 20, 2003, 11:13 PM

the tower looks good, though i am still distraught about only one optical drive.

Mr.Hey

Aug 20, 2003, 11:16 PM

Originally posted by Fender2112
hmmm...I may have to rethink this whole G5 thing. I really don't want to get a Dell. I'm sure I can figure out something. Do you thing I'll void the warranty if I cut the off the handles with a hacksaw?

Cut your damn desk not the G5...WHAT THE HELL IS WRONG WITH YOU :confused: :mad: :rolleyes: ;) :) :D

notEven

Aug 20, 2003, 11:18 PM

[QUOTE]Originally posted by Longey Nowze
[b]for those who think benchmarks are useless: why do car manufactures tell you how fast their car goes from 0-100km/h? and then tell you how fast it goes from 100Km/h-0? why do they give you fuel consumption numbers?

i think thats a faulty argument.. or analogy.. but i agree specs are very important. i think specs in general are only helpful to the well informed - and thats only assuming the specs themselves are acurate. we absolutely need them ...or else we'd have nothing to base our purchases except for blind faith and brand loyalty.

but anyway, general reception of the G5 (apple demos, pixar switch, etc.) seem real positive. and personally, im getting a dual as soon as financially possible.
i cant possibly see a downside to that purchase. ill gladly pay for 10.3 (update or full) - i think apple has been trying as hard as possible to get us the G5s.... "smeagol" might seem to be a compromise on the surface.. but to me, thats a sign of customer satisfaction.

holmesf

Aug 20, 2003, 11:19 PM

Originally posted by scem0
yes, a lot of surfing.

But a lot of programming too (the more the better ;)). I know C++, Im learning basic and java. I hope to be using all 3 of those languages a lot.

And with the amount of time I spend on my computer even if all I did was surf it wouldnt be overkill. If I save a hundredth of a second on every webpage I went to I'd save a BUNCH of time.

scem0

Why would you learn basic AFTER C++?

btw, using Java with Project builder is a really poor implimentation. If that's what you're doing, I'd suggest Cocoa.

adamfilip

Aug 20, 2003, 11:20 PM

im having trouble understanding the benchmark results..
why are there two sets of number 1) and 2) which machine is which?

where can i find cinebench results to compare too

ColdZero

Aug 20, 2003, 11:23 PM

Originally posted by mim
Stella's correct. The memory is running at full bus speed (or vise-versa), making L3 irrelevent. Those 8Gb's ARE your L3 :D

The memory isn't running at full bus speed. The FSB is 1/2 the clock speed of the processor. This FSB speed is to the chipset, not to memory. Memory will run at 400MHz or whatever it runs at to the chipset, not the processor.

porovaara

Aug 20, 2003, 11:25 PM

That people will complain about anything.

Witness:

The new LOW END machine is overall faster than the previous top of the line machine and people are complaining about the speed.

You do realize this is the first time Apple (or any company I can think of save Digital with the Alpha) has released a new line of machines that are bottom to top faster than their previous line. Don't you people remember the slow g4s? How on earth can you complain about the new low end box in un-optimized benchmarks besting the previous top of the line?

If anything this benchmarks shows the 1.6ghz part isn't just trash like most people had assumed.

MacRETARD

Aug 20, 2003, 11:32 PM

Originally posted by BrandonRP0123
Forgive my ignorance, but the DDR RAM Stevie presented as the G5 using at WWDC is DDR 400Mhz. At best (assuming that DDR means you get double the bandwidth out of it), that only gives you 800Mhz of bandwidth to play with. While this is fine for the 1.6 its not as fast as the system bus in the 1.8 (900Mhz) and the 2.0 (1000Mhz)

Anyone know how this works?

DDR 400 memory is 200 mhz that is effectively 400 because access is 2 times per cycle.

The G5 uses a dual channel memory architecture so it will be 800 mhz. The 1ghz bus has some overhead so even though the memory is not running at 1ghz you should still see some performance gain in memory related tests.

BTW - since it is dual channel you will have to upgrade memory 2 sticks at a time.

adamfilip

Aug 20, 2003, 11:34 PM

okay well here is a comparison of the G5 and my Athlon 1.4 running Cinebench 2003

G5 1.6 on Left Athlon 1.4 on right
Single CPU Render test:
158.2 sec 152.2sec for PC

Open GL Hardware Lighting test:

1) 13.16 sec; 7.8 fps 11.89; 8.6 fps
2) 4.18 sec; 21.5 fps 8.69, 10.4 fps

1466559 polygons/sec 706440 p/sec

Open GL Software Lighting test:

1) 19.96 sec; 5.1 fps 15.96; 6.4fps
2) 9.30 sec; 9.7 fps 10.54; 8.5 fps

659952 polygons/sec 582421 p/sec

Cinema4D Shading test:

1) 50.19 sec; 2.0 fps 62.23; 1.6fps
2) 24.62 sec; 3.7 fps 24.70; 3.6fps

249291 polygons/sec 248514 p/sec

overall my new g5 1.6 isnt very much better (speed wise) to my current Athlon 1.4.

luckily i can look forward to running jaguar and panther and not XP anymore.. but i was hoping for more of a performance Jump

mt pc is also running an All in Wonder Raedon 7500

Longey Nowze

Aug 20, 2003, 11:42 PM

notEven why is it faulty :( I thought i did a great job :( I was proud of myself :( no I'm sad :( at least we agree :)

the right benchmarks will mean a lot to the buyer, someone who makes his money on PS will not make much use of a SPEC benchmark, the SPEC results could be bad but PS filters could be scream! real world tests are more important to me.

We don't have Apple Stores here, I was just wondering if someone could walk into one and take his PS project or anything else Maya FCP or even code to compile and see how fast it is on the machines there compared to what ever he has back home. that would be the ultimate benchmark to anyone.

porovaara that's a good point! and i don't know how that passed right by me! but I seem to remember that the 733Mhz G4 was the top of the line, then Apple made it the bottom of the line, I don't remember any other machines that this happened to, I only remember this cuz my brother bought the 733 G4.

BrandonRP0123

Aug 20, 2003, 11:42 PM

Originally posted by MacRETARD

BTW - since it is dual channel you will have to upgrade memory 2 sticks at a time.

I never noticed that - but sure enough - you cannot get a single G5 configuration without 2 chips (2 x 128MB for the base configuration 1.6 and so forth).

Sucky sucky :)

Mudbug

Aug 20, 2003, 11:47 PM

just a note on the simple observation side of things (benchmarks excluded) that thing is huge. the side-by-side shot of the G5 next to the G4 is really eye opening as to the overall size...

I have to go and clear a little more space of my desk, and make sure there's no low-lying shelves in the vicinity.

macrumors12345

Aug 20, 2003, 11:49 PM

Originally posted by scem0
how about programs like Internet Explorer/safari/any browser?

I spend most of my time on the internet. I want to know if dual helps page rendering a lot. If It doesnt I might get the single 1.8.

I don't think it's even the case that Omniweb (or Safari, or other browsers) start a new thread for each window, and it certainly isn't the case that the rendering within a window is multithreaded. The bottom line is that if all you are doing is web browsing, then dual processors will not help you, especially if you are only loading one page at any given time.

But geez, even my iBook 500 loads pages at a decent speed (using Safari). It's just when I am trying to load a page *and* do something else that it really starts to choke (and that is a case in which dual processors is great).

How about program compiling time differences? Especially with Java compilers. I plan on doing a good amount of java on this computer, mostly for school.

Yeah, that would be a better use of the G5's horsepower. Nobody knows at the moment, but it is safe to say that a single G5 will be significantly faster in the long run than a single G4 (dual G4 vs. single G5 is more questionable).

Honestly, these benchmarks don't look that great.

My iBook 500 renders the Cinebench scene in 10 minutes even and scores a 44. That means that (scaling linearly), in this benchmark, the 1.6 Ghz G5 is equivalent to a 2 Ghz G3, and scores around 170 (i.e. equivalent to a 1.7 Ghz Pentium 4). My conclusion would be that the G5 is not making full use of its two double-precision FPUs: Maxon needs to recompile the app (at the very least).

My iBook 500 completes the Cinebench 4D Shading Tests in 208 sec and 87 sec respectively. That would put the 1.6 Ghz G5 on par with a 2.1 Ghz G3 and 1.8 Ghz G3 respectively (linearly scaled). It's not clear how the score is calculated between the two scenes, but assuming that the G5 works out to about a 1.95 Ghz G3, then its score would be equivalent to 200, i.e. a 2 Ghz Pentium 4.

The OpenGL Cinebench tests are not indicative of processor performance (since they rely to varying degrees on the graphics card), so I wouldn't use them for any comparisons.

I would also be very careful about trying to infer anything from the Xbench numbers. Xbench is a highly synthetic benchmark, and as such it is actually more sensitive to minor differences in processor architecture than most real world apps are. It is not a good indication of real world performance, and in fact the results don't even make sense between G5s. For example, the Dual 2 Ghz G5 scored 391/371 on the fp test, whereas the 1.6 Ghz G5 scored only 207/232. It is my understanding that the fp test is NOT multithreaded, so the Dual 2 Ghz and single 1.6 Ghz scores should actually be quite close together (but they're not, which means someone's benchmark is messed up). Also, the Dual G5 scored only 57 in Altivec Basic whereas the 1.6 Ghz G5 scored 95 on the same test! Needless to say, that is NOT reflective performance (the Dual 2 Ghz should never be SLOWER than the Single 1.6 Ghz). So, like I said, you really should just competely ignore the XBench numbers.

Actually, I guess in retrospect this is a pretty strong showing for a brand new CPU. At least it is actually showing a 10 to 20 percent INCREASE in performance per clock cycle on unoptimized apps (i.e. Cinebench), as compared to the G3 and the P4. Both the MPC 7450 (i.e. "G4e") and the Pentium 4 were actually anywhere from 30 to 50 percent SLOWER per clock cycle on unoptimized apps when compared to the processors that they replaced (the MPC 7400, i.e. "G4", and the Pentium III, respectively).

Once more software is compiled for the G5, it will really be on fire. I suspect it will easily be 50% faster per clock cycle than the G4 (and also the P4) in many apps. A friend of mine at Adobe said that they actually ran a massive Photoshop benchmark on the G5 that took hours to complete, and it really was twice as fast as the G4 (presumably Dual 1.42). The thing that is really impressive about this is that the Altivec unit on the G5 is actually slightly *weaker* than that of the G4 (though it does have superior memory bandwidth to compensate), so that indicates that the scalar code was running *more* than twice as fast on the 2 Ghz G5 as compared to the 1.42 Ghz G4.

notEven

Aug 20, 2003, 11:54 PM

Originally posted by Longey Nowze
notEven why is it faulty :( I thought i did a great job :( I was proud of myself :( no I'm sad :( at least we agree :)

the right benchmarks will mean a lot to the buyer, someone who makes his money on PS will not make much use of a SPEC benchmark, the SPEC results could be bad but PS filters could be scream! real world tests are more important to me.

We don't have Apple Stores here, I was just wondering if someone could walk into one and take his PS project or anything else Maya FCP or even code to compile and see how fast it is on the machines there compared to what ever he has back home. that would be the ultimate benchmark to anyone.

porovaara that's a good point! and i don't know how that passed right by me! but I seem to remember that the 733Mhz G4 was the top of the line, then Apple made it the bottom of the line, I don't remember any other machines that this happened to, I only remember this cuz my brother bought the 733 G4.

ok then, hows about.. "technically faulty".. or "will not stand in court - faulty".. ;)

Edot

Aug 20, 2003, 11:57 PM

This is a cross platform optimized application that also reflects real world performance!! Where are the tests using it? Why run apps that are known to be unoptimized when an optimized app already exists? I don't understand it! Unless the person with the G5 doesn't have Photoshop, but I suspect a G5 owner and internet benchmark poster would have it! All the excitment has me worked up. But really, I would love to see a Photoshop benchmark.

Longey Nowze

Aug 21, 2003, 12:01 AM

Originally posted by notEven
ok then, hows about.. "technically faulty".. or "will not stand in court - faulty".. ;)

no that wont do I'm still sad :( only a DP G5 with 8GB RAM and the ATI 9800 would make me happy again, the two 23" HD Cinema displays would return my pride :p

Thank You
MaT

garymm

Aug 21, 2003, 12:15 AM

Originally posted by BrandonRP0123
Forgive my ignorance, but the DDR RAM Stevie presented as the G5 using at WWDC is DDR 400Mhz. At best (assuming that DDR means you get double the bandwidth out of it), that only gives you 800Mhz of bandwidth to play with.

I'll forgive your ignorance, and tell you that DDR400 is actually 200MHz RAM doubled to 400, not 400 doubled to 800.

mim

Aug 21, 2003, 12:19 AM

Originally posted by MacRETARD
DDR 400 memory is 200 mhz that is effectively 400 because access is 2 times per cycle.

The G5 uses a dual channel memory architecture so it will be 800 mhz. The 1ghz bus has some overhead so even though the memory is not running at 1ghz you should still see some performance gain in memory related tests.

BTW - since it is dual channel you will have to upgrade memory 2 sticks at a time.

The main point being that accessing memory doesn't saturate the bus - so you (theoretically) have access at full speed to all your installed ram.

Sorry my other post was a big vauge.

stingerman

Aug 21, 2003, 12:19 AM

In no way do those benchmarks represent the performance of Apps run with GCC 3.3 with the G5 profile flag set. You people need to understand that the Compiler organizes and aligns the instructions in the object code to match the profile of the processor. So here you have this highly parallelized, super scalar G5 running instructions optimized for a G4 pipeline and functional units. It's like putting two 14 wheelers on the road with only one pickup trucks load. It's a ridiculous benchmark to judge G5 native apps. Though it is fine to see how current Apps will run even before they are recompiled.

That is why the demos shown at the WWDC keynote were of quickly recompiled Apps with no further tweaking done. That is a two times performance gain over a G4 native App! So take a look at these benchmarks and multiply or divide by two (whatever being the case.) According to Apple, Cocoa apps that dynamically link to the public frameworks will automatically benefit from 10.2.7's and 10.3's up-to-date and properly profiled code. However, re-compiling is not the end all in performance gains. Programmers relooking at their code to take advantage of the G5 and its architecture further could see 10X performance gains. Apple demonstrated increading the performance of an OpenSource app by 24X simply by optimizing it for the G5 at one of their WWDC sessions.

eric_n_dfw

Aug 21, 2003, 12:23 AM

Originally posted by garymm
I'll forgive your ignorance, and tell you that DDR400 is actually 200MHz RAM doubled to 400, not 400 doubled to 800. ...and since it's dual channel (gotta add those DDR modules in pairs) you get 800MHz performance out of it. (roughly)

P4's do the same thing, but they don't have the FSB that the G5 has which exceeds that memory speed either.

garymm

Aug 21, 2003, 12:23 AM

Originally posted by stingerman
In no way do those benchmarks represent the performance of Apps run with GCC 3.3 with the G5 profile flag set. You people need to understand that the Compiler organizes and aligns the instructions in the object code to match the profile of the processor. According to Apple, Cocoa apps that dynamically link to the public frameworks will automatically benefit from 10.2.7's and 10.3's up-to-date and properly profiled code. However, re-compiling is not the end all in performance gains. Programmers relooking at their code to take advantage of the G5 and its architecture further could see 10X performance gains. Apple demonstrated increading the performance of an OpenSource app by 24X simply by optimizing it for the G5 at one of their WWDC sessions.
does this mean that every app has to be recompiled in a special G5 version to take advantage of it? the PS update isn't a complete recompile, yet it still adds tonnes of speed, supposedly.

BrandonRP0123

Aug 21, 2003, 12:25 AM

Let's just take a minute and think about all we're discussing here. Seriously. Do this before you quickly hit ``Reply'' or ``Quote'' on what I'm about to say.

First -

We're arguing over inaccurate benchmarks. Think about this for a second. Does anyone remember when the Power PC came out? Remember the difference in performance there? Remember how HORRIBLY slow 68k coded apps were on the original PPC, yet everyone pitched a fit because the PPC didn't seem to be a major jump in performance. Why? Simply because the benchmarks weren't designed to take advantage of the PowerPC architecture.

The bottom line on the first point:
This is the same thing.

For those of you that speak PC, this happens every time Intel releases a new improved chip. It always has functions and features unheard of, and thus not coded for. This will CONTINUE to happen for years to come - every time a new chip is developed. Developers write apps for what they have access to. Developers haven't written software to take advantage of the G5, and thus the advantages of its new and improved technology won't be realized until this happens.

Second:
If you're asking how much faster a G5 draws web pages and surfs the internet over a G4 - I suggest you do further research before you completely decide to spend $1999+ on a new G5. Same applies if you currently own a PC and you are planning on making the switch solely on the G5 hype, and because ``everyone else is doing'', please consider carefully. While Apple will happily take your money, sure, your false expectations I fear will lead you in the complete opposite direction.

Furthermore, if you own a machine capable of it, perhaps its time to upgrade that ol' graphics card to work with Quartz Extreme?

Macs for years have been made bigger and better for those musicians, photographers, desktop publishers, (now) gamers, and video artists (for lack of a better word). Think about Jobs' very own Pixar. This is the union of the products his two companies, as Apple has finally produced a machine suitable for their use (and Mac OS is graduating to a state along with said hardware to be blessed with Renderman).

If you do any of the above - my hat goes off to you. You will soon know (if you have ordered one) the truth - and while we may piss on each other's comments to our heart's content you (as the creative mind behind the Mac) will know the difference.

Apple wouldn't spend millions developing a COMPLETELY new inside and out flagship computer if just to make it ``slightly'' faster. Think G3 to G4. Same case, different color. Think 604e to G3. We were beige until our younger brother was born into a colorful case. The G5 is entirely different inside and out - between materials used, technology developed, and time spent in developing it, this is the first machine in years that Apple has tossed out all notions of what they USED to sell. Think January 24th, 1984. Think the birth of the Macintosh. Think different.

This is the FIRST day that apple has sprung the G5 from their loins to the public. Don't be so quick to judge. See it with your own eyes and by all means go with what you believe. Believe it when you see it, instead of relying on inaccurate data.

alset

Aug 21, 2003, 12:27 AM

That's the tallest Mac I have ever seen. It even beats out my old 9500. Sad. Pathetic. I was so happy, now I'm depressed.

Oh, well... I'll still buy one when the time comes.

Dan

garymm

Aug 21, 2003, 12:29 AM

I'd agree that this is a huge change, but I don't think that the G3-G4 change was as comprable. the G4 was much more similar to the g3 (from what little I know) with the exception of AltiVec, which is very useful in some apps, and completely useless in others.

jason166

Aug 21, 2003, 12:29 AM

After seeing all sorts of people complaining about the performance on the benchmarks and people saying they will be re thinking the purchase of a G5 I was gonna say something to the tune of

These programs are not good indicators of G5 performance

stingerman has explained this so I don't have to.

Well said.

And no, this does not mean that every program will have to be recompiled to get decent performance. Only apps that make use of procedures with DST Altivec calls and those that have code that is highly optimized for the G4 scheduler.

It is important to note that all the libraries in 10.2.7 have been recompiled to work well with the G5 so apps that make standard API calls will benifit allready.

Jason

sparkplug

Aug 21, 2003, 12:29 AM

Here is another benchmark result, this one from nasa on the 2ghz dual running fluid dynamics simulations in JET3D

Nasa test dual 2ghz G5 (http://members.cox.net/craig.hunter/g5/)

Here is the bottom line

Based on raw scalar floating point performance in Jet3D, a 2GHz G5 system can match a 2.66GHz P4 system, and this is a dramatic improvement from earlier tests where G4 systems lagged behind higher clock speed P4 systems. Based on an extrapolation of current P4 results, the 2GHz G5 would lag newly announced 3.2GHz P4 systems in Jet3D scalar floating point performance by about 20%, but this kind of comparison is best deferred until G5-aware compiler tools become available (since a 20% performance gain is well within the potential of compiler optimization).

The G5 is a nice machine, the term "fastest" however is an embarresment.

aje

Aug 21, 2003, 12:33 AM

i'm desperate to leave the world of windows but as a flash developer macs totally suck at compiling and rendering flash movies. 1000% slower generally on the g4's i've been playing with (not QS).

i would love to know if any flash developers who share my pain have had a chance to render/compile a decent size flash application/movie on the new G5's - and if so was it at the least as fast as a decent pc?

i suspect from my google searches that its more of a architecture problem than anything, so i'm wondering if the new g5's architecture will speed up the calculations for flash movies??

BrandonRP0123

Aug 21, 2003, 12:35 AM

Originally posted by garymm
does this mean that every app has to be recompiled in a special G5 version to take advantage of it? the PS update isn't a complete recompile, yet it still adds tonnes of speed, supposedly.

To my knowledge, no. We must realize that between Mhz, the ceiling gets higher.

Between generations of architecture (think the ``G'' in G5) the ceiling not only gets higher, the room gets wider.

For a wider room, unless you have a wider ladder (or move the ladder) you won't be able to touch the ceiling in every spot of the room

halabar

Aug 21, 2003, 12:40 AM

Originally posted by aje
i'm desperate to leave the world of windows but as a flash developer macs totally suck at compiling and rendering flash movies. 1000% slower generally on the g4's i've been playing with (not QS).

That sounds like the Flash "compiler" is not optimized for the Mac at all.

We'll never get around that. If the software is written poorly, performance will always suck.

-- Karl

stingerman

Aug 21, 2003, 12:43 AM

Originally posted by garymm
does this mean that every app has to be recompiled in a special G5 version to take advantage of it? the PS update isn't a complete recompile, yet it still adds tonnes of speed, supposedly.

Re-compiling for the G5 is fast and easy. That was one of the points that was illustrated during the WWDC keynote. The photoshop demo during WWDC was of a simple re-compile and yet it blew away the dual XEON.

As far as this PS update notice what Adobe says:

• Modifies many Photoshop operations to fully utilize the G5 processor capabilities
• Replaces the Adobe Color Engine component (ACECarbonLib) with a new version that is designed for the G5 processor

ACE effects every part of PS. The Plugin does a lot more than you would think a plugin would do.

Midiplaya

Aug 21, 2003, 12:46 AM

So, I do quite a bit of audio work and run Virtual PC for a variety of apps. Virtual PC performance is pathetic on a 1GHz machine with 1GB Ram. I am anxiously hoping that the G5 machine will run Virtual PC fast enough to "feel" like a decent PC. If not, I am going to have to switch back. :mad:

BrandonRP0123

Aug 21, 2003, 12:50 AM

Longey Nowze

Aug 21, 2003, 12:51 AM

Originally posted by sparkplug
Here is another benchmark result, this one from nasa on the 2ghz dual running fluid dynamics simulations in JET3D

Nasa test dual 2ghz G5 (http://members.cox.net/craig.hunter/g5/)

Here is the bottom line

The G5 is a nice machine, the term "fastest" however is an embarresment.

if you notice, they only used one of those 2GHz CPUs, and it still beat the 2.66GHz P4, read that again please.

Thank You
MaT

edit: from the NASA tests

Though dual processor benchmarks are not presented in detail here, it is worth noting that the G5 system benchmarked at 498 MFLOPS and 0.125 MFLOPS/MHz for scalar Jet3D performance when two processors were used.

the singe 2GHz G5 had 254 MFLOPS and 0.125 MFLOPS/MHz the P42.66GHz had 255 MFLOPS and 0.096 MFLOPS/MHz... you do the math :)

edit: one more very important thing to remember that test was done on pre-release machines runing BETA software the results of the final products should be much better.

stingerman

Aug 21, 2003, 12:55 AM

One more thing to keep in mind why new benchmark software is needed and you should ignore these benchmarks: The new ASIC Controller actually steps down the CPU speed and then steps it back up slowly depending upon the current load. So those benchmarks may actually be showing results of the processor running at a slower average clock. The benchmarks need to be coded to tell the system that these are benchmarks and give us everything you got, as if you were under a heavy load. This is a major advance of the G5 architecture.

