PDA

View Full Version : Mac vs. Window's mhz


dylomel
May 14, 2005, 11:42 PM
My friends owns a PB G4 867 mhz, and is wondering what's the difference between Mac and Window's mhz/ghz ('coz I heard Mac's 1ghz would be faster than Win's 1 ghz). Thank you !

strider42
May 14, 2005, 11:50 PM
My friends owns a PB G4 867 mhz, and is wondering what's the difference between Mac and Window's mhz/ghz ('coz I heard Mac's 1ghz would be faster than Win's 1 ghz). Thank you !


It toaly depends on what you areusing them for. Different processors are faster at different things. Generally speaking, a PPC processor should vbe more efficient that an x86 processor, and so will perform faster than an x86 at the same speed. But perhaps the biggest difference is a mac doesn't have windows bogging it down. My computer at work is 10 times faster in terms of mhz than my old iMac, but I get more done on my iMac because windows isn't in the way. But I'm not using it to do any heavy lifting.

Abstract
May 14, 2005, 11:59 PM
A PB 867MHz G4 will probably feel like a 2.4 - 2.6 GHz P4. However, when it comes to the hardcore intensive stuff, the P4 will be faster at the processing. But again, in terms of "feel", it'll feel like the P4 I mentioned.

It may also feel a bit slower than the first Centrino chip released. Forgot what it was (a 1.3GHz?).

nightdweller25
May 15, 2005, 12:20 AM
Correct me if I'm wrong, but it's kind of like multiplying the speed x2 for the mac. So a 1.5 would be a 3.0 P4...

toothpaste
May 15, 2005, 01:20 AM
Can't find the article, but PPC chips are Risc while x86 are Cisc. The increased performance is on the way that the chips handle their instruction set. Risc is roughly 2.5 times more effiecent. But the mzh is used by microsoft and x86 chip makers as a propaganda tool to confuse. Don't read into it too much. PPC chips are better than x86.

Chrispy
May 15, 2005, 01:28 AM
I have a windows 3.0 GHz P4 HT Dell and a 400MHz G3 that I am going to upgrade to a 128MB Graphics card, a 700MHz G4 and an 80GB Hard Drive. Other than games I will use this computer for almost everything. Even without any upgrades the computer still runs Panther really well. I am truely amazed at how well Macs utilize their power. Score a big one for Apple on this one.

tech4all
May 15, 2005, 01:29 AM
Correct me if I'm wrong, but it's kind of like multiplying the speed x2 for the mac. So a 1.5 would be a 3.0 P4...

In that case, the 2.7Ghz G5 would be equivalent to a 5.4Ghz PC.....? :eek: :confused:

I think it's just a bit lower than 2x though. But not sure....

Mav451
May 15, 2005, 01:40 AM
In that case, the 2.7Ghz G5 would be equivalent to a 5.4Ghz PC.....? :eek: :confused:

I think it's just a bit lower than 2x though. But not sure....

Nah, 2x was with G4's vs. P4s. I'm not the expert here (some others could help me here), but I believe the G5 is much less efficient compared to the G4 (instructions per clock).

tech4all
May 15, 2005, 01:44 AM
Nah, 2x was with G4's vs. P4s. I'm not the expert here (some others could help me here), but I believe the G5 is much less efficient compared to the G4 (instructions per clock).

Hmmm interesting. Probably a weird question (for anyone) then why didn't Apple just give the G4 higher "Ghz" since it does more per clock cycle?

I wonder what the formula is for comparing a G5 to a G4 processor?

dylomel
May 15, 2005, 01:44 AM
thank you for all the kind help ! my friend wanna switch because he is really into games, and right now there are more out there for Win than Mac.

eXan
May 15, 2005, 01:50 AM
In that case, the 2.7Ghz G5 would be equivalent to a 5.4Ghz PC.....? :eek: :confused:

+ you have 2x 2.7 GHz G5 in PowerMac :D

hjhhjh
May 15, 2005, 02:27 AM
the sad but true thing is

ppcs areent that much faster per clock than x86 chips

its about a 1 : 1.55 ish ratio


not 2 ..

the g4 and the g5 arent much different .. and when the g4 reaches higher speeds .. the g5 is much more powerful

2 ghz g4 vs 2 ghz g5 .. the g5 wins

1.6 ghz g5 vs 1.67 ghz g4 .. the g4 wins

now ..

a dual 2.7 would be like a 4.2 ghz pentium 4 (around where AMDs top computers are at right now)

an 867 mhz chip is like 1.35 ghz pentium 4 .. but if it had l3 cache like the 867 mhz quicksilvers did .. then it would be around 1.5 ghz penitum 4

but in the end it all depends on what programs are being tested ..

hjh*

Platform
May 15, 2005, 02:31 AM
Nah, 2x was with G4's vs. P4s. I'm not the expert here (some others could help me here), but I believe the G5 is much less efficient compared to the G4 (instructions per clock).

Well that is bad then they sould work more on the G4's and get them to 64Bit...and re-name them G6 ;)

aussie_geek
May 15, 2005, 03:15 AM
Here is a link for the MHz myth as told by Apple a few years ago.

Click the link and save it to disk if you don't want it to stream instantly.

Mhz Myth Vid (http://pulsar.esm.psu.edu/Faculty/Gray/graphics/movies/mhz_myth_320f.mov)

Hope this sorts your friend out regarding Apples, PC's and Hertz. ;)


aussie_geek

After G
May 15, 2005, 03:26 AM
I don't know, even if the actual multiplier is something like 1.55, as mentioned previously, it always felt like a G4 was equivalent to Intel hardware up to a P4 with at least twice the clock, at least when running Windows on said hardware.

The thing with Windows, in my experience, is that the UI feels fast but then bogs down when processing during application use, etc. Mac OS X, on the other hand, feels a tiny bit slower on some UI things, but stays at roughly the same speed even when doing things in the background (except when beach-balling).

What I am wondering, is how this has changed when comparing to Centrinos, newer AMD chips, etc. Anyone care to comment?

aussie_geek
May 15, 2005, 03:51 AM
I don't know, even if the actual multiplier is something like 1.55, as mentioned previously, it always felt like a G4 was equivalent to Intel hardware up to a P4 with at least twice the clock, at least when running Windows on said hardware.

The thing with Windows, in my experience, is that the UI feels fast but then bogs down when processing during application use, etc. Mac OS X, on the other hand, feels a tiny bit slower on some UI things, but stays at roughly the same speed even when doing things in the background (except when beach-balling).

What I am wondering, is how this has changed when comparing to Centrinos, newer AMD chips, etc. Anyone care to comment?

A few months ago one of my colleagues brought their new Sony VAIO in for show and tell. I know it had a Centrino chip - there was an Intel sticker on the palmrest :rolleyes:.

Application launch times were pretty much the same as my PowerBook when it opens an app for the first time after reboot. One thing about Apple's have is an application cache. I don't think PC's have this feature. When I opened Safari the second time it beat the VAIO hands down in opening IE.

Another workmate has a 3.2Ghz Toshiba that is really bogged down with tonnes of apps and probably heaps of spy / mal ware. Considering what has been said in the other posts it should have eaten my PowerBook. The PowerBook beat it hands down.

