PowerPC 970 up to 2.5GHz?

[MacRumors]

This IBM press release (http://www-5.ibm.com/de/pressroom/cebit2003/en/highlights/powerpcblade.html) on their new PowerPC Blade servers reveals details on upcoming PowerPC 970 processors.


The new IBM PowerPC 970 is the heart of the PowerPC Blade. It is based on the 64-Bit Power 4 architecture which is also used in the processors of the IBM eServer pSeries. The 64-bit microprozessor

Offers full symmetrical multi-processing
Has a high reliability (with parity L1, ECC L2 and parity checked system bus)
Is manufactured in the latest 0,13 micrometer Copper/SOI CMOS technology
Runs at frequences ranging from 1.8 GHz - 2.5 Ghz

Therefore the IBMPowerPC 970 is the fastest PowerPC so far.



The IBM PowerPC 970 was originally detailed (http://www.macrumors.com/pages/2002/11/20021109083345.shtml) to start at a speed of 1.4 to 1.8GHz. No timeframe is provided for the delivery of these speeds (1.8GHz-2.5GHz)

[BWhaler]

Boy, would that be some good news. It would remove a big perception gap out there, and I bet with Apple's engineering of the new 64 bit motherboard and OS, the effective speed would make it the fastest computer on the market.

[Nipsy]

I'll have you all know that when the 970 was announced, I believed it would debut at speeds higher than those announced.

Let's hope we Apple folk get some trickle down, quick.

[Actuary]

The good thing about another vendor using the same processor as Apple is that if Apple doesn't use the best available, we all will know it (unless the other vendor holds back as well).
ie. IBM G3

[NicoMan]

If that CPU is anything it's hyped up to be, AND if it is running at those frequencies, wow that is pretty good news for the Apple community. So so sweet.

Last question: when are we gonna see those babies in a mac???

NicoMan

[gaomay]

If I recall correctly, IBM at MPF 2002 said that the clockspeed announced there (1.8Ghz) was a conservative estimate.

Seems it was very conservative.

Bring on a 2Ghz 17" Powerbook!

[MrMacMan]

2.5 GHZ nice.

Okay IBM you have the hype around the chip. Time to realease it. :D
Giving all the specs of a chip is very un-PowerPC like.

I like to here the news like that but can someone please give me the chip already? :rolleyes:

If the 1.8 chip is fall(2003)-2004. The 2.5 chips well be (...?) well into 2004-and 2005.

Ech... :(

This chip NEEDS to be put out now.

Edit: it seems I was unclear about the 1.8 chip, sorry.

[jettredmont]

Originally posted by MrMacman
If the 1.8 chip is fall-2004. The 2.5 chips well be (...?) well into 2004-and 2005.

Ech... :(

This chip NEEDS to be put out now.

Ummm ... the 1.8 chip is supposed to be fall-2003 to Q1 2004, and that date estimate seems to be moving more towards Fall 2003.

If IBM is claiming the 970 in their soon-to-debut PowerPC Blade servers runs up to 2.5GHz, then you can expect that at most there will be a 3-month delay between general introduction and 2.5GHz parts being available. IBM tends to not play games with press releases.

Therefore, assuming that this press release was not a typo, we can expect 2.5GHz parts no later than this time next year (March 2004).

If the 1.8GHz part gets the estimated SPEC scores, the 2.5GHz part should get approximately (although not exactly) 40% higher scores. This is some very good news. If nothing else, IBM's blade servers should be selling like hotcakes!

[MacsRgr8]

OMG.....

Can anybody imagine what will happen if Apple weren't to use these procs????? The hype (we all made) around the PPC 970 is so huge....

[Inunyan]

Very cool.

I would finally be able to put my goodol' DEC XP1000 to rest, and do serious number-crunching on my desktop OSX!

Wish they come up with cooler (underclocked?) version. I would put one in my Cube.

[Mr. MacPhisto]

I hadn't said anything, but had been told that the initial clock speeds would be higher. I'm glad to see it was true here. When IBM switches over to .09 micron we should se a greater increase.

The thing to wonder is how much do these puppies draw? The 1.2 GHZ chip was at 19 watts. Wonder if IBM drove down some of the others or not. It'd certainly be nice to have a 1.8 or even 2 GHZ chip in a Powerbook. That would throw a definite advantage in speed to the Mac.

Actually, 2.5 GHZ may favor the Mac with the architecture of the 970. The only thing superior is the multi-threading found in Intel's high end P4s. Interesting.

[groov']

Further technical highlights of the PowerPC 970:

· Onchip 512 KB L2 Cache
· Altivec ™ Vector/SIMD unit
· 6,4 GB/s I/O system bus throughput

[Mr. Anderson]

Nice, now deliver them Apple!

Damn, that would be a nice machine in dual or possibly more cpu cores. :D

That's the order of magnitude processing power that I've waiting for.

D

[G4scott]

YEAH!!! ***** YEAH!!!!

This is what we needed. A miracle! The 2.5GHz PPC970, if used by Apple (which I am pretty sure they will) will show up any, and I mean any pentium system. Go Apple!!!

This is gonna rock...

But I'm going ot have to change my user name...

[ibjoshua]

Bring em on!!

i_b_joshua

[Vanquish]

Apple can innovate all they like (an innovate they do - like no other company I've seen!) but this is what they really need - SPEEED!!!!

Apple have everthing else in place. When these processors arrive, they will be the industry leader on ALL counts!

I can't wait! :D


-
After reading MacRumors for almost 2 years now, I thought I'd finally register! So, here's a big hello to all! :)

[littlerich]

I am looking to get a new PowerBook in about 2 - 3 months, obviously these processors or any fast processors will not be in the powerbooks but would you say it is better to wait a while. Until something better/cheaper comes out? Like I said it will be 3 months from now so??

[Dont Hurt Me]

Originally posted by Mr. MacPhisto
I hadn't said anything, but had been told that the initial clock speeds would be higher. I'm glad to see it was true here. When IBM switches over to .09 micron we should se a greater increase.

The thing to wonder is how much do these puppies draw? The 1.2 GHZ chip was at 19 watts. Wonder if IBM drove down some of the others or not. It'd certainly be nice to have a 1.8 or even 2 GHZ chip in a Powerbook. That would throw a definite advantage in speed to the Mac.

Actually, 2.5 GHZ may favor the Mac with the architecture of the 970. The only thing superior is the multi-threading found in Intel's high end P4s. Interesting. This is great news, I think this chip has apple written all over it. the 1.8 ghz was pulling 40 watts if im not mistaken. Now the question is how soon for apple to use these and it what products? I would love a shocker this summer like apple saying its moved all the new architecture machines to a 970 in one giant swoop. 1.2 in the the imac/powerbooks and 1.4- 2.0 in the powermacs. Any chance of this becoming a reality this year?

[-hh]

Originally posted by MacsRgr8
OMG.....

Can anybody imagine what will happen if Apple weren't to use these procs????? The hype (we all made) around the PPC 970 is so huge....


That's what I'm starting to think about. I hate to be a pessimist or a curmudgen, but the 970 rumors have been floating around on the rumors beach like some dead seaweed, and its starting to have a distinct odor of decay.

Yeah, I know that things take time to develop, but supposedly, the IBM project had been hush-hush working on things all through 2002 (at least). I'm beginning to be very much inclined to say "enough with the R&D, where's the product roll-out?"

Say what we want about product rolloouts that are logical with IBM's new fab plant and so forth, but the sad reality is that Apple probably needs to announce the 970 no later than July, or else they'll miss the time window in which the Educational Market finalizes their 03/04 school year purchase plans, etc. Otherwise, they miss the budget window of this customer.

FWIW, I just went through this last fall with a piece of test equipment. We knew the manufacturer had a new/better model coming "any day now", but our money was going to go belly-up, so we had to make a decision...


-hh

[MisterMe]

Originally posted by G4scott
YEAH!!! ***** YEAH!!!!

This is what we needed. A miracle! The 2.5GHz PPC970, if used by Apple (which I am pretty sure they will) will show up any, and I mean any pentium system. Go Apple!!!

This is gonna rock...

But I'm going ot have to change my user name... There is no miracle here. It is simply quality engineering by a reputable company.

[Dont Hurt Me]

I wonder if the architecture changes made in the powerbooks/xserve/powerrmac and 1 gig imac were prepping those machines for this cpu? any comments? Also here is a chip that is allready scaling up and will so even more with smaller process. Yet you look at motorola and what they have done and you have to wonder?Maybe they really never did care for the chip they were making for Mac.

[cr2sh]

It's intersting to note... that these CPU's will be shipping around the same time the 17" Powerbooks do... Now if we only knew when that was... :rolleyes:

2.5GHz would be amazing, but with the full symmetrical multi-processing you have to wonder if they'll be doing dual systems.
drool....

[Stormcrow]

Hey guys. I'm new to the forum, but I have been lurking for a while due to my interest in macs. I have considered getting a mac for quite a while, but the cpu/speed issue has held me back. IMO, macs need these chips now! If macs get a 2.5 ghz chip in 2004, so what? Macs will be in the same boat as they are now. The P4s will probably be clocking at 4-4.5 ghz. 2.5 ghz with a development path that is at least at parity with AMD and Intel would be good NOW, not next year.

Still, I like macs. I think they are pretty cool. I want macs to do well. I want there to be at least some competition for M$. We need it. I would love to see Macs trounce any PC they go up against in terms of performance no matter who does the testing. For macs to compete, we need action, not anticipation and waiting around.

I dont want to be a stick in the mud with this post, but I just really would like to see these chips in Macs NOW.

[MacManiac1224]

If IBM were to release a 2.0 Ghz chip in July, then would Apple actually use it, or move up slowly to 2.0Ghz? I am not so sure Apple would use the highest speed chips?

What do you guys think?

[cubist]

Welcome Stormcrow! The 970 is 64-bit, so it should have no trouble outperforming a P4 running at double the clockspeed. The P4 is not as fast as the MHz numbers would indicate; they just have a really, really long prefetch queue. This has already been discussed at length in many other threads on this site.

It is still an open question whether Apple will use the chip at all. To say the chip "has Apple written all over it" is a bit hubristic. We are hopeful, however.

[starflyer]

Originally posted by Stormcrow
The P4s will probably be clocking at 4-4.5 ghz. 2.5 ghz with a development path that is at least at parity with AMD and Intel would be good NOW, not next year.

4-4.5ghz 32-bit chips...laugh laugh laugh...would any of the companies have any 64-bit chips to compete?

[blueBomber]

ah, but your comparing a 32-bit x86 chip to a 64-bit PPC chip... apples and oranges (no pun intended)

2.5 Power4 will more than likely cripple the fastest of the Pentium chips, even if Intel makes it to 4ghz next year. Since your relativly new, I'm guessing your still seeing the "Megahertz Myth".

[rjsweb]

Hey guys. I'm new to the forum, but I have been lurking for a while due to my interest in macs. I have considered getting a mac for quite a while, but the cpu/speed issue has held me back. IMO, macs need these chips now! If macs get a 2.5 ghz chip in 2004, so what? Macs will be in the same boat as they are now. The P4s will probably be clocking at 4-4.5 ghz. 2.5 ghz with a development path that is at least at parity with AMD and Intel would be good NOW, not next year.

It is true that by the time IBM has the 970 ready, Intel will be shipping in the 4-5 GHz range. However, those chips from Intel will probably remain at 32 bit (at least in the desktops/notebooks) until AMD puts out their 64 bit chip and Intel faces market pressure. It is probable that if Apple uses the 970 like so many people think they will, Apple will be able to hype the 64 bit even if the clock speed is slower than what Intel is shipping. Once Intel moves to 64 bit, then Apple will have marketing problems, unless IBM has a trick up their sleeve when ramping up the clock speed of the 970.

Just my thoughts...

[usersince86]

Originally posted by Stormcrow
If macs get a 2.5 ghz chip in 2004, so what? Macs will be in the same boat as they are now. The P4s will probably be clocking at 4-4.5 ghz. 2.5 ghz with a development path that is at least at parity with AMD and Intel would be good NOW, not next year.

EXACTLY.

I've being using Macs since 1986, and we've always been behind in the MHz war.

You can't compare future, undelivered chips to ones that are currently on the market and say, "See, we're just as fast!"

We're not. Don't give me all the "MHz myth" stuff; you can spin it however you want, and even though the gap isn't as big as it looks, it's still a gap. Period.

If it wasn't for Apple's terrific engineering and true innovation, the ship would have sunk a long time ago. It's just frustrating to know the best OS and the best software don't run on the fastest chips, and it doesn't look like that will change anytime in the forseeable future.

Maybe the OSX on Intel/AMD project is for real, and not just another "Megahertz Myth". Maybe IBM will catch-up or pass Intel/AMD. Here's hoping!

[GPTurismo]

My dream of a quad 2 gigahertz processor 17 inch powerbook is one step closer :|

[DakotaGuy]

Hey I am just waiting for Steve Jobs to do his Photoshop test with a Dual 2.5Ghz PowerMac 970!

[Frobozz]

Originally posted by Macrumors

The IBM PowerPC 970 was originally detailed to start at a speed of 1.4 to 1.8GHz. No timeframe is provided for the delivery of these speeds (1.8GHz-2.5GHz)



Well, this is VERY promising news. I have to say, however, that I am not surpsrised that Apple would debut at higher frequencies than IBM _orginally_ announced. If we look at what is currently shipping, Motoroloa does not mention it's chips even being capable of such frequencies or bus speeds.

Could this mean we will see a low end 1.8, a mid-range 2.2, and a top end 2.5 in the July/Aug/Sept time frame? Will they be duals?

[Dont Hurt Me]

For usersince86 i feel your pain. But if and the word IF is a bif IF apple announces 970 this summer at 2 gigs they will have a machine to smoke or at least match a p4 running in the low 3-3.6 gigs. Also the new process would scale the 970 up to perhaps 3 gigs or better if they are allready talking of 2.5 from the current process.This would mean the best software on the fastest machine. the current g4 just cant keep up and its time for retirement.

[AidenShaw]

Originally posted by blueBomber
ah, but your comparing a 32-bit x86 chip to a 64-bit PPC chip... apples and oranges (no pun intended)

2.5 Power4 will more than likely cripple the fastest of the Pentium chips, even if Intel makes it to 4ghz next year. Since your relativly new, I'm guessing your still seeing the "Megahertz Myth".


Since OS X is 32-bit, and all your applications are 32-bit - what difference would it make that the 970 has a 64-bit mode?

The budding "64-bit Myth" is even sillier than the MHz Myth. All chips today have wide internal busses (both the P4 and the G4 use up to 256-bit wide internal datapaths), they all support 64-bit double float natively, etc, etc.

Unless your individual programs need more than 2GB of RAM each, it's really unlikely that you'll see any benefit to 64-bit. Sorry 'bout that....

(The 970 will be a big improvement (if Apple uses it) because it will be so much faster than the G4 -- but it's faster because it's faster, not because it has 64-bit addressing.)

[Zband]

Originally posted by MacsRgr8
OMG.....

Can anybody imagine what will happen if Apple weren't to use these procs????? The hype (we all made) around the PPC 970 is so huge....

I totally agree with your above statement - there just is no proof about the 970 actually ending up in a PM. But if I were apple at this point, I would know that my customer base was expecting this chip in the newer machines - so I would have made a statement denying the 970's use in PM- just to drive current PM sales [sic]. SO I hope that I am wrong.

One more thing, please remember people - 64 bit DOES NOT make a processor twice as fast as a 32 bit processor! This has got to be understood.