The dual G5's actually step down to 1.3GHz under light loads, using 1/6th the power. The fans don't even need to spin at all to keep them cool, though Apple keeps them spinning at a trickle in readiness. As the load on the processors increases the ASIC slowly steps up the clock, monitoring each processor independently as far as heat goes, each fan spinning for its own CPU's needs and at a possibly different rpm. If there is a fan failure or the door isn't closed right, the CPU's will run only at their slowest speeds which will make sure the do not melt.

I hope this helps put the benchmarks in a better light.

BrandonRP0123

Aug 21, 2003, 12:55 AM

Originally posted by Longey Nowze
if you notice, they only used one of those 2GHz CPUs, and it still beat the 2.66GHz P4, read that again please.

Thank You
MaT

Exactly. Plus the P4 is 660Mhz faster than the G5. While Mhz myth, sure - the fact remains that the G5 would have proven itself if they ran it against a 2.0Ghz P4

BrandonRP0123

Aug 21, 2003, 12:58 AM

Originally posted by stingerman
One more thing to keep in mind why new benchmark software is needed and you should ignore these benchmarks: The new ASIC Controller actually steps down the CPU speed and then steps it back up slowly depending upon the current load. So those benchmarks may actually be showing results of the processor running at a slower average clock. The benchmarks need to be coded to tell the system that these are benchmarks and give us everything you got, as if you were under a heavy load. This is a major advance of the G5 architecture.

The dual G5's actually step down to 1.3GHz under light loads, using 1/6th the power. The fans don't even need to spin at all to keep them cool, though Apple keeps them spinning at a trickle in readiness. As the load on the processors increases the ASIC slowly steps up the clock, monitoring each processor independently as far as heat goes, each fan spinning for its own CPU's needs and at a possibly different rpm. If there is a fan failure or the door isn't closed right, the CPU's will run only at their slowest speeds which will make sure the do not melt.

I'd love to see a Dell do that. Or a Compaq. Or an IBM. Heck, even our brand new $8,000 Dell PowerEdge in the NOC doesn't do that (or even come close).

Midiplaya

Aug 21, 2003, 01:01 AM

Turning the color down to 16bit doesn't help. I am doing things where I need to open and close a lot of files fast. I also find that anything CPU intensive takes 3-4X's as long as a regular PC. I still prefer using OS X as my normal desktop and putting up with the performance as necesary. I really hope that I can get better performance though and snappiness out of the PC desktop.

That aside, if you can get the educational discount and ditch the modem and DVD-burner for a DVD-ROM, then you can get a 1.6 for $1590 or a 1.8 for $1900.

That is a bargain.

BrandonRP0123

Aug 21, 2003, 01:08 AM

Powerbook G5

Aug 21, 2003, 01:12 AM

VPC is not going to fly no matter what you do, it's just a fact of life trying to emulate a whole computer enviroment. It's about the convenience than performance with it, which it does have.

MrMacMan

Aug 21, 2003, 01:33 AM

Yeah that is nice.

I'm sure more info will be flowing in soon.

And when my Apple store gets them, I think I might have to stea... i mean *borrow* one... forever.

;)

So far so good.

wizard

Aug 21, 2003, 02:09 AM

If your going to be doing a serious amount of programming get the dual machine. Both Java and the gcc environment can take advantage of duals. Gcc (on linux at least) can, via make, take advantage of two processors when building a project.

Thanks
Dave

Originally posted by scem0
yes, a lot of surfing.

But a lot of programming too (the more the better ;)). I know C++, Im learning basic and java. I hope to be using all 3 of those languages a lot.

And with the amount of time I spend on my computer even if all I did was surf it wouldnt be overkill. If I save a hundredth of a second on every webpage I went to I'd save a BUNCH of time.

scem0

wizard

Aug 21, 2003, 02:25 AM

This won't help my perception of the machine, especially the 1.6 GHz machine, as I honestly believe that performanc of the hardware is going to be found to be lacking in many areas. That doesn't mean we won't see good results in specific areas, but the online information that I've seen and the write up on Ars simple indicates that the 970 is a processor that will just keep parity with the current Intel hardware.

But that isn't what I wanted to comment on anyways, what is interesting is that you described an almost ideal chip set for a LAPTOP. If the description is accurate this is a bit of a surprise in a desktop. It does appear that these could be very energy effiecent machines. So to the people receiving the machines; if anybody has the abiltiy to measure energy draw on these machines please do so and post it.

Thanks
dave

Originally posted by stingerman
One more thing to keep in mind why new benchmark software is needed and you should ignore these benchmarks: The new ASIC Controller actually steps down the CPU speed and then steps it back up slowly depending upon the current load. So those benchmarks may actually be showing results of the processor running at a slower average clock. The benchmarks need to be coded to tell the system that these are benchmarks and give us everything you got, as if you were under a heavy load. This is a major advance of the G5 architecture.

The dual G5's actually step down to 1.3GHz under light loads, using 1/6th the power. The fans don't even need to spin at all to keep them cool, though Apple keeps them spinning at a trickle in readiness. As the load on the processors increases the ASIC slowly steps up the clock, monitoring each processor independently as far as heat goes, each fan spinning for its own CPU's needs and at a possibly different rpm. If there is a fan failure or the door isn't closed right, the CPU's will run only at their slowest speeds which will make sure the do not melt.

I hope this helps put the benchmarks in a better light.

HopefulSwitcher

Aug 21, 2003, 02:48 AM

For all you people out there that are contemplating getting a mac or a dell... get a mac...

For those thinking between dell and another pc... get a mac

It's not war veterans that make me proud to be living in this country.. its companies like Apple lol XD

I love you G5... you sexay thang you!

Now... Steve Jobs... I command you to release the new powerbooks now!

nagromme

Aug 21, 2003, 03:04 AM

Originally posted by aje
i'm desperate to leave the world of windows but as a flash developer macs totally suck at compiling and rendering flash movies.

I don't care as much about compile speed--it's never bothered me n my G4--but Flash playback isn't up to speed. Macromedia has acknowledged their failing in this regard, and said to expect big Mac speed optimizations in future versions. (The commend was made specifically in regard to the player.)

There's no flaw in Macs that makes them unsuitable for vector animation--Macromedia simply has an issue to address.

mim

Aug 21, 2003, 03:10 AM

Originally posted by HopefulSwitcher
Now... Steve Jobs... I command you to release the new powerbooks now!

Heh! Nice try.

<<Cut to dusty Biblical scene. Steve's in a toga, surrounded by nubile young ladies not wearing that much. He surveys the crowd from his podium>>

Steve: Who shall we welease?
Crowd: Welease the powewbooks! (giggling)

Steve confers with advisers: What? There is no Powewbook?
Steve to crowd: There is no "Powewbook". Who shall we welease then?
Crowd: Welease Woderwick!

Steve: Welease Woderwick!

<<Cut to classic white Apple slick add>>

Powerbook style voice over: Woderwick. The next big - and small - thing - because that's all there is - from Apple.

hvfsl

Aug 21, 2003, 04:33 AM

Originally posted by mim
Heh! Nice try.

<<Cut to dusty Biblical scene. Steve's in a toga, surrounded by nubile young ladies not wearing that much. He surveys the crowd from his podium>>

Steve: Who shall we welease?
Crowd: Welease the powewbooks! (giggling)

Steve confers with advisers: What? There is no Powewbook?
Steve to crowd: There is no "Powewbook". Who shall we welease then?
Crowd: Welease Woderwick!

Steve: Welease Woderwick!

<<Cut to classic white Apple slick add>>

Powerbook style voice over: Woderwick. The next big - and small - thing - because that's all there is - from Apple.

More monty python than biblical.

kaos_de_moria

Aug 21, 2003, 04:34 AM

That doesn't mean we won't see good results in specific areas, but the online information that I've seen and the write up on Ars simple indicates that the 970 is a processor that will just keep parity with the current Intel hardware.

wait... that was exactly what everyone expected from apple... right?

ppl seem to forget fast what they expected before the G5 was announced!

mainstreetmark

Aug 21, 2003, 04:36 AM

Geezz... bitch bitch bitch...

if the G5 isn't the fastest computer overall, it's certainly the fastest in a few big categories. The new PCs may be faster at some things, but by a very small fraction now.

The G5 is good news all around. (Yes, it would have been nicer without the 'World's Fastest' claim, and that poor kid getting blown through his house)

sparkplug

Aug 21, 2003, 05:00 AM

Longey Nowze writes

if you notice, they only used one of those 2GHz CPUs, and it still beat the 2.66GHz P4, read that again please

Yes I did notice this, how exactly is this relevant? I allso noticed that they did not test a dual athlon, opteron or xeon machine, just a single g4, single g5 and single p4, does this somehow invalidate the results? no.

By all means elaborate as to how 254 for the G5 "beat" 255 for the P4 ? you yourself quoted these mflop scores in closing. Unless you are referring to the scalar performance, ie mflops per clock, Indeed the G5 is faster per clock, but this does not mean it is "faster" certainly not "faster" in any usefull terms. IE actual purchasable today performance. Seeing as one cant actually purchase a 2.6ghz G5, nor indeed a 3.25 for that matter.

The nasa bench quite clearly shows that the 2ghz G5 chip does about the same work (1 point slower) that a 2.6ghz P4 does running their real word code. Maybe you should read it again.

jettredmont

Aug 21, 2003, 05:24 AM

Originally posted by scem0
how about programs like Internet Explorer/safari/any browser?

I spend most of my time on the internet. I want to know if dual helps page rendering a lot. If It doesnt I might get the single 1.8.

Why about just opening finder windows. Simple tasks like that, which you use all the time.

Photoshop benchmarks dont help me all that much, although I would love to see them.

How about program compiling time differences? Especially with Java compilers. I plan on doing a good amount of java on this computer, mostly for school.

scem0

edit:

I agree. I think it is performing very well. I expected less from a single processor system.

Safari renders? I'm not sure, but I suspect not. Safari seems oddly single-threaded in places; not sure if that extends to the WebCore framework or not. I somewhat doubt that Internet Explorer takes advantage of multiple procs.

A dual-processor system will definitely speed up "everyday" GUI responsiveness (ie, opening Finder windows, new apps, etc).

Using gcc (or gjc) and Project Builder/XCode, compile times roughly double with a dual-proc system, although makefiles may or may not work as well (my makefiles didn't use dual procs or XCode parallel compiling, which was the final straw in me switching everything over to a Project Builder project on OSX and leaving the makefiles to Solaris/Linux only); different 'make' implementations may vary though).

Javac? I'm not sure (I don't do any Mac-based Java; all my Java is done on an old Windows box). I would suspect it uses dual procs (I believe it does on, for instance, Solaris), but can't say for certain.

jettredmont

Aug 21, 2003, 05:34 AM

Originally posted by Powerbook G5
Well, using both OS 9 and XP, I've found that nothing is as snappy as a well maintained Mac OS 9 running on my G3...I can only imagine how fast it would be on a newer G4 system. It makes XP seem slow, for sure, though. If Panther is as snappy as my PowerBook G3 400 MHz running at 400 MHz, I'd be really happy. If a G5 cannot produce those results, then there is something wrong...

Yes, however, Windows 95 on an old 400MHz P2 is positively "snappy" as well. OS 9 is "snappy" for a reason.

IMHO, OS X 10.2.6 on my 733MHz G4 machine is as "snappy" as OS 9 running on the 500MHz G4 next to it. OS 9 is faster on identical hardware, but the difference isn't nearly as large as you are making it out to be!

jettredmont

Aug 21, 2003, 05:43 AM

Originally posted by BrandonRP0123
I never noticed that - but sure enough - you cannot get a single G5 configuration without 2 chips (2 x 128MB for the base configuration 1.6 and so forth).

Sucky sucky :)

That's why you have eight slots for memory (okay, four in the 1.6, but is that also dual-channel?)

It's generally cheaper buying two 1GB sticks than one 2GB stick ...

kristianm

Aug 21, 2003, 05:45 AM

Originally posted by sparkplug
Longey Nowze writes
Yes I did notice this, how exactly is this relevant? I allso noticed that they did not test a dual athlon, opteron or xeon machine, just a single g4, single g5 and single p4, does this somehow invalidate the results? no.

I am quite happy with equal performance, and I can not belive that there is someone not thrilled with this. But, I think G5's scale better with 2 procs than the P4 xeon.

alex06

Aug 21, 2003, 05:46 AM

[QUOTE]Originally posted by macrumors12345
[B]I don't think it's even the case that Omniweb (or Safari, or other browsers) start a new thread for each window, and it certainly isn't the case that the rendering within a window is multithreaded. The bottom line is that if all you are doing is web browsing, then dual processors will not help you, especially if you are only loading one page at any given time.

But geez, even my iBook 500 loads pages at a decent speed (using Safari). It's just when I am trying to load a page *and* do something else that it really starts to choke (and that is a case in which dual processors is great).
.... <removed> ....
[QUOTE]

OmniWeb rules !
Let's read on http://www.omnigroup.com/applications/omniweb/

"The only web browser on any platform designed to harness multiple processors for the fastest web browsing possible.":D :D

jettredmont

Aug 21, 2003, 06:03 AM

Originally posted by BrandonRP0123
Exactly. Plus the P4 is 660Mhz faster than the G5. While Mhz myth, sure - the fact remains that the G5 would have proven itself if they ran it against a 2.0Ghz P4

Um, no it wouldn't have. By all projections, the 970 is NEVER set to approach more than 2/3 the top speed of the Pentium 4/5.

A key component of processor design is the tradeoff of clock speed versus calculations per clock cycle. The only measure for a chip is not "operations per cycle" but "operations per unit time".

While it would be great to have a 3.2 GHz G5 today to beat the pants off the top-of-the-line P4, that's not going to happen. Today we have 2.0GHz G5s and 3.2GHz P4s. A year from now we will have 3GHz G5s but 4.8GHz (if I recall the Intel roadmap correctly) P5s.

Of course, cost-wise you can't get a top-of-the-line dual-proc Xeon for the same money as a dual-2.0GHz G5 (assuming you want all the same features on the Xeon), but that only forces you down by a few hundred MHz, not into the equal-clockspeeds range.

jettredmont

Aug 21, 2003, 06:07 AM

Originally posted by Midiplaya
Turning the color down to 16bit doesn't help. I am doing things where I need to open and close a lot of files fast. I also find that anything CPU intensive takes 3-4X's as long as a regular PC. I still prefer using OS X as my normal desktop and putting up with the performance as necesary. I really hope that I can get better performance though and snappiness out of the PC desktop.

As I've said before, I completely fail to see the logic of VirtualPC to begin with. Just buy a cheapo Windows box, and a USB KVM switch (<$100) to switch between the Mac and PC. A much more flexible solution, plus you get to use a PC that is about 3-4 times as fast as anything you'll get with VPC, and you can use your Mac at the same time as your PC is churning through calculations (just switch using the KVM). Cost is slightly higher, but performance is at least passable!

jettredmont

Aug 21, 2003, 06:14 AM

Originally posted by kristianm
I am quite happy with equal performance, and I can not belive that there is someone not thrilled with this. But, I think G5's scale better with 2 procs than the P4 xeon.

That will be true in an FSB-limitted situation. If you are processor-limitted, then the Xeon scales quite well.

OTOH, that's just the problem the G4 has had forever, so if your app scales well on Xeon (meaning 1->2 CPU) then it would have performed reasonably well on the G4 too ...

jettredmont

Aug 21, 2003, 06:22 AM

Originally posted by alex06

OmniWeb rules !
Let's read on http://www.omnigroup.com/applications/omniweb/

"The only web browser on any platform designed to harness multiple processors for the fastest web browsing possible.":D :D

I find this claim somewhat dubious, as "harness[ing] multiple processors" is a very vague term. Does this mean that rendering is multi-threaded (not likely; it uses the same WebCore framework as Safari, so if Safari's not multi-threaded by OmniWeb's definition neither is OmniWeb!), or that the HTTP communications is multi-threaded, or that UI widgets are multi-threaded, or ...?

tazznb

Aug 21, 2003, 07:00 AM

mvc

Aug 21, 2003, 07:16 AM

Originally posted by tazznb
Looks like it's time to get a new desk;)

Yep, they used to call IBM, the "big iron", I guess since Apple joined forces with them they are now the "big aluminium". :rolleyes:

Actually, I suspect that the enclosure is nice and big so that everything will still fit when the Rev.C model comes out with quad processors, 32GB of ram and 47 fans!

adamfilip

Aug 21, 2003, 07:40 AM

OK so i want more benchmarks.. to get a better picture

i want photoshop, folding, boot time,.. i donno anything!

Dont Hurt Me

Aug 21, 2003, 07:41 AM

just saw this, i think bench marks are a good way to get a rough idea on performance from a machine. just in case anyone may be interested a single 1.4 g4 in a quicksilver gave me a xbench of 114. that doesnt compare that bad with the new 1.6 which had a score of 128.

justytylor

Aug 21, 2003, 07:59 AM

Originally posted by yoshi1013
I'm looking forward to the speed because of video applications. I'm sick and tired of waiting until I won't be using my computer just to encode maybe five minutes of video which will take around an hour or so also depending on if I use 2-pass VBR encoding.

While this is true (I have a pet project and it around 20 hours to encode 2 hours of video at the 1-pass VBR MPEG-2 setting in FCP4's Compressor on my PowerBook 667) I've noticed that Compressor plays very well with OS X's multitasking. The UI remains quite responsive and as long as I'm not doing anything else *too* processor intensive (I think I could play music, but I wouldn't) I can get other work done. I can't wait to see what the 1.6 GHz G5 can do on this score, considering that my 667MHz G4 does pretty well.

scem0

Aug 21, 2003, 08:03 AM

Originally posted by holmesf
Why would you learn basic AFTER C++?

btw, using Java with Project builder is a really poor implimentation. If that's what you're doing, I'd suggest Cocoa.

I know..... It sounds wierd

But it's because my school is retarded.

And yeah, I want to learn cocoa too. And other languages too. By the end of senior year I want to be able to make a decent program for either mac or PC, doing something that everyone needs.

scem0

soggywulf

Aug 21, 2003, 08:23 AM

Originally posted by The Ancients
Reliability, that's why!

At least the Fiesta would get you there, AND back...

Hey, the Lotus is PERFECLY reliable! Just buy some antifreeze and oil at the grocery store for the trip back.

whooleytoo

Aug 21, 2003, 08:29 AM

Originally posted by sparkplug
Yes I did notice this, how exactly is this relevant? I allso noticed that they did not test a dual athlon, opteron or xeon machine, just a single g4, single g5 and single p4, does this somehow invalidate the results? no.

But why did you ignore the Altivec version? It's far, far faster than the 2.66GHz P4 with or without SSE2.

Mike.

jayscheuerle

Aug 21, 2003, 08:38 AM

So real life benchmarks don't live up to Steve's Reality Distortion Field.

And you people act like this is the first time? Gimme a break! The only thing of note here is that the plastic on the inside is a much uglier putty grey/beige than I'd ever imagine Apple to produce.

soggywulf

Aug 21, 2003, 08:39 AM

Originally posted by Longey Nowze
that's kinda my point, the G5 is a Pro machine, I think it will be used by professionals more than consumers, and some car programmes have lap times rather than from the 0-100 like BBC's Top Gear which gives you a better way to compare cars, so PS FCP and Maya and the like should be better tests to run compiling software code is too, but isn't that what some benchmarks do?

I disagree. Benchmarks suck, because they can never really represent real-world hi-power usage. At best benchmarks can give a very vague indication of relative performance. And that only if they are app benchmarks--those silly CPU exercising tests are less than useless for users.

In the real world, when you are doing a photoshop filter and a 3D render and a video stream encoding and a DVD burn and surfing the web at the same time, then I think the 1.6 G5 will annihilate the Dual 1.42.

twinturbo

Aug 21, 2003, 08:42 AM

Originally posted by Dont Hurt Me
just in case anyone may be interested a single 1.4 g4 in a quicksilver gave me a xbench of 114. that doesnt compare that bad with the new 1.6 which had a score of 128.

Yah, but shouldn't the G5 be ALOT faster than a G4. The difference between the two scores seems to be soley dependant on faster clock speeds, not on any new architecture. I would have expected all the new stuff with the G5 to at least make it a ton faster than the G4 but I guess it doesn't. Which basically means that the top-end G4 whoops up on a G5 1.6Ghz. It also means that the G5 really isn't al that revolutonary if all it can do is simply outperform a G4 by the sole virtue of having faster clock speeds. That's like the difference between a G3 and G4 (which wasn't much of a difference unless you were running altivec enabled apps). I guess I'm just really disappointed with these results, and everybody here is trying to make up some poor excuses on "optimization", when in fact the G5 is not the wonder chip that we dreamed it would be. When I sit people down and show them that Photoshop doesn't run that much faster and then tell them some dumb "unoptimized" excuse, guess what, they won't listen-they could come up with just as many excuses themselves on the P4 performance tests if the G5 were to beat it. So forget it. Seems like the G5 sucks for now. I'm kinda scared to benchmark it against our dual 3.0 Xeon, cause it might get it's butt kicked, and I don't want to be the fool to have to say "unoptimized apps". Ha!!!

John Q Public

Aug 21, 2003, 08:50 AM

ummmm...no...that 1.6Gz G5 won't out power the Dual 1.42 on 3D Rendering...

...with the exceedingly pathetic nVidia FX5200 that the 1.6 ships with (and the Dual 1.42's Stock ATi 9700 Pro) people seem to forget the GPU on the graphics card is a big factor in rendering performance...

...not to mention Adobe Pornoshop is SMP Aware...Altivec Enhanced...NOT optimised for the G5's 64bit processor...

mkaake

Aug 21, 2003, 09:03 AM

Originally posted by BrandonRP0123

Apple wouldn't spend millions developing a COMPLETELY new inside and out flagship computer if just to make it ``slightly'' faster. Think G3 to G4. Same case, different color. Think 604e to G3. We were beige until our younger brother was born into a colorful case. The G5 is entirely different inside and out - between materials used, technology developed, and time spent in developing it, this is the first machine in years that Apple has tossed out all notions of what they USED to sell. Think January 24th, 1984. Think the birth of the Macintosh. Think different.

This is the FIRST day that apple has sprung the G5 from their loins to the public. Don't be so quick to judge. See it with your own eyes and by all means go with what you believe. Believe it when you see it, instead of relying on inaccurate data.

dude, you just seriously sounded like a pr guy for apple... maybe you could see if they need someone over there :o)

seriosly, that got me all tingly...

matt

twinturbo

Aug 21, 2003, 09:09 AM

Originally posted by John Q Public
not to mention Adobe Pornoshop is SMP Aware...Altivec Enhanced...NOT optimised for the G5's 64bit processor...

It is as of yesterday.
I'm not expecting much from it's optimization though. Seems as though all the naysayers where right when they were making fun of me saying that Apple's lying, and like a fool I was coming up with all these explanations on why the tests are they way they are, and blah blah this and that. I'm still getting a dual G5, but lets just say I'm really disappointed.

jayscheuerle

Aug 21, 2003, 09:14 AM

Originally posted by twinturbo
Seems as though all the naysayers where right when they were making fun of me saying that Apple's lying, and like a fool I was coming up with all these explanations on why the tests are they way they are, and blah blah this and that.

You should have chosen the red pill...

nagromme

Aug 21, 2003, 09:21 AM

For te record, there are NO Mac apps that don't benefit to SOME degree from dual CPUs.