Mac's are meant to multitask and people are doing this more often these days. It is not uncommon for someone to be doing more than 2-3 things at a time. You should consider this when comparing the two systems.

aussie_geek

Maxiseller
May 15, 2005, 04:26 AM
QUOTE:

e thing with Windows, in my experience, is that the UI feels fast but then bogs down when processing during application use, etc. Mac OS X, on the other hand, feels a tiny bit slower on some UI things, but stays at roughly the same speed even when doing things in the background (except when beach-balling).



Absolutly right.

It took me a few weeks to get used to this change in particular, and it can be disconcerting having a slower user interface, but all the processing in the background seems much more powerful.

On a windows PC if you ask it to do 5 things at once, you'll have hangs, slowdowns, moans and groans. If you do the same thing on my emac (Low spec comparitivly speaking) it happily churns away at these tasks.

I love the Mac OS. It is at the moment how a computing experience should be. It makes you blissfully unaware of the processing, and lets you actually be productive.

Applespider
May 15, 2005, 04:57 AM
thank you for all the kind help ! my friend wanna switch because he is really into games, and right now there are more out there for Win than Mac.

If your friend is really into games (ie likes playing the latest/greatest with high FPS) then tell him to stick with PCs. Only a few games are ported over to Mac, and they don't tend to be optimised for the Mac so the FPS aren't great.

Macs are great and, on current offerings, I'm unlikely to buy a PC for the house again but the one class of computer user that they're still not great for is the hardcore gamer.

wrldwzrd89
May 15, 2005, 06:28 AM
If your friend is really into games (ie likes playing the latest/greatest with high FPS) then tell him to stick with PCs. Only a few games are ported over to Mac, and they don't tend to be optimised for the Mac so the FPS aren't great.

Macs are great and, on current offerings, I'm unlikely to buy a PC for the house again but the one class of computer user that they're still not great for is the hardcore gamer.
You're right when it comes to commercial games. However, for shareware games, the situation is the exact opposite. The Mac shareware game community has always been quite lively; if you search well enough, you're bound to find something you like. Many of these shareware games are Mac-only, too.

Platform
May 15, 2005, 07:35 AM
You're right when it comes to commercial games. However, for shareware games, the situation is the exact opposite. The Mac shareware game community has always been quite lively; if you search well enough, you're bound to find something you like. Many of these shareware games are Mac-only, too.

Yes that is so good...many more than for XP :rolleyes: :D

mad jew
May 15, 2005, 07:42 AM
As everyone says, it depends on what you're doing as to which processor is more efficient. As an example of the PPC being better than a P4, my 800MHz G4 iBook rips music from CDs at about the same speed as my 1.7GHz P4 desktop. That's using iTunes, which is obviously much better on a Mac, it chugs on PCs. :(

Just thought you might like a piece of solid evidence. :)

36183
May 15, 2005, 07:50 AM
it speeds also depend on how the software you are using has been written. almost all games will run faster on a pc. my dell machine with a 2.4 ghz porcessor 128mb 5200 graphic card and 768 ram performs much better at UT2004 than my powerbook does (1.5ghz 64mb 9700 512 ram).

but on my powerbook moive encoding and multitasking is faster than my pc. but comparing disk instesive tasking with my laptop mac and my desktop pc is not a very fair test, but i do find my powerbooks disk performance far better when it is reading files off my external firewire drive.

wrldwzrd89
May 15, 2005, 08:01 AM
Yes that is so good...many more than for XP :rolleyes: :D
Fair enough. However, the Mac doesn't have so much of the lousy shareware programs and has plenty of good ones.

Platform
May 15, 2005, 09:32 AM
Fair enough. However, the Mac doesn't have so much of the lousy shareware programs and has plenty of good ones.

That was what I said :confused: :rolleyes:

Abstract
May 15, 2005, 09:59 AM
.its about a 1 : 1.55 ish ratio
.a dual 2.7 would be like a 4.2 ghz pentium 4 (around where AMDs top computers are at right now)


Yes, but notice how you need DUAL 2.7GHz chips for a 1.55 factor for this difference in the speed of the system (using your factor of 1.55).

The G5 is not as efficient as the P4 or Centrino. A single 2.7GHz PPC 970 chip would probably not be as fast as a single AMD chip rated at 2.7GHz.

iQuit
May 15, 2005, 10:33 AM
Well...PowerPC processors are built a whole lot different and run more effecient and smoother. On general things such as video editing and web and etc. the difference is the PPC is multiplied by 2.2. Also Mac OSX runs a whole lot smoother considering it is built better and on UNIX. Generally though PPC is faster except for gaming,not because the Mac's suck at games...just they are ported over and don't run too well on a Mac. If there were developers that were more interested in Mac's..they have much more potential in running games smoother with there faster processors. PowerPC's are really great and will smoke a PC in video editing,and graphic editing.

http://www.asia.apple.com/g4/myth/

Go there Here,it will help explain more in a short video

wrldwzrd89
May 15, 2005, 10:36 AM
That was what I said :confused: :rolleyes:
I thought you meant that Windows has more programs than the Mac in ALL areas (knocking out part of my point), so I came back with the argument that Mac cuts out all the garbage :confused:

BornAgainMac
May 15, 2005, 10:43 AM
I noticed a Pentium III at even 450 Mhz doesn't "feel" any slower than a Pentium 4 running at 2 Ghz with the same memory and both running XP. I tried a Celeron 2.8 Ghz machine running XP Home and it seemed much slower than my old 233 Mhz running Win2000. Maybe today's PCs are downloading too much spyware and with the addition of all the addon programs you need to have running to keep from getting a virus and keeping PC running good the additional load is killing the performance. XP isn't helping anything either because of it's memory footprint.

I did have a friend encode uncompressed AVI files to MPEG-1 on his Athlon 1.3 Ghz to compare with my Pentium 3 450 Mhz PC and it was only a 30% improvement in performance. I couldn't control what was installed on his PC or memory configuration so I can't conclude any thing from that benchmark.

Platform
May 15, 2005, 10:45 AM
Yes, but notice how you need DUAL 2.7GHz chips for a 1.55 factor for this difference in the speed of the system (using your factor of 1.55).

The G5 is not as efficient as the P4 or Centrino. A single 2.7GHz PPC 970 chip would probably not be as fast as a single AMD chip rated at 2.7GHz.

Yes that is bad,but that is proberly the reason why apple sell Dualies :p :rolleyes: Anyway SP G5 2.0Ghz and above is enough power for most people ;)

malevolent_i
May 15, 2005, 10:47 AM
The speed difference really depends on the application. I think many mac apps are just poorly optimised (esp. games like d3).
Mac mhz = 2x windows/x86 mhz?? no way!
just look at the benchmarks at barefeats.com.
I would say a G5 is comparable to athlon's 64 fx line in terms of mhz/performance ratio. If u look at cinebench ratings, a g5 1.8 (mono) compares well with a p4 2.5 w/out hyperthreading.
A g4, on the other hand...is significantly slower. it's hampered by its puny bus speed of 167 mhz. The g4's bus, coupled with slower floating point calculations means it does worse than the g5 in games. However, the g4 has a more straightforward vector calculations unit and so i think it does better than the g5 in those types of calculations. Vector calculations are only used in many productivity applications that take advantage of it (not in games though)--photoshop does a good job in that, for example.

btw, I dunno about u guys, but i always thought windows' UI felt faster! unless u have lots of spyware and open apps bogging the system down, don't u feel like windows resizing and moving is faster in xp?