[ryan]

Originally posted by Dont Hurt Me
I wonder if the architecture changes made in the powerbooks/xserve/powerrmac and 1 gig imac were prepping those machines for this cpu? any comments? Also here is a chip that is allready scaling up and will so even more with smaller process. Yet you look at motorola and what they have done and you have to wonder?Maybe they really never did care for the chip they were making for Mac.
In all fairness to Motorola (even though they don't really deserve it) they are shipping G4s, unlike IBM with their 970.

[AidenShaw]

Originally posted by Abercrombieboy
Hey I am just waiting for Steve Jobs to do his Photoshop test with a Dual 2.5Ghz PowerMac 970!


Fine, as long as it's against a dual 3.8 GHz Xeon with HyperThreading!

Enough with the "our dual matches your single" nonsense - compare duals to duals (or quads to quads, or octos to octos, or hexas to hexas).

Ooops, sorry - Apple doesn't make any 16-way systems to test against 16-way Intel boxes! Guess the testing has to stop at duals.

[jethroted]

Well this is very good news. It was begining to look like the 970 was gonna fall flat on it's face as far as Ghz are concerned. Now it looks way more promising, and way more of a x86 crusher! I'm sure they will be able to push it alot farther than that as time goes by.

[Cappy]

Originally posted by AidenShaw
Fine, as long as it's against a dual 3.8 GHz Xeon with HyperThreading!

Enough with the "our dual matches your single" nonsense - compare duals to duals (or quads to quads, or octos to octos, or hexas to hexas).

Ooops, sorry - Apple doesn't make any 16-way systems to test against 16-way Intel boxes! Guess the testing has to stop at duals.

Technically you have a point on the comparisons but from a market standpoint you're not as close. Remember that everything comes down to markets and sales. Price out a dual xeon system that might be used against a dual mac and you might find that for the money the Mac might hold up pretty well. There's advantages to both platforms but what Apple has done in their comparisons are tackle the more common P4 which cannot be placed in a dual config. Technically it's an apples to oranges comparison but for their market there's nothing wrong with comparing like so.

[ultrafiel]

Originally posted by usersince86
EXACTLY.

I've being using Macs since 1986, and we've always been behind in the MHz war...

Obviously you have a short memory. Remember the snail ads during the mid-90's mocking the Pentium 3? Really it hasn't been until the last 5 years or so when Intel/AMD ramped up their Mhz speeds and left Macs in the dust (you don't need to explain the Mhz myth to me, I understand it).

Moving on though, I agree that the 970 rumors have been hyped through the roof. Yes, I want one, but if something doesn't turn out right, we we all be disappointed because of rumors. Case in point, people already are dissatisfied on the clock speeds announced, or aren't happy with only 1 chip on a non-existent machine. I want dual, quad, etc. First to add more chips it would be more expensive, and if you're like me, every dollar counts. I don't care if it is a single cpu, I'm on a 450 Mhz G4, and if a 970 came out (any 970) I would be overwhelmed by the speed that I wouldn't care about anything else. <sarcasm>Forget Photoshop tests, just think how fast my email will be.</sarcasm>

[Cappy]

I hope people remember that Apple has not officially announced that they're going with the 970. Many seem to have forgotten already how lots of folks including the media thought that Apple was going to select the BeOS for their next OS. You've probably figured out by now that they went with NeXT. Apple could still do the same with the cpu and choose something else. I've yet to see a clear win in going to the 970 over an Intel or AMD cpu. All of them have clear advantages over the other.

BTW, a conspiracy theory is forming that despite the fact that many have believed that MS is only interested in VirtualPC for server consolidation, it is being thought that MS can adapt and integrate the code in a similar fashion to what Apple did and use that layer for backward compatibility in a new 64bit version of their OS. This could potentially be a greater advantage for MS in dealing with the different directions Intel and AMD are going. Essentially leveling the playing field which will likely keep Intel in control of the market. Rumor is that Intel might even be pumping money into MS for this to happen.

In other words don't count anyone out and Apple in just because on the surface Intel looks like they're shooting themselves in the foot.

[Gus]

Originally posted by usersince86
EXACTLY.

I've being using Macs since 1986, and we've always been behind in the MHz war.


While, I get your point, this is not true. Before the "Year of 500Mhz", we actually still had the speed advantage over Intel and AMD. That year was and is the reason we are still behind.

AidenShaw, you are starting to sound belligerent. Are you here to support Macs/educate posters/troll or what?

Regards,
Gus

[drastik]

I think these sound like perfectly fine speeds.

Wha you have to ask yourself: does my Mac do what I want it too as quickly as I need it done? The answer, for most people is yes. I want everyone to own a Mac so they can experience it, but its just fantasy. If you have a render farm, you're going to use wintell boxes clustered up. Everything has its proper use, and in the average to prosumer desktop market, a Mac will work better and more reliable no matter what the speed. Everyone complains about the pices, but no one considers the specs. Sure, you can get a $600 PC, but it has SDRAM not DDr, no firewire, CD-ROM instead of superdrive, etc, etc. It could go to 10ghz, the $600 machine is still an email station for all the good it dos you for pro work. Not to mention the clunky applications that drag the processor speeds dawn anyway.

Some of you out there have this wierd thing about being faster, bigger, better all the time. I don't know if you're 14 or you just have a complex, but please relax. If your mac does what you want and allows you to be more efficient, who cares how fast the clock is?

And the most important part: enjoying use. If you like using a Mac, (very important if you actually have to use one all day everyday, so that you and your ffamily can eat. If the Mac makes all of that easier, and dare I say it, fun, who cares about speed?

[AidenShaw]

Originally posted by Gus
AidenShaw, you are starting to sound belligerent. Are you here to support Macs/educate posters/troll or what?

Gus,

Sorry to come off as "belligerent" - I try to keep things brief and concise, sometimes that's too close to "curt", then "belligerent".

I'm here to learn about Macs, and to share my experience with high performance computing, storage, networking and 64-bit computing.

I'll usually just read, but when I see things that IMO are wrong or misleading I'll make an entry. In particular, since I have a lot of experience in x86 and 64-bit systems I'll often respond there.

The comments that you can't compare a 32-bit x86 with the rumours of a 64-bit 970 are such triggers. Maybe these folks didn't see the story that Pixar was abandoning the 64-bit SPARC systems in favor of 32-bit Xeons.... Yes, they can be compared! And yes, sometimes the 32-bit system is better and faster!

[Frobozz]

Originally posted by Zband
I totally agree with your above statement - there just is no proof about the 970 actually ending up in a PM.

True. But let's look at the VERY CONVINCING arguements for it:

1) It has AltiVec. IBM decided not to do AltiVec before because it was Motorola's baby. AltiVec won't do much in a server environment. The 970 has AltiVec and it marks a big step for IBM if it were not being prodded by Apple.

2) Steve Jobs mentioned late last year that 2003 would be the biggest year ever for Apple. In that same statement he said Apple's options were open for new processors in 2003. He also said there would NOT be a move to x86. What does that leave? Either Motorola or IBM producing the "G5" or whatever you want to call it... which brings me to my next point...

3) Motorloa has no plans for a G5 desktop chip. IBM seems to with the 970. Why make a desktop version of a multi-core server chip without guarateed sales? You could make the arguement for their low end UNIX servers and Blade servers (see press release), but that is NO WHERE near the volume needed to recoup R&D. Apple will be.

[Dont Hurt Me]

Originally posted by ryan
In all fairness to Motorola (even though they don't really deserve it) they are shipping G4s, unlike IBM with their 970. Thats true but the 970 is a all new cpu so give IBM a break. Also they have been shipping g3's to mac for years and have probably sold more g3's to apple then motorola sold g4's.

[kumichou]

Originally posted by cr2sh
It's intersting to note... that these CPU's will be shipping around the same time the 17" Powerbooks do... Now if we only knew when that was... :rolleyes:

I just ordered my 17" Powerbook a few days ago, and according to my order it'll be shipped on or before 4/8/03. Here's to hoping! :D I know if I were one of the folks that ordered during MacWorld, I'd be pretty bent about the lag between announcement and shipping product.

[iAlan]

Firstly, I luv Apple. I always have. I have been using Macs since 1990. Back then they were so much better. However, Windows has closed the gap since. You cannot just say that MS 'stole' their OS from Apple, I think they did, but they did the right thing from the start and allowed other companies to install it on their machines. Great leaders have also said that if you see a great idea, steal it. MS did. Apple screwed up with their licensing to Power Computing, et al. Too little too late. Apple will however always innovate, pushing other companies to do the same - even if this means Apples's ideas are copied or stolen.

Now to the 970. Apple could also screw the pooch with this if they do not clearly let the world know their development pipeline. Apple too often keep everything quiet until the very last minute. This usually means we all hype up what we think is going to happen, and then feel let down when we don't get what we want. Well Apple, let us know you are going to use the new IBM chips. Fine, don't tell us which speed, single or duel, new product design, etc...but keeping quiet is only making things worse. To my knowledge, no one waits and speculates on any other computer company like they do with Apple. It might make for a good MW Wherever Keynote by King Steve, but most pro-users need to budget ahead of time and cannot budget based on rumors or speculation.

Sory for the long post, but wouldn't it be funny if Apple were in fact going to go with Intel or AMD and IBM is just sticking it to us all - if so, no wonder Apple is being quiet.

[zulgand04]

Originally posted by rjsweb
It is true that by the time IBM has the 970 ready, Intel will be shipping in the 4-5 GHz range. However, those chips from Intel will probably remain at 32 bit (at least in the desktops/notebooks) until AMD puts out their 64 bit chip and Intel faces market pressure. It is probable that if Apple uses the 970 like so many people think they will, Apple will be able to hype the 64 bit even if the clock speed is slower than what Intel is shipping. Once Intel moves to 64 bit, then Apple will have marketing problems, unless IBM has a trick up their sleeve when ramping up the clock speed of the 970.

Just my thoughts...

a artical on cnet.com states intell will not havce a 64bit processor for the consummer untill around 2010, due to they suposivly don't think the market needs it.


neal

[arn]

Originally posted by Cappy
Apple could still do the same with the cpu and choose something else. I've yet to see a clear win in going to the 970 over an Intel or AMD cpu. All of them have clear advantages over the other.

No "clear win"?

How about 100% application compatability vs 0% ?

arn

[Hawthorne]

Now to the 970. Apple could also screw the pooch with this if they do not clearly let the world know their development pipeline. Apple too often keep everything quiet until the very last minute. This usually means we all hype up what we think is going to happen, and then feel let down when we don't get what we want.
I think by playing their cards close to their chest, Apple is trying to avoid the opposite effect, known as the Osbourne Effect (no, not the drug-addled rock and roll star:) ). Way back in the early 80's, the Osbourne was the first successful portable computer (gad, did I really put up with a Z-80 chip back then?). The successor to it was supposed to be the hottest thing ever, a computer that would blow the doors of anything else. And we waited for it, and waited, not buying more of the current machine because the new one was supposed to be so good. Which killed the cash flow of Osbourne, and the company went out of business.

[howard]

i really hope apple doesn't start these off at a lower clockspeed than they can get them...but i don't think they will... partly cause i don't think that they will use them as dual. so that they'll release them as 1.8 or 2 instead of 1.4 but only as one processor. then its a big jump..but not as big. i really wish they would just release them as fast as they can with dual processor...that would seem the logical thing to do. to put out the best product that you can. but whatever

[iShater]

Originally posted by howard
i really hope apple doesn't start these off at a lower clockspeed than they can get them...but i don't think they will... partly cause i don't think that they will use them as dual. so that they'll release them as 1.8 or 2 instead of 1.4 but only as one processor. then its a big jump..but not as big. i really wish they would just release them as fast as they can with dual processor...that would seem the logical thing to do. to put out the best product that you can. but whatever

What would be better is that they get to a true Built-To-Order system. The Fast, Faster, Fastest crap sux. They should use those 3 preconfigs to make it easier to build, but a true customize option would rock.

Think about it. If they do not cripple any of the PowerMacs, they can offer multiple chips speeds, RAM, HDs, etc, etc.

I would love to be able to get a top of the line CPU without having to pay for a SuperDrive if I don't need it. That is a true BTO system. :rolleyes: Before anybody starts flaming, I am using this as an example, so don't tell me "but u can get top of the line without the SD!"

[moon.drop]

Originally posted by dukestreet
Nice, now deliver them Apple!

Damn, that would be a nice machine in dual or possibly more cpu cores. :D

That's the order of magnitude processing power that I've waiting for.


It's not completely up to Apple here. Remember, IBM is developing these processors. They need to hurry up and blaze past Motorola, get these processors to Apple at least in prototype form, and then Apple can develop systems which they can then ship to us!

[ffakr]

Originally posted by dukestreet
Nice, now deliver them Apple!

Damn, that would be a nice machine in dual or possibly more cpu cores. :D

That's the order of magnitude processing power that I've waiting for.

D

An order of magnitude would indicate that this processor would be 10x times faster then existing G4s.
... riiight.

[Dont Hurt Me]

Ishater is right a build to order computer, a lot of the stuff i dont need either. Let us order the cpu,video card,hard drive etc to suit us. I dont think we will see this though. sort of like the imac philosophy here it is take it or leave it. If the customer is king then how about acknowledge that. I love apple but sometimes feel like they are king and we the subjects.

[zoetropeuk]

Everybody is forgeting one thing. The majority of mainstream applications for the Mac are poorly ported PC versions. The main reason the G4 continually gets hammered in benchmarks is not due to the processor but lazy programmers. Show me one single application that has been fully optimised for the G4.
Not even OSX had been, otherwise it wouldn't run on the G3. If Adobe were to realise a version of photoshop that was fully optimised for Altivec and dual processors then not even a 4.5Ghz P4 would see the light of day.
We are yet to see a program make the best use of the G4 so what makes you think that by moving across to the 970 things will be any different.
Just take a look at the first batch of PS2 games that were released and compare them to say Tony Hawk 4. The hardware is the same, it's just that the programmers know how to push the CPU and GPU's to the limit.
Apple needs to forget this stupid G3, G4 set-up and just move everything across to ONE platform. Apple also needs a department that will optimise all major software releases for any new CPU and system that it chooses the implement.
Otherwise we are going to see the same old crappy slow applications that we do now.

[eric_n_dfw]

Originally posted by Dont Hurt Me
Ishater is right a build to order computer, a lot of the stuff i dont need either. Let us order the cpu,video card,hard drive etc to suit us. I dont think we will see this though. sort of like the imac philosophy here it is take it or leave it. If the customer is king then how about acknowledge that. I love apple but sometimes feel like they are king and we the subjects.
Apple used to do this - when I bought my B&W G3 400, I seem to remember being able customize any part of it. I think they stopped being that flexible when the SuperDrive PowerMacs came out.

[macmax]

I only know i want to see an add that starts with the word:


BEHOLD!!!

And then i hope the aim for the heart and head, hehhehehhehehheheheh:D

go mac!!!!!!

[iShater]

Originally posted by eric_n_dfw
Apple used to do this - when I bought my B&W G3 400, I seem to remember being able customize any part of it. I think they stopped being that flexible when the SuperDrive PowerMacs came out.

You might be right, at one point they had a really good BTO system, then they made it a Build-To-Margin system. How to force the customer to choose the more expensive system because of one friggin thing they need. I mean it is great for Apple's bottom line, but it is not good for building market share and being able to compare to other makers.

[eric_n_dfw]

Originally posted by drastik
Sure, you can get a $600 PC, but it has SDRAM not DDr, no firewire, CD-ROM instead of superdrive, etc, etc.
Just for acuracy, DDR is SDRAM. "SD" stands for Syncronous Dynamic, not Single Data rate as many seem to think.

Your point is that the cheapo PC's come with non-DDR ram where Mac's do. Problem with that arguement is that untill we get processors with DDR capable front end's it really doesn't help all that much. (Sure the system controller can make some use of it by spreading other DMA requests arround from the other busses, but computationally, it means squat.)