Mac OS X will, at the very least, devote one CPU to the current app, letting the OS and other apps run on the second CPU. So your single-CPU app has the WHOLE CPU to itself, and thus will run faster.

Other apps will of course take FULL advantage of duals.

And in defense of VPC: the benefit of it over a real PC is convenience. (And also price, now that it's bundled with Office Pro.) You can share yuur screen between Mac and Windows and imply drag and drop. You can access Windows without any boot time with one click. And you don't need to find space for a second pile of hardware. And if you are a portable user, carrying two laptops is no solution! VPC to the rescue.

I got a PC, then I got VPC... I use VPC for convenience, and let the real one collect dust.

Plus, the cheapest PCs tend to be very unreliable. If your cheap PC dies and leaves you with nothing, you haven't saved.

(PS It looks from MS's stated requirements that VPC will ONLY run on a G3 or G4 at the moment. Hopefully a solution will come from MS--or Real PC--soon.)

MacsRgr8

Aug 21, 2003, 09:23 AM

I would like some simple speed performance benchmarks:
Nascar Racing 2003: 1600 x 1200, extreme res, all grfx settings as high as possible, Coca Cola oval, exiting turn 4 while being the last car ..... How many FPS?
:D

twinturbo

Aug 21, 2003, 09:27 AM

Originally posted by jayscheuerle
You should have chosen the red pill...

lol :-)

John Q Public

Aug 21, 2003, 09:31 AM

Originally posted by twinturbo
It is as of yesterday.
I'm not expecting much from it's optimization though...I'm still getting a dual G5, but lets just say I'm really disappointed.

That would be why when my faithful B&W G3 finally died last month I went out and bought one of the last Dual 1.25 FW800 PowerMacs in my city...and not waiting on the G5...seeing that I got a helluva deal (last one CompUSSR had...talked them down to $1400)...dropped 2GB of CAS 2 PC2700...another 80GB Barracuda (and 2 more 40's in RAID 0 on the ATA66 chain)...PLUS with the money I saved over the Dual 1.8 (or 2.0) G5...let me justify a 23" Cinema HD http://forums.macrumors.com/images/smilies/biggrin.gif

...and as another sidebar...the G5 1.6 scores a 128 on XBench...and the Twin 2.0 scores 164...so I'm happy with the 146 mine rates...and really happy for NOT paying G5 prices (ok...I really like my Flat-Panel too)

NavyIntel007

Aug 21, 2003, 09:39 AM

Originally posted by adamfilip
a painted Delorean.. thats criminal!

Well... It was in Little Havana...

cubist

Aug 21, 2003, 09:50 AM

... of the keyboard, mouse (I assume they're the new "keyboard" and "mouse"), and how about some shots of the packing materials?

What does "about this macintosh" show?

Enough bickering about the benchmarks already. Four pages worth, sheesh. MR ought to recycle the posts from all the old benchmark-bickering threads.

sebaz

Aug 21, 2003, 09:53 AM

i heard it comes with a sightly modified emac keyboard and mouse...so much for the bluetooth or metal finish keyboards...:(

jason166

Aug 21, 2003, 10:02 AM

I keep seeing posts referencing the Xbench benchmarks...

this app is broken on the G5... hell the 1.6 is getting faster scores than the 2.0s in some cases, surely this can't be correct.

These benchmarks are totally unrepresentative of G5 performance. Being proud that your g4 has comparable xbench performance to the g5 is a moot point.

Jason

jayscheuerle

Aug 21, 2003, 10:04 AM

Interesting that on Apple's own page, they had to scale up a PC to appear to be as large as the G5. Even then, the PC still has 2 optical drives, a floppy/zip drive, space for 2 hard-drives...

Half of the G5's space appears to be taken up by fans and heat-sinks...

zygoat

Aug 21, 2003, 10:07 AM

What I really want to know is... what keyboard and mouse ship with the thing?

Are they bluetooth? are they the "non-pro" white versions that ship with the iMacs?

(Web boards are such a PITA to use, I apologise for not already searching through screens of comments for the answer)

-b

curious0

Aug 21, 2003, 10:12 AM

You know, as much as I want the spiffy Dual G5 2Ghz doing so would require that I get all new PCI cards and as an audio guy I don't want to have to reinvest into all my PCI hardware. So even though the Single 1.6 may not be the end all be all of computers it may be the right computer for me and others like me.

BTW, MOTU just annouced a trade in program for their 424 PCI cards, check out their site. Othewise if you have a UAD or a PCI Powercore or any PCI audio interface card (like myself) then my understanding is that they won't work on the new 3.3V PCI-X architecture, but will work just fine in the 1.6Ghz machine because it has the older architecture PCI slots.

Time for either firewire powercore or firewire something else I 'spose. I really do think the 1.6Ghz would be a great machine for audio with its increased bus speed and superior hard drive storage architecture and RAM. I want someone to post some info on how many virtual synths it can play and tracks it can record etc etc. Mostly the recording side is of interest to me.

peace,

James

twinturbo

Aug 21, 2003, 10:17 AM

Originally posted by jason166
I keep seeing posts referencing the Xbench benchmarks...

this app is broken on the G5...

What about Cinebench? Is everything "broken" on the G5? I'm guessing I'll just have to see for myself once I get my own machine. Hopefully it'll be as fast as all those bakeoffs showed it to be (and the Pixar Renderman test), but the 1.6 doesn't look too promising.

Anybody get a 1.8 yet to benchmark? Maybe the 1.6 is hobbled in a couple more areas than we though.

Dave K

Aug 21, 2003, 10:22 AM

Originally posted by Macrumors
A few caveats however: There's perhaps one other, or a corolary to caveat 1 at least:

Cinebench uses 64 Bit double precision floating point math which, once combined with their opinion that raytracing tends to not be vector friendly, sees to it that they make no use of SIMD extensions (Altivec/SSE/etc.), on either the PC or Mac platform.

That makes Cinebench a pure test of raw Floating Point performance using code compiled with the 7450 G4 as the target. Thus, looping back to caveat 1: this test is likely not making full/efficient use of the G5's FP units at this time.

This Ars Post (http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=48409524&m=5790920455&r=7280994275#7280994275) from a Maxon developer for further reference

soggywulf

Aug 21, 2003, 10:25 AM

Originally posted by John Q Public
ummmm...no...that 1.6Gz G5 won't out power the Dual 1.42 on 3D Rendering...

Maybe you missed the part about "at the same time".

Originally posted by John Q Public
...with the exceedingly pathetic nVidia FX5200 that the 1.6 ships with (and the Dual 1.42's Stock ATi 9700 Pro) people seem to forget the GPU on the graphics card is a big factor in rendering performance...

This is entirely untrue. Non-real-time 3D rendering is not affected at all by the GPU.

Now if you are talking games, then yes of course the 9700 is much better than the 5200. But you can get a 9800 for the G5, and with that addition games will be much faster on the G5 than the Dual G4. Games make heavy use of the memory bus.

Originally posted by John Q Public
...not to mention Adobe Pornoshop is SMP Aware...Altivec Enhanced

And these will only make a difference if you are CPU-limited, not bus-limited. How hard is it to saturate a 167 MHz bus?

Originally posted by John Q Public
...NOT optimised for the G5's 64bit processor...

As someone else pointed out, it is now. And more and more will be optimized with time.

xtekdiver

Aug 21, 2003, 10:27 AM

Originally posted by BrandonRP0123

The fact that the G5 is 20.1" tall with non removable handles dashes my hopes of spending 1999 on a 1.6 and using it as a rack-mouted server versus 3000 on an xserve.

Why do you think they made it that way? They can't afford people doing just that.

Cubeboy

Aug 21, 2003, 10:29 AM

Just a rather crude comparison but this should explain at least to some extent why the G5 won't significantly outpace a comparative G4 in some benchmarks:

Legend:
G4: Motorola 7455
K7: AMD Athlon
P4: Intel Pentium 4
G5: IBM PPC970

Process Technology:
G4: .18 micron, SOI
K7: .13 micron
P4: .13 micron
G5: .13 micron, SOI

Clockspeed:
G4: 1400 MHz
K7: 2200 MHz
P4: 3200 MHz
G5: 2000+ MHz

Nothing really new here, most of this is already known to all of us so I'll just leave this as it is.

Pipeline Stages:
G4: 7 stages
K7: 10 stages (?)
G5: 16 stages
P4: 20 stages

Branch History Table:
G4: 2048 entires
K7: 2048 entries
P4: 4096 entries
G5:16000 entries (!)

Average Hitrate:
G4: 93% (?)
K7: 93%
P4: 96%
G5: 97%+

Penalty Estimations:
G4: (.07*7= .49)
K7: (.07*10=.7)
P4: (.04*20)= .8)
G5: (.03*16)= .48) or with a 98% Hitrate (.02*16)=.32)

I didn't find any data regarding the G4's branch prediction although it'll probably have a hitrate similar to the Athlon. The G5 is going to have excellent branch prediction, 16000 entries is simply amazing, this is going to more than make up for it's greater branch mispredict penalty.

Integer Instruction Issue Rate (IPC):
G4: 3 instructions/cycle
K7: 3 instructions/cycle
P4: 4 instructions/cycle
G5: 2 instructions/cycle

FP Instruction Issue Rate (IPC):
G4: 1 instruction/cycle
K7: 3 instructions/cycle
P4: 1 instruction/cycle
G5: 2 instructions/cycle

Vector Instruction Issue Rate (IPC):
G4: 2-4 instructions/cycle (Altivec)
K7: 4 single instructions/cycle (SSE)
P4: 4 single/ 2 double instructions/cycle (SSE2)
G5: 2-4 instructions/cycle (Altivec)

Max instructions execute+retire (IPC):
G4: 3 instructions/cycle
K7: ??????????????? (if anyone knows, please post it)
P4: 3 micro-ops/cycle
G5: 4 instructions/cycle

OoO Window:
G4: 16 instructions
K7: 72 instructions
P4: 126 instructions
G5: 215 instructions

This is the important part, while the G5 will be able to extract more ILP out of code, and execute+retire more instructions/cycle, it's two ALUs limits it to only two integer instructions/cycle, this is perhaps the main reason why we might not see the G5 perform as impressively in certain benchmarks. Most likely, it will achieve parity to a similarly clocked G4 in typical integer code. However, the G5 should perform excellently in floating point code, largely due to it's dual double precision FPUs. Theres no doubt that the G5 will easily exceed the G4 in this aspect and probably the Athlon (triple FPU/extended precision:80 bit) as well due to it's ability to extract more ILP out of code (that contains more ILP and less load-stores). Vector performance should be up there with the G4 which is, to say the least, quite impressive.

L1 Cache (Instruction / Data):
G4: 32 KB / 32 KB
K7: 64 KB / 64 KB
P4: 12 KB / 08 KB
G5: 64 KB / 32 KB

L2 Cache:
G4: 256 KB (256 Bit)
K7: 512 KB (64 Bit)
P4: 512 KB (256 Bit)
G5: 512 KB (256 Bit)

Front Side Bus:
G4: 166 MHz
K7: 400 MHz
P4: 800 MHz
G5: 700-1000 MHz

Not much else to say here that isn't already known, it should be noted that the G4 also has a L3 cache that alleviates some of the bandwidth bottleneck, the fast fsb will definitely help the G5 in memory intensive apps. I haven't seen anything about the G5's cache latencies or hitrates so I can't comment on them yet.

The main point is, the G5 probably won't look as impressive in some benchmarks (heavy in integer operations) as it does in other benchmarks (heavy in floating point operations). Perhaps Cinebench employs rendering techniques (i.e raytracing) that are written in integer code. If thats the case, it should come as little surprise that the G5 doesn't perform as well in this benchmark relative to others we've seen so far.

SlowX

Aug 21, 2003, 10:29 AM

AngryAngel

Aug 21, 2003, 10:45 AM

Originally posted by Cubeboy
L1 Cache (Instruction / Data):
G4: 32 KB / 32 KB
K7: 64 KB / 64 KB
P4: 12 KB / 08 KB
G5: 32 KB / 32 KB

Correction: G5 has 64k level 1 instruction cache (preliminary developer note, page 24).

Cubeboy

Aug 21, 2003, 10:49 AM

Originally posted by AngryAngel
Correction: G5 has 64k level 1 instruction cache (preliminary developer note, page 24).

Thank you AngryAngel, it's been corrected.

stingerman

Aug 21, 2003, 10:52 AM

Originally posted by Midiplaya
So, I do quite a bit of audio work and run Virtual PC for a variety of apps. Virtual PC performance is pathetic on a 1GHz machine with 1GB Ram. I am anxiously hoping that the G5 machine will run Virtual PC fast enough to "feel" like a decent PC. If not, I am going to have to switch back. :mad:

With Audio work, you are talking about a lot of Math functions so it would be best to migrate over to native OS X apps when you can, since the G5 has much better math processing capability than any Intel/AMD processor-Audio benefits from a lot of vector math too. In the meantime, even though you have 1MB of RAM, make sure you allocated the max 512MB to the VPC session. Then after starting your Windows session, if you are running XP, turn off the Aqua look and go to the Classic look. For XP and win2k: Turn off anti-aliasing and Show Window when dragging. Turn off shadowing, animations and for the most part 16-bit color is more than enough. Under networking turn off the firewall if it is on and turn off File and print sharing in the protocol stack.

Whenever possible use Win2k over XP and Home over Pro. Win2k and XP are functionally identical at the core as far as any current user needs to be concerned. XP is basically v5.1 and Win2K is v5.0. But in MSFT's case 5.1 is more bloated and slower than 5.0.

I am running VPC with Win2Kpro on a G4/500MHz AGP with 512MB allocated to the VPC session and it is pretty snappy. It is actually faster in many areas to my Compaq 500Mhz with 25MB and just a little slower than my 1GHz Pentium III Dell with 384 MB. I use for VS Studio 2003 development, if that isn't a kick in the pants.

The Grimace

Aug 21, 2003, 11:15 AM

Originally posted by jettredmont
As I've said before, I completely fail to see the logic of VirtualPC to begin with. Just buy a cheapo Windows box, and a USB KVM switch (<$100) to switch between the Mac and PC. A much more flexible solution, plus you get to use a PC that is about 3-4 times as fast as anything you'll get with VPC, and you can use your Mac at the same time as your PC is churning through calculations (just switch using the KVM). Cost is slightly higher, but performance is at least passable!

And has been stated before, some of us don't have this option. I will personally spend most of this summer/fall out of cell-phone range, let alone near any type of 120v AC source. In fact, I drove over 350 miles yesterday to come home, and will drive it again tomorrow. And I need Windows, as much as I may dislike MS products.

(tig)

macrumors12345

Aug 21, 2003, 11:29 AM

Originally posted by twinturbo
When I sit people down and show them that Photoshop doesn't run that much faster and then tell them some dumb "unoptimized" excuse, guess what, they won't listen

Photoshop will run a lot faster (with the G5 plugin) on the Dual 2.0 than it did on the Dual 1.42. Like, twice as fast. And yes, it will also easily beat the Xeon.

The same will be true of many other apps when they are recompiled for the G5.

Once again, the XBench numbers are worthless. Ignore them. I would also not worry too much about the Cinebench numbers until the app has been recompiled for the G5 (though I don't know how soon that will happen since apparently Maxon uses CodeWarrior, not GCC). The fact that it is beating the G4 by 10 to 20 percent per clock cycle in Cinebench is actually pretty impressive given that the code is compiled with the MPC 7450 as the target processor. Other apps will show much bigger gains, especially when they have been recompiled for the G5.

Everyone seems to be forgetting how the MPC 7450 was considered a dud at its release because it was performing 20% slower than the MPC 7400 core, clock for clock (see, for example, http://www.xlr8yourmac.com/G4ZONE/G4_733mhz_review/apple_G4_733_tests.html OR http://www.barefeats.com/G4733.htm). But today the 7450 core matches the 7400 core clock for clock (the 7400 is no longer produced, but it is literally the same core as the G3 750FX, so you can compare the iBook 900 to the PB 12" 867 on non-Altivec tasks to see what I am saying). We are also forgetting how the 1.5 Ghz Pentium 4 was considered an abject failure on release day because it could barely keep up with a 1 Ghz Athlon, but today the 3 Ghz P4 will spank a 2 Ghz Athlon (1.5/1 is the same ratio as 3/2, if that isn't clear).

Actually, you REALLY ought to take a look at the second link (http://www.barefeats.com/G4733.htm)! The Cinebench (Cinema 4D) benchmark in it "proves" that the new 733 Mhz G4 is actually SLOWER than the 533 Mhz G4 it replaced! Yes, that's right, the single G4/533 ran the test in 44 seconds vs. the single G4/733 time of 47 seconds. In other words, before Cinema 4D was recompiled for the MPC 7450, it actually ran 33% SLOWER clock for clock on the MPC 7450 than it did on the MPC 7400. But today, with Cinebench recompiled for the MPC 7450, it runs at the same speed clock for clock on an MPC 7450 and an MPC 7400 (in other words, there was a speed gain of almost 50% from recompiling for the 7450). So, like I said, it is really impressive that the non-recompiled Cinebench app is running 10-20% FASTER clock for clock on the G5 versus the G4.

The moral of the story is DON'T judge a new processor before the before binaries have been recompiled for it.

tazznb

Aug 21, 2003, 11:49 AM

Originally posted by macrumors12345
Photoshop will run a lot faster (with the G5 plugin) on the Dual 2.0 than it did on the Dual 1.42. Like, twice as fast. And yes, it will also easily beat the Xeon.

The same will be true of many other apps when they are recompiled for the G5.

Once again, the XBench numbers are worthless. Ignore them. I would also not worry too much about the Cinebench numbers until the app has been recompiled for the G5 (though I don't know how soon that will happen since apparently Maxon uses CodeWarrior, not GCC). The fact that it is beating the G4 by 10 to 20 percent per clock cycle in Cinebench is actually pretty impressive given that the code is compiled with the MPC 7450 as the target processor. Other apps will show much bigger gains, especially when they have been recompiled for the G5.

Everyone seems to be forgetting how the MPC 7450 was considered a dud at its release because it was performing 20% slower than the MPC 7400 core, clock for clock (see, for example, http://www.xlr8yourmac.com/G4ZONE/G4_733mhz_review/apple_G4_733_tests.html OR http://www.barefeats.com/G4733.htm). But today the 7450 core matches the 7400 core clock for clock (the 7400 is no longer produced, but it is literally the same core as the G3 750FX, so you can compare the iBook 900 to the PB 12" 867 on non-Altivec tasks to see what I am saying). We are also forgetting how the 1.5 Ghz Pentium 4 was considered an abject failure on release day because it could barely keep up with a 1 Ghz Athlon, but today the 3 Ghz P4 will spank a 2 Ghz Athlon (1.5/1 is the same ratio as 3/2, if that isn't clear).

Actually, you REALLY ought to take a look at the second link (http://www.barefeats.com/G4733.htm)! The Cinebench (Cinema 4D) benchmark in it "proves" that the new 733 Mhz G4 is actually SLOWER than the 533 Mhz G4 it replaced! Yes, that's right, the single G4/533 ran the test in 44 seconds vs. the single G4/733 time of 47 seconds. In other words, before Cinema 4D was recompiled for the MPC 7450, it actually ran 33% SLOWER clock for clock on the MPC 7450 than it did on the MPC 7400. But today, with Cinebench recompiled for the MPC 7450, it runs at the same speed clock for clock on an MPC 7450 and an MPC 7400 (in other words, there was a speed gain of almost 50% from recompiling for the 7450). So, like I said, it is really impressive that the non-recompiled Cinebench app is running 10-20% FASTER clock for clock on the G5 versus the G4.

The moral of the story is DON'T judge a new processor before the before binaries have been recompiled for it.

When I used Photoshop on a G5 at the expo it was a lot faster than my dual Xeon machine I use.

Example: I resized a.psd file from a 56Mb (can't remember the exact size right now) all the way to 2.5Gb I was able to see it finish in under a minute.

I went home to try it on my Dell Xeon (with UW SCSI 3),and ended up walking away so it can finish it.

allpar

Aug 21, 2003, 12:00 PM

Comparing the 1.6 G5 to a dual 1.25 G4 is very appropriate since they cost roughly the same amount. (The dual g4 is a bit cheaper.)

With an unoptomized system and not-G5-aware XBench, the G5 was slower than the dual G4 1.25 - faster on memory, slower on hard drive, slower on CPU. This isn't very surprising.

The G4's dual processor design also means it is sometimes very snappy where the G5 might not be.

It is not really fair at this point to compare a dual 2.0 to a dual 1.25 G4, since the dual 1.25 G4 is selling for $1,500 or so and the dual 2.0 for nearly $3,000. The current comparison should be single 1.6 to dual 1.25...where the G5 is only a couple of hundred dollars more expensive.

I'll add that the smaller cases on the G4s have room for two optical drives (but not cartridge DVD-RAM!!!) and four hard drives! And the current models are pretty darned quiet. Case design is STUPID though with no reset button, and the on button ONLY on the case (unless you have an Apple monitor) and no possibility of a cartridge optical drive or zip drive. My $50 PC case has four external bays and two internal bays - I have in it a catrdige drive ($8), a DVD-RAM I used to have in my B&W G3 which I got "new-used" for $80 some time ago, a CD-writer, and a CD-reader. Yup, using 'em all. I'd rather have the DVD-RAM in the Mac...but it's not going to happen. (The second optical slot will eventually go to a Superdrive when they reach $100.)

HOWEVER, of course, the G5 has a lot more room to grow - as more programs are optimized for the G5 it will become faster.

twinturbo

Aug 21, 2003, 12:10 PM

Originally posted by macrumors12345
We are also forgetting how the 1.5 Ghz Pentium 4 was considered an abject failure on release day because it could barely keep up with a 1 Ghz Athlon, but today the 3 Ghz P4 will spank a 2 Ghz Athlon (1.5/1 is the same ratio as 3/2, if that isn't clear).

This was a really great explaination, but for those of us who've bought the G5 what does it mean? Will we have to wait awhile with underpar performance, waiting for optimized and recompiled apps? For those people who bought the 3Ghz P4 I'm sure life is good compared to their AMD brethren, but what about those on the 1.5 P4-did they get screwed (or did the new line of applications that push the 3Ghz P4 past the Athlon also help the 1.5P4 against the 1Ghz Athlon?)?

allpar

Aug 21, 2003, 12:18 PM

In the case of the 1.6 and 1.8, I believe these will provide out of the box performance somewhat exceeding the dual G4 (feel sorry for those who just spent $3,000 on a dual G4!).

In time, while the G4 seems to get slower, the G5 will remain current or get faster as 10.3 comes out, and other app makers recode. Examples - photoshop (already done!), Safari, VirtualPC.

The G5 is NOT slower than the dual G4 in real life, I'd guess, but the 1.6 is probably similar - and since the price is similar, that makes sense.

But there ARE advantages to dual processors!

PieMac

Aug 21, 2003, 12:29 PM

Originally posted by allpar
In the case of the 1.6 and 1.8, I believe these will provide out of the box performance somewhat exceeding the dual G4 (feel sorry for those who just spent $3,000 on a dual G4!).

In time, while the G4 seems to get slower, the G5 will remain current or get faster as 10.3 comes out, and other app makers recode. Examples - photoshop (already done!), Safari, VirtualPC.

The G5 is NOT slower than the dual G4 in real life, I'd guess, but the 1.6 is probably similar - and since the price is similar, that makes sense.

But there ARE advantages to dual processors!

So the standing question is do the advantages of the current dual 1.25 mhz G4 processor out weigh the 1.6 G5 single processor? I spoke with an Apple rep the other day and she said that the only time there is an advantage to the dual is when you are multitasking. If you don't run a lot of apps at once, it's not worth it and that even the single 1.25 would be perfectly suitable.

reflex

Aug 21, 2003, 12:45 PM

eric_n_dfw

Aug 21, 2003, 12:51 PM

Originally posted by BrandonRP0123
Furthermore, if you own a machine capable of it, perhaps its time to upgrade that ol' graphics card to work with Quartz Extreme?Amen - I did this on a my Dual G4 500, made a world of difference!