Applespider
May 15, 2005, 11:10 AM
btw, I dunno about u guys, but i always thought windows' UI felt faster! unless u have lots of spyware and open apps bogging the system down, don't u feel like windows resizing and moving is faster in xp?

Ah..that well known snappiness™ which is the Holy Grail of every Mac OS update. Windows does resize windows faster but it is apparently all to do with how the two systems draw their windows which would take someone far more technical than me! IIRC it's something along the lines of the Mac redrawing each pixel of a move, whereas Windows doesn't. It's basically also the reason that the Mac has attractive drop shadows around windows whereas XP is 'flatter'. You makes your choice and you takes your trade offs.

I know what you mean about feeling faster. I used to think that my Windows (work) PC opened applications faster than my Mac at home does until I actually timed it from clicking on the app to it being available to use and discovered that they're pretty much identical. The difference I was 'seeing' I think came from the way that with Windows, the program identifier came flashing up on the screen and then the program loaded. In the Mac, it bounces a couple of times in the Dock and then opens. The bouncing isn't as noticeable as that big icon in the middle of the screen.

Mav451
May 15, 2005, 11:27 AM
Ah..that well known snappiness™ which is the Holy Grail of every Mac OS update. Windows does resize windows faster but it is apparently all to do with how the two systems draw their windows which would take someone far more technical than me! IIRC it's something along the lines of the Mac redrawing each pixel of a move, whereas Windows doesn't. It's basically also the reason that the Mac has attractive drop shadows around windows whereas XP is 'flatter'. You makes your choice and you takes your trade offs.

I know what you mean about feeling faster. I used to think that my Windows (work) PC opened applications faster than my Mac at home does until I actually timed it from clicking on the app to it being available to use and discovered that they're pretty much identical. The difference I was 'seeing' I think came from the way that with Windows, the program identifier came flashing up on the screen and then the program loaded. In the Mac, it bounces a couple of times in the Dock and then opens. The bouncing isn't as noticeable as that big icon in the middle of the screen.

Opening apps has little if any to do with the OS. Its all about the I/O connection >> mainly the hard drive itself. SCSI? 5400/7200? SATA vs. ATA? WD Raptor 15ks? FW800? RAID? You get the idea. RAM plays a part, afterwards, but the initial launch is all in your hard drive.

maxvamp
May 15, 2005, 11:30 AM
Maybe today's PCs are downloading too much spyware and with the addition of all the addon programs you need to have running to keep from getting a virus and keeping PC running good the additional load is killing the performance. XP isn't helping anything either because of it's memory footprint.

I did have a friend encode uncompressed AVI files to MPEG-1 on his Athlon 1.3 Ghz to compare with my Pentium 3 450 Mhz PC and it was only a 30% improvement in performance. I couldn't control what was installed on his PC or memory configuration so I can't conclude any thing from that benchmark.


You cannot go out and directly compare an Apple OSX system to a Windows system. There are too many variables.

I could go out and make two different benchmark suites right now that would show two different result on which was faster. This is where Steve gets in trouble because he does exactly this.

As mentioned above, there are many more factors that affect a PC's performance. The simplest one is comparing which Microsoft OS is on your system.

Mac users may say that OS X.4 feels a little sluggish on a G3 233 compared to OS 9, but they can still be productive. A PC person running a Pentium 233 would never be able to be productive on a system upgraded from Windows 98 to the latest WindowsXP revision. The OS has way too much bloat from 98 to XP. This bloat slows down the procs.

Another thing that skews a result in the two processors is if the applications are written and optimized for the host processor. If you take photoshop that will use Altivec, a clearly superior SIMD engine, then a PPC always wins on certain filters. However, if you take generic open source programs that generally have little to no platform tweaking ( such as DB4 ), then clock speed becomes a factor and the x86 platform can clearly win.

In short you cannot say one proc is x faster than another, because even within the families, this is not always true. G4 Altivec vs G5 Altivec, PIV against P-M, etc.

Max.

shane1
May 15, 2005, 12:04 PM
A PB 867MHz G4 will probably feel like a 2.4 - 2.6 GHz P4. However, when it comes to the hardcore intensive stuff, the P4 will be faster at the processing. But again, in terms of "feel", it'll feel like the P4 I mentioned.

It may also feel a bit slower than the first Centrino chip released. Forgot what it was (a 1.3GHz?).

Yikes. I like my mac as much as the next guy, but that's fanboyism at its finest. You're not doing anyone a favour by blatantly lying.

vatel
May 15, 2005, 12:32 PM
As has already been said you really have to consider what OS and software your running in addition to just how fast is the G4 vs P4. For example the extremely poor cinebench and SPEC performance of G4 machines doesn't really reflect in my real world computing experience

I find that for most 'normal' tasks (email, surfing, text editing, spreadsheet) there isn't any difference between my PB and my Dell or my friends' Centrino laptops. The difference comes up when running computationally intensive code. For these tasks I've found as a rule of thumb the G4 is roughly equivalent to a first gen Centrino 1.6 (Banias), and significantly slower than either a straight P4 or the newer Centrino's (Dothan), these difference seem to be due mainly to the higher cache in the Dothan chips.

Workflow is also an important consideration. For example the much maligned window resizing issue, doesn't bother so much because I rarely resize or even move windows on PB desktop, especially as compared to how often I perform the same tasks on my Dell. I also find Expose allows me to handle multitasking faster and more efficiently than on my Dell.

I wonder what the formula is for comparing a G5 to a G4 processor?[/QUOTE]

G4 and G5's are about equal (on a per GHz basis) for tasks which:
1. Don't require significant I/O from memory
2. Don't involve double precision floating point code
(or non-vectorized single precision code)

Once memory bandwith or floating point math become factors, a G5 can run around 2x faster than a G4
(You can see this in barefeats' comparison of iMac vs PB performance)

As long as memory bandwith is not a factor, the G4 can actually run vectorized code faster than a G5.

zach
May 15, 2005, 01:31 PM
the g4 and the g5 arent much different .. and when the g4 reaches higher speeds .. the g5 is much more powerful

2 ghz g4 vs 2 ghz g5 .. the g5 wins

1.6 ghz g5 vs 1.67 ghz g4 .. the g4 wins

hjh*

That's purely conjecture.

You are correct in your latter, however there IS no G4 2 GHz chip, therefore you cannot make that statement.

And from what I know of the G4's architecture versus the G5's, the 2 GHz G4 would in fact beat the 2 GHz G5.

auxplage
May 15, 2005, 06:06 PM
The iMac 1.6ghz does do video much more efficiently than the 1.67ghz Powerbook as mentioned somewhere above.