I've said it hear a 100 times, give the PowerPC a DDR FSB and watch it screem - ESPECIALLY on AltiVec enhanced applications. AltiVec is memory starved at speeds well below 1Ghz.

[Dont Hurt Me]

Originally posted by zoetropeuk
Everybody is forgeting one thing. The majority of mainstream applications for the Mac are poorly ported PC versions. The main reason the G4 continually gets hammered in benchmarks is not due to the processor but lazy programmers. Show me one single application that has been fully optimised for the G4.
Not even OSX had been, otherwise it wouldn't run on the G3. If Adobe were to realise a version of photoshop that was fully optimised for Altivec and dual processors then not even a 4.5Ghz P4 would see the light of day.
We are yet to see a program make the best use of the G4 so what makes you think that by moving across to the 970 things will be any different.
Just take a look at the first batch of PS2 games that were released and compare them to say Tony Hawk 4. The hardware is the same, it's just that the programmers know how to push the CPU and GPU's to the limit.
Apple needs to forget this stupid G3, G4 set-up and just move everything across to ONE platform. Apple also needs a department that will optimise all major software releases for any new CPU and system that it chooses the implement.
Otherwise we are going to see the same old crappy slow applications that we do now. I agree with a lot of what you have said. they do need to go with 1 cpu across the whole line then they could optimize for that 1 unit and then get software makers do the same. Right now we have g3,g4 altivec,os9,osx what they need is 970 and optimized X and give software makers help in utilizing that to the fullest. Bravo post!

[yzedf]

Just for those among you that don't know what a Blade is...

http://www.pc.ibm.com/us/eserver/xseries/bladecenter_family.html

To Arn:

Regarding your "clear win" statement, the BSD base for OS X makes the OS, as well as the programs that run on the OS very portable. x86, x86-64, PPC, are just as easy to do, for the most part. PPC kinda sucks due to Altivec, same as P4 with HT.

2 hardware generations from now we could very well laugh and say "remember when were stuck with PPC?! Thank goodness that junk is gone!"

[MacBandit]

Originally posted by zoetropeuk
Everybody is forgeting one thing. The majority of mainstream applications for the Mac are poorly ported PC versions. The main reason the G4 continually gets hammered in benchmarks is not due to the processor but lazy programmers. Show me one single application that has been fully optimised for the G4.
Not even OSX had been, otherwise it wouldn't run on the G3. If Adobe were to realise a version of photoshop that was fully optimised for Altivec and dual processors then not even a 4.5Ghz P4 would see the light of day.
We are yet to see a program make the best use of the G4 so what makes you think that by moving across to the 970 things will be any different.
Just take a look at the first batch of PS2 games that were released and compare them to say Tony Hawk 4. The hardware is the same, it's just that the programmers know how to push the CPU and GPU's to the limit.
Apple needs to forget this stupid G3, G4 set-up and just move everything across to ONE platform. Apple also needs a department that will optimise all major software releases for any new CPU and system that it chooses the implement.
Otherwise we are going to see the same old crappy slow applications that we do now.


The reason they can run the G3 and the G4 at the same time is that the G4 is basically a G3 with an integer unit. The base architecture is the same. Even if a program was optimized to work to it's fullest on a G4 it would most likely still work on a G3 and vice versa. The real slowdown right now is that the system and all the software is not only OS9 friendly but still has code in there for 604 and older processors.

I agree on one point for sure though. Apple needs to put out a better compiler with more automation for Altivec optimization.

[zoetropeuk]

Even if a program was optimized to work to it's fullest on a G4 it would most likely still work on a G3 and vice versa.

This is not true. If you compile a small app to add a set of integers together that only uses alitvec then it WILL NOT WORK ON A G3, at all.
And we don't want to see pansy automated compilers, we want programmers to do what their paid to do and program. We want hand optimised altivec applications.

I'd pay 3 or 4 times the price for apps that are 4-8 times faster.

[os4]

Originally posted by Dont Hurt Me
Ishater is right a build to order computer, a lot of the stuff i dont need either. Let us order the cpu,video card,hard drive etc to suit us. I dont think we will see this though. sort of like the imac philosophy here it is take it or leave it. If the customer is king then how about acknowledge that. I love apple but sometimes feel like they are king and we the subjects.

There are two concepts at work with Apple's ordering process:

1) the "3 Choices rule" - people are most comfortable with three choices.

2) Apple needs to control its operations. The best way to do that is to simplify as much as possible. Each additional choice multiplies the logistical issues involved in putting the machines together. It would be nice if Apple added a "Power User" section to the Apple Store to allow for more choice; however, I am sure Apple has studied this issue and decided that the incremental revenue is less than the incremental costs.

[nuckinfutz]

Photoshop BTW is not a PC port. It was born on the Mac.

Altivec is going to take some time for developers to master like any tool.

There are plenty of apps that show Altivecs power.

Photoshop
iDVD
Altiverb
Blast

Just wait until the Altivec isn't starved by slow bus throughput. You're seeing the crippling effects of a bus that cannot let Altivec shine.

I see a nice 2004 with the fastest G4's going into iBooks and iMacs and PPC 970's in the Pro Series. Just like it's supposed to be.

[os4]

Originally posted by Dont Hurt Me
Ishater is right a build to order computer, a lot of the stuff i dont need either. Let us order the cpu,video card,hard drive etc to suit us. I dont think we will see this though. sort of like the imac philosophy here it is take it or leave it. If the customer is king then how about acknowledge that. I love apple but sometimes feel like they are king and we the subjects.

There are two concepts at work with Apple's ordering process:

1) the "3 Choices rule" - people are most comfortable with three choices.

2) Apple needs to control its operations. The best way to do that is to simplify as much as possible. Each additional choice multiplies the logistical issues involved in putting the machines together. It would be nice if Apple added a "Power User" section to the Apple Store to allow for more choice; however, I am sure Apple has studied this issue and decided that the incremental revenue is less than the incremental costs.

[nuckinfutz]

This is not true. If you compile a small app to add a set of integers together that only uses alitvec then it WILL NOT WORK ON A G3, at all.

That's generally a non factor. OSX makes it easy to include both Altivec and non altivec binaries in a Package. The end user never needs to know.

[law guy]

Originally posted by Frobozz
True. But let's look at the VERY CONVINCING arguements for it:

3) Motorloa has no plans for a G5 desktop chip. IBM seems to with the 970. Why make a desktop version of a multi-core server chip without guarateed sales? You could make the arguement for their low end UNIX servers and Blade servers (see press release), but that is NO WHERE near the volume needed to recoup R&D. Apple will be.

Frobozz, the PC Magazine feb. 4 issue with the processor roadmap stated that Motorola would have a new processor in Q3 aimed at being the G4's successor and competing with the 970 for that role. Has Motorola cancelled their (what must be very far along) development?

[Raiden]

My question is, if apple annownces the 970 at MWNY, saying the new processor will be availible in 2-3 months, what will motorola do? If I were motorola, I would be really pissed off that one of my biggest clients is jumping ship to a company that offers a better product than me. Therefore would motorola stop making G3s and G4s?

I guess im wondering is apple keeping their use of the 970 secret because they are afraid motorola will get angry? Does motorola know about apple and their supposed use of the 970?

sorry for all the questions...

[nuckinfutz]

Therefore would motorola stop making G3s and G4s?


No not at all. Apple has sourced both companies for G3's and mainly Moto for G4's.

Motorola designes the G3/G4 for embedded applications.

Cisco uses a G4 in a Router. There are more markets than just Apple for Moto G4/G4 processors.


The PPC 970 is squarely aimed at Desktops however. Apple will be able to use both for some time.

[law guy]

Originally posted by Raiden
My question is, if apple annownces the 970 at MWNY, saying the new processor will be availible in 2-3 months, what will motorola do? If I were motorola, I would be really pissed off that one of my biggest clients is jumping ship to a company that offers a better product than me. Therefore would motorola stop making G3s and G4s?

I guess im wondering is apple keeping their use of the 970 secret because they are afraid motorola will get angry? Does motorola know about apple and their supposed use of the 970?

sorry for all the questions...

From a deal point of view, one would think that Apple is talking with whoever it intends to buy their next gen chips from at this point - unless they're mocking up systems for both a Moto and an IBM G5 and still haven't made any decisions as to which is best. I imagine that Moto / IBM must have some inclination of what's going to be happening with orders three months from now.

[hacurio1]

Originally posted by arn
No "clear win"?

How about 100% application compatability vs 0% ?

arn

Couldn’t agree more Arn!!!

People here who question whether Apple will use the 970 or not aren’t very perceptive of what is going on right now! First, what options does Apple have? Is there any other PPC chip coming? NO!!! Will they go x86? NO!!!!! I know, “970’s in Macs” are just rumors but it makes absolute sense. Just look at the facts, Apple just went trough a big transition (OS9 to OSX.) Developers are barely finishing porting their software!!! Apple will not go x86!!! What other PPC option do they have? Motorola announced they will stop developing desktop processors, so I ask again, what other option does apple have? The 970 promises to be an excellent processor and it will definitely be as fast or faster that any P4 released by the end of the year. At 1.8 GHz the specint scores of the 970 are very close to the 3 GHz P4’s scores. At 2.5 GHz it will be as fast or faster and like many people said, it will be 64 bits. Contrary to what some uninformed people posted above, 64bit architecture is faster than a 32bit. Memory is only one advantage and it all depends on the OS and the design of the chip. In some instances, a 64bit processor can process two 32bit data paths a once and as mentioned above, it all depends on the OS and the chip. Since the core of OSX is 64 bit I doubt apple will have any trouble porting OSX to 64bit and optimizing it so that it takes advantage of the 64 bit ints and data paths. Since the 970 is fully backwards compatible with 32bits, I see it as the best option Apple has. IBM has even announced plans to develop the 980’s and Motorola hasn’t mentioned any plans for desktop chips.

Just my 2c

[shadowfax]

Originally posted by nuckinfutz
That's generally a non factor. OSX makes it easy to include both Altivec and non altivec binaries in a Package. The end user never needs to know.

but if you don't take advantage of that, your program won't be G3 compatible, right?

[beatle888]

Originally posted by Nipsy
I'll have you all know that when the 970 was announced, I believed it would debut at speeds higher than those announced.

Let's hope we Apple folk get some trickle down, quick.



oh wow, do you have inside information?

[pgwalsh]

RUMORS:
Apple could go with IBM
Apple could go with AMD
AMD Processor are RISC with CISC emulation
AMD is collaborating with IBM
AMD is entering 64/32 bit arena
Apple May use Nvidia Chipset
NeXT could run on Multiple Platforms
NeXT could run same programs on Multiple Platforms.
Apple is developing Marklar or an x86 version of OS X

FACT:
IBM is delivering a 64 bit PPC 970
AMD is delivering a 64 Bit Processor
AMD, nVidia, ATI, HP, and Apple are part of the HyperTransport consortium.
Apple has worked with all these companies except AMD.

CONCLUSION:
I don't know what to make of it, but it seems like Apple is investing time into multiple areas. It would make sense for Apple to go with IBM and AMD. Since IBM and AMD are working together, there's may be some cohesion here.

What if and this is a big what if, Apple were to bring OS X Server to the new AMD 64 bit crowd. They bring this with HyperTransport and allow people to build there own boxes while providing a list of compatible 3rd party products. Why would they do this? Well, more and more companies are moving to cheap WinLin products Apple wants people to taste OS X. Those IT guys like to tinker at home on cheap boxes. This opens the door. They could still run Linux/Windows and now OS X. Of course I'm speculating big time, but it's food for thought.

[shadowfax]

Originally posted by nuckinfutz
No not at all. Apple has sourced both companies for G3's and mainly Moto for G4's.

Motorola designes the G3/G4 for embedded applications.

Cisco uses a G4 in a Router. There are more markets than just Apple for Moto G4/G4 processors.


The PPC 970 is squarely aimed at Desktops however. Apple will be able to use both for some time.

the G3 is an IBM processor.

[Frobozz]

Originally posted by yzedf

2 hardware generations from now we could very well laugh and say "remember when were stuck with PPC?! Thank goodness that junk is gone!"

Apple/Steve clearly stated there is not even a chance for a processor-platform change for 3 years. Aside from that, it makes no sense to start over right after a very large and public transition to OS X. The only thing apple would gain from such a crazed action would be Ghz speed. They would lose their advanatges in marketing (differentiation from the PC market), lose advantage in power dissapation, lose altivec (which makes drastic improvements in number crunching apps), lose the ability to put it's chips in laptops, and lose ability to run any currnt software.

They'd gain nothing. Absolutely nothing. Ugh. When will people let the x86 thing go? The 970 at 2.5 Ghz, when the P4 3.x is out, will be very competitive.

[nuckinfutz]

Originally posted by Shadowfax
but if you don't take advantage of that, your program won't be G3 compatible, right?

You know I thought I rememberd reading that Altivec code should run unmodified on G3's but I'll have to see if I have any docs.

[shadowfax]

Originally posted by hacurio1
Couldn’t agree more Arn!!!

People here who question whether Apple will use the 970 or not aren’t very perceptive of what is going on right now! First, what options does Apple have? Is there any other PPC chip coming? NO!!! Will they go x86? NO!!!!! I know, “970’s in Macs” are just rumors but it makes absolute sense. Just look at the facts, Apple just went trough a big transition (OS9 to OSX.) Developers are barely finishing porting their software!!! Apple will not go x86!!! What other PPC option do they have? Motorola announced they will stop developing desktop processors, so I ask again, what other option does apple have? The 970 promises to be an excellent processor and it will definitely be as fast or faster that any P4 released by the end of the year. At 1.8 GHz the specint scores of the 970 are very close to the 3 GHz P4’s scores. At 2.5 GHz it will be as fast or faster and like many people said, it will be 64 bits. Contrary to what some uninformed people posted above, 64bit architecture is faster than a 32bit. Memory is only one advantage and it all depends on the OS and the design of the chip. In some instances, a 64bit processor can process two 32bit data paths a once and as mentioned above, it all depends on the OS and the chip. Since the core of OSX is 64 bit I doubt apple will have any trouble porting OSX to 64bit and optimizing it so that it takes advantage of the 64 bit ints and data paths. Since the 970 is fully backwards compatible with 32bits, I see it as the best option Apple has. IBM has even announced plans to develop the 980’s and Motorola hasn’t mentioned any plans for desktop chips.

Just my 2c

Motorola may not have plans for any desktop chips, but then, they never really made desktop chips, did they? the G4 is an embedded chip, and apple is right at home putting it on a desktop. and it really works pretty nicely, all things considered. Motorola does, i believe, have a roadmap for taking the G4 up and over 2 GHz now, so it's conceivable that Apple might stick with them at least for the rest of the year. would this be stupid? absolutely. but Apple may not want the inconvenience of an architecture change, which is inevitable, but they still might try to avoid it. i dunno about the processor price differences between the 970 and the G4, but it is conceivable that it could cost even more than the G4; it would certainly be justifiable.

but, i certainly agree that apple would really be going to the mattresses if they went with Marklar and intel/AMD x86 chips. I think that's just way out of left field.

[reyesmac]

By the time Apple releases these chips, how will they compare to what will be on the PC side for the following 6 months? How fast from the time of their announcement will people be able to own one? And how much of a premium is Apple going to charge its customers for the privalege of owning a Mac thats as fast as it should have been a year ago.
I will be asking myself these questions until it comes out.

[Frobozz]

Originally posted by nuckinfutz
Photoshop BTW is not a PC port. It was born on the Mac.

Altivec is going to take some time for developers to master like any tool.

There are plenty of apps that show Altivecs power.

Photoshop
iDVD
Altiverb
Blast

Just wait until the Altivec isn't starved by slow bus throughput. You're seeing the crippling effects of a bus that cannot let Altivec shine.