Dont Hurt Me

Aug 21, 2003, 12:59 PM

I have to say that the high end upgrades for older machines are a very good alternative's for the g5. Yes i will be getting a g5 but it may be a year or 2? Heck max out that ram, video card, cpu upgrade you still save a grand. Now a year from now apps will be coming out that know how to use that g5 then its abilities will be shown but for now it is not being used to its most, sort of like Mac games that dont know what altivec is. Anyways some great posts!

macrumors12345

Aug 21, 2003, 01:07 PM

Originally posted by twinturbo
This was a really great explaination, but for those of us who've bought the G5 what does it mean?

Basically it means that your computer will actually get faster as it ages (though of course it will still be slower than the new machines that replace it), much like OS X 10.2 is faster than OS X 10.0 (if you can remember that) on the same hardware.

But it is important to keep in mind that even on most non-recompiled apps, the G5 will still outperform the current high end G4's, due to its higher clock speed, faster memory bus, and superior architecture (the exception would be apps that make heavy use of both G4's and are not memory bound...then a Dual 1.42 will outperform a Single 1.6). And it's not like the current G4's are slow, so I wouldn't worry about it. Furthermore, the most processor intensive apps (e.g. Photoshop, Final Cut Pro, Logic, Lightwave, etc.), which are the ones where you will really need all the speed you can get, are the ones that are most likely to be recompiled first.

jkojima

Aug 21, 2003, 01:08 PM

Do you thing I'll void the warranty if I cut the off the handles with a hacksaw?

<begin heart attack> AARRGGGMMMPH!!! </end heart attack>

Dont Hurt Me

Aug 21, 2003, 01:36 PM

I can relate my desk area is just big enough for the quicksilver with 1/2 inch to spare ,a g5 would need a new desk or a hacksaw. Did they need to make it so big? a 1.6 & 1.8 have a lot of wasted space.

Makosuke

Aug 21, 2003, 01:43 PM

Validity of benchmarking in general put aside, I'm actually very happy with these numbers, so long as you take into account the large list of caveats. I'm repeating what others have said in different words, but:

-The G5 is architecturally very different from the G4; as big a jump in the Mac since at least the G3 vs. 604e. This means that without performance tuning (at least a recompile), you have no idea of what it's capable of, only how it does with code tuned for an entirely different processor.

-XBench is, I think, at least somewhat G4 optimized; at the very least, it's completely un-G5 optimized.

-Benchmarks only offer the most general idea of performance; actual app performance is what really counts.

So basically, what we can take away from numbers like these is, in very rough terms, how the G5 is going to compete with a G4 when running code that's optimized for the G4 (old, non-updated programs). Meaningful, but hardly the G5's strength.

And taking that into account, I read three things from the benchmarks:

1) The G5, at worst, is running at about the same speed as a comperably clocked G4 if you extrapolate.

2) In the only "real world" test, Cinebench, the G5 is around 40% faster, per clock, than the G4.

3) In some tests, the G5 runs circles around a dual CPU G4 at only slightly lower clock speed.

What this means is that, in the real world, the absolute worst you're going to get out of your new G5 is the same performance as a similarly clocked G4. That means that, at the very least, a 40% speed increase if you go dual 2Ghz. Nothing to sneeze at.

In some cases--still running unoptimized code--the low-end G5 is significantly faster than the top-of-the-line G4. Though degree of multiprocessing aware-ness obviously makes a big difference, that points to a huge advantage for the G5.

On top of all that, most people in the know seem to think that G5 optimization can at least double speed vs a G4 optimized program. That points to a speed jump of aproximately 300% between the two generations of Macs. That makes me happy.

There, I'm done.

twinturbo

Aug 21, 2003, 02:00 PM

Originally posted by Makosuke

1) The G5, at worst, is running at about the same speed as a comperably clocked G4 if you extrapolate.

That means being kicked by the Pentium/Athlon camp

2) In the only "real world" test, Cinebench, the G5 is around 40% faster, per clock, than the G4.

That means getting performance in the midrange of the current x86 lineup. Still not that great if you ask me.

3) In some tests, the G5 runs circles around a dual CPU G4 at only slightly lower clock speed.

I hope this is were most apps perform at, cause that's going to place the G5 in the world-class performance category. 300% increase is good, but from what we've seen so far, it just ain't there yet. At least Photshop is now optimized, and I'm sure Final Cut is. Not sure about After Effects, but hopefully that is too. With Maya and a couple other apps optimized, that would be sufficient for us in the content-creation market. But I really, really wanted this to be the one time where we could really stomp all over x86 without any questions or caveats. At this point, it doesn't seem as though it's looking like that will happen. Can't wait to see the dual vs. dual that Charlie White will run once the dual 2Ghz comes out. And it's those kind of benchmarks that people will notice. If the G5 performs like crap on that, then I don't know what to say, cause the excuse-making at that point in time ain't gonna work.

jaedreth

Aug 21, 2003, 02:16 PM

Actually, Photoshop is not *optimized* in the truest sense of the word. Just a plug in can't do that. It now *supports* the hardware, and thus won't treat the hardware like it would a G4.

Big difference.

Final Cut Pro optimized for G5? No. Not yet.

The FCP box would bear a silvery "sticker" that says Optimized for G5".

In fact, I bet they don't optimize it for G5 until PowerBooks and iMacs have gone G5.

Jaedreth

macrumors12345

Aug 21, 2003, 02:59 PM

Originally posted by jettredmont
Um, no it wouldn't have. By all projections, the 970 is NEVER set to approach more than 2/3 the top speed of the Pentium 4/5.

Untrue (even given that you mean clock speed, as opposed to actual speed).

Today we have 2.0GHz G5s and 3.2GHz P4s. A year from now we will have 3GHz G5s but 4.8GHz (if I recall the Intel roadmap correctly) P5s.

Hmmm...well, either you recall incorrectly, or you saw a incorrect roadmap. Currently the G5 is running at 2 Ghz and the P4 is running at 3.2 Ghz, so there is a 60% clock speed advantage for Intel. Apple/IBM have committed to increasing clock speed to 3 Ghz within one year, which presumably means no later than Q3 2004. Intel is not committing to anything, but their official roadmap has the Pentium 4/5 at "3.4 Ghz or greater" through June 2004 (Q2 2004) (http://www.intel.com/products/roadmap/index.htm?iid=ipp_browse+process_roadmap&). There is some upside potential in the "or greater" part (though definitely not 1 Ghz worth of upside potential!), but the recently confirmed power dissipation problems of this processor are likely to limit the clock speed increases. Certainly 3.4 Ghz is the fastest projected speed through Q1 2004 (http://www.theinquirer.net/?article=11092), so at best we are likely to see is 3.6 Ghz in Q2 and 3.8 Ghz in Q3. The highest possible projection I have seen for Q3 2004 (from before the power dissipation problems were apparent) is 4 Ghz - that is unlikely to be attained (http://pc.watch.impress.co.jp/docs/2003/0812/kaigai01l.gif).

So the bottom line is that in one year the G5 should be at 3 Ghz, whereas Intel should be between 3.6 to 3.8 Ghz, with 4 Ghz at the upper end of possibilities (but increasingly unlikely). So Intel's clock speed advantage will be reduced to between 20 to 33 percent (as opposed to 60 percent right now) - not bad! And of course the G5 will be getting much more done per clock cycle, so in real world performance it should have a definite advantage.

sparks9

Aug 21, 2003, 03:22 PM

The G5 bucket is gigantic, huh?

eric_n_dfw

Aug 21, 2003, 03:33 PM

Originally posted by twinturbo
This was a really great explaination, but for those of us who've bought the G5 what does it mean? Will we have to wait awhile with underpar performance, waiting for optimized and recompiled apps? For those people who bought the 3Ghz P4 I'm sure life is good compared to their AMD brethren, but what about those on the 1.5 P4-did they get screwed (or did the new line of applications that push the 3Ghz P4 past the Athlon also help the 1.5P4 against the 1Ghz Athlon?)? Acutally, yes. P3's were faster at many things than P4's were untill P4 optimized code started showing up.

You've got to realize that while the G5 may run any PPC binaries, it will not be able to shine unless those bits are arranged in a way that takes advantage of the G5. It's like trying to run a top fuel drag car on low octane gasoline - it might run but you're not going to win any races with it!

So yes, early adopters of G5 machines will have to live with relatively mediorcre performance until their software is recompiled for the G5; just like the first Power Mac users did on the 601, just like G4 users did before Alti-Vec app's came out and just like Windows users did on the P4.

(Disclaimer: Intel did have some design issues in the early P4's which contributed to it being slower than P3's at lower clock frequencies - but the main problem was still that existing code was not taking advantage of P4 technologies like NetBurst and whatnot - it still is a problem with the new HyperThreading P4's as app's are slowing catching up to use it)

John Q Public

Aug 21, 2003, 03:46 PM

Originally posted by allpar
The G5 is NOT slower than the dual G4 in real life, I'd guess, but the 1.6 is probably similar - and since the price is similar, that makes sense.

But there ARE advantages to dual processors!

...and would partially explain why Apple dropped the Dual 1.42 FW800 from the line...opting for the original MDD (with only FW400) so as not to outperform and of the G5's and steal their thunder for less money...

Dont Hurt Me

Aug 21, 2003, 03:49 PM

Still pretty amazed at the size difference, the g5 is gigantic in that photo compared to the mdd.

PieMac

Aug 21, 2003, 03:55 PM

I remember seeing a thread with a link to photos of a G4 and a G5 and the G5 didn't look near as large in comparison in those pictures. I think I located the link, but the photos have since been removed. Any body know of any other comparison photos?

wilburpan

Aug 21, 2003, 04:26 PM

Regarding the photo of the G5 interior:

Comparing the photo of an actual G5 to the one on the Apple web site (http://http://www.apple.com/powermac/design.html), I see that the shipping units have the cool pink and blue wavy line inserts removed. :)

John Q Public

Aug 21, 2003, 04:33 PM

Originally posted by Dont Hurt Me
Still pretty amazed at the size difference, the g5 is gigantic in that photo compared to the mdd.

yup...and with all that extra size...half the expandability...since SATA drives aren't significantly faster than the ATA100 on current G4's...and you can only have 2 HDDs...and one optical drive...

in a sense...Apple is screwing people into buying a G5 though...if want a new Mac and want FireWire800...you'll have to buy a G5 (seeing that they dropped the FW800 MDD's and opted to sell last-year's-model as the "New" G4)...

...and if you look at the Apple Store site...you have to look WAY down to the bottom...below the iPod....below the "Apple Software" to even find it on the page...

Apple would rather you buy an iMac or eMac than a G4 Tower...personally I'd have rather seen them completely drop the G4 Tower than treat it as the "Red-Headed Step-Child" the marketing people have done...

After my B&W G3 finally died last month...I did the research...read the articles...crunched the numbers...weighed how much I hate the design of the G5...looked at expandability...and finally thought about what goodies I could buy with how much I saved in buying one of the last FW800's...the decision was natural and easy...

...screw what Apple wants me to do...regardless of the opinions...the G4 is in many ways (from my POV) an excellent machine...and in quite a few ways superior to the way Apple now wants me to use my computer...

I've built many PC's (from 386's to a 3.0 Gz Barton Cored Athlon) over the years (rather than buy prebuilt)...for the same reasons I WON'T buy a G5...to an extent Apple already decides for you what you can or can't do with your shiny new toy...I can live with that...but to purposely create a machine that limits your choices as much as the G5 does...PCI-X doesn't have industry wide support yet...SATA is a nice technology but doesn't live up to it's promises...and limiting expansion inside the case...my G3 had a ComboDrive and 7 HDD's (smallest being 40GB 2:ATA33 1:Slave on the CD Chain...4 on a Sonnet TempoTrio)...when I moved to G4 I gave up 3 drives (less expandability...but I was still able to use the old ATA100 drives from the G3)

I guess for me...not going to the G5 meant not giving up being able to make MY machine MY machine...that and being able to afford extra goodies (like GeForce4 Ti...and 23" Cinema HD Display)

In a non-typically long winded fashion...I'm saying the choice is yours...personal preference is more a factor than electronic d*** measuring contests...I switched fully (ok...my laptop is still Windoze...can't be perfect...seeing that I make my living coding for companies using Micro$haft OS's) to Mac because I wanted to...I've always had one around (much to my SO's displeasure...although she likes having a Pentium-Rated 3Gz Athlon on her desk to play with now)...I like my Mac...and I don't care who knows it or what they thnk (especially old friends who now call me a traitor) :P

Dont Hurt Me

Aug 21, 2003, 04:51 PM

Johnq i hear you, the g5 is and will be a excellent machine but it does have a lame fx5200 and you have to start over with memory. I to have reached the conclusion that my quicksilver wasnt called that for nothing so when and if i max the machine out ill know. ati 9800 is sounding so sweet,so is another 512 of memory. maybe a superdrive?? or a ati9800 all in wonder for mac if ATI ever gets off there ---.

daveg5

Aug 21, 2003, 05:02 PM

i think the reason for the L3 cache in 7440 g4 and above is because moto could not deliver the faster L2 cache in any size larger then 256k which is 400% smaller then the 1MB L2 cache in the previous g4s' 7400/7410
a test at insidemacgames between an agp powermac g4 with the standard 500 g4 1MBL2 cache and the same machine at 800 with a 256k L2 cache showed the 500 faster due to the larger L2 cache and smaller pipeline 4vs7
however with an 800 processor with an added 1MB L3 cache in addition to the 256k L3 the 800 was then slightly faster then the 500.
the original G4 was one of the fastest processors apple ever had based on performance per clock cycle, however it could not scale, because of a small 4 pipeline stages and large L2 cache.
the g5 vs g4 debate is similar, what is it 29 pipeline stages to 7, L2 and L3 cache to L2 alone, simply put the g4 is a more efficient cheap in a few areas with a number of major flaw no bandwidth for fsb, memory, and just about everything else, limited to 256k L2 cache which made L3 necessary, extremely slow scaling in clock speed, or should i just say moto has not got a clue.
i dont know this to be true but is the g4 L2 cacheing more advanced then the G5 8way as opposed to 4 way and its altivec units are superior also supposedly.
however that wont do any good stuck at 1.4 and 167 fsb

daveg5

Aug 21, 2003, 05:05 PM

if you have an older g3/g4 and so is sata pci cards

PieMac

Aug 21, 2003, 05:12 PM

Originally posted by John Q Public
yup...and with all that extra size...half the expandability...since SATA drives aren't significantly faster than the ATA100 on current G4's...and you can only have 2 HDDs...and one optical drive...

in a sense...Apple is screwing people into buying a G5 though...if want a new Mac and want FireWire800...you'll have to buy a G5 (seeing that they dropped the FW800 MDD's and opted to sell last-year's-model as the "New" G4)...

...and if you look at the Apple Store site...you have to look WAY down to the bottom...below the iPod....below the "Apple Software" to even find it on the page...

Apple would rather you buy an iMac or eMac than a G4 Tower...personally I'd have rather seen them completely drop the G4 Tower than treat it as the "Red-Headed Step-Child" the marketing people have done...

After my B&W G3 finally died last month...I did the research...read the articles...crunched the numbers...weighed how much I hate the design of the G5...looked at expandability...and finally thought about what goodies I could buy with how much I saved in buying one of the last FW800's...the decision was natural and easy...

...screw what Apple wants me to do...regardless of the opinions...the G4 is in many ways (from my POV) an excellent machine...and in quite a few ways superior to the way Apple now wants me to use my computer...

I've built many PC's (from 386's to a 3.0 Gz Barton Cored Athlon) over the years (rather than buy prebuilt)...for the same reasons I WON'T buy a G5...to an extent Apple already decides for you what you can or can't do with your shiny new toy...I can live with that...but to purposely create a machine that limits your choices as much as the G5 does...PCI-X doesn't have industry wide support yet...SATA is a nice technology but doesn't live up to it's promises...and limiting expansion inside the case...my G3 had a ComboDrive and 7 HDD's (smallest being 40GB 2:ATA33 1:Slave on the CD Chain...4 on a Sonnet TempoTrio)...when I moved to G4 I gave up 3 drives (less expandability...but I was still able to use the old ATA100 drives from the G3)

I guess for me...not going to the G5 meant not giving up being able to make MY machine MY machine...that and being able to afford extra goodies (like GeForce4 Ti...and 23" Cinema HD Display)

Many excellent points. I have looked high and low both locally and on line to try and find a firewire 800 G4 that might fit the bill for myself, but unfortunately, these are few and far between. I need one with the superdrive as well which really limits your choices....there are very few superdrive model G4's left at any resellers, period. So it's either order "last years model" from Apple or go with the G5 and I have a feeling I will be choosing the G5 (and I personally prefer the actual design of the G4 as well). As much as I don't like feeling like I am being coerced into this decision by Apple's marketing tactics, I have to give them credit...Steve isn't stupid.

jettredmont

Aug 21, 2003, 05:19 PM

Originally posted by macrumors12345
Hmmm...well, either you recall incorrectly, or you saw a incorrect roadmap.
[...]
Intel is not committing to anything, but their official roadmap has the Pentium 4/5 at "3.4 Ghz or greater" through June 2004 (Q2 2004) (http://www.intel.com/products/roadmap/index.htm?iid=ipp_browse+process_roadmap&).

[...]

The highest possible projection I have seen for Q3 2004 (from before the power dissipation problems were apparent) is 4 Ghz - that is unlikely to be attained (http://pc.watch.impress.co.jp/docs/2003/0812/kaigai01l.gif).

First, Intel official roadmaps are historically quite conservative more than a month out. They are for planning more than six months out. For a year out, you need to tap into more "dubious" sources ... :)

Second, yes that last was where I got the "4.8 IIRC" figure ("4.4 GHz" was the figure that had stuck in my mind, Tejas debut projected, which had morphed with "3.8 GHz", Prescott, mid Q1 2004).

What power dissipation problems are you talking about? Haven't heard of anything major or unexpected.

jettredmont

Aug 21, 2003, 05:29 PM

Originally posted by jayscheuerle
Interesting that on Apple's own page, they had to scale up a PC to appear to be as large as the G5. Even then, the PC still has 2 optical drives, a floppy/zip drive, space for 2 hard-drives...

Half of the G5's space appears to be taken up by fans and heat-sinks...

Funny, I have that particular Gateway (1.5GHz P4 from a few years back) beast sitting in my office right here next to a 733MHz G4 and ... well, it looks pretty much th size relative to the old G4 as the G5 does. On the other hand, newer Gateways are a good 2" shorter than the old ones ...

That PC photo may have been blown up slightly, but definitely not a huge amount.

macrumors12345

Aug 21, 2003, 05:38 PM

Originally posted by jettredmont
What power dissipation problems are you talking about? Haven't heard of anything major or unexpected.

They first surfaced a month or two ago, but they were only recently confirmed by internal Intel docs. See http://www.theinquirer.net/?article=11092

Basically, Prescott is dissipating 103 watts at 3.4 Ghz. Now, it is not big news that a processor is dissipating around 100 watts - I believe that some P4s and Athlons have neared 100 watts, and the Opteron may actually be over 100 watts. The big news is that it is dissipating 103 watts on the *new* 90 nm process, i.e. it is doing so at the lowest clock speeds they had planned for that process!! If it were burning 103 watts on the 130 nm process, that wouldn't be such a big deal, because you would expect the power consumption to fall down to more reasonable levels when they move to the cooler 90 nm process, allowing for further frequency scaling. But the fact that it's at 103 watts at the *beginning* of its life cycle on this process, rather than at the end, is potentially a much bigger problem. Needless to say, Intel is going to have to make some revisions if they want to be able to significantly scale the clock speeds on this process. Otherwise the power consumption is going to get out of control.

And needless to say, systems built around this chip are not going to be quiet. ;-)

job

Aug 21, 2003, 06:28 PM

This is the one thing I can't understand.

Before the G5 was released, all the Mac users were screaming for SATA, AGP 8x, dual-channel DDR400, faster buses, and in a few cases, PCI-X.

Now that we have this tech to buy, people are beginning to say that they really don't need the technology that they have been waiting for years?! I don't think we could have expected that people would be more willing to buy old G4s when the G5s were made available.

Am I missing something here?

Also, lots of people complaining about performance. "compares only to mid-range current x86 hardware, yadda, yadda, yadda..."

Need I remind anyone that this is the baseline G5 which is competing with the current mid-range x86 hardware?

daveg5

Aug 21, 2003, 07:36 PM

Originally posted by job
This is the one thing I can't understand.

Before the G5 was released, all the Mac users were screaming for SATA, AGP 8x, dual-channel DDR400, faster buses, and in a few cases, PCI-X.

Dave G5: those people wanted it in a similar case design with 2 optical 4 harddrives 4 pci and the ability to use 5v pci cards and 4x agp cards basically they waanted all new but still wanted full expansion and capatability with all thier pci and agp stuff and the great g4 case.

Now that we have this tech to buy, people are beginning to say that they really don't need the technology that they have been waiting for years?! I don't think we could have expected that people would be more willing to buy old G4s when the G5s were made available.

Dave G5: now that we have it they dont see a major performance benefit in the g5 over the g4 at least not yet as it will take time for optimizing just like with the g4 vs the g3. performance wise in osx when multitasking the dual G4 does seem to be the best performance buy the most upgradeable and most compatable especially for those with older pro sound cards.. this will of course change as apps are optimized except while multitasking 2 is better then 1

Am I missing something here?

Dave G5: no, they want thier cake and to eat it too. they want everything thats good with the new, while still keeping everything whats good with the past

Also, lots of people complaining about performance. "compares only to mid-range current x86 hardware, yadda, yadda, yadda..."

DaveG5: we will always complain until we can whoop intels a!s at all price points, i so wished apple could have made all g5's duals at the same price points and there would be very little complaints, this may yet happen

Need I remind anyone that this is the baseline G5 which is competing with the current mid-range x86 hardware?

DaveG5: this is true i cant wait for more real world test

John Q Public

Aug 21, 2003, 07:38 PM

Originally posted by PieMac
Many excellent points. I have looked high and low both locally and on line to try and find a firewire 800 G4 that might fit the bill for myself, but unfortunately, these are few and far between. I need one with the superdrive as well which really limits your choices....there are very few superdrive model G4's left at any resellers, period. So it's either order "last years model" from Apple or go with the G5 and I have a feeling I will be choosing the G5 (and I personally prefer the actual design of the G4 as well). As much as I don't like feeling like I am being coerced into this decision by Apple's marketing tactics, I have to give them credit...Steve isn't stupid.

SuperDrive shouldn't be an issue...I bought mine without one...and NewEgg has Pioneer DVR-105BK (the same 4X DVD-RW as Apple used in FW800's) for $154.99 (with free FedEX Ground Shipping)...not to mention CAS 2 PC2700 memory is MUCH cheaper than high quality PC3200 that the G5 needs...and if I remember correctly...PowerMax (http://powermax.com) should still have FW800's...

ColdZero

Aug 21, 2003, 07:44 PM

Its a matter of price, the G5 1.6 is not a very attractive option at $1999 when the $700 I just spent on upgrading my g4 to dual 1ghz can still compete with it. Not as fast, but not $1299 slower.

PieMac

Aug 21, 2003, 07:44 PM

Originally posted by John Q Public
SuperDrive shouldn't be an issue...I bought mine without one...and NewEgg has Pioneer DVR-105BK (the same 4X DVD-RW as Apple used in FW800's) for $154.99 (with free FedEX Ground Shipping)...not to mention CAS 2 PC2700 memory is MUCH cheaper than high quality PC3200 that the G5 needs...and if I remember correctly...PowerMax (http://powermax.com) should still have FW800's...