Maedus
May 15, 2005, 07:00 PM
A PB 867MHz G4 will probably feel like a 2.4 - 2.6 GHz P4. However, when it comes to the hardcore intensive stuff, the P4 will be faster at the processing. But again, in terms of "feel", it'll feel like the P4 I mentioned.

It may also feel a bit slower than the first Centrino chip released. Forgot what it was (a 1.3GHz?).

Yikes. I like my mac as much as the next guy, but that's fanboyism at its finest. You're not doing anyone a favour by blatantly lying.

He's not saying a PB 867MHz G4 is as fast as a 2.4 - 2.6 GHz P4 but that when you're doing simpler tasks, it will feel as fast as a 2.4 - 2.6 GHz P4. This is because with simpler tasks, neither processor is really being pushed to their limits and so they "FEEL" similar. Its when you're doing hardcore processing where the G4 will max out before the P4 that you can tell the P4 is faster. So no, he's not being a fanboy nor is he lying.

I notice it on my computer. My pitiful 266 MHz G3 sometimes FEELS just as fast as the AMD Athlon XP 3200+ on some tasks but there's no question that the AMD is a more powerful processor which becomes noticeable as soon as I do anything remotely complicated on the AMD.

mklos
May 15, 2005, 07:05 PM
Nah, 2x was with G4's vs. P4s. I'm not the expert here (some others could help me here), but I believe the G5 is much less efficient compared to the G4 (instructions per clock).


The G5 is way more efficient than the G4. The G5 was perform 215 instructions per clock cycle vs 16 on the G4. HUGE difference there!

Maedus
May 15, 2005, 07:12 PM
Correct me if I'm wrong, but it's kind of like multiplying the speed x2 for the mac. So a 1.5 would be a 3.0 P4...

I think the 2x multiplier originated around the times of the G3's and the early G4's to get the equivalent pentium of the time. Probably a P3 but I don't remember offhand.

And I think the 2x also was used not because it was an exact conversion but because it was easy to tell somebody "multiply the mac's processor speed by 2 to get a rough idea how it stacks up to a pentium" rather than creating a more accurate conversion rate which might end up being something like "multiply the mac's processor speed by 1.87593 to see how it'd measure up to a pentium" or more than likely "use this formula [insert complex formula] with x being your mac's processor speed and y being your mac's bus speed and z being your memory and so on and so on in order to a mutli-faceted answer that tells you what kind of pentium, motherboard, chipset, memory, etc it equals. Gah.

So the 2x comes from an older age when 2 was an easy to remember and easy to calculate number that gave you a ballpark answer for converting a mac processor to the contemporary pentium processor. Today I doubt you could devise a number that could easily do this since bus speeds, memory speeds, GPU speeds, etc play a much more noticeable role in computers overall speed than they did back then.

minimax
May 15, 2005, 08:14 PM
A PB 867MHz G4 will probably feel like a 2.4 - 2.6 GHz P4. However, when it comes to the hardcore intensive stuff, the P4 will be faster at the processing. But again, in terms of "feel", it'll feel like the P4 I mentioned.

It may also feel a bit slower than the first Centrino chip released. Forgot what it was (a 1.3GHz?).

hahaha LOL what drugs are you taking?

solvs
May 15, 2005, 09:15 PM
The G5 is not as efficient as the P4 or Centrino. A single 2.7GHz PPC 970 chip would probably not be as fast as a single AMD chip rated at 2.7GHz.
Depends on the AMD chip. I have a 2GHz Athlon XP that doesn't seem as fast as my friends 1.8GHz G5 iMac, but the newer AMDs are supposed to be really fast. Those dual core ones are way faster than the G5 (and the dual core P4), but they're expensive. The G5 is better than even a higher clocked P4, so I don't know what you mean by more efficient. The high end P4 have more cache, but are really hot, and not that much faster, they just reach higher clock speeds. The Centrinos are really efficient, more like the old P3, but also aren't clocked as high. The G4 is a pretty good chip, and even more efficient (and cooler temp. wise), but it's hampered by poor memory bandwidth. The highest G4 FSB being sold by Apple right now is a paltry 167MHz, and it barely supports DDR RAM. FSBs and memory bandwidth are a lot more important than clock speed.

Plus, it's hard to measure anyway. This has been discussed ad nauseam, if you do a search in this forum you'll see. Benchmarks don't mean much, it's all about what applications you use, and some will be better on one platform than another. It's all a matter of perception, and has a lot to do with Windows and OS X as well.

auxplage
May 15, 2005, 10:16 PM
The whole FSB is without doubt important, but let us remember that most of the current P4 chips being sold really only have a FSB of 200mhz which is only a 20% improvement over the present G4. Of course Intel says the FSB is 800mhz because of the whole 4 instructions per cycle, but that is only theoretical. The FSB is nowhere near 800mhz – it is most likely closer to 400mhz. The G4 should be able to move up to a 200mhz FSB with the 7448.

Mav451
May 16, 2005, 01:40 AM
The G5 is way more efficient than the G4. The G5 was perform 215 instructions per clock cycle vs 16 on the G4. HUGE difference there!

Hrm...guess I was wrong.

paperinacup
May 16, 2005, 02:22 AM
This is VERY confusing. So is what I'm hearing tat the G4 is better than the G5???!!! Then what's the point of waiting for a new PB?

Excuse me if I sound confused.... becuase I am.

mad jew
May 16, 2005, 02:26 AM
This is VERY confusing. So is what I'm hearing tat the G4 is better than the G5???!!! Then what's the point of waiting for a new PB?

Excuse me if I sound confused.... becuase I am.


I think the arguments are that a processor with very similar specifications (clock speed etc.) the G4 would be better than a G5. I'm not saying this is true or anything, but it's kind of irrelevant because the G5 is better than the G4 at the moment. Whether this is a result of the larger R & D the G5 has received in recent years or because of their different architectures is kind of a moot point.

anubis
May 16, 2005, 02:31 AM
Apple learned about 15 years ago that when it comes to computer processors, "less is more". If my memory serves me correctly (not even going to google it. going to go out on a limb here), Apple introduced its first "Power"Mac in 1992, based on the PowerPC architecture. This architecture (still found in the G5 today) is called RISC - "reduced instruction set code". A long time ago, Apple realized that if you took a computer chip and stripped it down so that it could only perform a few very basic calculations, then you could reduce its die size, power requirements, and clock speed while drastically increasing performance. Macs, and indeed all RISC-based architectures, run software that is highly optimized, coded and compiled using fewer possible types of instructions than the CISC based architectures. With a simpler, smaller instruction set, RISC processors can whip through data a lot faster than its more primative and complicated CISC counterpart (which has been around since the 70s, btw).

Think of it this way. If your G5 can only "understand" and perform 30 unique operations, it can perform more calculations per clock cycle (say 1 complete calculation in 5 cycles) compared to the P4, which can perform say 100 unique operations, but each complete calculation takes 15 cycles because the chip is larger and more complicated. That's why a 3GHz P4 runs like a piece of crap compared to a 2GHz G5.