I see a nice 2004 with the fastest G4's going into iBooks and iMacs and PPC 970's in the Pro Series. Just like it's supposed to be.

I agree! Lets not forget iTunes. I can rip an entire CD from disc into 160 bit mp3's in a matter of a couple minutes on my Mac. Most of my friends PC's will take 45 minutes to do that.

[shadowfax]

Originally posted by reyesmac
By the time Apple releases these chips, how will they compare to what will be on the PC side for the following 6 months? How fast from the time of their announcement will people be able to own one? And how much of a premium is Apple going to charge its customers for the privalege of owning a Mac thats as fast as it should have been a year ago.
I will be asking myself these questions until it comes out.

I think one thing you can count on through all of this is that apple will be vain about their hardware. they are, for sure, and they have been for as long as i can remember. you'll pay a considerable premium, but probably not much more of one than you're paying now for the G4s. :D

[shadowfax]

Originally posted by Frobozz
I agree! Lets not forget iTunes. I can rip an entire CD from disc into 160 bit mp3's in a matter of a couple minutes on my Mac. Most of my friends PC's will take 45 minutes to do that.

what kinda PCs were they running? i burned CDs at 128 on my 933 MHz P3 in like 15-20 minutes. i can't imagine what it's like on a P4 at 2.5... don't get me wrong, i think that iTunes and the G4 does some really fast encoding (i do mine at 256:D), but i think you've just exaggerated way too excessively. i guarantee you a P4 2.5 GHz will do a full CD at 160 in under 8 minutes... probably a lot under. anyone know?

[beatle888]

Originally posted by ryan
In all fairness to Motorola (even though they don't really deserve it) they are shipping G4s, unlike IBM with their 970.

um, the 970 is still under development. right?

[swdrumcp]

We must also remember that when steve was asked about this been "the year of the laptops" he said: "we have things for desktops up our sleaves too".

1.6 G4s by years end with a 400mz fsb is not anything to brag about. Apple has to have something new comming down the line, something that will make people take notice, however, this been apple I dont see this product until next year. Perhaps then well have "the year of the desktop"

later

[occam]

Ok, instead of 1.2-1.8GHz, IBM is releasing at 1.8-2.4GHz. Let's hope Apple gets access to the same range of chips at reasonable prices. Great!

However...

We also must hope that Apple has been designing motherboard(s) for the 970. If so, do they have to scrap their current mobo's to upgrade to the new 1.8-2.4 GHz range? I hope not. Let's hope they've been working on 2.4GHz compatible motherboard(s) from the beginning. Otherwise, it could be a long wait for any 970 machines. (ouch)

Cross your fingers...

[GPTurismo]

Man, if Megahertz is your only worry about computers please go to comp usa and by a sony, hp or even an alienware. There are more things to a computer than that and if all you care is about speeds you need to take a reality pill and just get off the mac platform and quit acting like you know anything about technology.

Unless you want to run a POSIX variant like BSD or Linux on a pentium chip you have to run Windows, and sorry, I like MS for some things but their os's are just crap.

[ktlx]

Originally posted by Shadowfax
what kinda PCs were they running? i burned CDs at 128 on my 933 MHz P3 in like 15-20 minutes. i can't imagine what it's like on a P4 at 2.5... don't get me wrong, i think that iTunes and the G4 does some really fast encoding (i do mine at 256:D), but i think you've just exaggerated way too excessively. i guarantee you a P4 2.5 GHz will do a full CD at 160 in under 8 minutes... probably a lot under. anyone know?

I have a Pentium 4 2.53Ghz with 512MB of memory and a Western Digital 80GB with 8MB cache. In my experirence ripping WMAs and burning CDs is limited by the speed of my Lite-On 48x CD-RW and not by the machine. I can rip and burn and still do other stuff on my PC as if nothing was going on in the background. When I have 48x capable CD-Rs, I can burn at that speed and keep the buffers nearly full.

[iShater]

Originally posted by os4
There are two concepts at work with Apple's ordering process:

1) the "3 Choices rule" - people are most comfortable with three choices.


I don't believe that is true. I don't see other manufactureres going with just 3 choices. Besides, the machine is aimed at PRO users, who understand what their needs are and what kind of system addons they will need.

They can still offer 3 Configurations to make selecting a machine easier, however, they should allow us to customize everypart that is available to all three configs.



2) Apple needs to control its operations. The best way to do that is to simplify as much as possible. Each additional choice multiplies the logistical issues involved in putting the machines together. It would be nice if Apple added a "Power User" section to the Apple Store to allow for more choice; however, I am sure Apple has studied this issue and decided that the incremental revenue is less than the incremental costs.

All manufacturers control their operations, yet they still figured out how to make true BTO, so how come Apple can't figure it out? why do I have selections available on one system and not the other?

[macphoria]

Watch out Intel & AMD! When IBM 64bit PowerPC 970 2.5 Ghz comes to PowerMac, the game is ON!

[hacurio1]

Originally posted by Shadowfax
Motorola may not have plans for any desktop chips, but then, they never really made desktop chips, did they? the G4 is an embedded chip, and apple is right at home putting it on a desktop. and it really works pretty nicely, all things considered. Motorola does, i believe, have a roadmap for taking the G4 up and over 2 GHz now, so it's conceivable that Apple might stick with them at least for the rest of the year. would this be stupid? absolutely. but Apple may not want the inconvenience of an architecture change, which is inevitable, but they still might try to avoid it. i dunno about the processor price differences between the 970 and the G4, but it is conceivable that it could cost even more than the G4; it would certainly be justifiable.

but, i certainly agree that apple would really be going to the mattresses if they went with Marklar and intel/AMD x86 chips. I think that's just way out of left field.

Yes I agree, but then, at least IBM has plans. After Motorola begun to lose millions and millions of dollars, they closed several plants and many pople lost their jobs. I also remember reading that they sold some of their semicoductor divisions. Motorola has alot of finanical problems. Apple invested alot on the G4 but that is a sunk cost. Apple will make a desision based on what is better for the company and thir customers. What is cheaper, what is efficient, what is doable. The 970 and the G4 are both PPC. The cost of going from G4 to 970 will be minimal.

[ktlx]

You know, I think a lot of this discussion is completely missing the point. It is not a question of who will have the fastest CPU? in their boxes.

Apple has to look at its target audience. For the PowerMacs that audience is the consumer who needs more computer than the iMac can provide and creative professionals.

Anyone who thinks that Mac OS X and the PowerMac will take the corporate world by storm is on drugs. The corporate world will continue to buy tons and tons of Windows/Intel PCs for the foreseeable future.

The gamers will not move onto Mac equipment in droves because the games are not there. The game developers will not spend the resources on the Mac because the money is not there.

The overclocker/enthusiasts will not go with Macs because Apple will never give them the hooks that Gigabyte, Abit, etc. do to build 1.7x overclocked, watercooled, glowing neon, personalized PCs.

I don't think Apple needs have the fastest CPU for its PowerMac market. I does however need to be price competitive and provide value to its customers. There seems to be a magic number of high end desktops being priced in the $2500 to $3000 range. I checked a bunch of on-line retailers and everyone (not just Apple) seems to shoot for this price range.

I think Apple needs to produce a machine that is not humilated in non-Apple benchmarks by machines that fall into this class. For example, there should be no benchmarks anywhere that show Photoshop on a $2700 Dell performing significantly better than a $2700 Apple. I do not think the G4 can claim this anymore and it certainly will not be able to claim this once Prescott arrives. However the IBM PowerPC 970 appears to be able to.

Once that happens, then Apple can start using their software strength to show that a $2700 PowerMac is a far better machine than a $2700 Dell or HP. Comparing Mac OS X to Windows XP does not cut it. Creative professionals use an application to do their work and not an OS. Besides AppleScript being superior to VBScript, can anyone who uses Adobe Photoshop a lot really say there is much difference between it under either a Mac or a PC?

In my opinion, Apple needs to provide a price competitive high end desktop (not necessarily the fastest CPU) and then leverage its strengths in software such as being easier to make DVDs, being easier to create movies (in either iMovie or FCP), easier workflow from raw camera images to 16-bit TIFFs in Photoshop, etc.

[beatle888]

Originally posted by yzedf

http://www.pc.ibm.com/us/eserver/xseries/bladecenter_family.html



MY GOD. have you seen that disgusting site?
and the blade server looks like something found
at a garage sale....hahaha did you see those cheapo handles on the side. oh man...how can ibm let themselves be identified with such crap. god their website is so sad. anyway. looking forward to the 970

[ryme4reson]

From a business standpoint, do you really think Apple is going to release new Towers in 4 months and say "Here is the new Apple Tower ruinning @ 2.5GHz, and all this for the same price as curent towers. The new Dual 2.5Ghz is 400% faster than the previous model."

Well the simple answer is HELLLL NO!!! What they will most likely do is start at somewhere around the current 1.4 Ghz speed, and just come out with updates at a faster rate. That will keep demand up. When the G4 came out it was at 350Mhz, while the G3 was at 400 (correct me if I am wrong there)

It has been rumored(sorry proven) that the G3 has been at 1Ghz for what a year and a half, well where is the chip? Its being used, but not by Apple. The goal is to give speed bumps that are JUST ENOUGH to keep you customers happy ENOUGH, and keep demand up for each new update.

If they had 2.5Ghz towers, who would buy a PowerBook or an IMac, (or any ofther computer in the line up) I sure wouldnt, because I would be expecting (demanding) such a big update to the respective line that it would make NO sense to buy before the update. Hopefully all of you agree Apple couldnt survive long with only strong Tower sales, or Ford selling only mustangs, while everything else is sitting around.

When the G4 came out, it was faster than the G3, but not "mind boggling" fast.

I want Apple to have faster machines just like the rest of all of you do, but something I want more is for Apple as a company to survive, and as a company, they must fallow the simple rules of supply and demand.

[nuckinfutz]

The cost of the actualy chip should not be much higher than a G4.

The PPC 970 won't require expensive L3 Cache to maximize perfromance

However Apple will have to redesign their motherboards of course. There's been rumors of Apple creating a Proc to memory controller codenamed ApplePi.

The retooling necessary to test and deliver supporting chipsets is going to be more expensive. We'll see if Apple decides to go for margin or volume once they announce.

My guess is they stick to initial price points and perhaps go Dual proc on the Top system.

[ktlx]

Originally posted by iShater
All manufacturers control their operations, yet they still figured out how to make true BTO, so how come Apple can't figure it out? why do I have selections available on one system and not the other?

My understanding is that you have to have high volumes in order to make money at BTO systems. Even now, it is not clear from the quarterly earnings reports that anyone other than Dell can make money this way.

It would not suprise me in the least that Apple has looked at store.apple.com and determined that its volume through that channel simply will not profitably support a BTO model like Dell's. I don't think Apple has any other channel where BTO could be an option.

[mustang_dvs]

Originally posted by Raiden
My question is, if apple annownces the 970 at MWNY, saying the new processor will be availible in 2-3 months, what will motorola do? If I were motorola, I would be really pissed off that one of my biggest clients is jumping ship to a company that offers a better product than me. Therefore would motorola stop making G3s and G4s?

I guess im wondering is apple keeping their use of the 970 secret because they are afraid motorola will get angry? Does motorola know about apple and their supposed use of the 970?

Well, the fact of the matter is that IBM manufacturers the majority of the G3-series processors on the market these days.

If Apple still uses the current series of G4 processors (PPC 7xxx) -- which Moto exclusively holds the right to manufacture (b/c they hold the patents to AltiVec) -- after the shift to a new processor type, I highly doubt Moto will effectively sever a profitable relatioinship. I mean, it's not like they're run by their own Mercurial iCEO(tm). :D

Anyway, Moto has been looking to restructure their CPU division to focus more on embedded products for a while, and their new PPC series (8xxx) is aimed solely at that market. I would guess that Moto would be happy to not have Steve Jobs on their back.

[applejilted]

For a change Apple should abandon it's obsessively secretive ways and pre-announce that the PPC 970 will ship in systems on ...(insert date)..... Use a little FUD which M$FT is so good at ! By pre-announcing they would be CONSERVING sales as it would stop the exodus of pro users migrating to the DARK side....it wouldn't affect current sales all that much in that people aren't exactly breaking Apple's doors to buy their current relatively slow Powermacs. And that's all I have to say about that !

[mustang_dvs]

Originally posted by nuckinfutz
The cost of the actualy chip should not be much higher than a G4...

Actually, due to the small die size of the 970 and the resultingly-high per-wafer yield, the projected unit cost of a 970 is expected to be cheaper than the current G4 (PPC 7xxx), which has significant production difficulties. The 970 is pegged, conservatively, somewhere in the sub-$400 range, according to current estimates that I've seen.

[eric_n_dfw]

Originally posted by beatle888
MY GOD. have you seen that disgusting site?
and the blade server looks like something found
at a garage sale....hahaha did you see those cheapo handles on the side. oh man...how can ibm let themselves be identified with such crap. god their website is so sad. anyway. looking forward to the 970
It goes in a server room, who cares.

[nuckinfutz]

Originally posted by mustang_dvs
Actually, due to the small die size of the 970 and the resultingly-high per-wafer yield, the projected unit cost of a 970 is expected to be cheaper than the current G4 (PPC 7xxx), which has significant production difficulties. The 970 is pegged, conservatively, somewhere in the sub-$400 range, according to current estimates that I've seen.

OMG I'm sooooo dense. I just realized that IBM's specing the 2.5Ghz at .13 Micron!!! :eek:

I imagine they easily hit 3Ghz at .09 Micron and have better yields. I hear they're be fabbing on 200mm wafers initially and them move to 300mm.

Methinks you are correct. The PPC 970 will be cheaper for Apple. They need to go for broke if they can get enough chips from IBM. Price the systems to move and eventually the chips costs will drop and profits will rise.

8 Billion in revenue is going to take getting systems out the warehouses and into our greedy grubby little hands!

[law guy]

Originally posted by ryme4reson
From a business standpoint, do you really think Apple is going to release new Towers in 4 months and say "Here is the new Apple Tower ruinning @ 2.5GHz, and all this for the same price as curent towers. The new Dual 2.5Ghz is 400% faster than the previous model."

Well the simple answer is HELLLL NO!!! What they will most likely do is start at somewhere around the current 1.4 Ghz speed, and just come out with updates at a faster rate. That will keep demand up. When the G4 came out it was at 350Mhz, while the G3 was at 400 (correct me if I am wrong there)


What I find discouraging is that I just called the Apple store in Tysons Corner, VA and they tell me at least four (4) weeks until they begin to get 1.42 inventory. So, we won't even be seeing the 1.42 G4 machines until April!! This I find particularly sad as in a different thread I mention that I'm debating a Dell precision workstation vs. a 1.42 G4... I'm leaning towards the apple, but one could have a dual 3.06 Xeon machine in 9 business days. It's that kind of currently available PC hardware vs. dreams of new macs (or even announced macs) that is something of a letdown. I'm cheerleading for apple here, but lets say we see an announcement in Sept. and new machines in stores by Christmas... how long until software that's optimized for the chip, esp. the OS?

[applejilted]

Originally posted by law guy ... I'm leaning towards the apple, but one could have a dual 3.06 Xeon machine in 9 business days. It's that kind of currently available PC hardware vs. dreams of new macs (or even announced macs) that is something of a letdown. I'm cheerleading for apple here, but lets say we see an announcement in Sept. and new machines in stores by Christmas... how long until software that's optimized for the chip, esp. the OS? [/B]


Precisely !!!! Read my post on previous page....Apple is losing lots of sales by NOT pre-announcing .....people can make do with current systems if the promise of a blazingly new fast system is just around the corner. I for one would even thank Apple as I wouldn't buy a slow system now, but I wouldn't abandon ship either...