I was wanting to purchase the superdrive built in because I want to be able to use iDVD. Thanks for the suggestions...I think I've already been to those sites but I will double check...you never know!

soggywulf

Aug 21, 2003, 08:15 PM

Originally posted by job
Need I remind anyone that this is the baseline G5 which is competing with the current mid-range x86 hardware?

The G5 is a good machine. But the baseline G5 compares in price closer to the mid-range PC hardware. And it is priced with hi-range PC hardware if you DIY.

Originally posted by DaveG5
now that we have it they dont see a major performance benefit in the g5 over the g4 at least not yet as it will take time for optimizing just like with the g4 vs the g3. performance wise in osx when multitasking the dual G4 does seem to be the best performance

I don't think so. Not unless you are running (some) continuous benchmarks. If you are multitasking, and actually doing heavy processing with your background apps (as opposed to just opening a bunch of apps and switching between them), then I think the fast memory will make the G5 machine much faster, snappier, and more productive.

John Q--looks like you are defending a purchase. Be objective. The Dual 1.4 you got may suit your needs just fine, but it's not a miracle-machine.

Forget benchmarks. Benchmarks are just for d***-measuring. Except game FPS benchmarks, these actually accurately represent how the computer is used--if you play games. I suppose if you are just doing continuous PS filters on the machine, benchmarks are useful for that too. All other cases, they are pretty useless.

John Q Public

Aug 21, 2003, 08:18 PM

Originally posted by job
This is the one thing I can't understand.

Before the G5 was released, all the Mac users were screaming for SATA, AGP 8x, dual-channel DDR400, faster buses, and in a few cases, PCI-X.

...personally...I don't think SATA is a great thing...it's only marginally faster than ATA100 (and the drives are more expensive naturally because of the new Techno-Babble Buzzwords)...

...AGP 8x...really isn't...it's only roughly 3-4% faster than AGP 4X...and unless you're running a machine with more than 1GB RAM...it's likely you AGP can only palletize 64-128MB of system memory anyway...

...DDR400...although fast...isn't the "End-All" of power and speed...yes it's nice...but in the real world how fast can your machine cache a web page or process what's coming across the LAN without creating a bottleneck at the processor's bus is a more likely case than would be through synthetic benchmarks...besides...it's always a case where more memory is better than faster memory (that's one of the things I'll concede to the G5...but I really don't have a need for more than the 2GB that's in my machine)

...lastly...PCI-X...where to begin with that one...they've been on PC's for the past 2 years (server/workstation logic boards)...but like the 64bit 33Mz PCI that's been on the Mac since '99...there's little industry support...that part will changein the next couple years as AGP falls out of favor entirely for the greater bandwidth afforded by PCI-X...but that won't be for another year or so...

The advertising should be (insert Jeff Goldblum voiceover) "...the new PowerMac G5...with it's PCI-X...AGP 8X...Serial ATA interface...SuperDrive and Dual-Channel DDR400 Memory Bus...It's the World's-Fastest-Fully-Buzzword-Compliant Personal Computer ever made..."

I have a PC that is "Buzzword Compliant" with a AthlonXP 3000+...it's in my SO's posession...'cause I like my pathetic little FW800 Dual 1.25Gz (or so some would try to have me believe'cause I don't want the over-hyped industrial-grade-ugly G5)

...the only thing I've ever asked for from Apple was to bring back the "Cube Era" black-keyed Apple Pro KB...the white keys are a pain-in-the-ass to keep clean...

daveg5

Aug 21, 2003, 08:29 PM

Originally posted by soggywulf
The G5 is a good machine. But the baseline G5 compares in price closer to the mid-range PC hardware. And it is priced with hi-range PC hardware if you DIY.

I don't think so. Not unless you are running (some) continuous benchmarks. If you are multitasking, and actually doing heavy processing with your background apps (as opposed to just opening a bunch of apps and switching between them), then I think the fast memory will make the G5 machine much faster, snappier, and more productive.

DaveG5:wow i never heard that one so i single 1.6 will out do a dual g4 1.25 when multitasking and perhaps even on multi aware apps. i thought one of osx's best attributes was its ability to work natively with duals. if i hear you the 1.6 g5 will multi task better then a dual G4 comparably equipped simply because of the bandwith advantages.
now you may be right but we still dont have any benchmarks to support that at all. i would love to impot in itunes while exporting in final cut, while rendering in imovie and playing backs some tracks in soundtrake, while working on filters in photoshop.
if the single g5 does multi tasking better and is faster when using multi aware apps then the dual 1.25-1.42 then that is indeed remarkable and i want one.
i just thought multitasking and multi aware apps would be the few areas where the old duals would surpass the single g5s' equally equipped. i hope i am wrong though. i know bandwidth is remarkable for memory based things like logic and 3d video. but for simple things like mp3 encoding, 2d rendering, and playing back tracks with no dsp the g4's bandwidth is never reached.

John Q--looks like you are defending a purchase. Be objective. The Dual 1.4 you got may suit your needs just fine, but it's not a miracle-machine.

wizard

Aug 21, 2003, 08:52 PM

Well all of thses things would be very helpful if they where connected to a processor that could make use of them. In the case of the 970 we have machine that simple doesn't cut the mustard so to speak. It's nice that the 970 excels at FP but FP is only applicable to a small segment of the user base.

While many people are jumping the gun with respect to their reactions to these first postings, it is probally an unadvised reaction. Apples own publicly posted documentation indicates clearly where the 970 comes up short, why anybody would be surprised is beyond me. In any event I can't see why anybody would puchase the 1.6 GHz machine, it is just an odd implementation relative to its sister machines.

I'm not really sure what the 1.6 GHz machine is, the baseline is really the 1.6 GHz machine. The 1.6 is just to differrent from the rest of the line to be considered baseline - it is its own line. Like someone else alluded to it is the G5 'yikes' machine.

I lok at it this way the G5 offers the Mac consumer everything they asked for except for a high performance CPU. Hopefully that will arrive around January. I will also admit that the GPUs are a bit lacking also.

Thanks
Dave

Originally posted by job
This is the one thing I can't understand.

Before the G5 was released, all the Mac users were screaming for SATA, AGP 8x, dual-channel DDR400, faster buses, and in a few cases, PCI-X.
...snipped
Also, lots of people complaining about performance. "compares only to mid-range current x86 hardware, yadda, yadda, yadda..."

Need I remind anyone that this is the baseline G5 which is competing with the current mid-range x86 hardware?

garymm

Aug 21, 2003, 09:04 PM

Originally posted by wizard

I will also admit that the GPUs are a bit lacking also.

Umm, the 9800 pro is there if you want it. That's about as good as it gets.

wizard

Aug 21, 2003, 09:10 PM

Your statements below are all well and good but you seem to mis one important thing here. The 970 does not have the hardware to optimize for, it only has 2 integer units. The VMX unit barely competes with the one in the G4. What the 970 does have is FP performance and more bandwidth, these advantages only come into play for certain code bases so you do not get a uniform increase in performance.

People should not get their hopes up to far. For some code bases the 970 will excel and simply be fantastic. You will never see a fantastic performance increase in run of the mill software. If your lucky you will get clock rate increase and maybe 10% over the G4.

Dave

Originally posted by eric_n_dfw
Acutally, yes. P3's were faster at many things than P4's were untill P4 optimized code started showing up.

You've got to realize that while the G5 may run any PPC binaries, it will not be able to shine unless those bits are arranged in a way that takes advantage of the G5. It's like trying to run a top fuel drag car on low octane gasoline - it might run but you're not going to win any races with it!

So yes, early adopters of G5 machines will have to live with relatively mediorcre performance until their software is recompiled for the G5; just like the first Power Mac users did on the 601, just like G4 users did before Alti-Vec app's came out and just like Windows users did on the P4.

(Disclaimer: Intel did have some design issues in the early P4's which contributed to it being slower than P3's at lower clock frequencies - but the main problem was still that existing code was not taking advantage of P4 technologies like NetBurst and whatnot - it still is a problem with the new HyperThreading P4's as app's are slowing catching up to use it)

rog

Aug 21, 2003, 09:16 PM

Originally posted by job
"

Need I remind anyone that this is the baseline G5 which is competing with the current mid-range x86 hardware?

Except that the low end G5 sells for more than most high-end PCs with much better video cards. It's too bad Apple is letting the slower G5s trickle out 1st. People will be disappointed and underwhelmed, and when the DP 2GHz comes out, people won't care much. If that came out first, or at the same time, everyone would be going on and on about how fast it is. Once again, Apple turns what should have been a successful rollout into a disappointing mess.

wizard

Aug 21, 2003, 09:16 PM

Yes his points are excellent. I'm wondering if you guys have considered the X-Serve as possibly the best choice for a G4 based machine at the moment?

Something that has been running throu my head recently.

Dave

Originally posted by PieMac
Many excellent points. I have looked high and low both locally and on line to try and find a firewire 800 G4 that might fit the bill for myself, but unfortunately, these are few and far between. I need one with the superdrive as well which really limits your choices....there are very few superdrive model G4's left at any resellers, period. So it's either order "last years model" from Apple or go with the G5 and I have a feeling I will be choosing the G5 (and I personally prefer the actual design of the G4 as well). As much as I don't like feeling like I am being coerced into this decision by Apple's marketing tactics, I have to give them credit...Steve isn't stupid.

John Q Public

Aug 21, 2003, 09:16 PM

Originally posted by daveg5
John Q--looks like you are defending a purchase. Be objective. The Dual 1.4 you got may suit your needs just fine, but it's not a miracle-machine.

...actually no...not defending a purchase...I made the rational decision to buy a G4 before the G5's hit stores rather than waiting for a machine I wouldn't have been happy with...just because Uncle Steve says it's the best thing since buttered-bread...

...and trying to illustrate that although Apple NEEDS for you to buy into all the hype and buy a G5...you don't need to buy it...

leicaman

Aug 21, 2003, 09:52 PM

Originally posted by John Q Public
...actually no...not defending a purchase...I made the rational decision to buy a G4 before the G5's hit stores rather than waiting for a machine I wouldn't have been happy with...just because Uncle Steve says it's the best thing since buttered-bread...

...and trying to illustrate that although Apple NEEDS for you to buy into all the hype and buy a G5...you don't need to buy it...

I think the hype here is all the noise about how G5s are flim-flam. Have you studied the architecture? Have you seen the videos at the ADC site? (You have to be at least a free member to see them.)

http://developer.apple.com/adctv/

I'd say a whole lot of horse puckey has been blown around here by people who have no idea what the final performance of any G5 is going to be. Rationalizations for sure.

Of course G5s will be faster with optimizations. But the people who have actually used them in real world applications are blown away with the performance. People looking at numbers on paper, on the other hand, seem to feel free to pontificate all they want to feed their own prejudices and cynicism. :rolleyes:

mim

Aug 21, 2003, 09:55 PM

I question the need for the internal expansion slots. They're expensive to add, and will increase the heat load on the inside of the case. This may not be an issue now, but if this case is to last you 2 or 3 years then it may become a problem (esp. with a super hot running graphic card upgrade + 2 more 10000rpm drives :( ).

I also think that you'll get the same speed out of external FW800 devices as you would plugging them into the internal connections. Maybe slightly higher latency...but then you could just swap the new and old drives around.

I read recently that FW800 external drives were getting close to the theoretical maximum read/write speed of the fastest available drive (I'll look for the link).

Personally I think it's no bad thing for Apple to leaverage FW800 a bit more by reducing the number of bays inside the case - especially when it allows for the real crux of "life extentions"... frying hot processor upgrades.

daveg5

Aug 21, 2003, 10:09 PM

Originally posted by mim
I question the need for the internal expansion slots. They're expensive to add, and will increase the heat load on the inside of the case. This may not be an issue now, but if this case is to last you 2 or 3 years then it may become a problem (esp. with a super hot running graphic card upgrade + 2 more 10000rpm drives :( ).

DaveG5:wow please dont say that to creative pros that build thier studios around protools, and those with special purpose cards for vido and what not. you dont want us all with non upgradable imacs and emacs so you. all with g3/g4 can add usb2 firewire 800 newer video cards, processor upgrade cards bandwidth heavy sound and special purpose cards and i think that is a great selling point. when you buy a pci based mac you know you can have usb 3 firwire 1600 sata 300 and new agp video cards and processor cards in the future.so it can still be on your desk and fast 4 years from now if you get a closed system you are stuck out of any new technical changes and standard

I also think that you'll get the same speed out of external FW800 devices as you would plugging them into the internal connections. Maybe slightly higher latency...but then you could just swap the new and old drives around.

DaveG5: technically you do, you can get about 60+ MBS with one, two will over saturate it and require a pci firewire 800 card to over come that, a scsi 320 or dual 320 can do 640MB per second, but is overkill except if you need it for your buisness as are the 10000-15000 rpm scsi drives with thier close to 80MBS speed and under 4ms access speed.
many people want everything in one box, less cables, clutter etc. and many g4 systems have worked well fully loaded
also those 800 external cases add about $100+ per drive

I read recently that FW800 external drives were getting close to the theoretical maximum read/write speed of the fastest available drive (I'll look for the link).

DaveG5:actually its faster. storage review has ide doing 60+ tops and scsi doing 80+per single drive, but be careful some external firewire cases dont have fans, making the drive cooler in the machine

Personally I think it's no bad thing for Apple to leaverage FW800 a bit more by reducing the number of bays inside the case - especially when it allows for the real crux of "life extentions"... frying hot processor upgrades.

DaveG5:i agree, but i dont thinl apple did this on purpose. it the size of the g5 chips that made this a necessity. you can bet as the g5 chips get smaller cooler that these same g5 cases will have another optical drive and 2 extra hard drive bays. they cant at this time.[FONT=arial][SIZE=1]

mim

Aug 21, 2003, 10:30 PM

DaveG5-

Sorry, I didn't mean NO expansion slots :p

I was talking more about the need (or desire) to have 4 rather than 3, etc. Even with your example there are many things (like pro soundcards) that are starting to use firewire as an interface.

Thanks for the info on the tech specs. And ofc ourse you're right about it nice being able to have everything inside the case - but all I'm doing is questioning how often all those expansion bays and slots get used.

Mind you, I personally have got very good at minimizing my needs given the shoeboxes I've been living/working in for the past few years. It's very easy to get used to being very particular about what you "need" when you've got so little room to play with ;)

Apple's hidden social strategy.

John Q Public

Aug 21, 2003, 11:11 PM

Originally posted by leicaman
I think the hype here is all the noise about how G5s are flim-flam. Have you studied the architecture? Have you seen the videos at the ADC site? (You have to be at least a free member to see them.)

http://developer.apple.com/adctv/

I'd say a whole lot of horse puckey has been blown around here by people who have no idea what the final performance of any G5 is going to be. Rationalizations for sure.

Of course G5s will be faster with optimizations. But the people who have actually used them in real world applications are blown away with the performance. People looking at numbers on paper, on the other hand, seem to feel free to pontificate all they want to feed their own prejudices and cynicism. :rolleyes:

I do agree with waiting for the final analysis before declaring G5 faster or slower...my chief complaint (other than being uglier-than-mass-produced-sin and lacking the aesthetics of the MDD G4's) about the G5 was lack of options (ie internal expansion) and price...by opting for the older (and cheaper) machine I afforded the opportunity to get my hands on a 23" Cinema HD that I wouldn't have allowed myself to spend otherwise...and the ability to use some of the ATA100 drives out of my now dead G3...

leicaman

Aug 21, 2003, 11:26 PM

Originally posted by John Q Public
I do agree with waiting for the final analysis before declaring G5 faster or slower...my chief complaint (other than being uglier-than-mass-produced-sin and lacking the aesthetics of the MDD G4's) about the G5 was lack of options (ie internal expansion) and price...by opting for the older (and cheaper) machine I afforded the opportunity to get my hands on a 23" Cinema HD that I wouldn't have allowed myself to spend otherwise...and the ability to use some of the ATA100 drives out of my now dead G3...

If i works for you, great. But I think that with the airflow setup and the G5 being quieter, I don't see any reason to worry about internal expansion for hard drives. One would do better to have a nice Firewire 800 RAID outside the box. Check Barefeets to see the numbers. Pretty compelling.:)

daveg5

Aug 22, 2003, 12:30 AM

WM.

Aug 22, 2003, 12:46 AM

Originally posted by curious0
if you have a UAD or a PCI Powercore or any PCI audio interface card (like myself) then my understanding is that they won't work on the new 3.3V PCI-X architecture, but will work just fine in the 1.6Ghz machine because it has the older architecture PCI slots.
NOT TRUE!!!

It's not the PCI-X that's the problem, it's just that Apple has decided to go wtih 3.3V-only slots--on all three models of the G5.

In other words, contact MOTU to make ABSOLUTELY SURE your cards will work before you blow two grand on a G5 that (AFAIK) won't work with them.

HTH
WM

Mav451

Aug 22, 2003, 02:20 AM

Originally posted by John Q Public
...personally...I don't think SATA is a great thing...it's only marginally faster than ATA100 (and the drives are more expensive naturally because of the new Techno-Babble Buzzwords)...

...AGP 8x...really isn't...it's only roughly 3-4% faster than AGP 4X...and unless you're running a machine with more than 1GB RAM...it's likely you AGP can only palletize 64-128MB of system memory anyway...

...DDR400...although fast...isn't the "End-All" of power and speed...yes it's nice...but in the real world how fast can your machine cache a web page or process what's coming across the LAN without creating a bottleneck at the processor's bus is a more likely case than would be through synthetic benchmarks...besides...it's always a case where more memory is better than faster memory (that's one of the things I'll concede to the G5...but I really don't have a need for more than the 2GB that's in my machine)

...lastly...PCI-X...where to begin with that one...they've been on PC's for the past 2 years (server/workstation logic boards)...but like the 64bit 33Mz PCI that's been on the Mac since '99...there's little industry support...that part will changein the next couple years as AGP falls out of favor entirely for the greater bandwidth afforded by PCI-X...but that won't be for another year or so...

The advertising should be (insert Jeff Goldblum voiceover) "...the new PowerMac G5...with it's PCI-X...AGP 8X...Serial ATA interface...SuperDrive and Dual-Channel DDR400 Memory Bus...It's the World's-Fastest-Fully-Buzzword-Compliant Personal Computer ever made..."

I have a PC that is "Buzzword Compliant" with a AthlonXP 3000+...it's in my SO's posession...'cause I like my pathetic little FW800 Dual 1.25Gz (or so some would try to have me believe'cause I don't want the over-hyped industrial-grade-ugly G5)

...the only thing I've ever asked for from Apple was to bring back the "Cube Era" black-keyed Apple Pro KB...the white keys are a pain-in-the-ass to keep clean...

AGP 8x? One word. Games. You'd be suprised how well a 5900/9800pro can saturate that AGP bus. Understandably, 3 years ago i had a motherboard capable of using AGP4x. Guess what? I had a 5 year old G400--reknown for its "image quality". Needless to say, an 8500 replaced that (sadly) only 9 months ago.

PCI-X: for the future...but i agree with you. Right now, most of your everyday consumers will not need it. Did PC users REALLY need SATA 2 years ago? No, but people still wanted it anyway.

DC DDR memory? Um, anyone using FCP/FCE would gladly like that kind of memory interface to FEED that monstrous FSB. Any less would be uncivilized :)

John Q Public

Aug 22, 2003, 05:56 AM

[QUOTE]Originally posted by Mav451
AGP 8x? One word. Games. You'd be suprised how well a 5900/9800pro can saturate that AGP bus. Understandably, 3 years ago i had a motherboard capable of using AGP4x. Guess what? I had a 5 year old G400--reknown for its "image quality". Needless to say, an 8500 replaced that (sadly) only 9 months ago.

glad to see I'm not the only one who bought a G400...my G400MAX is currently resting in my "Retro Box" PIII 667 on a BX mobo (AGP mildly overclocked to 83Mz) but insofar as 4X vs 8X...it only yeilds 3-5% improvement based on identical hardware...

PCI-X: for the future...but i agree with you. Right now, most of your everyday consumers will not need it. Did PC users REALLY need SATA 2 years ago? No, but people still wanted it anyway.

I never wanted SATA...I wanted Ultra320LVD SCSI :p

DC DDR memory? Um, anyone using FCP/FCE would gladly like that kind of memory interface to FEED that monstrous FSB. Any less would be uncivilized :)

DC/DB DDR ain't as fast as some people believe...though an improvement over the status quo...again only marginal improvement over running high quality/low latency DDR Interleaved...as far as FPM goes...CAS2 PC133 SDRAM gives up only a little over PC2100 DDR SDRAM because of the design...DDR is a MUCH faster memory chip...but suffers from higher latency timings...

oh...btw...the G5's "Monstrous FSB" you mention is only the "Processor Bus"...the SystemBus is significantly slower...and the MemoryBus is only 400Mz (thus the usage of PC3200)

If Apple designed the G5 in a manner consistant with the past PowerMac...the actual SystemBus is only 200Mz...

and BTW...I only see a PC listed there...where's YOUR Mac...LOL...

eric_n_dfw

Aug 22, 2003, 06:05 AM

Originally posted by PieMac
I was wanting to purchase the superdrive built in because I want to be able to use iDVD. Thanks for the suggestions...I think I've already been to those sites but I will double check...you never know! It will. I added a Pioneer DVR104 to my old Blue & White G3 (with a G4 ZIF upgrade) and iDVD installed and worked just fine.

John Q Public

Aug 22, 2003, 06:16 AM

Originally posted by eric_n_dfw
It will. I added a Pioneer DVR104 to my old Blue & White G3 (with a G4 ZIF upgrade) and iDVD installed and worked just fine.

I miss my old B&W...adequately powered for surfing the net...but I still prefer my G4...although I wish it didn't put out so much heat :(

eric_n_dfw

Aug 22, 2003, 06:21 AM

Originally posted by John Q Public
...DDR400...although fast...isn't the "End-All" of power and speed...yes it's nice...but in the real world how fast can your machine cache a web page or process what's coming across the LAN without creating a bottleneck at the processor's bus is a more likely case than would be through synthetic benchmarks...besides...it's always a case where more memory is better than faster memory (that's one of the things I'll concede to the G5...but I really don't have a need for more than the 2GB that's in my machine)If all you are doing with your machine is web surfing or processing data over the network, then hell, buy an iBook, but us FCP and Photoshop users (not to mention you gamers) have been FSB bottlenecked for soooo long. Dual channel DDR400 is going give us whiplash, especially if you can afford gobs of RAM. Imagine being able to cache 10+ min's of DV video in RAM and letting those AltiVec units go full throttle against 'em on renders and such.

eric_n_dfw

Aug 22, 2003, 06:23 AM

Originally posted by John Q Public
I miss my old B&W...adequately powered for surfing the net...but I still prefer my G4...although I wish it didn't put out so much heat :( I thought I'd keep mine forever - loved her dearly. Then I got a used Dual G4 500. Kicked the B&W to the eBay curb in a NY Minute! :eek: (Got $400 for her plus $140 for the 400 Mhz G4 too!)

eric_n_dfw

Aug 22, 2003, 06:33 AM

Originally posted by wizard
Your statements below are all well and good but you seem to mis one important thing here. The 970 does not have the hardware to optimize for, it only has 2 integer units. The VMX unit barely competes with the one in the G4. What the 970 does have is FP performance and more bandwidth, these advantages only come into play for certain code bases so you do not get a uniform increase in performance.

People should not get their hopes up to far. For some code bases the 970 will excel and simply be fantastic. You will never see a fantastic performance increase in run of the mill software. If your lucky you will get clock rate increase and maybe 10% over the G4.