And wouldn't you know it, Intel is finally starting to realize that, but only by accident. The Pentium M Centrino line of processors was initially designed with power conservation and small size in mind. But in the process, they found that a 1.5GHz Centrino is faster than any other 1.5GHz chip they made. That is, when they finally decided to make a chip simpler rather than more complicated (each incarnation of the Pentium line has a larger instruction set than the previous one), they found that it was actually better. That's why you hear everyone talking about how Intel is going to start pushing for the Centrino in desktops as a P4 replacement.

Now of course, this is a gross oversimplification of what is really going on. Other factors include caching, the tree structure, incorrect estimation penalty, and other nuances of modern CPU design that vary greatly between the x86 line and the PowerPC line.

Here's one last idea for you to chew on: the world's fastest supercomputer, the 70+ teraflop IBM bluegene, only runs at 700MHz. ;)

aswitcher
May 16, 2005, 02:45 AM
It would be nice if there was some sort of website that indepdently compared G4/G5s versus pentiums for different programs etc. Good place to point wouldbeswitchers

jiggie2g
May 16, 2005, 03:04 AM
The G4 is about 1.5x faster then a P4 at same speed.

1Ghz G4 = 1.5Ghz P4 = 1.1Ghz Athlon XP
2Ghz G5 = 2.2Ghz G4 = 3Ghz P4 = 1.8Ghz Athlon 64

Thats my Formula based on all the benches i've seen.

solvs
May 16, 2005, 03:24 AM
It would be nice if there was some sort of website that indepdently compared G4/G5s versus pentiums for different programs etc.
http://www.arstechnica.com and http://www.barefeats.com are good places to start.

jiggie2g
May 16, 2005, 04:16 AM
http://www.arstechnica.com and http://www.barefeats.com are good places to start.


Or here www.systemshootouts.org


By the way does anyboy else think it's amazing that there is a 1200mhz Performance difference between the Athlon 64(939) vs. the P4 at the same clock speed. :eek:

rmanger
May 16, 2005, 04:29 AM
The formulas I use have never failed me once:

G4 vs. G5:

There are two formulas, one for lower speeds, and one for higher speeds.

Lower clock speeds = 1.0GHz and less
Higher clock speeds = 3.0GHz and more
Between = Your best guess in the middle of the two...

At lower clock speeds, a G4 will beat a G5 by roughly 25%. This is because a G4 is more efficient than a G5 when it comes to actual processor cycles. Also, lower clock speeds tend to negate the G5s superior memory bandwidth.

Lo Clock G4 -> G5 Conversion: 1.0GHz G4 = 1.25GHz G5 (In Theory)

Higher clock speeds favor the G5 by 50%. At this point, data is usually forced to move in and out of memory faster because of the higher clock. Since the G4s memory bus is so much slower, it will start yielding diminishing returns at higher clock speeds. In other words, a 3GHz G4 will usually be much slower than a 3GHz G5 due to memory bandwidth!

Hi Clock G4 -> G5 Conversion: 3.0GHz G4 = 2.0GHz G5 (In Theory)

If you want some proof of this, try overclocking (experts only!) a G4 some time and comparing the speed to a comparably clocked iMac G5. Then you'll see why the G4 wasn't designed for higher clock speeds.

G4 vs. P4:

As jiggie2g and others have already mentioned, yes, a G4 at the same clock speed as a P4 is roughly 50% faster. This has been observed on lots of hardware from my PC using friends, so this formula is the most reliable.

G4 -> P4 Conversion: 1GHz G4 = 1.5GHz P4

So why does the P4 handily beat the G4? Brute force of a higher clock. The G4 is struggling to hit 2.0GHz, while the P4EE is 3.8GHz. But don't feel too bad, G4 users. The P4EE is very expensive, and is seems to be overclocked as it is. Why else would you need overclock quality components to use them?

G5 vs. P4:

For some reason, the G5 doesn't seem to be that much faster than the P4 in terms of raw clock speed. The G5 is fast, but its not THAT fast. It just feels like a 33% speed advantage to the G5.

G5 -> P4 Conversions:

2.0GHz G5 = 2.66GHz P4
2.5GHz G5 = 3.33GHz P4
2.7GHz G5 = 3.6GHz P4

Of course, a Power Mac G5 will naturally be faster in most tasks simply because it uses two processors. Multiprocessing is its main advantage. In other words, don't expect a G5 to beat a PC in games (mostly single-threaded). It won't.

AMD Processors:

The clever marketing gurus at AMD are actually pretty accurate about making their processor ratings, but only if you knock off 5% of their number. Yes, AMD is very clever to knock Intel's marketing back in their face.

4000+ AMD = 3.8GHz P4
3200+ AMD = 3.04GHZ P4

I don't even bother looking at the GHz of an AMD processor, because I consider it a waste of time. Using the P4 speed name -5% is easier and more accurate to me. Also note that the AMD costs much less than the equivalent P4.

jiggie2g
May 16, 2005, 04:46 AM
I have to agree , Single G5's are not efficient CPU's and would benifit huge form a Dual core conversion. as yoy see it works every well in Dual CPU Config , but is not very affordable to most people in this setup.

I think once Apple gets a single DC PM G5 out they will sell like crazy provided apple doesn't start price gouging.

G5 is great for apple but is not the Atlhlon 64's Equal clock 4 clock.

Mord
May 16, 2005, 04:52 AM
Nah, 2x was with G4's vs. P4s. I'm not the expert here (some others could help me here), but I believe the G5 is much less efficient compared to the G4 (instructions per clock).

it was with the g4 Vs the p4 not the g4e which has a 7 stage pipeline compared with the 4 stage pipeline, it's slightly made up by the on die cache and L3. but it still is less efficient clock for clock, which is why a dual 533MHz g4 beats a dual 733MHz G4e in a few things. (see www.xlr8yourmac.com there is an article somewhere about it).

with the g4 a 500MHz g4 was about the same as a 1GHz p3 or a 1.3GHz p4 or a 633MHz G4e.

however the g4's bus has not scaled well at all a 1.5GHz G4e is not twice as fast as a 750MHz G4e thus reducing the amount the g4 is faster clock for clock, it's still way faster than the p4 clock for clock but no where near as much as it was.

now the G5 has a 20 stage pipeline but is massively more efficient in other ways, http://www.barefeats.com/g4up.html here is a 2GHz G4 Vs 2GHz G5 benchmark but it's more complicated than that, with a 2MB L3 cache the g4 dose much much better as it makes to slow ass bus less of an issue.

Platform
May 16, 2005, 06:59 AM
For some reason, the G5 doesn't seem to be that much faster than the P4 in terms of raw clock speed. The G5 is fast, but its not THAT fast. It just feels like a 33% speed advantage to the G5.

G5 -> P4 Conversions:

2.0GHz G5 = 2.66GHz P4
2.5GHz G5 = 3.33GHz P4
2.7GHz G5 = 3.6GHz P4




2.0 Ghz G5 = a 2.66 P4 :eek: :eek: That I don't believe ;)

Platform
May 16, 2005, 07:00 AM
I thought you meant that Windows has more programs than the Mac in ALL areas (knocking out part of my point), so I came back with the argument that Mac cuts out all the garbage :confused:

Oh...well I meant that Mac has a lot and good FREE games :p

Platform
May 16, 2005, 07:01 AM
That's purely conjecture.