[nuckinfutz]

When you're CFO says they want to increase Marketshare by %66
I think it's safe to assume that they plan on moving alot of Hardware.

People that purchase G4's now NEED them and cannot wait. I'm sure Apple is preparing for the onslaught that's coming.

[DakotaGuy]

Originally posted by Shadowfax
the G3 is an IBM processor.

Actually I have read before that at one time the G3 was actually a Motorolla and IBM built processor, back when the first ones were released. IBM took over as Apple's G3 supplier when the introduced the "copper" G3 in 1999. Later versions including the 750 "copper", 750cx, and 750fx have all been IBM chips. My iMac DV and iBook are both running on IBM chips.

[Dont Hurt Me]

Originally posted by nuckinfutz
Photoshop BTW is not a PC port. It was born on the Mac.

Altivec is going to take some time for developers to master like any tool.

There are plenty of apps that show Altivecs power.

Photoshop
iDVD
Altiverb
Blast

Just wait until the Altivec isn't starved by slow bus throughput. You're seeing the crippling effects of a bus that cannot let Altivec shine.

I see a nice 2004 with the fastest G4's going into iBooks and iMacs and PPC 970's in the Pro Series. Just like it's supposed to be. I hope not, get rid of the g4 and give me a low end 970. I'd be happy with a 1.2 ghz 970 in the imac. this would make it simple if all lines had the 970. Apple could buy these chips in large volumes since all the machines would be using them! I noticed no answers on wether the current architecture is set up for the 970. Xserve,powermac,powerbook, and now imac 1 giger. would the current structure and system controler accept the 970 as is in these models?

[PretendPCuser]

Originally posted by AidenShaw
The comments that you can't compare a 32-bit x86 with the rumours of a 64-bit 970 are such triggers. Maybe these folks didn't see the story that Pixar was abandoning the 64-bit SPARC systems in favor of 32-bit Xeons.... Yes, they can be compared! And yes, sometimes the 32-bit system is better and faster! [/B]

Hi. I've found your posts to be educational and often going against the tide of some of the mac faithful here, but i have a serious question for you Aiden regarding the Pixar comparison. Since i have no interest in buying a machine for 3d rendering, and i don't feel like doing the leg work (i know you'd give accurate price comparisons), what is the price difference between comparable 64-bit SPARC systems and 32-bit Xeon system? What about the processors themselves?

[applejilted]

Originally posted by PretendPCuser
Hi. I've found your posts to be educational and often going against the tide of some of the mac faithful here, but i have a serious question for you Aiden regarding the Pixar comparison. Since i have no interest in buying a machine for 3d rendering, and i don't feel like doing the leg work (i know you'd give accurate price comparisons), what is the price difference between comparable 64-bit SPARC systems and 32-bit Xeon system? What about the processors themselves?

Intel rightly considers that 64 bit computing will not take off in consumer systems until at least 2008 and the reason for that is that 4 gb of RAM is prohibitively expensive. Studies show that high end PCs will only ship with 1gb of RAM by 2005. The primary advantage that 64 bit computing gives you is accessing over 4 gb of RAM....apart from huge data bases and perhaps video that require huge amount of RAM there is no compelling reason to go to 64 bit...Say what you want about Intel but nobody ever accused them of being stupid

[Frobozz]

Originally posted by law guy
Frobozz, the PC Magazine feb. 4 issue with the processor roadmap stated that Motorola would have a new processor in Q3 aimed at being the G4's successor and competing with the 970 for that role. Has Motorola cancelled their (what must be very far along) development?

I never saw that, but it's interesting. In my experience, though, Motorola says one thing and does another. IBM says something and does it.

Either way I suppose that would be good news for Apple, since they would FINALLY have competition for the chips they put in their machines. I would see this as a very positive turn of events.

[pseudobrit]

Originally posted by littlerich
I am looking to get a new PowerBook in about 2 - 3 months, obviously these processors or any fast processors will not be in the powerbooks but would you say it is better to wait a while. Until something better/cheaper comes out? Like I said it will be 3 months from now so??

The newer processors usually hit the PowerMacs first. I would wager it would take about 10 - 18 months to get the 970 in a PowerBook, along with a total redesign.

[Frobozz]

Originally posted by Shadowfax
what kinda PCs were they running? i burned CDs at 128 on my 933 MHz P3 in like 15-20 minutes. i can't imagine what it's like on a P4 at 2.5... don't get me wrong, i think that iTunes and the G4 does some really fast encoding (i do mine at 256:D), but i think you've just exaggerated way too excessively. i guarantee you a P4 2.5 GHz will do a full CD at 160 in under 8 minutes... probably a lot under. anyone know?

I wasn't exaggerating... I had to wait that long. Seriously. Either way my friends would never pay for a P4 2.5.... but I see your point. I don't know what a more robust system would do. I would expect that, in any case, the Mac is seriously faster. Perhaps 2 or 3 times as fast.

[BenRoethig]

Motorola MPC 7xxx series: Embedded processor. Low power usage is the main concern. Beating Intel/AMD processors is not really a concern since they are not a direct competitor in the embedded market.

IBM PPC970: Designed first and foremost as a desktop/ server chip. Must be able to compete with the Xeon on a price and performance level. Benefits from a company that knows exactly what they're up against.

[arn]

Originally posted by yzedf
JTo Arn:

Regarding your "clear win" statement, the BSD base for OS X makes the OS, as well as the programs that run on the OS very portable. x86, x86-64, PPC, are just as easy to do, for the most part. PPC kinda sucks due to Altivec, same as P4 with HT.


That still doesn't change the fact that the PowerPC 970 will have 100% Mac Application binary compatability while the x86 will have 0%.

People don't care how "portable" apps are... people care what will run.

arn

[Dont Hurt Me]

For applejilted there is a reason to go 64 bit and here it is. with the ability to use ram galore you could have a solid state hard drive(memory bank) no more waiting for the hard drive to go find something. I could see hard drives being obsolete in the next few years. thats if memory can be made larger and cheaper. Please anyone tell me if the newer architecture will support the 970 as is?

[Frohickey]

Originally posted by zoetropeuk
Apple needs to forget this stupid G3, G4 set-up and just move everything across to ONE platform. Apple also needs a department that will optimise all major software releases for any new CPU and system that it chooses the implement.
Otherwise we are going to see the same old crappy slow applications that we do now.

Pie in the sky.

When developers start handing out source code like candy to end users, then you will have the Apple-Software-Optimizing-Department that does what the label says it does... otherwise, that ASOD will just be full of blood-sucking lawyers arguing about terms of use, intellectual property and the size of their company's patent pool.

[Catfish_Man]

One advantage of the 970 is that it will handle poorly optimized code better (because of its extensive out of order execution capabilities). I would say that a 2.5GHz 970 is going to be AT LEAST on a par with Intel's Prescott when it's released. I would guess it'll have a SPECfp of 1300-1400 (IBM announced that a 1.8GHz 970 should hit at least 1050 SPECfp) which will beat AMD's Athlon64 (the current 3.06GHz P4 gets about 1100, the Athlon64 is said to get about 1200-1250, the G4 gets about 300-500). SPECint should be about on a par with a 3ish GHz Pentium 4 (just over 1000). If IBM really can deliver it at a lower price than the G4 (which I doubt), and vends a nice memory controller along with it, then I think Apple's got a winner. Even if it's a bit more expensive, that might be countered out by not doing dual processors yet.

[GeneR]

Just bring it on! The 2.5GHz really sounds great. "Sounds" great. "Sounds"... sounds...

hmmm...

;)

[hacurio1]

Originally posted by applejilted
Intel rightly considers that 64 bit computing will not take off in consumer systems until at least 2008 and the reason for that is that 4 gb of RAM is prohibitively expensive. Studies show that high end PCs will only ship with 1gb of RAM by 2005. The primary advantage that 64 bit computing gives you is accessing over 4 gb of RAM....apart from huge data bases and perhaps video that require huge amount of RAM there is no compelling reason to go to 64 bit...Say what you want about Intel but nobody ever accused them of being stupid
Funny!!! I still remember studies saying that the transition from 16 to 32 bits would take longer of what it actually took. People use to say, "there is no need for 32 bits yet, 16 is fine." and here we are!! The truth is that there will always be the need for more. When 16 bits was mainstream there was no need for 32 because the programs were small and people weren’t interested on editing videos nor had 30gb music libraries on their computers. What would be out in a year from now? Two years from now? Probably something that will benefit from 64 bits? The problem with studies is that they assume programs and uses for computes constant. When I bought a Power Mac 9600 in 1997 I was called crazy many times. I spent more than 5 grand on that computer and people use to say that it was such a machine that there is no way that I could use all its power. Today, that computer can’t even run OSX. I’m writing from the first dual processor Power Mac G4 with 2GB of RAM. So far so good, but sometimes I feel I need more RAM. The 500Mhz G4 is no big deal today either, but 4 years ago some of you might of asked, why do you need so much power? :confused:

[szark]

Originally posted by Dont Hurt Me
Please anyone tell me if the newer architecture will support the 970 as is?

It is extremely unlikely that Apple could have built any kind of support for the 970 in the current machines. The 970 has a completely different bus, which would require more signal lines between the processor and chipset. That would almost certainly require a new daughterboard connector.

[DavPeanut]

How fast are the buses for the 970's supposed to be?

[applejilted]

Don't look mow but MacCentral is trying to steal this awesome site's thunder....they just posted this story. And I thought that MacCentral never delved into rumoors or unannounced Apple products...will Apple now be withdrawing their MacWorld credentials ?

All I can say is that Macrumors wrox !!!! Keep up the great work Arn !

[nuckinfutz]

Originally posted by DavPeanut
How fast are the buses for the 970's supposed to be?


We'll have bragging rights. The 970 bus runs at half the core frequency. PPC 1.8Ghz has a 900Mhz FSB. 100Mhz faster than Prescott ;)

[DavPeanut]

Arn, where do you actually get all this stuff? How do you find it? Do you go through everything about a product on Google or something?

[ibookin']

Originally posted by pseudobrit
The newer processors usually hit the PowerMacs first. I would wager it would take about 10 - 18 months to get the 970 in a PowerBook, along with a total redesign.

I remember that G3-based desktops and laptop were released around the same time, due to the low heat and power consumption of the G3. If the PPC 970 is in any way similar in the respect of heat and power, we may see a PowerMac and a PowerBook using the 970 released at the same time.

I hope...

But then again, I may be wrong...

[jettredmont]

Originally posted by Dont Hurt Me
Please anyone tell me if the newer architecture will support the 970 as is?

No.

The System Controller, at the least, would have to be replaced. It is possible that other systems on the motherboard would also have to be re-arranged, but replacing the SC is the big hurdle in implementing any new chip FSB architecture (the 970 has a radically different "front-side" bus from the G3 and G4).

[-hh]

Originally posted by Abercrombieboy
Hey I am just waiting for Steve Jobs to do his Photoshop test with a Dual 2.5Ghz PowerMac 970!


I'm sitting there waiting for Steve walk out and unveil the new PowerMac's. He starts to go through his normal routine of showing how the new Mac beats the PC in the usual Altivec Photoshop routines. Yeah, yeah, seen it before (polite applause).

Steve then sets up the Mac vs. PC for an extensive performance benchmark test that goes across a whole bunch of applications - - not just the usual Photoshop BS - - - and explains that it will take awhile to run this benchmark, so he goes on to whatever other product announcements...

An hour later, Steve is wrapping things up, and he wanders back over to the Mac -vs- PC benchmark test. Hey! Good news: the Mac ran this app faster, matched the PC for that app, etc...a very even performance match all the way across the board. Steve smiles, says "thankyouverymuch..." etc.

As the lights come up in the auditorium, it is heard: "Oh, one more thing..."

Steve goes back over to the Mac. He reveals that it had run all the benchmarks as Windows Applications under VirutalPC!

The audience is confused. They stand there, silent and stunned.

Steve sits back down, closes VPC and says "Let's run those benchmarks again, shall we?" :D


Suddenly, I wake up in a start, in an ice cold sweat. Dang, its just that same recurring dream again.



-hh

[hacurio1]

Originally posted by hacurio1
Funny!!! I still remember studies saying that the transition from 16 to 32 bits would take longer of what it actually took. People use to say, "there is no need for 32 bits yet, 16 is fine." and here we are!! The truth is that there will always be the need for more. When 16 bits was mainstream there was no need for 32 because the programs were small and people weren’t interested on editing videos nor had 30gb music libraries on their computers. What would be out in a year from now? Two years from now? Probably something that will benefit from 64 bits? The problem with studies is that they assume programs and uses for computes constant. When I bought a Power Mac 9600 in 1997 I was called crazy many times. I spent more than 5 grand on that computer and people use to say that it was such a machine that there is no way that I could use all its power. Today, that computer can’t even run OSX. I’m writing from the first dual processor Power Mac G4 with 2GB of RAM. So far so good, but sometimes I feel I need more RAM. The 500Mhz G4 is no big deal today either, but 4 years ago some of you might of asked, why do you need so much power? :confused:

Of course 4Gb of memory is expensive today!!!! About $1,000, but wait two years, heck…just one more year and 4Gb of memory will cost about $500. When I bought my G4 on November 1999, 1.5Gb or Ram cost about $1,500…today…2Gb of PC133 will cost only $319.

[DavPeanut]

Originally posted by nuckinfutz
We'll have bragging rights. The 970 bus runs at half the core frequency. PPC 1.8Ghz has a 900Mhz FSB. 100Mhz faster than Prescott ;)
That means that the processors should be 1.35 Ghz, 1.8 Ghz, 2.25 Ghz, and 2.7 Ghz. Also, the Prescott will be 200x4 if i'm not mistaken, like the P4's "bus speed".

[Dont Hurt Me]

HH that was Great! maybe you should plan his next appearence.

[Frohickey]

Originally posted by applejilted
Precisely !!!! Read my post on previous page....Apple is losing lots of sales by NOT pre-announcing .....people can make do with current systems if the promise of a blazingly new fast system is just around the corner. I for one would even thank Apple as I wouldn't buy a slow system now, but I wouldn't abandon ship either...

People who are perpetually waiting for the next big thing are not people that really need the computer to do real work with. Its more along the lines of a 'status symbol' for them.

Apple cannot pre-announce. At least, not without cannibalizing existing sales, which tends to hurt the bottom line. About the only time Apple pre-announced, was for the first iMac, and the iPod. At that point, there was no existing sales to cannibalize, and the only purchases that Apple hurt was the sales to the iMacs/iPod's competitors.

The people that need a pre-announcement in order to keep from 'defecting' are the people that were going to 'defect' anyway. Apple is like the US, we'll take 'defectors' but we won't stop anyone that has already made up their mind to 'defect' to another. I guess that makes Microsoft, the commies. :D

[szark]

Originally posted by DavPeanut
That means that the processors should be 1.35 Ghz, 1.8 Ghz, 2.25 Ghz, and 2.7 Ghz. Also, the Prescott will be 200x4 if i'm not mistaken, like the P4's "bus speed".

The Prescott will have (I believe) one 64-bit wide bi-directional bus running at 800 MHz.

The 970 will have two 32-bit wide uni-directional buses (one from the chipset, one to the chipset) running at 900 MHz each.

[Catfish_Man]

Originally posted by DavPeanut
That means that the processors should be 1.35 Ghz, 1.8 Ghz, 2.25 Ghz, and 2.7 Ghz. Also, the Prescott will be 200x4 if i'm not mistaken, like the P4's "bus speed".

erm... I think you misunderstood. The bus frequency is half the core frequency. So, for a 2500MHz 970, it'll have a 1250MHz bus. For a 2000MHz 970, it'll have a 1000MHz bus. A 1.8GHz 970 and Prescott have the same effective bus speed (800MHz. The 970's is actually 900MHz, but there's some overhead). A 2.25GHz 970 would have a 1.125GHz bus, which seems a bit unwieldy.