Dave Software that remains compiled for only the G4 scheduler will probably not see huge increases, but I personally think that the huge increase in FSB speed will speed up the whole OS so much, as with most comercial software being recompiled for the G5 , that the overall impression will be that it is much faster.

As to the "barely competes" comment about the AltiVec/VMX units - Hannible at ArsTechnica agreed with you untill he got the dirt on the G5 from IBM:

http://www.arstechnica.com/cpu/03q2/ppc970-interview/ppc970-interview-1.html
...the 970's VMX (a.k.a. Altivec) unit, which turns out to be a lot more flexible and robust than what I'd initially described in my articles. ...

... Contrary to what I stated in Part II of the 970 article, any vector instruction can dispatch from any of the four non-branch dispatch slots. ... This makes the 970 look a lot better for Altivec code.(Read the link for all the interview questions and whatnot.)

He goes on to talk about how the VMX units look a lot worse on paper than they will perform in real life and the IBM guys basically say that it will be much better in future chips.

PieMac

Aug 22, 2003, 06:36 AM

Originally posted by eric_n_dfw
It will. I added a Pioneer DVR104 to my old Blue & White G3 (with a G4 ZIF upgrade) and iDVD installed and worked just fine.

You're right, thanks for reminding me. It's the externals that won't run iDVD. How difficult is it to install? I've done very little upgrading (except for RAM).

daveg5

Aug 22, 2003, 06:38 AM

Originally posted by mim
DaveG5-

Sorry, I didn't mean NO expansion slots :p

I was talking more about the need (or desire) to have 4 rather than 3, etc. Even with your example there are many things (like pro soundcards) that are starting to use firewire as an interface.

Thanks for the info on the tech specs. And ofc ourse you're right about it nice being able to have everything inside the case - but all I'm doing is questioning how often all those expansion bays and slots get used.

Mind you, I personally have got very good at minimizing my needs given the shoeboxes I've been living/working in for the past few years. It's very easy to get used to being very particular about what you "need" when you've got so little room to play with ;)

Apple's hidden social strategy.
well iguess i agree with you then, firewire is the way to go but has two pitfalls. one is firewire based exspansion is of course higher priced and adds clutter. #2 according to motu and others you can get mega bandwidth thru the pcix slots which means many more simutaneous 24 bit /192hz stereo tracks which is important to many pro audio guys but for joe public i agree firewire may be the way to go.

eric_n_dfw

Aug 22, 2003, 06:42 AM

Originally posted by PieMac
You're right, thanks for reminding me. It's the externals that won't run iDVD. How difficult is it to install? I've done very little upgrading (except for RAM).
Pretty easy - depends on the model of G3/G4.

For Blue & Whte G3's and Graphite G4's here's a link: http://xlr8yourmac.com/G3-ZONE/hival52x/install.html
(QuickSilve G4's would probably be about the same I'd think)

Don't have a link for Mirror Drive door G4's but I'd presume they'd be even easier.

Basically, you pull the old drive, pop the DVD-R drive in and install iDVD.

On mirror drive machines, you wouldn't even have to remove the old drive as they can hold 2

PieMac

Aug 22, 2003, 06:44 AM

Definitely will look into it...thanks!

John Q Public

Aug 22, 2003, 07:10 AM

Originally posted by eric_n_dfw
If all you are doing with your machine is web surfing or processing data over the network, then hell, buy an iBook, but us FCP and Photoshop users (not to mention you gamers) have been FSB bottlenecked for soooo long. Dual channel DDR400 is going give us whiplash, especially if you can afford gobs of RAM. Imagine being able to cache 10+ min's of DV video in RAM and letting those AltiVec units go full throttle against 'em on renders and such.

the net-surfing and processing data over a network was an analogy...admittedly I mostly use mine for coding windoze apps in VS.NET (VPC: Win2000)...that was one of the biggest selling points for me (going to Mac completely) was Visual Studio running well enough under VPC...and a little video editing for tutorials in the software I'm write...but seeing that Reason and CuBase are Mac compatible I'll probably be hooking up my MIDI controller to the Mac too...

John Q Public

Aug 22, 2003, 07:15 AM

Originally posted by PieMac
You're right, thanks for reminding me. It's the externals that won't run iDVD. How difficult is it to install? I've done very little upgrading (except for RAM).

dropping a CD/DVD drive is a breeze...Apple has all the instructions (considering there are differences in each generation of cases) D/L'able in PDF format...only takes about 10 minutes after you lower the drawbridge to finish the project...in most cases you'll only need a #1 and #2 Phillips Screwdriver (maybe tweezers if you want to change the jumper settings on the drive)...

CooCooCaChoo

Aug 22, 2003, 10:37 AM

Originally posted by BrandonRP0123
Forgive my ignorance, but the DDR RAM Stevie presented as the G5 using at WWDC is DDR 400Mhz. At best (assuming that DDR means you get double the bandwidth out of it), that only gives you 800Mhz of bandwidth to play with. While this is fine for the 1.6 its not as fast as the system bus in the 1.8 (900Mhz) and the 2.0 (1000Mhz)

Anyone know how this works?

Mhz have little or no impact on performance, plain and simple.

I'll give you an example. SGI O2 uses an UMA (Unifed Memory Archictecture) structure of having one large pool of memory and everything that connects to that pool of or memory is hooked up via high speed connectors.

The memory is running a 66Mhz, HOWEVER, the bandwidth of the memory is 2100mbps.

It ulimately doesn't matter what the memory clock speed is, what is important is banwidth and if the memory bandwith is 13.7Gbps (IIRC), then who gives a flying continental whether or not the memory runs at 400mhz. All that happens is that 400Mhz moves more data per-clock cycle.

Alexevst

Aug 22, 2003, 12:01 PM

Been lurking for awhile but the about post about Snappiness rang so true I had to register. I use ICQ, Safari, and occasionally Photoshop and Dreamweaver for light weight stuff. What is my next purchase? Dual G5 with Panther pre-installed. It's all about perceived speed.

jettredmont

Aug 22, 2003, 02:44 PM

Originally posted by PieMac
I was wanting to purchase the superdrive built in because I want to be able to use iDVD. Thanks for the suggestions...I think I've already been to those sites but I will double check...you never know!

I thought iDVD would work with an add-in DVD burner, so long as you added one of the models Apple shipped with their systems.

No practical experience here, but it seems like the only way Apple can keep iDVD working only on built-in drives would be to either use the model number (which is the same as the retail version; it doesn't have to be -- OEM's can add a "A" or "M" to the model number reported for high-volume customers ... but Apple doesn't appear to have taken this route, and the model number is identical to the retail version) OR Apple could "brand" the built-in drives with custom silicon (much easier to just ask Pioneer to add an extra letter to the model number though!) OR Apple could "brand" the motherboard of computers shipped with a built-in drive (which would be sinister, odd, and, again, far more costly than the first option above).

IMHO, it is quite unlikely that iDVD would know the difference between a retail drive and a built-to-order drive, so long as the model numbers match.

WM.

Aug 22, 2003, 02:48 PM

Originally posted by John Q Public
[QUOTE]oh...btw...the G5's "Monstrous FSB" you mention is only the "Processor Bus"...the SystemBus is significantly slower...and the MemoryBus is only 400Mz (thus the usage of PC3200)

If Apple designed the G5 in a manner consistant with the past PowerMac...the actual SystemBus is only 200Mz...
WTF? What do you mean by "SystemBus"? Look at the diagram at www.apple.com/powermac/[something].html or in the preliminary developer note. Apple did not "design the G5 in a manner consistent with the past Power Mac"--that's why everyone is so excited about it. The bus between each processor and the U3 runs at 800, 900, or 1000 MHz (effective; 32 bits wide in each direction). There are two buses between U3 and RAM (thus the term "dual channel"); each runs at 400 MHz effective, 64 bits wide.

So I guess I've described the "ProcessorBus" and the "MemoryBus", to use your terms. You also refer to the "SystemBus"; I'm not sure what else is left--the HyperTransport links to the PCI bridge (most likely made by AMD) and the I/O controller (K2)? Those all run at 800 MHz effective, and they're 16 and 8 bits wide, respectively, in each direction.

Not to be rude, but...please try to know what you're talking about.

HTH
WM

jettredmont

Aug 22, 2003, 02:55 PM

Originally posted by John Q Public
...personally...I don't think SATA is a great thing...it's only marginally faster than ATA100 (and the drives are more expensive naturally because of the new Techno-Babble Buzzwords)...

For a Pro rig, however, one that is going to last a few years, SATA is the way to go. You won't see any difference today, but as HD speeds increase (or cache RAM sizes increase :)) you'll see the extra bandwidth being used.

Also, the nice skinny SATA cables are much better for cooling airflow, and for assembling (or user modifying) than the flat IDE ribbon cables (and the round ones have data corruption issues ... acceptable for a throw-away box but not a PRO workstation!).

SATA drives are more expensive because they aren't used across the entire industry yet. They will go down in price as demand increases. SATA as an interface isn't more expensive than IDE as an interface for the HDD manufacturer.

...AGP 8x...really isn't...it's only roughly 3-4% faster than AGP 4X...and unless you're running a machine with more than 1GB RAM...it's likely you AGP can only palletize 64-128MB of system memory anyway...

AGP 8x is soon going to be the industry standard. It's hard enough finding AGP 1x cards out there already ...

...DDR400...although fast...isn't the "End-All" of power and speed...yes it's nice...but in the real world how fast can your machine cache a web page or process what's coming across the LAN without creating a bottleneck at the processor's bus is a more likely case than would be through synthetic benchmarks...besides...it's always a case where more memory is better than faster memory (that's one of the things I'll concede to the G5...but I really don't have a need for more than the 2GB that's in my machine)

1) The processor bus bottleneck has been taken care of quite nicely, thank you very much!

2) More memory (8GB, say) is better than faster memory if and only if you are running out of main memory (hence using much slower virtual memory). If your system only uses 256 MB of memory at any point in its lifetime then having 1 or 2 GB of RAM isn't going to help you at all.

3) Even if more memory is better, more and faster memory is a godsend!

...lastly...PCI-X...where to begin with that one...they've been on PC's for the past 2 years (server/workstation logic boards)...but like the 64bit 33Mz PCI that's been on the Mac since '99...there's little industry support...that part will changein the next couple years as AGP falls out of favor entirely for the greater bandwidth afforded by PCI-X...but that won't be for another year or so...

And you plan on replacing your computer in a year? If so, then you are correct. Otherwise, you just proved the point you were trying to discredit.

jettredmont

Aug 22, 2003, 03:04 PM

Originally posted by rog
Except that the low end G5 sells for more than most high-end PCs with much better video cards. It's too bad Apple is letting the slower G5s trickle out 1st. People will be disappointed and underwhelmed, and when the DP 2GHz comes out, people won't care much. If that came out first, or at the same time, everyone would be going on and on about how fast it is. Once again, Apple turns what should have been a successful rollout into a disappointing mess.

No, $1799 is mid-range for a PC (1.6GHz G5 with the combo drive instead of super drive is $1799). $1999 (the 1.6GHz G5 as advertized) is upper-mid-range for a PC.

$4000 is high-end for a PC.

This has been fairly true for over a decade. The only thing that has changed recently is that the low-end PC market has gotten really low-end.

If I were to buy a development workstation today I'd spend upwards of $3000. If I were to buy a secondary computer for home, I'd spend in the $1500-2000 range. If I were to buy a cheapo computer with no bells or whistles for internet access and the occaisional word processing document, I'd spend less than $1000.

jettredmont

Aug 22, 2003, 03:14 PM

Originally posted by PieMac
You're right, thanks for reminding me. It's the externals that won't run iDVD. How difficult is it to install? I've done very little upgrading (except for RAM).

Two cables and 2-3 screws, once you flip open the side of the box (and make sure you are grounded/static-free!). Not as easy as plugging in new RAM, but certainly not brain surgery either.

Replacing an optical drive is very easy to do, as there are about three steps involved, and you "undo" them to remove the old drive, then "redo" them the other way around to put the new one in.

My father's replaced his own CD drive (on a Windows box, even!). That would tell you how easy it is if you knew my father and electronics :)

John Q Public

Aug 22, 2003, 04:44 PM

Originally posted by WM.
WTF? What do you mean by "SystemBus"? Look at the diagram at www.apple.com/powermac/[something].html or in the preliminary developer note. Apple did not "design the G5 in a manner consistent with the past Power Mac"--that's why everyone is so excited about it. The bus between each processor and the U3 runs at 800, 900, or 1000 MHz (effective; 32 bits wide in each direction). There are two buses between U3 and RAM (thus the term "dual channel"); each runs at 400 MHz effective, 64 bits wide.

So I guess I've described the "ProcessorBus" and the "MemoryBus", to use your terms. You also refer to the "SystemBus"; I'm not sure what else is left--the HyperTransport links to the PCI bridge (most likely made by AMD) and the I/O controller (K2)? Those all run at 800 MHz effective, and they're 16 and 8 bits wide, respectively, in each direction.

Not to be rude, but...please try to know what you're talking about.

HTH
WM
...where to begin...

Processor Bus (aka: Front Side Bus) only dictates Processor to "North-Bridge" of the motherboard's chipset.

Memory Bus is what the traces running from the Memory Hub/Controller (usually part of the "Northbridge" of the system's controller chipset)

System Bus is where the motherboard's Controller Chipset talks to everything else.

System Bus is NOT 800, 900, or 1000Mz.

System Bus is broken down based on Archetecture...if a system has a seperate "South Bridge" some of the functions are offloaded (ie PCI, ISA and IDE controllers).

If the System Controller is integrated where the functions of the traditional North/South Bridge archetecture are carried on to one location so as to make a cheaper alternative (usually only older/slower systems do this because of the obvious bottlenecks it creates...but I digress, I have worked with fast machines where integration helped rather than hindered but the exception rarely the rule)...

Maximum speeds you are able to acheive are thus:

Processor Bus: Based solely on processor and chipset manufacturer in the case of IBM...the PPC970's Processor Bus is half of the rated processor speed.

Memory Bus: Partially determined by type of memory used DDR speed ratings are a misnomer...the actual speed is half the rating because of latency issues. Which is why PC2100 (DDR266) speeds are roughly equivalent to PC133 SDRAM...PC3200 is actually equivalent to 200Mz considering the archetecture. In the case of "Dual Channel" DDR, you're spreading/splitting instructions across two seperate memory busses, increasing bandwidth (which naturally increases speed), this is something that Intel got into trouble with using RAMBUS on the first generation Pentium4 (the memory hub for early RAMBUS wasn't fully up to the task...not to mention a 20 stage pipeline and inferior FPU in the processor didn't really help matters)

NOW we get to the System Bus...speeds on the system bus are limited to what they are connceted to.

ISA=8Mz
EISA=16Mz
PCI=33Mz
PCI (Enhanced)=66Mz (which was used on second generation G3's and Yikes G4 for video)
AGP=66Mz (it doesn't matter which flavor 1/2/4/8x...it's only 66Mz...where the differences in AGP formats resides...the width of the data path and how the Northbridge saturates the bus)
PCI-X=133Mz (although PCI-X has been out for nearly 2 years on Workstations and Servers, as a replacement for the aging PCI (Enhanced), there is still not adequate support...and unfortunately won't be until "Mainstream" PC manufacturers begin to build it into their systems...as a sidebar...the next generation of Video Card will be PCI-X because it allows higher bandwidth than the proposed AGP 16x)

And I'm not going to muddy the waters further with NuBus, Micro-Channel or VESA...

...to wrap...a CPU is an expensive calculator and it crunches numbers increasingly faster each generation...but to do its job, relies on components to perform optimally...there has been no machine produced since the first generation PPC's or Pentium (and clones) where the Processor and memory bus performed in parallel...likewise the System Bus in and of its components hasn't performed in parallel with the processor since 68030 and 80386...Likewise...the G5 is not using a Gz speed memory or system bus...although it's processors can each speak to the north-bridge at up to 1Gz (thus the Dual-FSB)...it's still the System Bus Chipset that allocates instructions to every other part of the system...the processor is the brain...but the System Bus is the Central Nervous System.

and yes...I did find your accusation of a lack of knowledge exceedingly rude...

ffakr

Aug 22, 2003, 05:30 PM

Originally posted by John Q Public
...where to begin...

Processor Bus (aka: Front Side Bus) only dictates Processor to "North-Bridge" of the motherboard's chipset.

Memory Bus is what the traces running from the Memory Hub/Controller (usually part of the "Northbridge" of the system's controller chipset)
you do know that most current chipset designs don't use the traditional Northbridge/Southbridge design philosophy anymore, right? P4 doesn't. The latest G4s don't. The G5 doesn't.
Here's a quote from a PC Tech site to illustrate "The 820 chipset employs the Accelerated Hub Architecture that is offered in all Intel 800 series chipsets - the first chipset architecture tomove away from the traditional Northbridge /Southbridge design."
System Bus is NOT 800, 900, or 1000Mz..... er, sorry, it is 800MHz.
The HyperTransport protocol integrates thePower Mac G5’s I/O subsystems and connectsthem to the system controller. Serial ATA, GigabitEthernet, FireWire, USB 2.0 and optical digital andanalog audio are all integrated through twobidirectional 16-bit, 800MHz HyperTransportinterconnects for a maximum throughput of 3.2GBper second. (http://www.apple.com/powermac/architecture.html) The system bus connects components with 800MHz HT busses. They have half the bandwidth of the processor busses as they are only 16 bit instead of 32bit.

Memory Bus: Partially determined by type of memory used DDR speed ratings are a misnomer...the actual speed is half the rating because of latency issues. Which is why PC2100 (DDR266) speeds are roughly equivalent to PC133 SDRAM...PC3200 is actually equivalent to 200Mz considering the archetecture.
You aren't helping your stance that you know what you are talking about.
a) the actual speed is half the DDR speed not because of latency, but because that's the actual speed.
b) ddr speeds are NOT equivilent to SDR speeds. There is a reason to use DDR over SDR, the effective bandwidth IS double what you get in SDR.
c) The latency IS generally higher in ticks, but that doesn't mean that the latency is longer in real time. If the memory controller would take 1 second (just for clear argument) to fetch an instruction.. the reasoning is, it will take twice as many ticks to get the data with DDR as SDR because with DDR there are twice as many ticks in a second. BUT, it still takes one second.
... and, you should probably be a little more clear when you discuss memory speed. Are you refering to frequency, bandwidth, latency? they are all different things.

NOW we get to the System Bus...speeds on the system bus are limited to what they are connceted to.

ISA=8Mz
EISA=16Mz
PCI=33Mz
PCI (Enhanced)=66Mz (which was used on second generation G3's and Yikes G4 for video)
AGP=66Mz (it doesn't matter which flavor 1/2/4/8x...it's only 66Mz...where the differences in AGP formats resides...the width of the data path and how the Northbridge saturates the bus)
PCI-X=133Mz (although PCI-X has been out for nearly 2 years on Workstations and Servers, as a replacement for the aging PCI (Enhanced), there is still not adequate support...and unfortunately won't be until "Mainstream" PC manufacturers begin to build it into their systems...as a sidebar...the next generation of Video Card will be PCI-X because it allows higher bandwidth than the proposed AGP 16x)

why is most of this relevent? ISA and EISA? HUH? who gives a crap? EISA didn't even make it on the PC.
And I'm not going to muddy the waters further with NuBus, Micro-Channel or VESA...
don't worry, you've done a good job of that already.
back to your points above...
PCI-X isn't mainstream yet, but it is completely compatible with PCI. Compatability and the fact that these are considered 'pro' machines are the reason why Apple went PCI-X. There are PCI-Express devices out now, but generally they are only running on PCI-X because they _actually_ need the bandwidth. By putting PCI-X into the G5s, Apple has not only extended the useable life of the chipset, but they've also prepped the machine for the use of high end interface cards like GigE, FibreChannel, Myrinet...
Another point. Could you provide some backup to your contention that newer AGP ports are actually only run at 66MHz? I was always under the impression that an AGP 4x slot actually ticked off at 266 million cycles per second.

...to wrap...a CPU is an expensive calculator and it crunches numbers increasingly faster each generation...but to do its job, relies on components to perform optimally...there has been no machine produced since the first generation PPC's or Pentium (and clones) where the Processor and memory bus performed in parallel...likewise the System Bus in and of its components hasn't performed in parallel with the processor since 68030 and 80386...Likewise...the G5 is not using a Gz speed memory or system bus...although it's processors can each speak to the north-bridge at up to 1Gz (thus the Dual-FSB)...it's still the System Bus Chipset that allocates instructions to every other part of the system...the processor is the brain...but the System Bus is the Central Nervous System.
huh?
the first pentiums ran on 60 and 66MHz sys buses... only the first two (complete crap) models ran on a 1:1 ratio. The first PPC chips ran at 1:2 from the start. My Powermac 6100/66 had a 33MHz system bus. There has NEVER been a PPC that ran at a 1:1 cpu to system multiple.

The G5 is not using GHz memory, but it is using 400MHz memory on a 128 bit bus. The processors (in the 1.8GHz for example) run on two, one way, 32 bit 900MHz buses. The effective bandwidth of the 1.8 GHz processor bus is essentially equal to the memory bus (due to routing overhead on the HT BASED processor bus). And, because of the one way, 32 bit nature of the processor buses... the CPU can actually only read data at half the rate of the memory (same with exclusive writes).
and yes...I did find your accusation of a lack of knowledge exceedingly rude...
I don't like name calling (unless the person is REALLY an idiot and they just keep coming ;-)
I'm not sure you've put up a good rebuff for that other person's slam yet though.

ffakr

Aug 22, 2003, 05:50 PM

Originally posted by John Q Public
...personally...I don't think SATA is a great thing...it's only marginally faster than ATA100 (and the drives are more expensive naturally because of the new Techno-Babble Buzzwords)...
SATA is hot swappable. It has one cable and it's faster. The drives will eventually be cheaper when volume goes up because the interface is physically more simple. SATA also does away with the slow master/slave arrangement. Every device gets its own data channel

...AGP 8x...really isn't...it's only roughly 3-4% faster than AGP 4X...and unless you're running a machine with more than 1GB RAM...
this is very misleading. In real world tests, agp 8x cards aren't much faster than 4x cards. This doesn't mean, in any way, that the 8x agp standard is not twice as fast as 4x AGP. In the opinion of people like Tom Pabst (tom's hardware.com), one big issue is related to poor driver performance... the current vid cards just don't take advantage of the available bandwidth.
Other related point... the speed of the AGP bus typically serves two purposes... as a buffer, you have to move texture and frame buffer data to the vid card. The more complex the textures and the scenes, the more bandwidth you need. AGP was also designed from the ground up to allow for DMA for caching in the system memory when the on board vid memory runs out. They doesn't typically happen with modern video cards since hardware vid card makers are usually ahead of the curve when it comes to how much on board memory the current (and near future) games need. This is why there is generally no performance difference between the high end 128MB and 256 MB gaming vid cards. This is also why, in real world tests, the GeForce4 64MB cards tend to be faster than the 128MB versions.. since the games today don't use more than 64MB and the 64MB versions of the GF4s generally have slightly faster clocked memory.