You are correct in your latter, however there IS no G4 2 GHz chip, therefore you cannot make that statement.

And from what I know of the G4's architecture versus the G5's, the 2 GHz G4 would in fact beat the 2 GHz G5.

Well then your statment don't work either because there is no 2.0Ghz G4 ;)

ewinemiller
May 16, 2005, 07:05 AM
Until recently I was running a dual PIII-800, a dual G4-800, a PIII-1ghz laptop, and a P4 2.26. There was no noticable difference for the things I did (3D modeling and rendering) between the two dual 800s. The PIII-1ghz was a little snappier for modeling, but slower to render. The P4 was snappier still and smoked everything at render time. However if you were doing email, web, itunes, word processing, you couldn't tell what you were working on.

Now we've added an ibook G4 to the mix, a Pentium M 1.6 and a dual Xeon 3ghz. The ibook is fine for modeling, but renders real slow, the Pentium M is snappy for both, beating the old P4 by quite a bit in the render dept, the dual Xeon destroys everything at render time.

Nearly anything built in the last 5 years with enough memory is fast enough to run a modern OS and do web, mail, listen to music, etc. just fine. We still have an 8 year old dual PII-266 with 128meg of RAM running win2k here being used for the basics. That is why you can say that a slow G4 FEELS like a faster P4, but when it comes to really crunch the numbers, MHZ does matter, don't let the RDF fool you.

There is no simple conversion, each platform has its strengths. The key is figure out the applications you want to run, if they are only for one platform, then your choice is easy. If they are available on both, find someone to run some benchmarks for you, figure out the total costs of each platform and do the math according to your needs (i.e. are you looking for best performance, best bang for the buck, etc).

Just a quick note here, contrary to what many of the folks here will tell you, an XP installation is not an expensive time drain to keep safe and secure. Get a virus scanner installed, run the built in firewall, keep up with your service packs and patches (they can be set to install automatically), don't execute email attachments, don't run IE as a browser, and you'll be just fine. You can get free virus scanners (AVG) for non-commercial use and Norton can be had for dirt cheap if you watch the sales and rebates. I paid $5 each machine for virus scanning this year. Last year I made money on virus scanning because of the rebates.

mad jew
May 16, 2005, 07:17 AM
Well then your statment don't work either because there is no 2.0Ghz G4 ;)


Check this (http://www.gigadesigns.com/products7A1620U.html) out. Cool, huh.

Platform
May 16, 2005, 07:22 AM
Check this (http://www.gigadesigns.com/products7A1620U.html) out. Cool, huh.

But I was thinking about Apple's products ;) And that one is OVERCLOCKED

" Powered by a Freescale (formerly Motorola) 1.6GHz 7447A Processors"

But it is good to see that it is comming up to speed ;)

mad jew
May 16, 2005, 07:26 AM
But I was thinking about Apple's products ;) And that one is OVERCLOCKED

" Powered by a Freescale (formerly Motorola) 1.6GHz 7447A Processors"

But it is good to see that it is comming up to speed ;)


Yeah, I figured you were. I just thought you might be interested. :)

Mord
May 16, 2005, 08:13 AM
The formulas I use have never failed me once:

G4 vs. G5:

There are two formulas, one for lower speeds, and one for higher speeds.

Lower clock speeds = 1.0GHz and less
Higher clock speeds = 3.0GHz and more
Between = Your best guess in the middle of the two...

At lower clock speeds, a G4 will beat a G5 by roughly 25%. This is because a G4 is more efficient than a G5 when it comes to actual processor cycles. Also, lower clock speeds tend to negate the G5s superior memory bandwidth.

Lo Clock G4 -> G5 Conversion: 1.0GHz G4 = 1.25GHz G5 (In Theory)

Higher clock speeds favor the G5 by 50%. At this point, data is usually forced to move in and out of memory faster because of the higher clock. Since the G4s memory bus is so much slower, it will start yielding diminishing returns at higher clock speeds. In other words, a 3GHz G4 will usually be much slower than a 3GHz G5 due to memory bandwidth!

Hi Clock G4 -> G5 Conversion: 3.0GHz G4 = 2.0GHz G5 (In Theory)

If you want some proof of this, try overclocking (experts only!) a G4 some time and comparing the speed to a comparably clocked iMac G5. Then you'll see why the G4 wasn't designed for higher clock speeds.

G4 vs. P4:

As jiggie2g and others have already mentioned, yes, a G4 at the same clock speed as a P4 is roughly 50% faster. This has been observed on lots of hardware from my PC using friends, so this formula is the most reliable.

G4 -> P4 Conversion: 1GHz G4 = 1.5GHz P4

So why does the P4 handily beat the G4? Brute force of a higher clock. The G4 is struggling to hit 2.0GHz, while the P4EE is 3.8GHz. But don't feel too bad, G4 users. The P4EE is very expensive, and is seems to be overclocked as it is. Why else would you need overclock quality components to use them?

G5 vs. P4:

For some reason, the G5 doesn't seem to be that much faster than the P4 in terms of raw clock speed. The G5 is fast, but its not THAT fast. It just feels like a 33% speed advantage to the G5.

G5 -> P4 Conversions:

2.0GHz G5 = 2.66GHz P4
2.5GHz G5 = 3.33GHz P4
2.7GHz G5 = 3.6GHz P4

Of course, a Power Mac G5 will naturally be faster in most tasks simply because it uses two processors. Multiprocessing is its main advantage. In other words, don't expect a G5 to beat a PC in games (mostly single-threaded). It won't.

AMD Processors:

The clever marketing gurus at AMD are actually pretty accurate about making their processor ratings, but only if you knock off 5% of their number. Yes, AMD is very clever to knock Intel's marketing back in their face.

4000+ AMD = 3.8GHz P4
3200+ AMD = 3.04GHZ P4

I don't even bother looking at the GHz of an AMD processor, because I consider it a waste of time. Using the P4 speed name -5% is easier and more accurate to me. Also note that the AMD costs much less than the equivalent P4.

the p4 EE is at 3.73GHz and the g5 is nearly exactly the same clock for clock as the K8, AMD's performance rating do tend to hold with the non EE p4's and the EE p4's are a fair bit faster with there 2MB L3 cache, a 2GHz G5 is about the same as a 2GHz (3200+) socket 939 K8, this is shown in various barefeats benchmarks, the k8 wins in some areas and the g5 wins in others.

now G5-p4
2GHz G5 = 3.2GHz p4
2.3GHz G5 = 3.6GHz p4
2.5GHz G5 = 3.9GHz p4
2.7GHz G5 = 4.2GHz p4

i know these p4's dont exist it's just a p4 over/underclocked would be about the same, if you want to compare duals you just multiply by .9 as the xeon has a huge cache which helps allot.

Platform
May 16, 2005, 09:43 AM
Yeah, I figured you were. I just thought you might be interested. :)

Yeah, just that if you include overclocked once then you will have P 4 at 5Ghz ;)

Platform
May 16, 2005, 09:47 AM
now G5-p4
2GHz G5 = 3.2GHz p4
2.3GHz G5 = 3.6GHz p4
2.5GHz G5 = 3.9GHz p4
2.7GHz G5 = 4.2GHz p4



Now those look realistic ;) :cool:

Mav451
May 16, 2005, 11:14 AM
Aish, the one problem I have with the OCed G4 is that the bus speed is still low as hell.