[praetorian_x]

Originally posted by beatle888
MY GOD. have you seen that disgusting site?
and the blade server looks like something found
at a garage sale....hahaha did you see those cheapo handles on the side. oh man...how can ibm let themselves be identified with such crap. god their website is so sad. anyway. looking forward to the 970

How charmingly provincial.

prat

[hacurio1]

Originally posted by szark
The Prescott will have (I believe) one 64-bit wide bi-directional bus running at 800 MHz.

The 970 will have two 32-bit wide uni-directional buses (one from the chipset, one to the chipset) running at 900 MHz each.
I thought the 970 had bi-directional buses

[phampton81]

This may be a long shot, but does anyone feel that with the move from duals across the line to a single low end and dual mid/high end models on the powermac that this might suggest that they will keep this configuration for if/when the 970 comes out. I think it makes sense, since I would be extremely happy with even just one 970 and others would be willing to shell out another grand or so for that second 970.

[mbos]

There's a news article by Heise, Germany stating that prototypes of the new Blade servers will be shown at the Cebit conference (12.3.2003-19.3.2003).
Article (http://www.heise.de/newsticker/data/ciw-27.02.03-000/)

P.S. Heise publishes the very famous and serious computer journal c't.

[szark]

Originally posted by hacurio1
I thought the 970 had bi-directional buses

Nope. See this PDF (http://www-3.ibm.com/chips/techlib/techlib.nsf/techdocs/A1387A29AC1C2AE087256C5200611780/$file/PPC970_MPF2002.pdf) on IBM's website (page 12).

[FlamDrag]

Originally posted by Raiden
My question is, if apple annownces the 970 at MWNY, saying the new processor will be availible in 2-3 months...

Then it will really be available in about 6 mos.

While I'd love to see apple jump to the 2 ghz range, they seem content to milk every dollar out of every .2 ghz climb. So - don't hold your breath.

[Frohickey]

IBM PowerPC 970 is 64bit...

Let's see... my The PowerPC Architecture: A Specification for a New Family of RISC Processors book is copyright 1993... in the book, they talk about 32 bit and 64 bit implementations.

Its now 2003... so 10 years for the 32bit to 64bit jump...

I guess it will be 2013 when the PowerPC runs out of steam, and we'd have to jump to 128bit processing, and/or a new architecture. :D

[jettredmont]

Originally posted by hacurio1
Funny!!! I still remember studies saying that the transition from 16 to 32 bits would take longer of what it actually took. People use to say, "there is no need for 32 bits yet, 16 is fine." and here we are!!

(Sorry if this is Intel-ish instead of Motorolla ... I wasn't on the "Mac scene" during Moto's 32-bit transition)

Really? I don't remember any studies saying that the 32-bit transition on Intel would take >7 years (from introduction of the 386 in ~1988, a 32-bit pure processor, to the introduction of Windows 95, the first 32-bit-ish OS from MS).

Yes, people will always underestimate the ability for us to use power, and hind sight will always show that such predictions were foolish. On the other hand, certainly by the time Windows 3.1 was out (1992) the computer industry as a whole had a very well-defined idea that 32 bit processing was necessary for many activities. Once Win 95 came out, yes, there were "experts" who claimed that 16-bit software was okay still, but the reality of mode-switching (the CPU had to switch between 32-bit and 16-bit code, which is reportedly not the case on the 970 between 64-bit and 32-bit code) meant that running 16-bit software on a 32-bit OS was overall a pretty bad idea.

Now, contrasting 32-bits to 64-bits: when Intel switched over to 32 bits on their processors it was because users were already bumping into the limitations of 16 bits (memory allocation and management had been kludged to be larger than 16 bit pointers allowed, but even those kludges' boundaries were being bumped into by the majority of developers). By and large, developers are happy with 32-bits today, except for a few areas of development (video processing and databases forefront amongst them). When MS switched over to 32-bits on their OS, many applications which required 32-bit calculations had already been written to thunk the processor into 32-bit mode while they were in control and thunk it back down to 16-bit mode when they surrendered CPU control, which was both incredibly bad for OS stability (often applications left the CPU in a bad state) and for multitasking (apps such as this tended to relinquish control sparingly to avoid the performance hit of thunking/dethunking the processor when no other app needed the timeslice). While there are certainly apps that make use of 64-bit ints today (although obviously not 64-bit memory addressing etc), 64-bit processing support amidst regular 32-bit code is quite well supported on most processors today (not efficient - operations take 5x as long as 32-bit int operations - but such operations can not leave the CPU in a bad state nor do they require expensive mode-switching which makes developers want to hoard the CPU for as long as possible while they have it).

The 128-bit discussion here is more ludicrous: 64-bit pointers can address more memory than has been produced in the entire history of the personal computer (16 billion billion bytes), and it will be a while before we have enough memory on our desktop to even take full advantage of 64-bit addressing; I can think of no applications today that use 128-bit integer calculations - though I am sure they exist, they certainly are nowhere near mainstream. Will 128 bits be needed eventually? Eventually. I just don't see it as being as quick as the 16->32 bit transition (~10 years) or even the 32->64 bit transition (~16 years). Integer/pointer bit width, once a bottleneck and constraining factor for the majority of computing applications, is no longer the bottleneck.

[praetorian_x]

Originally posted by hacurio1
Funny!!! I still remember studies saying that the transition from 16 to 32 bits would take longer of what it actually took. People use to say, "there is no need for 32 bits yet, 16 is fine." and here we are!! The truth is that there will always be the need for more. When 16 bits was mainstream there was no need for 32 because the programs were small and people weren’t interested on editing videos nor had 30gb music libraries on their computers. What would be out in a year from now? Two years from now? Probably something that will benefit from 64 bits? The problem with studies is that they assume programs and uses for computes constant. When I bought a Power Mac 9600 in 1997 I was called crazy many times. I spent more than 5 grand on that computer and people use to say that it was such a machine that there is no way that I could use all its power. Today, that computer can’t even run OSX. I’m writing from the first dual processor Power Mac G4 with 2GB of RAM. So far so good, but sometimes I feel I need more RAM. The 500Mhz G4 is no big deal today either, but 4 years ago some of you might of asked, why do you need so much power? :confused:

Lets do a little math, shall we?

2^32 = 4294967296 bits = 4 Gigs

2^64 = 18446744073709551616 bits = 17592186044 Gigs

Consider that an entire film can usually fit on a 4 gig DVD disk.

The point that I am trying to make is that the "bitness" of an OS goes up exponentially, whereas other system measures, such as clock speed, go up sub-linearly (A 1000 Mhz chip is usually not 2x as fast as a 500 Mhz chip in practice). So the argument that "people said we'd never need 32 bits", while historically accurate, doesn't take into account the interaction of the exponential curve of addressable space and the relatively linear increase in need for addressable space (out side of databases).

In reality, it is RAM prices that will constrain memory size in PCs for the next few years rather than system bitness.

All of this is not to say that the 970 won't be a kick ass chip, but it is to say that 64-bitness will not be a significant contributer to that. The system bus and deep-and-wide approach that the 970 brings are far more important.

Cheers,
prat

[jettredmont]

Originally posted by DavPeanut
How fast are the buses for the 970's supposed to be?

From IBM's press release:

6.4 billion bytes per second.

That is, on a 64-bit wide bus (not sure if it's 64-bits wide or 128-bits wide), 800MHz.

This is identical to the original press release which touted an 800MHz bus on the 1.8GHz part.

In contrast, the fastest PC bus today is the Pentium-4, 533MHz (4.2GB/s). However, Intel is developing an 800MHz FSB as well, so if IBM gains the FSB speed crown it will be a short-lived triumph.

[hacurio1]

Originally posted by szark
Nope. See this PDF (http://www-3.ibm.com/chips/techlib/techlib.nsf/techdocs/A1387A29AC1C2AE087256C5200611780/$file/PPC970_MPF2002.pdf) on IBM's website (page 12).

my mistake!!!

[jettredmont]

Originally posted by nuckinfutz
We'll have bragging rights. The 970 bus runs at half the core frequency. PPC 1.8Ghz has a 900Mhz FSB. 100Mhz faster than Prescott ;)

Where are you getting that? IBM's press release says (after you run the numbers) that the FSB is 800MHz, not 900MHz, and does not hint that the 2.5GHz part might have a faster FSB than the 1.8GHz part.

Generally, the FSB rate is not tied to CPU core frequency directly, although the CPU core frequency is usually a basic multiple (ie, not necessarily integral, but not a highly complex multiple either) of the FSB.

[jettredmont]

Originally posted by hacurio1
Of course 4Gb of memory is expensive today!!!! About $1,000, but wait two years, heck?just one more year and 4Gb of memory will cost about $500. When I bought my G4 on November 1999, 1.5Gb or Ram cost about $1,500?today?2Gb of PC133 will cost only $319.

Note also that the large parts (1GB memory sticks) do not have enough buyers to attain mass-market pricing, and are thus more expensive than they would be if everyone and their cousin wanted 4GB of RAM on their machine (which, of course, would first and foremost require that their machines could accept 4GB of RAM, which is a rarity even in the Intel world still ...)

Memory prices, from a non-scientific guesstimate, appear to be following about an inverse-Moore's law, if not better. I'd say that roughly doubling the capacity of memory which can be bought for a specific sum from year to year is about right. Which would jive with the figures above: 4GB of memory being mass-market affordable (~$250 for memory) within two years.

However, when there is a barrier to people using an amount of memory, the more dense sticks tend to be significantly more expensive to purchase than the less dense sticks. For instance, just a few years back a single 128MB stick of SDRAM was running about 1.5-2x the cost of two 64MB sticks. If you are going to be putting 4GB in a PC, you'll need at least 1GB per stick, possibly 2GB per stick (depending on how many slots the motherboard has for RAM), which puts you in premium-priced memory densities. This will only change when the mass market moves to 64-bit processors ... unfortunately, probably a year or so after the introduction of the 970.

[Frohickey]

Originally posted by jettredmont
The 128-bit discussion here is more ludicrous: 64-bit pointers can address more memory than has been produced in the entire history of the personal computer (16 billion billion bytes), and it will be a while before we have enough memory on our desktop to even take full advantage of 64-bit addressing; I can think of no applications today that use 128-bit integer calculations - though I am sure they exist, they certainly are nowhere near mainstream. Will 128 bits be needed eventually? Eventually.

By the time we are using 128bit computing, Apple, IBM, Microsoft and Intel would have merged into one company, called Cyberdyne, and the first application will be the Model 101.

Hasta La Vista, Baby. :D :D :D

[Rocketman]

Originally posted by jettredmont
Ummm ... the 1.8 chip is supposed to be fall-2003 to Q1 2004, and that date estimate seems to be moving more towards Fall 2003.

If IBM is claiming the 970 in their soon-to-debut PowerPC Blade servers runs up to 2.5GHz, then you can expect that at most there will be a 3-month delay between general introduction and 2.5GHz parts being available. IBM tends to not play games with press releases.

Therefore, assuming that this press release was not a typo, we can expect 2.5GHz parts no later than this time next year (March 2004).

If the 1.8GHz part gets the estimated SPEC scores, the 2.5GHz part should get approximately (although not exactly) 40% higher scores. This is some very good news. If nothing else, IBM's blade servers should be selling like hotcakes!

This is HIGHLY CONSISTENT with Apple January release notices and late Feb, early Mar ship dates.

Seems the writing is on the wall indeed. Save your pennies, but whoard OS9 machines NOW.

Rocketman

[hacurio1]

Originally posted by praetorian_x
Lets do a little math, shall we?

2^32 = 4294967296 bits = 4 Gigs

2^64 = 18446744073709551616 bits = 17592186044 Gigs

Consider that an entire film can usually fit on a 4 gig DVD disk.

The point that I am trying to make is that the "bitness" of an OS goes up exponentially, whereas other system measures, such as clock speed, go up sub-linearly (A 1000 Mhz chip is usually not 2x as fast as a 500 Mhz chip in practice). So the argument that "people said we'd never need 32 bits", while historically accurate, doesn't take into account the interaction of the exponential curve of addressable space and the relatively linear increase in need for addressable space (out side of databases).

In reality, it is RAM prices that will constrain memory size in PCs for the next few years rather than system bitness.

All of this is not to say that the 970 won't be a kick ass chip, but it is to say that 64-bitness will not be a significant contributer to that. The system bus and deep-and-wide approach that the 970 brings are far more important.

Cheers,
prat

Your DVD example is out of point. Yes, the contents of a DVD will fit in about 4Gb, but compressed (MPEG2.) Now, how long will it take you to compress 17GB of uncompressed video into MPEG2? I know that it mostly depends on the CPU, but it will be really nice to load more than 4GB of data into RAM to keep the idle time of the CPU minimum. I know that today most people don’t do this on a daily basis, but five years ago for an average user to do this type of things was unimaginable (Average User.) tell me something? How much RAM did you have on your previous computer? How much RAM do you have on you current computer? That is my point, we are getting closer and closer to the point were 4GB would not be enough!!!! My first Mac, a PerformaCD, had 4mb of RAM, my second Mac, a Performa 6400 had 16mb of RAM, my Power Mac 9600 has 96Mb of RAM, and my G4 has 2GB or RAM. It has been about 10 years and my computers went form 4Mb to 2Gb.
And please explain, why do you think RAM prices will constrain memory size in the next few years?:confused:

[jettredmont]

Originally posted by jettredmont
Where are you getting that? IBM's press release says (after you run the numbers) that the FSB is 800MHz, not 900MHz, and does not hint that the 2.5GHz part might have a faster FSB than the 1.8GHz part.


Okay, re-examined IBM's original documentation on the 970 and see that the bus is "up to 900MHz" with an effective rate of "up to 6.4GB/s" ... which means that it's running at 900MHz but pumping data through like a 100% efficient 800MHz bus.

However, the "always half the CPU core frequency" thing I can't find any mention of, anywhere. Can you point me to something from IBM that says the bus is 1/2 of the CPU frequency for future (ie, non-1.8GHz) speed bumps?

[Rocketman]

Originally posted by Inunyan
Very cool.

I would finally be able to put my goodol' DEC XP1000 to rest, and do serious number-crunching on my desktop OSX!

Wish they come up with cooler (underclocked?) version. I would put one in my Cube.

If they just come out with a quad 970 in a cube you can have a desktop fanless terra-computer.

Probably practical too with 1000 Ethernet, FW800 and Fiberchannel. Store the raid and memory array on the floor :)

Rocketman

[hacurio1]

Originally posted by jettredmont
(Sorry if this is Intel-ish instead of Motorolla ... I wasn't on the "Mac scene" during Moto's 32-bit transition)

Really? I don't remember any studies saying that the 32-bit transition on Intel would take >7 years (from introduction of the 386 in ~1988, a 32-bit pure processor, to the introduction of Windows 95, the first 32-bit-ish OS from MS).

Yes, people will always underestimate the ability for us to use power, and hind sight will always show that such predictions were foolish. On the other hand, certainly by the time Windows 3.1 was out (1992) the computer industry as a whole had a very well-defined idea that 32 bit processing was necessary for many activities. Once Win 95 came out, yes, there were "experts" who claimed that 16-bit software was okay still, but the reality of mode-switching (the CPU had to switch between 32-bit and 16-bit code, which is reportedly not the case on the 970 between 64-bit and 32-bit code) meant that running 16-bit software on a 32-bit OS was overall a pretty bad idea.