Apple's choice of AGP 8X is a smart move. It may not be needed now, but there is room to use it in the future. Why? Because vid chipsets evolve VERY quickly, because you'll be able to fetch textures and data from 128bit wide DDR 400 to the on board vid memory, and because Apple is beginning to court the higher end market... Apple mentioned support for 'pro' video cards comming to the G5. We might see Mac Quatros soon.
[b]
...DDR400...although fast...isn't the "End-All" of power and speed...yes it's nice...but in the real world how fast can your machine cache a web page or process what's coming across the LAN without creating a bottleneck at the processor's bus is a more likely case than would be through synthetic benchmarks...besides...it's always a case where more memory is better than faster memory (that's one of the things I'll concede to the G5...but I really don't have a need for more than the 2GB that's in my machine)
There is a reason why Apple has a consumer line. Just because you don't need 6.4 GB/sec of memory bandwidth or more than 2GB doesn't mean that there aren't a whole lot of people who do. Open up a 500MB image, work on it for a while with multiple levels of undo, then do a 22 degree rotate on it and see how fast it is on a G4.[b]
...lastly...PCI-X...where to begin with that one...they've been on PC's for the past 2 years (server/workstation logic boards)...but like the 64bit 33Mz PCI that's been on the Mac since '99...there's little industry support...that part will changein the next couple years as AGP falls out of favor entirely for the greater bandwidth afforded by PCI-X...but that won't be for another year or so...
Actually, AGP is being supplanted by PCI-Express, which to the best of my knowledge, is a different standard than PCI-X.

I have to say that you've made a pretty silly argument against the G5. I can't believe that anyone has their panties in a bunch because Apple made the G5 too nice.
There are other machines available for you if you don't want that much power or bandwidth. Get an iMac or an eMac... Apple still makes the G4 for god sake.
That doesn't mean that there isn't a market for scary fast computers. No one [at my job] would have asked my opinion on the G5 if it would have been launched under the slogan "The G5, just 'good enough' in every way"

WM.

Aug 22, 2003, 06:03 PM

OK, let's go point-by-point here...
Originally posted by John Q Public
...where to begin...

Processor Bus (aka: Front Side Bus) only dictates Processor to "North-Bridge" of the motherboard's chipset.
Yes.
Memory Bus is what the traces running from the Memory Hub/Controller (usually part of the "Northbridge" of the system's controller chipset)
Yes.
System Bus is where the motherboard's Controller Chipset talks to everything else.
OK, that's not the term I usually see used, but hey, people can have different terms for the same thing...so I'll give you this one.
System Bus is NOT 800, 900, or 1000Mz.
Actually, it is. It's a 16-bit HyperTransport link running at 800 MHz effective (400 MHz DDR). (That's 16 bits each direction, to be exact.)

System Bus is broken down based on Archetecture...if a system has a seperate "South Bridge" some of the functions are offloaded (ie PCI, ISA and IDE controllers).
OK, yes.
If the System Controller is integrated where the functions of the traditional North/South Bridge archetecture are carried on to one location so as to make a cheaper alternative (usually only older/slower systems do this because of the obvious bottlenecks it creates...but I digress, I have worked with fast machines where integration helped rather than hindered but the exception rarely the rule)...
OK.
Maximum speeds you are able to acheive are thus:

Processor Bus: Based solely on processor and chipset manufacturer in the case of IBM...the PPC970's Processor Bus is half of the rated processor speed.
In the ArsTechnica interview with some IBM engineer-types, one of them pointed out that actually the FSB speed can be 1/2, 1/3, 1/4, or 1/6 of the processor speed (IIRC). But obviously in the current machines we only see the FSB running at 1/2 the processor speed, like you said.
Memory Bus: Partially determined by type of memory used DDR speed ratings are a misnomer...the actual speed is half the rating because of latency issues.
No, not because of latency issues. DDR400 RAM has an actual clock speed of 200 MHz, but it transmits data on both edges of the clock cycle--so the effective speed is 400 MHz (thus the 400 in DDR400).

Latency is another can of worms, and I believe you are correct that DDR memory can have higher latency. Also, the G5 architecture (with the U3 controller) has higher latency on RAM accesses than the G4 architecture (with U2) did.
Which is why PC2100 (DDR266) speeds are roughly equivalent to PC133 SDRAM...PC3200 is actually equivalent to 200Mz considering the archetecture.
No, I'm pretty sure you're wrong here. See above.
In the case of "Dual Channel" DDR, you're spreading/splitting instructions across two seperate memory busses, increasing bandwidth (which naturally increases speed),
Yes.
this is something that Intel got into trouble with using RAMBUS on the first generation Pentium4 (the memory hub for early RAMBUS wasn't fully up to the task...not to mention a 20 stage pipeline and inferior FPU in the processor didn't really help matters)
Can't comment on that.
NOW we get to the System Bus...speeds on the system bus are limited to what they are connceted to.

ISA=8Mz
EISA=16Mz
PCI=33Mz
PCI (Enhanced)=66Mz (which was used on second generation G3's and Yikes G4 for video)
AGP=66Mz (it doesn't matter which flavor 1/2/4/8x...it's only 66Mz...where the differences in AGP formats resides...the width of the data path and how the Northbridge saturates the bus)
I'm not sure about this "saturating the bus" part, but otherwise, it looks good to me...
PCI-X=133Mz (although PCI-X has been out for nearly 2 years on Workstations and Servers, as a replacement for the aging PCI (Enhanced), there is still not adequate support...and unfortunately won't be until "Mainstream" PC manufacturers begin to build it into their systems...as a sidebar...the next generation of Video Card will be PCI-X because it allows higher bandwidth than the proposed AGP 16x)
Can't speak to this last part, but that all sounds good too.
And I'm not going to muddy the waters further with NuBus, Micro-Channel or VESA...

...to wrap...a CPU is an expensive calculator and it crunches numbers increasingly faster each generation...but to do its job, relies on components to perform optimally...there has been no machine produced since the first generation PPC's or Pentium (and clones) where the Processor and memory bus performed in parallel...likewise the System Bus in and of its components hasn't performed in parallel with the processor since 68030 and 80386...
Not sure what you mean by "in parallel". I believe the 601 Power Macs generally had a FSB that ran at half or a third of the processor speed (my 6100's FSB, for example, runs at 30 MHz).
Likewise...the G5 is not using a Gz speed memory or system bus...although it's processors can each speak to the north-bridge at up to 1Gz (thus the Dual-FSB)...it's still the System Bus Chipset that allocates instructions to every other part of the system...the processor is the brain...but the System Bus is the Central Nervous System.
Yup.
and yes...I did find your accusation of a lack of knowledge exceedingly rude...
Sorry to hear that. I think I mainly take issue with your use of "System Bus" to describe the interface between the north and south bridges and your assertion that it's anything other than HyperTransport, which always (AFAIK) runs at 800 MHz.

It still seems to me that you haven't studied the block diagram and accompanying description in the Preliminary Developer Note. That was the main reason why I accused you of not knowing what you're talking about.

As I review my post and this thread, I see that while I was typing ffakr made several of my points for me. But, since I've invested a fair amount of time in this post, I think I'll hit "Submit" anyway. :)

HTH
WM

WM.

Aug 22, 2003, 06:22 PM

I appreciate the backup, man.

Originally posted by ffakr
you do know that most current chipset designs don't use the traditional Northbridge/Southbridge design philosophy anymore, right? P4 doesn't. The latest G4s don't. The G5 doesn't.
Really? I mean, OK, there are three main ASICs instead of two (U3, the AMD-made PCI[-X] bridge, and K2), but it doesn't seem all that different to me...
.... er, sorry, it is 800MHz.
The HyperTransport protocol integrates thePower Mac G5’s I/O subsystems and connectsthem to the system controller. Serial ATA, GigabitEthernet, FireWire, USB 2.0 and optical digital andanalog audio are all integrated through twobidirectional 16-bit, 800MHz HyperTransportinterconnects for a maximum throughput of 3.2GBper second. (http://www.apple.com/powermac/architecture.html) The system bus connects components with 800MHz HT busses. They have half the bandwidth of the processor busses as they are only 16 bit instead of 32bit.
To be really, super nit-picky, only the bus between U3 and the PCI bridge is 16-bit. The one between the PCI bridge and K2 is only 8-bit.
You aren't helping your stance that you know what you are talking about.
:D
[things that I all agree with]

Another point. Could you provide some backup to your contention that newer AGP ports are actually only run at 66MHz? I was always under the impression that an AGP 4x slot actually ticked off at 266 million cycles per second.
I think he may be right about the 66 MHz thing, or something like it. AGP 4x uses DDR, at the very least, and there may be some pretty fancy other stuff going on there too. This is kind of out of my league, though...
huh?
the first pentiums ran on 60 and 66MHz sys buses... only the first two (complete crap) models ran on a 1:1 ratio. The first PPC chips ran at 1:2 from the start. My Powermac 6100/66 had a 33MHz system bus.
To avoid any confusion: I was talking about a 6100/60 (actually a Performa 6115CD, but they were basically the same thing...).
There has NEVER been a PPC that ran at a 1:1 cpu to system multiple.

The G5 is not using GHz memory, but it is using 400MHz memory on a 128 bit bus. The processors (in the 1.8GHz for example) run on two, one way, 32 bit 900MHz buses. The effective bandwidth of the 1.8 GHz processor bus is essentially equal to the memory bus (due to routing overhead on the HT BASED processor bus). And, because of the one way, 32 bit nature of the processor buses... the CPU can actually only read data at half the rate of the memory (same with exclusive writes).
Yup.
I don't like name calling (unless the person is REALLY an idiot and they just keep coming ;-)
Me neither. I don't like to pick on people for no reason, but when people start spouting BS I feel like I have to step in.

And there's been plenty of other BS and/or FUD spouted in this thread (people crapping on the G5's performance when at this point there are essentially NO ACCURATE, USEFUL BENCHMARKS of it...); I guess this time I just kind of snapped. :)

Also note that I didn't actually call him a name--I was honestly making a suggestion that he do just a little more research before posting. Obviously it didn't come across that way, which isn't totally unexpected...
I'm not sure you've put up a good rebuff for that other person's slam yet though.
Heh :)

And one more thing: I've noticed that John Q Public uses "Mz" or "Gz" a lot, instead of "MHz" or "GHz". To me, that's not very accurate, or respectful to Mr. Hertz (http://www.bartleby.com/65/he/Hertz-He.html)...I'll grant people their shorthands, but really...

Thanks for being on my side ;)
WM

edit: added Hertz link and clarified stuff near the end

tychay

Aug 22, 2003, 06:52 PM

Originally posted by John Q Public
...and would partially explain why Apple dropped the Dual 1.42 FW800 from the line...opting for the original MDD (with only FW400) so as not to outperform and of the G5's and steal their thunder for less money...

Wow! someone is really bitter. Instead of spreading FUD and spite why don't you check your facts? The reason Apple is selling the MDD instead of the MDD FW800 is because MDD boots into OS 9 and the FW800 doesn't (http://docs.info.apple.com/article.html?artnum=25114). Considering that Apple stated that had no plans to support OS 9 in any computer this year, it's obvious that the MDD are still being released to satisfy the practical reality (that a lot of people use OS9) in the face of the official policy.

Besides, the MDD (http://www.everymac.com/systems/apple/powermac_g4/stats/powermac_g4_1.25_dp_mdd.html) gives up very little to the FW800 (http://www.everymac.com/systems/apple/powermac_g4/stats/powermac_g4_1.42_dp_mdd.html). The G5 design addresses all the people on the other spectrum as you who complained their the MDD so much that they dubbed it the "Wind Tunnel". (I'm sure the MDD's that Apple produces now have tweaks to address some of these issues.)

The new PowerMac G5 is a well-designed machine. The cable routing and compartmentalization show definite influences from the blade servers. I'd be surprised if the PC world doesn't start copying some of this in the near future, just like Dell and others took the latch down cases from the G4 and SFF PCs took the all-in-one motherboards from the iMac/Cube. They can start by using cheap aluminum instead of the huge one piece sheets of high quality aluminum that Apple uses. Nobody will notice in the PC world, in the Mac world everyone would claim how "discolored" or "flimsy" their new machine is (or some other assorted whine).

IMO, the dual-G4 MDD and FW800 are excellent machines. The case design alone has lasted since the B&W G3 (http://www.everymac.com/systems/apple/powermac_g3/stats/powermac_g3_300_bl.html)! Dual Opteron and Dual Xeon machines in the PC world are far more expensive (oh, you can try to put them together on the cheap, and you pretty much hit Apple's MDD price point when you're done). The only main weakness was that certain applications like video and audio tasks use a huge amount of bandwidth and the shared bus of the G4 and its tiny CPU memory bandwidth meant that the CPUs were data starved no matter how large you made the memory bus. Hence hacks like a L3 cache.

Every CPU has some give/take. I remember when the Athlon was considered a monster and then FSB speeds started to go through the roof and show its weaknesses. It still wins raw number crunching tests, but now is widely agreed to have been eclipsed by the latest P4s for all practical purposes except price/performance.

On a more relevant note, these benchmarks bode very well for the G5. I'm actually pretty shocked that a DP 1.45Ghz didn't beat the runt of the G5 litter (single CPU 1.6Ghz). There was a serious question about the hacked nature of the vector unit (Velocity, Altivec, or VMX depending on who you talk to) in the 970...

As for the 1 optical drive and 2 SATA limitation, study your history. Geez, it wasn't until QuickSilver that Macs supported 2 optical drives! Eventually, Apple will probably introduce G5s with more bays and until then you can buy a G4 via eBay or from Apple (MDD) or you can be a bit smarter and get another useless optical drive or drive array (FW800, SCSI with a card, NAS, or Fibre Channel (http://www.apple.com/xserve/raid/)). My bet is, even among graphics, video and scientific computation people, there is very little demand for two internal optical drives.

And there are about 10 reasons why SATA is better than IDE, and only about 1 reason why it isn't (about $20 saved/drive). I'm tired of rehashing them so you can just search the internet if you like. Even the PC world is moving to SATA, and that pace will quicken now that Apple has entered.

(Oh yes, I build my own AMD and Intel PCs too. 2/3 of the machines I personally own are x86 boxen and I work with them all the time. I fail to see how putting together some off-the-shelf parts and tweaking a bios is relevant to anything. Nobody is holding a gun to your head saying you have to buy a Mac.)

John Q Public

Aug 22, 2003, 08:34 PM

my apologies to WM. and ffakr for a couple mistakes...

The only early PPC's that I've been exposed to were the delapidated PM 6100's a company I worked for provided as reference machines for a couple pieces of software we used...hardly a machine to be appreciated when using PIII's...so I defer to your knowledge there where I should have noted 68040's instead...

DDR vs SDR...Bare Feats shows the difference between a DDR and SDR speeds in "Real World" apps...I'll let his tests prove that point...
http://www.barefeats.com/pmddr.html

SATA...The biggest argument against the G5 and SATA for me (you'll have to reference back a few days) is lack of expandability...and NOT being able to use the ATA100 drives out of my deceased B&W G3...and if you already have ATA Drives and want the new G5...there's no way you can add them...short of buying a FireWire or USB case for each of them...

This is why there is generally no performance difference between the high end 128MB and 256 MB gaming vid cards.

...actually the 256MB Gaming Cards are marginally slower than their 128MB counterparts...but only by a few fps (because of addressing issues)

why is most of this relevent? ISA and EISA? HUH? who gives a crap? EISA didn't even make it on the PC.

pardon me...but EISA is the 16bit slot you found in MOST PC's for many years...

ISA was the tiny 8bit slot about the same size as a PCI slot...

...and I do need to correct my own typos on a point...

ISA=8Mz
EISA=16Mz
PCI=33Mz
PCI (Enhanced)=66Mz (which was used on second generation G3's and Yikes G4 for video)
AGP=66Mz (it doesn't matter which flavor 1/2/4/8x...it's only 66Mz...where the differences in AGP formats resides...the width of the data path and how the Northbridge saturates the bus)
PCI-X=133Mz (although PCI-X has been out for nearly 2 years on Workstations and Servers, as a replacement for the aging PCI (Enhanced), there is still not adequate support...and unfortunately won't be until "Mainstream" PC manufacturers begin to build it into their systems...as a sidebar...the next generation of Video Card will be PCI-X because it allows higher bandwidth than the proposed AGP 16x)

that last reference to PCI (Enhanced) should have been PCI (Extended)...

Enhanced being the 32bit 66Mz
Extended being 64bit 33Mz

(and my apologies to Dr Hertz...several companies I've worked and consulted for over the years simply used "z" as shorthand for hertz)

...and other "Titles" I use (such as differenciating Processor, System and Memory busses) go back about 15 years
when I first started working in the industry...admittedly my bread and butter is still WinTel based (and I love the looks of horror when someone asks what kind of computer is on my desk at home)...I can add some knowledge on certain "Industry Standard" technologies that I work with extensively...

lastly ffakr...I didn't view WM.'s debate as a "Slam" as you've so stated...it's called intellectual discourse...where people agree and disagree until both sides have equal understanding of the other's viewpoint and logic...I simply found one statement a little offensive...

vniow

Aug 22, 2003, 08:46 PM

Originally posted by John Q Public
SATA...The biggest argument against the G5 and SATA for me (you'll have to reference back a few days) is lack of expandability...and NOT being able to use the ATA100 drives out of my deceased B&W G3...and if you already have ATA Drives and want the new G5...there's no way you can add them...short of buying a FireWire or USB case for each of them...

http://www.thechipmerchant.com/41593.html

John Q Public

Aug 22, 2003, 08:51 PM

[QUOTE]Originally posted by tychay
As for the 1 optical drive and 2 SATA limitation, study your history. Geez, it wasn't until QuickSilver that Macs supported 2 optical drives!...there is very little demand for two internal optical drives.

...you're talking cases here...there was support for multiple optical drives on Rev.2 G3's and higher...just required a little surgery to make a cage that would fit another optical drive...have a friend who did that to his...and as little as demand may be from where you see it...I typically (in my PC's now my Mac) run 2 burners (specifically 2 Combo Drives)...for me it's time saving to burn two-disks at once (without the expense or space required of buying a production duplicator)

And there are about 10 reasons why SATA is better than IDE, and only about 1 reason why it isn't (about $20 saved/drive).

I think SATA is a good thing (although I personally prefer Ultra320LVD SCSI)...but SATA is somewhat limiting as to internal expansion to two drives in the G5...where my G4 has 4 internal drives...and using a Sonnet Tempo used 7 drives in my G3 (2 on the primary IDE...one as slave to the combo-drive...and 4 more off the PCI card) the 6 on the bottom of the case were mounted on 3 double drive sleds...so when my G3 died I saw little reason to waste all those HDD's by waiting to buy a machine that they were incompatible with...and with which ultimately I would not have been happy...

John Q Public

Aug 22, 2003, 08:53 PM

Originally posted by vniow
http://www.thechipmerchant.com/41593.html

ok...I know there are adapters...but where would I mount more than 2 HDD's in the G5 case???

WM.

Aug 23, 2003, 12:08 AM

Originally posted by John Q Public
my apologies to WM. and ffakr for a couple mistakes...
Accepted, and I appreciate your getting the period at the end of my username right. :) It's only there because the MR forums require at least three characters.
The only early PPC's that I've been exposed to were the delapidated PM 6100's a company I worked for provided as reference machines for a couple pieces of software we used...hardly a machine to be appreciated when using PIII's...so I defer to your knowledge there where I should have noted 68040's instead...
Yes, I think the '040s did often have FSBs that ran at processor speed.
DDR vs SDR...Bare Feats shows the difference between a DDR and SDR speeds in "Real World" apps...I'll let his tests prove that point...
http://www.barefeats.com/pmddr.html
Um, that was on a machine that only had an SDR system bus--in other words, the FSB bandwidth was half the RAM bandwidth. This is not the case on the G5 or most PCs.
[responses to things I wasn't bothering you about]

(and my apologies to Dr Hertz...several companies I've worked and consulted for over the years simply used "z" as shorthand for hertz)
Well, I did say "I'll grant people their shorthands". :) I'm more forgiving of things that have been learned through years of professional experience rather than lack of knowledge (or proofreading ;) ).
...and other "Titles" I use (such as differenciating Processor, System and Memory busses) go back about 15 years
when I first started working in the industry...admittedly my bread and butter is still WinTel based (and I love the looks of horror when someone asks what kind of computer is on my desk at home)...I can add some knowledge on certain "Industry Standard" technologies that I work with extensively...

lastly ffakr...I didn't view WM.'s debate as a "Slam" as you've so stated...it's called intellectual discourse...where people agree and disagree until both sides have equal understanding of the other's viewpoint and logic...I simply found one statement a little offensive...
Again, I'm sorry that you found it offensive, but once again I'm still not convinced that you've read the Dev Note. All I'm trying to get here is accuracy. I was trying to correct your apparent misunderstandings of the G5's architecture, which hopefully wouldn't have occurred had you first read the Note (which I referred to as "knowing what you're talking about").

In other words, in my first post I was trying to figure out what you meant by "SystemBus", guessed that you were talking about the bus between the controller and the slots et al., and explained that that bus ran at 800 MHz. In your reply to that post you basically ignored me and/or told me that I was wrong. I doubt you would have done that had you read the Dev Note at that time.

It sounds like maybe you've been working for a long time on systems where, for example, the interface between the north and south bridges is always a PCI-type bus running at the same speed as the slots. Well, that's not the case in the G5, and it seems like that's been the source of the confusion here.

HTH
Sorry for any hurt feelings; I hope we can be friends of sorts (on these forums at least :) )
WM

John Q Public

Aug 23, 2003, 02:58 AM

WM., I actually have a serious question (something about the G5 that has caused a few questions)...I was looking at an Apple page (http://developer.apple.com/technotes/tn/tn2087.html) about the G5 and the differences to the G4...and something didn't quite seem right...

...about a year ago I watched Uncle Steve begin an Apple Keynote Speech about the "Megahertz Myth" (c. 2000), it was comparing the new (at the time) G4, Intel P4, MIPS and Athlon (I don't know if you're familiar with this presentation...it was archived on the ADC site until a couple weeks ago and vanished)...the argument for the G4 being more effiecient than the P4 was that it only had a more efficient Data Path because it used a "7 Stage Pipeline" rather than the P4's extremely inefficient "20 Stage Pipeline"...

...this address also "accidentally" pointed out (in an underhanded way by virtue of illustration) why the Athlon was faster (Clock-Cycle for Clock-Cycle) than the P4 also (the Athlon has a 10 Stage Data Path)...and from an archetectural standpoint makes beautiful logic...the address went on further to illustrate why it's so (and I really couldn't fault their logic on anything presented)...

...and that it was necessary for the P4 to run faster than the G4 to do the same work because of the longer Data Paths)...

now prefaced...

If the G4 is more efficient and faster (Cycle for Cycle) than the Pentium4 (because of 7-Stage vs. 20-Stage Pipelines)...and Uncle Steve is espousing the G5's improved efficiency over the G4 (and everything else for that matter).

Why then is the G5 using a Data Path with 23 Stages (and claimed higher efficiency), in contradiction to what they'd publicly stated a couple years ago when originally debunking the "Megahertz Myth"???

...now it seems there's been an "About-Face" in logic...saying "...we need more clock-cycles to keep up...and we're willing to sacrifice efficiency to get there...but seeing the overall result is faster...we'll tell everyone it's more efficient..."...it's just curious (to me at least)...

soggywulf

Aug 23, 2003, 08:40 AM

Originally posted by vniow
http://www.thechipmerchant.com/41593.html

Unfortunately, Apple has stated that these adapters won't fit in the G5 case. It is in the helper/FAQ pages on Apple's website.

What I am doing for my old ATA drives is sticking them in a spare ATX case, and running 36" round IDE cables to my PCI ATA adapter card in the Mac. Works like a charm. A bit messy looking though. :)

ColdZero

Aug 23, 2003, 09:33 AM

Originally posted by John Q Public
Why then is the G5 using a Data Path with 23 Stages (and claimed higher efficiency), in contradiction to what they'd publicly stated a couple years ago when originally debunking the "Megahertz Myth"???

...now it seems there's been an "About-Face" in logic...saying "...we need more clock-cycles to keep up...and we're willing to sacrifice efficiency to get there...but seeing the overall result is faster...we'll tell everyone it's more efficient..."...it's just curious (to me at least)...