* Tested at Guaranteed at 2.0GHz for 100MHz and 2.0GHz for 133MHz bus Power Macs


Such high clock speeds, and the memory bus remains unOCed. What the hell is the point then? Like the Athlon XP, multiplier OCing is ok, but when you are held back by craptastic memory bandwidth, it doesn't matter.

Remember when AMD finally got to 2Ghz? That was using 166Mhz x 2 = 333Mhz effective FSB. If AMD had done 2Ghz @ 100Mhz...those chips would have been downright worthless.

I would be a heckuva lot more interested in seeing a G4 when it is either at, the VERY least, 266Mhz or 333Mhz FSB. I don't care how exotic the cooling is on the northbridge/CPU. The fact is that all those benches don't matter until we finally see what a G4 can do with its legs stretched.

Right now, 2Ghz @ 100Mhz is like riding in coach class on an airplane.

Somone once told me a rule of thumb. When you divide clock speed by effective FSB, that number should never be higher than 9 or 10. Never. You can see with the last XP 3200+ (2.2Ghz / 400 = 5.5) that it followed that rule still. A64's? If you consider the effective FSB to be 1Ghz, the FX-55 (2.6Ghz) is doing just fine. P4's? Even at 3.8Ghz/800 = 4.75, they still follow that rule.

The 2.0 G4? Nope. Its not only much greater than 9 or 10, its double. Now a 266Mhz bus? That would reduce it to a factor of 7.52X from 20X. 333Mhz bus? Even better 6.006X.

rmanger
May 16, 2005, 05:59 PM
Originally Posted by Hector
now G5-p4
2GHz G5 = 3.2GHz p4
2.3GHz G5 = 3.6GHz p4
2.5GHz G5 = 3.9GHz p4
2.7GHz G5 = 4.2GHz p4

If you meant this...

now (Dual) G5-(Single) p4
(Dual) 2GHz G5 = (Single) 3.2GHz p4
(Dual) 2.3GHz G5 = (Single) 3.6GHz p4
(Dual) 2.5GHz G5 = (Single) 3.9GHz p4
(Dual) 2.7GHz G5 = (Single) 4.2GHz p4

then I'd believe the numbers. ;)

Mord
May 17, 2005, 03:10 AM
no i ment single to single, got some kind of benchmarks to prove it then go on http://www.barefeats.com/pentium4.html http://www.barefeats.com/macvpc.html


some cpu's have an advantage in some area's (xeons's in cinebench for expample), but it's hard to find software that is equally optimized for both platforms, you just have to do a few different tests and compare by what margin each platform wins by.

just like a 3GHz P4 Vs a 3000+ one cpu wins a bunch of tests when the other wins a different bunch of tests but in the end no one sits and applies photoshop filters all day, rendering is probably the most important test.

the fact of the matter is that there are not many benchmarks compareing these cpu's and if we look at singular tests they will be biased to the cpu that is most efficient at that task, like if toms hardware did a review of the X2 compared to the pentium extreme edition and focuses on cinebench.

i have had this arguement far too many times and as someone who has owned pc's and macs over the years i find myself useing my 600MHz ibook most of the time.

if you want to find someone with a 3GHz P4 and an imac G5 with equal specs bar the cpu feel free but it's harder than it sounds.

also those comparisons i gave were to the normal P4 not the EE just the vanilla prescott

Platform
May 17, 2005, 03:31 AM
If you meant this...

now (Dual) G5-(Single) p4
(Dual) 2GHz G5 = (Single) 3.2GHz p4
(Dual) 2.3GHz G5 = (Single) 3.6GHz p4
(Dual) 2.5GHz G5 = (Single) 3.9GHz p4
(Dual) 2.7GHz G5 = (Single) 4.2GHz p4

then I'd believe the numbers. ;)

Why....I know that the 2.0Ghz G5 "NOT DUAL" is a lot faster than a 2.8Ghz P4 at least so why not a 3.2 ;)

rmanger
May 17, 2005, 04:26 PM
Sorry, I should've been more specific.

I meant to compare the G5 to P4EE and the G4 to the vanilla P4.

Yes, the G5-vanilla P4 numbers would match up, but not to the P4EE. The P4EE just isn't that slow! But the vanilla P4 is...

As for Barefeat's testing methods, are the single processor tests turning off a processor, or is it just doing single-threaded apps?

Besides, I use conservative estimates to make up for the different software optimizations on the PC compared to the Mac. In ideal situations, the Mac can crush the PC, but the same holds true for the other side as well. Why would different benchmarks give different results? And how does Apple come up with their numbers?!!

Platform
May 17, 2005, 06:35 PM
Sorry, I should've been more specific.

I meant to compare the G5 to P4EE and the G4 to the vanilla P4.

Yes, the G5-vanilla P4 numbers would match up, but not to the P4EE. The P4EE just isn't that slow! But the vanilla P4 is...

As for Barefeat's testing methods, are the single processor tests turning off a processor, or is it just doing single-threaded apps?

Besides, I use conservative estimates to make up for the different software optimizations on the PC compared to the Mac. In ideal situations, the Mac can crush the PC, but the same holds true for the other side as well. Why would different benchmarks give different results? And how does Apple come up with their numbers?!!

Ok but the P4EE cost a looot more than the vanilla.....so :rolleyes: ;)

capone2
May 18, 2005, 01:03 PM
pc shootout.com it adds 7mhz to mac, like 1.5 G4 is a 2.2 P4!! Ive heard from hardcore techies that is pretty accurate.

So a 2.7mhz G5 would be a 3.4 P4

deputy_doofy
May 18, 2005, 01:13 PM
A PB 867MHz G4 will probably feel like a 2.4 - 2.6 GHz P4. However, when it comes to the hardcore intensive stuff, the P4 will be faster at the processing. But again, in terms of "feel", it'll feel like the P4 I mentioned.

It may also feel a bit slower than the first Centrino chip released. Forgot what it was (a 1.3GHz?).

I think that value may be stretching it a bit. I think an 867MHz G4 will be more like a 1300-1500MHz P4. Of course, it varies with the type of program running and whether or not it has been optimized for the chip on which it is running.
An 867 G4 will, however, be as fast or faster than a 2.4GHz Celeron. My 1GHz G4 w/512mb RAM is always faster than the 2.4 and 2.6GHz Celerons (at 768mb RAM) that I've personally used.

Mord
May 18, 2005, 01:18 PM
thats extremely inaccurate, even most hardcore techies dont know jack about most things they tend to be hardcore in there area, these comparisons should be in %, even then they are still inaccurate due to weather the bus and ram scales with clockspeed and weather tests are gfx card or cpu bootlenecked.

for the G5 Vs P4 EE i'll do a little research and edit the findings to this post

Mav451
May 18, 2005, 01:23 PM
thats extremely inaccurate, even most hardcore techies dont know jack about most things they tend to be hardcore in there area, these comparisons should be in %, even then they are still inaccurate due to weather the bus and ram scales with clockspeed and weather tests are gfx card or cpu bootlenecked.

for the G5 Vs P4 EE i'll do a little research and edit the findings to this post

Don't waste your time with that. I don't know a single person that would dare to waste their money on that overpriced chip...be it here, any techies I know, hell most of the PCPer and [H] community. Most know that paying about 70% gets them a much better FX chip.