Now, contrasting 32-bits to 64-bits: when Intel switched over to 32 bits on their processors it was because users were already bumping into the limitations of 16 bits (memory allocation and management had been kludged to be larger than 16 bit pointers allowed, but even those kludges' boundaries were being bumped into by the majority of developers). By and large, developers are happy with 32-bits today, except for a few areas of development (video processing and databases forefront amongst them). When MS switched over to 32-bits on their OS, many applications which required 32-bit calculations had already been written to thunk the processor into 32-bit mode while they were in control and thunk it back down to 16-bit mode when they surrendered CPU control, which was both incredibly bad for OS stability (often applications left the CPU in a bad state) and for multitasking (apps such as this tended to relinquish control sparingly to avoid the performance hit of thunking/dethunking the processor when no other app needed the timeslice). While there are certainly apps that make use of 64-bit ints today (although obviously not 64-bit memory addressing etc), 64-bit processing support amidst regular 32-bit code is quite well supported on most processors today (not efficient - operations take 5x as long as 32-bit int operations - but such operations can not leave the CPU in a bad state nor do they require expensive mode-switching which makes developers want to hoard the CPU for as long as possible while they have it).

The 128-bit discussion here is more ludicrous: 64-bit pointers can address more memory than has been produced in the entire history of the personal computer (16 billion billion bytes), and it will be a while before we have enough memory on our desktop to even take full advantage of 64-bit addressing; I can think of no applications today that use 128-bit integer calculations - though I am sure they exist, they certainly are nowhere near mainstream. Will 128 bits be needed eventually? Eventually. I just don't see it as being as quick as the 16->32 bit transition (~10 years) or even the 32->64 bit transition (~16 years). Integer/pointer bit width, once a bottleneck and constraining factor for the majority of computing applications, is no longer the bottleneck.

I agree, to think of 128bit today is ridiculous, but perhaps in 20-30 years who knows!!! My point is, however, that computer’s processing power advances so fast that memory limitation will soon be obvious in desktop computing!

[mbos]

Originally posted by mbos
There's a news article by Heise, Germany stating that prototypes of the new Blade servers will be shown at the Cebit conference (12.3.2003-19.3.2003).
Article (http://www.heise.de/newsticker/data/ciw-27.02.03-000/)


Correction: This is based on nearly the same IBM web page mentioned at the beginning of this thread. However, if one clicks the link "Photos" there (and agrees to the terms of use) one will find an image of the prototype to be shown at the bottom.

[Rocketman]

Originally posted by usersince86
EXACTLY.

I've being using Macs since 1986, and we've always been behind in the MHz war.

You can't compare future, undelivered chips to ones that are currently on the market and say, "See, we're just as fast!"

We're not. Don't give me all the "MHz myth" stuff; you can spin it however you want, and even though the gap isn't as big as it looks, it's still a gap. Period.

You know what? There * is * a Mhz gap in favor of wintel. However there is an integration gap and a software simplicity gap and a hardware reliability gap in favor of Macs.

Why are so many PC's sold?

1. Vertical market applications that dedicate a PC as a tool to a particular task.

2. Inertia arising from IBM-PC being on the market and clones being available for several years before the mac was originally released at HIGH prices with limited target marketing. That wave is still breaking in favor of Wintel. Microsoft just caught the wave by out moving IBM! And suckering Apple into a licensing agreement. Microsoft is not stupid. Intel is riding the Microsoft wave!

3. Microsoft was legally declared an illegal monopoly.

4. There used to be a cool computer called Amiga that had a primary benefit of optimized and specialized co-processors. It actually rocked. Mac outclassed it, out marketed it, out developed it, but never got any faster hardware.

5. Apple is about to have hardware as fast or faster than anything when used with available OS, software and for practical applications.

6. Apple already has server farms, clusters, arrays, and is about to triple the throughput of ech of those, albiet at a price comparable to 2/3 the highest prices it has ever charged.

But wait! Only 2/3 the peak price for 3x the performance of the best systems on the planet? That is VALUE.

People buy VALUE.

Hmmm.

Rocketman

[ryan]

Originally posted by Rocketman
[snip] Save your pennies, but whoard OS9 machines NOW.

Rocketman
Hoard soon to be obsolete machines to run an obsolete OS? I don't think so.

[AidenShaw]

Originally posted by PretendPCuser
what is the price difference between comparable 64-bit SPARC systems and 32-bit Xeon system? What about the processors themselves?

At http://www.specbench.org/cpu2000/results/ you'll see that on single CPU tasks (e.g. rendering) the P4 3.06 integer scores are about 1100 vs about 550 for a SPARC 1.05 GHz. Floating point is about 1100 to 700.

Price? An HPaq DL360/g3 1U system with dual 2.8GHz Xeon (P4) and 1GB lists for about $4400.

A SPARC 280R 4U dual 1.015GHz with 2GB is $18,000.

The Intel box is about twice the power in 1/4 the space for 1/4 the price....

[Rocketman]

Originally posted by ryan
Hoard soon to be obsolete machines to run an obsolete OS? I don't think so.

Backward compatibility workstation? Yes:

OSX
OS9
YDLinux
Tenon Unix
PC emulation
Win 95
Win 98
Win 2k
Win ME
DOS
etc . . .

Forward computers will rock but will not run all legacy applications that real people use.

Real people really suck :)

Rocketman

[ryan]

Originally posted by Rocketman
Backward compatibility workstation? Yes:

OSX
OS9
YDLinux
Tenon Unix
PC emulation
Win 95
Win 98
Win 2k
Win ME
DOS
etc . . .

Forward computers will rock but will not run all legacy applications that real people use.

Real people really suck :)

Rocketman
Funny, I've been using Mac's since '85 and was never able to do any "real" work on them until OSX came along. Try doing Java/DB (Oracle, MySQL, Postgre)/JBoss/Tomcat work on a Mac running anything pre-OSX, can't be done.

And, as arn has pointed out, the 970 will be able to run all existing OSX applications without modification just fine.

[edenwaith]

Originally posted by drastik

Some of you out there have this wierd thing about being faster, bigger, better all the time. I don't know if you're 14 or you just have a complex, but please relax. If your mac does what you want and allows you to be more efficient, who cares how fast the clock is?


Exactly!!! In reality, who needs 2 GIGAHERTZ TO RUN A FRICKIN' WORD PROCESSOR!!! I have a Mac SE with ClarisWorks 3 which runs just fine. The machine starts up in under 30 seconds. Okay, the screen might need a magnifying screen a la the computers in the movie Brazil, but the computer can still do a great job as a word processing machine.

If the 970 does come to pass, and let's just imagine that the next batch of PowerMacs get bumped up to 2.0 GHz. Would Apple be willing to just jump a bunch of their machines up so drastically, or would they do this in incremental steps? But having a dual 2.5 GHz machine would be so sweet. The machine I would like would be fast enough that I could even run Virtual PC and any current PC game, and still have no lag. I don't know if the 1.4GHz machines would do that for me or not...perhaps I should keep waiting for a little while. It would be nice to have a computer which lasts at least 3 years without becoming totally out dated. Granted, I'm running OS 10.2 on a 350 MHz iMac, and it works okay, but I certainly wouldn't mind having a 800 MHz or 1 GHz iMac! But to make a fully screaming machine, it needs to be optimized in about every way possible. Quick bus speeds, lots of RAM, plenty of on-chip L2 or L3 cache, speedy hard drives...however, there are still certain areas which just can't be sped up and are still bottle necks, such as drives (floppy, CD, hard).

Does anyone know why the processors from Motorola and IBM have gotten so behind the Intel and AMD chips?

Okay, enough babbling for now.

[Masker]

Originally posted by AidenShaw
At http://www.specbench.org/cpu2000/results/ you'll see that on single CPU tasks (e.g. rendering) the P4 3.06 integer scores are about 1100 vs about 550 for a SPARC 1.05 GHz. Floating point is about 1100 to 700.

Price? An HPaq DL360/g3 1U system with dual 2.8GHz Xeon (P4) and 1GB lists for about $4400.

A SPARC 280R 4U dual 1.015GHz with 2GB is $18,000.

The Intel box is about twice the power in 1/4 the space for 1/4 the price....

Wow, this is just not correct. Sparc boxes do one thing, and they do it really, really well: scale. Intel boxes do not have the same throughput when they're used as UNIX servers for many, many users. Sparc boxes blow Intel boxes away in terms of stability and responsiveness under high load. Try and run an ISP on all Intel boxes, and you'll crumple under the load. There's a lot more to a server than the CPU that's in it. Please provide links for those servers, since I'm SURE you're not giving us the biggest price/performance ratio at all. SPARC machines have dropped pretty drastically in price lately, so I'll believe it when I read it.

[Bregalad]

Originally posted by drastik
Sure, you can get a $600 PC, but it has SDRAM not DDr, no firewire, CD-ROM instead of superdrive, etc, etc. It could go to 10ghz, the $600 machine is still an email station for all the good it dos you for pro work. Not to mention the clunky applications that drag the processor speeds dawn anyway.

It seems I've joined this thread rather late, but I have to respond to the above crap.

What planet are you from?
I haven't seen a new PC motherboard that supports SDRAM in ages; they're all DDR or Rambus. The PC guys were using DDR more than two years before Apple did and their implementations actually send and receive data twice per clock cycle, something no currently shipping PowerPC can do.

For the price of a 1GHz PowerMac G4 tower ($1499) you could match the G4's specs, except for FireWire 800 which would be replaced by FireWire 400, USB 2.0 and 6 channel sound, and still have enough money left over for a 17" digital LCD display. Yes that includes Windows XP Professional, speakers, Microsoft optical mouse and USB keyboard.

To those who don't know or care about the advantages of OS X the PC looks like the real deal. To them Mac people like you and me are stupid because we seem to have paid more and gotten less. It's going to be tough to convince the PC majority that they're wrong.

As for Windows being clunky and slow, have you used 2000 or XP? Did you see how badly dual processor Macs fared in a recent video editing test against single processor AMD and Intel machines? Doesn't that count as "pro" work? Did you know that Steve Jobs' other company, Pixar, uses machines with Intel Xeon processors? Are animation studios not "pro"?

Yoda was wrong, the dark side is really easy to see if you open your eyes.

[Frohickey]

Originally posted by edenwaith
Does anyone know why the processors from Motorola and IBM have gotten so behind the Intel and AMD chips?


Easy. Back when Mr Gil Amelio was still in charge of Apple, and during the days of the Apple deathwatch, Mr. Gil Amelio was so inept that he did not do anything to assuage the fears of Motorola and IBM management that Apple will be around.

So, design momentum on the PowerPC was lost, and that lead to the IBM/Motorola split. No use having 2 chip vendors when your business is shrinking. Why chase good money after bad?

After the iMac success, Motorola/IBM were still skeptical, but they were more willing to listen to Apple's desire to get more resources for chip development. Plus, S. Jobs makes a pretty good sales pitch.

[Catfish_Man]

Originally posted by Masker
Wow, this is just not correct. Sparc boxes do one thing, and they do it really, really well: scale. Intel boxes do not have the same throughput when they're used as UNIX servers for many, many users. Sparc boxes blow Intel boxes away in terms of stability and responsiveness under high load. Try and run an ISP on all Intel boxes, and you'll crumple under the load. There's a lot more to a server than the CPU that's in it. Please provide links for those servers, since I'm SURE you're not giving us the biggest price/performance ratio at all. SPARC machines have dropped pretty drastically in price lately, so I'll believe it when I read it.

Sun boxes also do stability quite nicely. I have yet to hear of anyone else with hotswap processors. Basically though, SPARC is falling behind. The next battle is going to be POWER (and PowerPC) vs. IA-64 (Alpha would have been a competitor, but it got killed by HP).

[Freg3000]

Originally posted by edenwaith
Exactly!!!But having a dual 2.5 GHz machine would be so sweet. The machine I would like would be fast enough that I could even run Virtual PC and any current PC game, and still have no lag. I don't know if the 1.4GHz machines would do that for me or not...perhaps I should keep waiting for a little while.

I think the majority of problems with Virtual PC as per its slowness is due to the software. There is nothing special in it that can handle almost any type of 3D graphics. Until Connectix (now owned by Microsoft) implements a new way to emulate Windows based PCs, I am afraid that even a Dual 2.5 GHz PowerPC 970 won't make VPC seem fast enough to play games. :(

Just wanted to make sure nobody got their hopes up. I thought VPC going to fly on my new computer (Dual 1 GHz) but it doesn't. It probably won't on the Dual 2.5 Ghz PowerPC 970. Hopefully I?m wrong.

But hey-when Apple releases the new Powermacs, everyone will switch to Macs anyway, and there won't even be a need for a Windows emulator...right? :D

[end radical pipe dream]

[GeneR]

Me thinks Apple needs to come out with a GHz similar to whatever Intel's got when the 970 comes out. If not, it's still not going to look like Apple's doing well because the numbers to the average Joe still look relative between the two types of machines.

Hopefully, this number: 2.5GHz is real. Hopefully. But also hopefully, is that there is an even higher number to make Apple "move ahead" of Intel.

I guess we'll just have to wait and see.

:D

[shadowfax]

Originally posted by GeneR
Me thinks Apple needs to come out with a GHz similar to whatever Intel's got when the 970 comes out. If not, it's still not going to look like Apple's doing well because the numbers to the average Joe still look relative between the two types of machines.

Hopefully, this number: 2.5GHz is real. Hopefully. But also hopefully, is that there is an even higher number to make Apple "move ahead" of Intel.

I guess we'll just have to wait and see.

:D

i think that if the megahertz myth were demonstrably false on every count, that people would abandon it. if this thing really does outperform 4 GHz P4s, and with 900 MHz buses and true DDR, it's certainly poised to, Apple will make some true headway into the market.

as it is, the megahertz myth is both arguably true and arguably false, at a general level. apple has photoshop in its pocket, just barely. it needs to put a lot more there with it.

[BenRoethig]

Originally posted by edenwaith


Does anyone know why the processors from Motorola and IBM have gotten so behind the Intel and AMD chips?

Okay, enough babbling for now.

Economics. Motorola's main processor market is the embedded market. It's not financially viable for them to develop a desktop processor to compete with the x86 world. IBM is in a slightly different situation. They developed the PPC970 to compete with Xeon servers. They're not going to sell any if they're inferior to the Intel offering. Therefore, it's in their best financial interest to make a product that meets or exceeds the competition on price and performance.

[Choppaface]

w00t!

*starts saving*

[drastik]

Originally posted by Bregalad
It seems I've joined this thread rather late, but I have to respond to the above crap.

What planet are you from?
I haven't seen a new PC motherboard that supports SDRAM in ages; they're all DDR or Rambus. The PC guys were using DDR more than two years before Apple did and their implementations actually send and receive data twice per clock cycle, something no currently shipping PowerPC can do.

For the price of a 1GHz PowerMac G4 tower ($1499) you could match the G4's specs, except for FireWire 800 which would be replaced by FireWire 400, USB 2.0 and 6 channel sound, and still have enough money left over for a 17" digital LCD display. Yes that includes Windows XP Professional, speakers, Microsoft optical mouse and USB keyboard.

To those who don't know or care about the advantages of OS X the PC looks like the real deal. To them Mac people like you and me are stupid because we seem to have paid more and gotten less. It's going to be tough to convince the PC majority that they're wrong.

As for Windows being clunky and slow, have you used 2000 or XP? Did you see how badly dual processor Macs fared in a recent video editing test against single processor AMD and Intel machines? Doesn't that count as "pro" work? Did you know that Steve Jobs' other company, Pixar, uses machines with Intel Xeon processors? Are animation studios not "pro"?

Yoda was wrong, the dark side is really easy to see if you open your eyes.

Of course animation studios are Pro, if you read all of my post I made a point about render farms.

The whole point of my post was that you don't need the speed for most common applications. Every tool has its uses, personally I use PCs and Macs.