What I also hated about the "MHz Myth" was that Apple glanced over how fast each processor was running.

Say the P4 had 21 stages (to make my math easy). It is running at 3ghz, the G4 with 7 stages is running at 1ghz. In the same 1 second interval, the P4 has completed its pipeline 3x more than the G4 has. If they were running on an equal clock basis, the G4 would be more efficient, but they aren't. And if the P4 ran at 1ghz, it wouldn't have a 23 stage pipeline either.

tychay

Aug 23, 2003, 03:09 PM

Originally posted by John Q Public
DDR vs SDR...Bare Feats shows the difference between a DDR and SDR speeds in "Real World" apps...I'll let his tests prove that point...
http://www.barefeats.com/pmddr.html

Read this article again because you've misread it. The DDR vs. SDR comparison occurs only on PowerMac's not in general. DDR is still twice as fast in bandwidth, but the two G4's share the same SDR bus and thus become data starved. The only area where the DDR actually would improve performance is when the AGP graphics cards run out of memory.

This doesn't apply to the G5's. In fact the only case it might apply is the single CPU 1.6Ghz. If you look at the spec sheet, you'll see they use significantly cheaper PC2700 RAM instead of the overkill PC3400 RAM. I'll note that the fastest Pentium 4 have the same bandwidth on their FSB but you still see a lot of people putting PC3400 RAM in them.

That problem isn't Apple's fault. The G4 of that era simply do not allow independent CPU buses nor scale up to DDR speeds. This will start to be fixed in later Motorola 130nm G4s. AMD Athlons have similar problem on a microarchitecture scale (the tiny data width on memory access keeps the pipes from being fed) which is why they give up performance to P4s on the top end.

SATA...The biggest argument against the G5 and SATA for me (you'll have to reference back a few days) is lack of expandability...and NOT being able to use the ATA100 drives out of my deceased B&W G3...and if you already have ATA Drives and want the new G5...there's no way you can add them...short of buying a FireWire or USB case for each of them...

But since you don't have space in the case anyway, you can't use them. You might have a valid point that the case doesn't leave any room for legacy adapters but now we're quibbling over a cheap $150 drive (because the data can easily be transferred with Carbon Copy Cloner). On a more aesthetic note, putting a SATA adapted IDE drive in a G5 is positively sinful. What I'd do is buy a Firewire 800 case (they get cheap the more drives you have) or a Super Drive Dock and turn them into "removeables". Sure, that's more cables, but this has a lot of uses (such as portability) beyond which an internal gives you. SATA is the thing to put in: faster hard drives, faster bus, independant channels, smaller cables, hot swapabillity, single cable, jumperless, no interference issues, less voltage. If Apple put in IDE then people would really feel crippled in 5 years--Macs are used far into the future (or resale at high prices) and the design needs to think in terms of future standards, not current.

Arguing about not including the archaic IDE sounds to me much like when people complained that the iMac and the G4 PowerMac didn't have a 3.5" diskette drive.

Besides, for the vast majority of people, there is no need for a G5. For what I see more people use their computers for a iMac is good enough and a PC is probably better (since they seem to play a lot of video games). The MDD 2x1.25Ghz G4 is an excellent machine that boots OS 9. Other than FW800, I fail to see what it gives up to the FW800 model. Certainly not performance--you just pointed a series of benchmarks showing how starved the 2x1Ghz G4 already is!

Take care,

terry

John Q Public

Aug 23, 2003, 03:36 PM

[QUOTE]Originally posted by tychay
But since you don't have space in the case anyway, you can't use them. You might have a valid point that the case doesn't leave any room for legacy adapters but now we're quibbling over a cheap $150 drive (because the data can easily be transferred with Carbon Copy Cloner). On a more aesthetic note, putting a SATA adapted IDE drive in a G5 is positively sinful. What I'd do is buy a Firewire 800 case (they get cheap the more drives you have) or a Super Drive Dock and turn them into "removeables". Sure, that's more cables, but this has a lot of uses (such as portability) beyond which an internal gives you. SATA is the thing to put in: faster hard drives, faster bus, independant channels, smaller cables, hot swapabillity, single cable, jumperless, no interference issues, less voltage. If Apple put in IDE then people would really feel crippled in 5 years--Macs are used far into the future (or resale at high prices) and the design needs to think in terms of future standards, not current.

yes and no...those "Cheap" $150 drives add up when you're pulling 7 of them out of a deceased B&W (smallest drive removed after death was 40GB and yes...somewhat cheaper...but the 80's that found their way into the new G4 that replaced the B&W)

Arguing about not including the archaic IDE sounds to me much like when people complained that the iMac and the G4 PowerMac didn't have a 3.5" diskette drive.[B]

...I've never missed the floppy drive...the only reason I have a USB Floppy is for VirtualPC and installing Guest OS's that the install cds are not Torito Format (DOS, Win9X and OS/2)...

[B]Besides, for the vast majority of people, there is no need for a G5. For what I see more people use their computers for a iMac is good enough and a PC is probably better (since they seem to play a lot of video games). The MDD 2x1.25Ghz G4 is an excellent machine that boots OS 9. Other than FW800, I fail to see what it gives up to the FW800 model. Certainly not performance--you just pointed a series of benchmarks showing how starved the 2x1Ghz G4 already is!

my big issue with the "New" 2002 MDD machines isn't FW800 (although it will be nice when I get toys that support faster than FW400...but that you only have "AirPort" and not "AirPort Extreme" and seeing my intranet here at home (and home office) consists of most wireless machines (all supporting 802.11G)...the faster connections to other machines (namely my laptop) is a good thing.

...and to be honest...I really don't miss OS 9...most of the apps I use are Carbonized (if not Cocoa) and about the only thing I run that is still old-tech is Diablo and Starcraft as the occasional diversion...besides...if I really want to use OS 9...I'll resurrect the B&W (replace Logic Board and Power Supply) or buy a G3 AIO (always wanted one of those)...

the only thing that makes me jealous of the "New" G4 is it has 2MB L3 where my Dual 1.25 only has 1MB L3...

WM.

Aug 23, 2003, 06:41 PM

Originally posted by John Q Public
[how can Apple claim the G5 is more efficient than the G4 when its pipeline is so much longer and they've already told us how bad that is ("MHz Myth")]
I think one reason the G5 doesn't suffer from having such a long pipeline (which is still actually shorter than the P4's AFAIK) is that it has exceptional branch prediction logic. If I just made a fool of myself by using the wrong terms, well, you're really not asking the right person about this kind of stuff. All I can say is, read the ArsTechnica articles about the 970 (which I mostly haven't read) and wait until we get USEFUL, ACCURATE BENCHMARKS (of which there are none at this point because no one's tested Photoshop on the G5 with the optimization update thingy). Then we'll talk. :) But again, this really isn't something I know all that much about.

FWIW
WM

Cubeboy

Aug 23, 2003, 07:29 PM

Originally posted by WM.
I think one reason the G5 doesn't suffer from having such a long pipeline (which is still actually shorter than the P4's AFAIK) is that it has exceptional branch prediction logic. If I just made a fool of myself by using the wrong terms, well, you're really not asking the right person about this kind of stuff. All I can say is, read the ArsTechnica articles about the 970 (which I mostly haven't read) and wait until we get USEFUL, ACCURATE BENCHMARKS (of which there are none at this point because no one's tested Photoshop on the G5 with the optimization update thingy). Then we'll talk. :) But again, this really isn't something I know all that much about.

FWIW
WM

Yes, your absolutely right, while the G5 will suffer from a greater branch mispredict penalty relative to the G4, it's 16000 entry Branch History Table/Global Predictor Table/ Selector Table gives it far superior branch prediction. As I've shown in a table from my previous post in this thread:

Pipeline Stages:
G4: 7 stages
K7: 10 stages
G5: 16 stages
P4: 20 stages

Branch History Table:
G4: 2048 entires
K7: 2048 entries
P4: 4096 entries
G5:16000 entries (!)

Average Hitrate:
G4: 93% (?)
K7: 93%
P4: 96%
G5: 97%+

Penalty Estimations:
G4: (.07*7= .49)
K7: (.07*10=.7)
P4: (.04*20)= .8)
G5: (.03*16)= .48) or with a 98% Hitrate (.02*16)=.32)

Most likely the G5 will suffer even less from branch mispredicts than the G4.

Of course, Apple marketing didn't mention anything about branch mispredicts, their statement went something like this:

"The fewer the steps, the shorter — and more efficient — the pipeline. Thanks to its efficient 7-stage design (versus 20 stages for the Pentium 4 processor) the G4 processor can accomplish a task with 13 fewer steps than the PC. You do the math.

Which is to say the least, quite humorous. :D

wizard

Aug 23, 2003, 08:00 PM

Originally posted by eric_n_dfw
Software that remains compiled for only the G4 scheduler will probably not see huge increases, but I personally think that the huge increase in FSB speed will speed up the whole OS so much, as with most comercial software being recompiled for the G5 , that the overall impression will be that it is much faster.

Well yes I think we agree to a certain point here. The issue is that there will be little incentive for many developers to produce optimized G5 code when Apple is still selling G4 and G3 PC's. I can see the majority of the developers selecting optimizations that perform well on the entire processors base. The only vendors that will be generating 970 only code are those in the performance driven markets such as video and image processing.

While I really hate to put to much emphasis on the early bench marks floating around the net, I still believe that we will see a mix bag on existing code. Some reports seem to verify this. Support of existing code is very important especially in an economy in reccession, this is probally the 970's biggest flaw and also its best success.

As far as faster goes I think Apple has missed the boat here. A machine like the G5 deserves at least one or two models with high performance GPU's. It really boggles the mind to see that largest GPU memeory allotment is 64MB. With so much of the graphics, from the OS level, be processed on the GPU it would seem to me that maxing out the video memory would be the thing to do on the high end machines.

As to the "barely competes" comment about the AltiVec/VMX units - Hannible at ArsTechnica agreed with you untill he got the dirt on the G5 from IBM:

(Read the link for all the interview questions and whatnot.)

He goes on to talk about how the VMX units look a lot worse on paper than they will perform in real life and the IBM guys basically say that it will be much better in future chips.

I've read Hannibles write up it is very interesting. The problem is that VMX can choke on existing code which is bad. At the same time VMX is providing rather remarkable compatability so I geuss one should not complain to much.

I've heard much in the way of rumors about the enhanced VMX unit and VMX2. I must say this recent news gives me hope that PPC will ba around for awhile.

Thanks
Dave

wizard

Aug 23, 2003, 08:18 PM

Hi mim

While I'd have to think that here is some cost to an expansion slot, I wouldn't call it expensive. Just look at the expense of putting all of those holes into the G5! The reason to have a external drive slot is so that you can take advantage of new technology in 2 to 3 years. Not everybody goes wild over the latest graphics card, its also a mistake to assume that a drive slot will be used for a new optical drive.

Firewire is great technology, but the best place to add permanent technology to your computer is inside the case. To be honest a PCMCIA slot at the front of the machine could do wonders for expandability.

Apple needs to leverage firewire and every other technology built into the G5's. Its a shame they don't put more emphasise on the possibilities that FW offers the public. As to the processor upgrades it will be very interesting to see what happens with these. I would imagine with the case design of the G5 that one could move forward quite a bit with plug in processor upgrades. That is if Apple shares with the upgrade companies. With the case the way it is though these upgrades should have no impact on your graphics card nor vs versa. That is one of the good design features of the G5 case.

Dave

Originally posted by mim
I question the need for the internal expansion slots. They're expensive to add, and will increase the heat load on the inside of the case. This may not be an issue now, but if this case is to last you 2 or 3 years then it may become a problem (esp. with a super hot running graphic card upgrade + 2 more 10000rpm drives :( ).

I also think that you'll get the same speed out of external FW800 devices as you would plugging them into the internal connections. Maybe slightly higher latency...but then you could just swap the new and old drives around.

I read recently that FW800 external drives were getting close to the theoretical maximum read/write speed of the fastest available drive (I'll look for the link).

Personally I think it's no bad thing for Apple to leaverage FW800 a bit more by reducing the number of bays inside the case - especially when it allows for the real crux of "life extentions"... frying hot processor upgrades.

porovaara

Aug 23, 2003, 08:59 PM

Originally posted by ColdZero

Say the P4 had 21 stages (to make my math easy). It is running at 3ghz, the G4 with 7 stages is running at 1ghz. In the same 1 second interval, the P4 has completed its pipeline 3x more than the G4 has.
Think about this again.

John Q Public

Aug 23, 2003, 09:15 PM

[QUOTE]Originally posted by Longey Nowze
for those who think benchmarks are useless: why do car manufactures tell you how fast their car goes from 0-100km/h? and then tell you how fast it goes from 100Km/h-0? why do they give you fuel consumption numbers?

I generally read them...ignore them...and buy the vehicle that has jus the right "feel" to it...buying a car (or in my case Jeep) is rarely a rational decision...and to a point same with buying a computer...if it were...we'd all be using Amigas, driving Volvos and Hondas, and driving right past all those places that serve quick and easy artery-hardening burgers and fries...LOL...

WM.

Aug 23, 2003, 11:45 PM

Originally posted by Cubeboy
As I've shown in a table from my previous post in this thread:
That's how I knew all that stuff. :D It was one of those things where you know you know something, but you're not sure how...Good posts, both of them. :)

WM

Shrike_Priest

Aug 24, 2003, 05:53 AM

A thing about the benchmarks.

IIRC, Xbench hasn't been optimised in any way for these benchmarks. Now, they did update Xbench recently, but that was after the benchmarks. And as from Macrumors report, the only difference was that it wasn't using the vec_dst instruction. Although he clearly states that the G5s take big performance hit from the use of that. So you can add a few points just on that.

But here's the big issue with the Xbench benchmarks.

IIRC, the G5 has two FPUs, the G4 has one. An un-optimised app would only take advantage of one FPU, like the G4. An optimised app, like Photoshop, would run the FPUs in parallel, effectively doubling FP performance. So if you're not benchmarking using Photoshop, you're not gonna get anywhere near the optimal FP performance. To my knowledge, neither Cinebench or Xbench uses both FPUs. I might be wrong, but I haven't heard anything about it.

In short: Wait for the Photoshop benchmarks, then compare that to the old G4 Photoshop benchmarks. That way you'll get a good comparison. These are pretty much useless.

Cubeboy

Aug 24, 2003, 08:00 AM

Originally posted by Shrike_Priest
A thing about the benchmarks.

IIRC, Xbench hasn't been optimised in any way for these benchmarks. Now, they did update Xbench recently, but that was after the benchmarks. And as from Macrumors report, the only difference was that it wasn't using the vec_dst instruction. Although he clearly states that the G5s take big performance hit from the use of that. So you can add a few points just on that.

But here's the big issue with the Xbench benchmarks.

IIRC, the G5 has two FPUs, the G4 has one. An un-optimised app would only take advantage of one FPU, like the G4. An optimised app, like Photoshop, would run the FPUs in parallel, effectively doubling FP performance. So if you're not benchmarking using Photoshop, you're not gonna get anywhere near the optimal FP performance. To my knowledge, neither Cinebench or Xbench uses both FPUs. I might be wrong, but I haven't heard anything about it.

In short: Wait for the Photoshop benchmarks, then compare that to the old G4 Photoshop benchmarks. That way you'll get a good comparison. These are pretty much useless.

Careful now, the G5 can use it's dual FPUs in any app, no software can prevent it from doing so. Remember, it's not software that finds independent instructions that can be executed in parallel, thats the core's job!

Having dual FPUs doesn't automatically mean double the FPU performance, dependent instructions, consecutive load-stores, latency and memory accesses can all lower the actual IPC. This is why we don't see the Athlon, with it's triple FPUs, triple ALUs, and triple decoders, wipe the floor with other comparable chips; there are only "bursty" moments where the Athlon can actually execute 3 instructions/cycle.

Now the G5 will be much better at getting ILP out of code than the Athlon since:

1) It has a 215 entry instruction buffer (compared to the Athlons 72), allowing it to fill larger cycle gaps in each L1-cache access with other independent instructions. (principle of OOOE).

2) The code that it will be running will typically contain more ILP than x86 code, this includes much fewer load-stores.

3) The G5 has much faster memory access and lower latency than the Athlon. It won't have to stall for as many cycles with a L1/L2 cache miss, and it's won't have to spend as much time waiting to be fed new data (from the memory).

Despite all this, don't expect it to use the full power of it's dual FPUs all the time, even running code with high amounts of ILP, typical CPUs with the above charecteristics will only be able to utlize all it's execution slots around 60% of the time.

soggywulf

Aug 24, 2003, 08:30 AM

Originally posted by wizard
As far as faster goes I think Apple has missed the boat here. A machine like the G5 deserves at least one or two models with high performance GPU's. It really boggles the mind to see that largest GPU memeory allotment is 64MB. With so much of the graphics, from the OS level, be processed on the GPU it would seem to me that maxing out the video memory would be the thing to do on the high end machines.

The ATI 9800 Pro video card is available in 128 MB form as a BTO option on all the G5s. This is arguably the best consumer/semi-professional GPU on the market today.

However, I believe that OSX's use of OpenGL (Quart Extreme) alone is not going to max out the capabilitites of the 9800, not by a long shot. QE has to run on machines with much slower graphics hardware (iMacs, eMacs, etc). So the performance demands should be fairly limited, from the OS.

Games and 3D apps, on the other hand, will kick some serious butt with the 9800.

Shrike_Priest

Aug 24, 2003, 09:03 AM

Yeah, I was a bit optimistic there with the doubling :p

Still, in apps like Photoshop, which are the ones many of us care about, FPU performance is vital, and when a benchmark is made where it only uses one FPU you'll get no where near the actual performance peak.

Or did I mess up and any code run on the G5 uses both? I'm a bit uncertain about this.

Meh... I'll wait for the photoshop benches.

PieMac

Aug 24, 2003, 09:26 AM

Originally posted by jettredmont
Two cables and 2-3 screws, once you flip open the side of the box (and make sure you are grounded/static-free!). Not as easy as plugging in new RAM, but certainly not brain surgery either.

Replacing an optical drive is very easy to do, as there are about three steps involved, and you "undo" them to remove the old drive, then "redo" them the other way around to put the new one in.

My father's replaced his own CD drive (on a Windows box, even!). That would tell you how easy it is if you knew my father and electronics :)

Definitely sounds easy enough, but wouldn't this negate the Apple warrenty?

daveg5

Aug 24, 2003, 02:21 PM

Originally posted by PieMac
Definitely sounds easy enough, but wouldn't this negate the Apple warrenty?
of course not adding ram and drives and cards do not violate apples warranty on powermacs at least, thats why they made them open for expansion, thats why they give you a pdf on there site with the correct way to do it. now if you do it incorrectly with out being gronded etc. then you might be at risk

PieMac

Aug 24, 2003, 02:53 PM

That makes perfect sense.

*doh*

I've never really owned a Mac that was upgradable to any degree (I have an iBook).

Cubeboy

Aug 24, 2003, 04:48 PM

Originally posted by Shrike_Priest
Yeah, I was a bit optimistic there with the doubling :p

Still, in apps like Photoshop, which are the ones many of us care about, FPU performance is vital, and when a benchmark is made where it only uses one FPU you'll get no where near the actual performance peak.

Or did I mess up and any code run on the G5 uses both? I'm a bit uncertain about this.

Meh... I'll wait for the photoshop benches.

Read the first paragraph of my original response to you again.

As long as the code contains instructions that can be executed in parallel (on most programs this is a given), it will benefit from having multiple execution units. :)

mum

Aug 25, 2003, 01:13 AM

Originally posted by daveg5
of course not adding ram and drives and cards do not violate apples warranty on powermacs at least

Also my Powerbook's manual has instructions for replacing the HD as well as adding RAM.

ffakr

Aug 25, 2003, 11:01 AM

Originally posted by John Q Public
DDR vs SDR...Bare Feats shows the difference between a DDR and SDR speeds in "Real World" apps...I'll let his tests prove that point...
http://www.barefeats.com/pmddr.html
Barefeats benchmarks the performance of DDR vs. SDR Powermacs which are limited by SDR processor busses. This is NOT an indication of the performance difference between the SDR standard and the DDR standard.
Going over the posts a bit more, I've noticed that this has been pointed out to you a number of times but you still persist in this stance.
Your argument is essentially the same as saying, 'a two inch pipe won't move any more water at the same pressure than a 1 inch pipe...' but for comparison you are comparing a 2 inch pipe that is hooked up to a one inch source pipe vs. a one inch pipe. The DDR on a G4 system can never perform up to its full potential because it is faster than the rest of the system. The DDR on the G4 is NOT the limiting factor.
SATA...The biggest argument against the G5 and SATA for me (you'll have to reference back a few days) is lack of expandability...and NOT being able to use the ATA100 drives out of my deceased B&W G3...and if you already have ATA Drives and want the new G5...there's no way you can add them...short of buying a FireWire or USB case for each of them...

Well, you could put an ATA card in the machine. You would have to work a bit to route cables up to the top of the case but I'm sure it could be done. If you don't mind a kludge, you could also put the drives in the bottom of the case (for the single processor machines)
I think Apple produced a machine without compromise. If you need more drive space, Apple's answer is to buy more and move your data to the nice big SATA drives, rather than adding an additional ATA bus and adding to the production costs.

This is why there is generally no performance difference between the high end 128MB and 256 MB gaming vid cards.
...actually the 256MB Gaming Cards are marginally slower than their 128MB counterparts...but only by a few fps (because of addressing issues)
I'm not saying that there aren't addressing issues with 256MB cards (because I don't know) but this would seem odd since you would only need 32bit address space to directly access all of the bits in 256MB of ram.
Anyway, from what I've seen, the main cause of the speed disparity between high ram vid cards and higher ram vid cards is that the manufacturers tend to put slightly slower (and much cheaper) ram on the very high ram vid cards. This is certainly the case with the GF4 line. The lower ram cards are always marginally faster unless you benchmark them with HUGE textures. Rumor has it that Doom 3 will use up to 80 MB of texture in a scene so those 128 MB vid cards might actually pull ahead.

pardon me...but EISA is the 16bit slot you found in MOST PC's for many years...

ISA was the tiny 8bit slot about the same size as a PCI slot...

ISA is nearly twice the size of a PCI card. The connectors on the blade were much larger. http://www.nullmodem.com/images/conn_busisa16ma.gif
This is a 16 bit ISA (not EISA) card.
EISA is even larger.
Here is an overview of ISA and EISA card connectors that might help...
http://members.iweb.net.au/~pstorr/pcbook/showtell/show6.htm
I have to respectfully disagree that EISA was found on the majority of PCs.
I've been supporting PCs and Macs (professionally) for almost a decade (supporting lots of legacy hardware) and it has been my experience that it's actually unusual to find EISA slots in PCs.
This is what Techfest.com has to say on a page decribing the EISA technology... [i]"The EISA Bus originated in 1988 & 1989. It was developed by the so called "Gang of Nine" (AST, Compaq, Epson, Hewlett-Packard, NEC, Olivetti,Tandy, Wyse and Zenith) as an alternative to IBM's "patented" Micro Channel bus. It received limited use in 386 and 486 based Personal Computersthrough about 1995 before being obsoleted by the PCI bus as Pentium based systems were introduced."[i]
.. "received limited use"...
In fact, a quick search for EISA ISA on google provided me with links to EISA data structures and Digital Unix... over and over. Not much info for the PC. A few FDDI and ethernet card links, but very few.
It just hasn't been my experience that EISA was ever a pervasive technology. Most of the EISA Machines I've run across were PC Servers, specifically machines that would compete with IBM's microchannel based PC servers.
[/B][/QUOTE]