Mord
May 18, 2005, 02:35 PM
no **** sherlock, this is for comparisons sake, there still are people out there that refuse to work on a computer unless it has "an intel inside"

AMD owns intel thats pretty established, and compareing the G5 to AMD cpus is dead easy, they are about the same on average clock for clock.

strangely enough most of my techie friends currently have 2.8GHz northwood p4's at the moment because a friend of mine got a crate of them for what came out as £50 each.

i also have not seen personally one good OC attempt on a K8, best i have seen is a winnie 3000+ at

cube
May 18, 2005, 02:37 PM
The G5 is way more efficient than the G4. The G5 was perform 215 instructions per clock cycle vs 16 on the G4. HUGE difference there!

The G4 fetches upto 4 instructions per clock cycle. The G5, 8.

Mord
May 18, 2005, 02:45 PM
The G4 fetches upto 4 instructions per clock cycle. The G5, 8.


but the G5's pipeline is three times longer than the G4's, also altivec in the G5 lags a tad behind the G4.

not even if the G5 was 300 times faster would i let anyone pry my g4 cube from my cold dead fingers.

cant wait for 7448 upgrade cards 1MB full speed cache baby yeah.

Platform
May 19, 2005, 03:16 AM
pc shootout.com it adds 7mhz to mac, like 1.5 G4 is a 2.2 P4!! Ive heard from hardcore techies that is pretty accurate.

So a 2.7mhz G5 would be a 3.4 P4

People who graduate this and that uni with a computer degree often don't know anything at all about Mac's......this is from experience ;) so this is incorrect ;)

Bad_JuJu
May 20, 2005, 02:05 AM
The whole FSB is without doubt important, but let us remember that most of the current P4 chips being sold really only have a FSB of 200mhz which is only a 20% improvement over the present G4. Of course Intel says the FSB is 800mhz because of the whole 4 instructions per cycle, but that is only theoretical. The FSB is nowhere near 800mhz – it is most likely closer to 400mhz. The G4 should be able to move up to a 200mhz FSB with the 7448.

As stated above - you need to be careful on just stating specs of buses or cpu's - because they are based on simulations or mathematical calculations ...... not actual averages of activity during program and OS execution.

Without optimizations in compilers (tuned to the cpu's feature set) - and driver optimizations ---- GHz and bus speeds don't mean squat.

I worked in the group at Intel that developed the PCI spec. and the very first PCI motherboards --- I recall reading a PCWeek report about a high-end ISA (10Mhz) SCSI disk controller that out performed the newer PCI (133MHz) SCSI controller :eek: bottom line was the ISA controllers driver had been optimized for years to use every trick known to xfer data -- the PCI driver was a 1st generation piece of code on a new bus.

Same thing applies as to how efficient XP is on x86 (cache usage / thrashing) vs OS X on PPC -- and then you also need to look at memory page management - IO xfers + interrupt handling (both HW and OS)

cube
May 20, 2005, 03:44 AM
but the G5's pipeline is three times longer than the G4's, also altivec in the G5 lags a tad behind the G4.

not even if the G5 was 300 times faster would i let anyone pry my g4 cube from my cold dead fingers.

cant wait for 7448 upgrade cards 1MB full speed cache baby yeah.

The point was that the HUGE difference referred to was not as extremely great.

jiggie2g
May 20, 2005, 05:32 AM
no **** sherlock, this is for comparisons sake, there still are people out there that refuse to work on a computer unless it has "an intel inside"

AMD owns intel thats pretty established, and compareing the G5 to AMD cpus is dead easy, they are about the same on average clock for clock.

strangely enough most of my techie friends currently have 2.8GHz northwood p4's at the moment because a friend of mine got a crate of them for what came out as £50 each.

i also have not seen personally one good OC attempt on a K8, best i have seen is a winnie 3000+ at

Well On my Venice core 3000+ I can boot at 3ghz(334x9)+ upping the Voltage but is not stable , u need water or phase change. i've seen people keep Venice and San Diego Chips stable at 3.2ghz with Phase Change.

I have hit 2.7ghz on stock voltage :eek: and held it with Prime 95. I will keep it a 2.4ghz(3800+ speed) just for the sake of keeping things cool as i don't need all that speed. not bad for a $149USD chip. :D

Mav451
May 20, 2005, 09:28 AM
Damn dude. Since I'm done with finals, I've been contemplating the move to A64. Alot of people have success with CH-5 (pumping it to 3.2V) so I might not even have to upgrade RAM (only need CPU/Mobo/HSF/PSU). I presume you're running on TCCD (3200XL)?

Yeah, I just realized i need a new PSU fer PCI-E too...ehhh

*back on topic*
I think auxplage is confusing FSB with "effective FSB". Effective FSB is the one that is shown in benchmarks. Of course, higher FSB can only do so much (250FSB x 2 on Athlon XP vs. 250FSB x 4 on A64)...

(edit: Haha, I notice I said effective FSB...probably made it up from my geotech/soils class "effective stress" << ok, engineering joke over).

Bad_JuJu
May 20, 2005, 11:08 AM
back on topic*
I think auxplage is confusing FSB with "effective FSB". Effective FSB is the one that is shown in benchmarks. Of course, higher FSB can only do so much (250FSB x 2 on Athlon XP vs. 250FSB x 4 on A64).


This is true and Bmarks are a good tool to measure performance - but always keep in mind that they are "synthetic" in nature as they may/will perform operations that a normal user wouldn't do - or wouldn't do in that way or that quickly --- plus as has been said - its like comparing apples and oranges - ha ha ha ---- because you've got differnt OS's and HW

The closest thing I've seen to a cross platform Bmark was an Internet browsing Bmark that used code that was not optimized for either PPC or x86. Of course you would need to run the Bmark dozens of times and average it out to eliminate net based delays

Mord
May 20, 2005, 05:42 PM
I think auxplage is confusing FSB with "effective FSB". Effective FSB is the one that is shown in benchmarks. Of course, higher FSB can only do so much (250FSB x 2 on Athlon XP vs. 250FSB x 4 on A64


it's 200 x2 on the athlon XP and it was only the latter bartons that supported it, it was 166 x2 133 x2 and 100 x2 before it.

and the A64 is 200MHz on a 5x LTD multiplier on 939 and 4x on 754 (3x on a few early nvidia boards)

Mav451
May 20, 2005, 09:36 PM
it's 200 x2 on the athlon XP and it was only the latter bartons that supported it, it was 166 x2 133 x2 and 100 x2 before it.

and the A64 is 200MHz on a 5x LTD multiplier on 939 and 4x on 754 (3x on a few early nvidia boards)

Yeah, sorry about that. Had a brain fart while I was imagining what kind of overclocks I could be getting if I gave my Ch-5 more voltage. I guess we call that a Freudian?