BTW, I was talking about the bargi basement PCs, the $600 dollar style. People who buy these machines aren't dong pro work, hence my comments about the machines doing pro work.

Yes, I've used Xp and 2000, frequently. The installs take forever, files get shifted into random locations (this is important if you are typically filling drives and want to get rid of stuff), feels clunky and non-intuitive.

Read before you write. I will give you the memory thing, I just looked quickly at the Dell site and saw sdram, it really says DDR SDRAM, which I shold have noticed.;)

Don't be rude, calling posts crap is just ignorant.

[Hattig]

Originally posted by hacurio1
I thought the 970 had bi-directional buses
IIRC, the 970 has 4 of these busses, allowing glueless SMP capability up to 16 processors. However, this may have changed since last year, if 16-processor 970 based systems are not going to happen, or maybe IBM will make two variants of the 970, one for single or dual processor machines, and one for 4, 8 or 16 processor machines.

The bus is most likely similar in nature to AMD's HyperTransport - a very fast DDR bus, uni-directional (i.e., the bus is split into a part away from, and a part into the processor).

The latest data says 6.4GB/s for the bus data transfer. That will be 16-bits in each direction at 800MHz DDR (1.6 Giga transfers per second). I expect that the bus speed is decoupled from the core speed now. It is possible that it runs at 900MHz DDR like the original announcement said.

[ktlx]

Originally posted by jettredmont
The 128-bit discussion here is more ludicrous: 64-bit pointers can address more memory than has been produced in the entire history of the personal computer (16 billion billion bytes), and it will be a while before we have enough memory on our desktop to even take full advantage of 64-bit addressing; I can think of no applications today that use 128-bit integer calculations - though I am sure they exist, they certainly are nowhere near mainstream. Will 128 bits be needed eventually? Eventually. I just don't see it as being as quick as the 16->32 bit transition (~10 years) or even the 32->64 bit transition (~16 years). Integer/pointer bit width, once a bottleneck and constraining factor for the majority of computing applications, is no longer the bottleneck.

The one place where 128-bit processing would really be helpful is in network appliances that support IPv6. That way you could deal with IP addresses in one large gulp instead of having to break it down into smaller pieces. I am not sure if this will be important enough to justify a general purpose 128-bit CPU or if it will be cheaper to just add a specialized network processor.

[bertagert]

So where is the PC970 on this board?
Prototype Blade Server with PC970 (http://www-5.ibm.com/de/pressroom/cebit2003/i/highres/bladeprototype_300_Dpi.jpg)

[MacFan25]

2.5 GHz would be great.

[law guy]

Originally posted by drastik

The whole point of my post was that you don't need the speed for most common applications. Every tool has its uses, personally I use PCs and Macs.


Of course, and its a good, sane point you make. At some point, though, the speed of a machine becomes too slow to support a future operating system or future word processors or future spreadsheets or browsers, etc. Makes me think of when MS Word came on a few diskettes... anyway. There were all sorts of folks who did not NEED a Mac IIfx or a Quadra, or a PII 233, but those machines did make the software of the day work a little faster, were able to deal with more programs at once, and kept up for just a little longer.

I think when some folks want speed today for everyday software, they not only want the scroll bars to move quickly, for explorer to not crawl along, etc., but they want it to be a functional investment for just that much longer.

As new hardware comes out, people seem to write to it. Thus a word processing program that now takes up 40 megs of hd space. It seems wanting to buy speed for everyday computing is, at least in part, a desire to have a machine that can handle what is expected of a lower level machine of a year from now.

[shadowfax]

Originally posted by bertagert
So where is the PC970 on this board?
Prototype Blade Server with PC970 (http://www-5.ibm.com/de/pressroom/cebit2003/i/highres/bladeprototype_300_Dpi.jpg)

under that little heat sink?

[law guy]

Since there is a good of 970 vs. Intel discussion going, I thought I'd pass along this link - apparently PC magazine has the dirt, skinny, dig (not rumor) on Intel's roadmap for the next three years (which appears to mirror rumor). The link is:

http://www.pcmag.com/article2/0,4149,906060,00.asp

"Oh boy, do we ever have the goods on Intel.

Our crack reporter, Mark Hachman, got his hands on some killer information – basically Intel's entire roadmap for the next few years.

It's way too much for just one article, so we've got him slaving over the keyboard, banging out the information as fast as he can."

[bertagert]

Originally posted by Shadowfax
under that little heat sink?

I want to see what one looks like. Anyone have a link to a picture of the 970? Or, get your photoshop out and lift that heatsink off so i can take a peak.:D

Edited for spelling

[phampton81]

I'm curious as to whether there is RAM available right now that is able to take full advantage of the 800mhz bus of the 970, if so what kind is it pc4500? (just threw that out there) Also how expensive is this ram that is fast enough?

[law guy]

Originally posted by bertagert
I want to see what one looks like. Anyone have a link to a picture of the 970? Or, get your photoshop out and lift that heatsink off so i can take a peak.:D

Edited for spelling

For a PICTURE - see page five of this .pdf from the IBM Microelectronics page (takes awhile to load)

http://www-3.ibm.com/chips/techlib/techlib.nsf/techdocs/A1387A29AC1C2AE087256C5200611780/$file/PPC970_MPF2002.pdf

There appears to be a fair amount of information on IBM's site.

[Hattig]

Originally posted by phampton81
I'm curious as to whether there is RAM available right now that is able to take full advantage of the 800mhz bus of the 970, if so what kind is it pc4500? (just threw that out there) Also how expensive is this ram that is fast enough?
For a 6.4GB/s interconnect, dual channel PC3200 DDR would be required.

Or quad-channel PC800 RDRAM... heh.

That mock-up blade sucks... is that 133MHz SDRAM? If it works, it is going to be a poor performer, hence it is a mock-up - but it shows that you can implement the PPC 970 interconnect in a FPGA like the Xilinx chip on the board.

[AidenShaw]

Originally posted by Masker
Wow, this is just not correct. Sparc boxes do one thing, and they do it really, really well: scale.

Note that I quoted "single CPU tasks" specifically!

Render farms are not about scaling tasks in a single SMP box. They're about getting as much compute power as possible directed towards lots of smaller jobs.

Scaling is not as important as raw system power per CPU, and density of CPUs.

An E15K is awesome, almost as awesome as its price.

[law guy]

Originally posted by law guy
For a PICTURE - see page five of this .pdf from the IBM Microelectronics page (takes awhile to load)

http://www-3.ibm.com/chips/techlib/techlib.nsf/techdocs/A1387A29AC1C2AE087256C5200611780/$file/PPC970_MPF2002.pdf

There appears to be a fair amount of information on IBM's site.

UPDATE - there are some add'l really nifty PICTURES in the IBM .pdf. For whatever reason - the link above will just lead to an access denied screen. TRY: IBM.com - go to "products", then "microelectronics", then "powerpc", then under the choices to select an offering, go to "9xx 64-bit", the .pdf link comes up and is accessable (although large). I attempted to just attach a copy that I'd saved, but it was too large to post.

[bertagert]

Thanks for the link law guy.

I want you guys to know that I just copied that PC970 off the pdf and pasted it onto my mother board. Running at 1.8 now and things are sweet! Wooohoo...first to have the PC970!

[Frohickey]

Originally posted by bertagert
So where is the PC970 on this board?
Prototype Blade Server with PC970 (http://www-5.ibm.com/de/pressroom/cebit2003/i/highres/bladeprototype_300_Dpi.jpg)

Prolly under that big hunka metal heatsink. If it was made out of transparent aluminum, you'd see it. :D

[AidenShaw]

Originally posted by ktlx
The one place where 128-bit processing would really be helpful is in network appliances that support IPv6.

Is that why you have a 32-bit moniker? ("ktlx" fits in 4 8-bit bytes)

Why on earth would you think that you'd need a 128-bit computer to deal with longer strings?

OhMyGawd - I'd better change my postname - "AidenShaw" is 72-bits, no way anyone will be able to reply to me if they're on a 32-bit system.

Sheesh....

[scem0]

Originally posted by MacsRgr8
OMG.....

Can anybody imagine what will happen if Apple weren't to use these procs????? The hype (we all made) around the PPC 970 is so huge....


THis is all I have been thinking about ever since the 970 hyper began..... Even though apple would be insane not to use this, they might not.... ahhhhhhhh don't get my hopes down. :rolleyes: ;)

[bones]

Originally posted by law guy
I'm debating a Dell precision workstation vs. a 1.42 G4... I'm leaning towards the apple, but one could have a dual 3.06 Xeon machine in 9 business days.

Um.. huh? I really don't get this.. Surely the first consideration is "do i want to use XP or OS X?" and that will decide it.

I don't care if DELL has 50Ghz XP workstations while apply only has 5, i'll still get the mac, because i don't want to use Windows.

[bones]

Originally posted by applejilted
Intel rightly considers that 64 bit computing will not take off in consumer systems until at least 2008 and the reason for that is that 4 gb of RAM is prohibitively expensive....t...Say what you want about Intel but nobody ever accused them of being stupid


I read that same article. We all did. I did call them stupid, in an earlier post. As have several other people... Many folks, specifically the likes of IBM and AMD think intel is betting WAY wrong on this. 970 is a consumer chip, and is 64bit, and is coming out this year. Same with AMD's x86-64 offering.

4GB of RAM is not prohibitively expensive. The article says it costs $1k, well yeah, if you buy it as a single module. But I filled up my Dual 800 with 1.5GB of RAM for $150. And that was 2 years ago.

2008? Please. Thats a REALLY LONG TIME. Intel got this one wrong.

Mark this post, come back in 2008.

[mauricev]

Say PowerPC 970 is the next Power Mac processor. The most logical choice is Power Mac G5 even though the chip isn't really the G5.

[beatle888]

Originally posted by eric_n_dfw
It goes in a server room, who cares.




edit



yes it will be in a server room, but the fact remains, its ugly, and so is their site.

[shadowfax]

Originally posted by beatle888
did i ask if you cared? i was only stating a fact.
whats the problem, hard day at work?
but if you aren't interested in other people's opinions, why state your own? who cares about them? honestly, the aesthetic appeal of a computer is all but entirely subjective, so if his opinion doesn't matter, there's no way in hell that yours does. besides, he actually brought up a logical point that doesn't even touch on aesthetics like yours does. you weren't stating a fact. here's a fact. IBM's website is effectively designed, even though you think it's ugly. do you have any notion of the vast amount of information they purvey to the public from that site? IBM is a really stable, pretty successful company, one reason i hope apple switches to them instead of sticking with whatever the heck motorola pulls out their arses.

[beatle888]

Originally posted by Shadowfax
but if you aren't interested in other people's opinions, why state your own? who cares about them? honestly, the aesthetic appeal of a computer is all but entirely subjective, so if his opinion doesn't matter, there's no way in hell that yours does. besides, he actually brought up a logical point that doesn't even touch on aesthetics like yours does. you weren't stating a fact. here's a fact. IBM's website is effectively designed, even though you think it's ugly. do you have any notion of the vast amount of information they purvey to the public from that site? IBM is a really stable, pretty successful company, one reason i hope apple switches to them instead of sticking with whatever the heck motorola pulls out their arses.

hahaha, so i say ibm's servers are ugly like their website....*fact*....and now we're down to motorola sucks :D thats funny.

my opinion will always matter, it's mine :D i can show that box to anyone and bet that they would agree. oh and just so your not too far off track with the whole subjective thing....even the guy that said "who cares" didnt deny that it IS ugly. anyway i dont really care :D i just post here for fun :D

[kenohki]

Don't know if anyone posted this yet. (I'm a little late to the party here.) But, Real World Tech has an article with a nice chart of the SPECint and SPECfp of most contenders in the chip race. Granted, the 970 isn't on there but POWER4 is. There are some very real differences between them (cache size, bus architecture, VMX unit), but they are using the same core at least. Take it for what it's worth but it's interesting and maybe an indicator of how performance of the 970 will compete.

Notice how the 1.4GHz POWER4+ is a nice balance between the FP performance of the 1GHz Itanium and the integer performance of the 3GHz Pentium 4.

http://www.realworldtech.com/page.cfm?AID=RWT012603224711&p=2

[Dont Hurt Me]

Originally posted by kenohki
Don't know if anyone posted this yet. (I'm a little late to the party here.) But, Real World Tech has an article with a nice chart of the SPECint and SPECfp of most contenders in the chip race. Granted, the 970 isn't on there but POWER4 is. There are some very real differences between them (cache size, bus architecture, VMX unit), but they are using the same core at least. Take it for what it's worth but it's interesting and maybe an indicator of how performance of the 970 will compete.

Notice how the 1.4GHz POWER4+ is a nice balance between the FP performance of the 1GHz Itanium and the integer performance of the 3GHz Pentium 4.

http://www.realworldtech.com/page.cfm?AID=RWT012603224711&p=2 Actually the 970 spec is 1050 at 1.8 ghz so that puts next to the p4 at 3.06 and since they are allready testing at up to 2.5ghz that should be on par or better then what intel will have this summer.

[ktlx]

Originally posted by AidenShaw
Is that why you have a 32-bit moniker? ("ktlx" fits in 4 8-bit bytes)

Why on earth would you think that you'd need a 128-bit computer to deal with longer strings?

OhMyGawd - I'd better change my postname - "AidenShaw" is 72-bits, no way anyone will be able to reply to me if they're on a 32-bit system.

Sheesh....

Before you label someone as clueless and write a rude post, you ought to think a bit or at least even try to understand what is going on.

IPv6 uses 128-bit address fields. If your network appliance is acting like a router or a firewall, being able to do bit manipulation on the complete address all at once would mean you could process the address and network information much quicker than if you had a 32-bit or a 64-bit processor. That does not mean it cannot be done but it just is considerably slower and less efficient. By my calculations a 64-bit processor would need to be around four times as fast and a 32-bit processor around 16 times as fast to do routing and filtering compared to a 128-bit processor.

That said, I am not sure the network appliance market is large enough to support a 128-bit general purpose CPU. That is why I said a network coprocessor would probably be a better option in the same way that you have graphics coprocessors.

Next time think before being a prick.

[AidenShaw]

Originally posted by ktlx

[ktlx]

Originally posted by AidenShaw


Do you have any idea how much a custom ASIC costs? And what happens to that development if you make a boo-boo? Very few vendors can afford that development and risk.

That is why network processors are becoming more interesting.

[AidenShaw]

Originally posted by ktlx
Do you have any idea how much a custom ASIC costs?

No, but I can guess based on the prices Cisco charges for their network gear! ;)

http://www.xilinx.com/xlnx/xil_prodcat_landingpage.jsp?title=Virtex-II+Pro+FPGAs

[Frobozz]

Originally posted by bertagert
So where is the PC970 on this board?
Prototype Blade Server with PC970 (http://www-5.ibm.com/de/pressroom/cebit2003/i/highres/bladeprototype_300_Dpi.jpg)

Right underneath the suspiciously large heat sink.

[Frobozz]

Originally posted by bertagert
I want to see what one looks like. Anyone have a link to a picture of the 970? Or, get your photoshop out and lift that heatsink off so i can take a peak.:D

Edited for spelling

Just type "IBM PowerPC 970" into Google. I came up with this:

The page URL (http://www-3.ibm.com/chips/products/powerpc/newsletter/dec2002/newproductfocus2.html). And the image:

http://www-3.ibm.com/chips/products/powerpc/newsletter/dec2002/images/970modagr.jpg

[backspinner]

it looks like this (using transparant heatsink mode)

[backspinner]

- on the left is the processor under the heatsink
- below it is a very large and expensive Xilinx programmale chip, which acts as the memory interface in this design (in later stages, these normally will be finalized in cheaper asic chips)
- above the heatsink the voltage regulators for the processor core
- under the harddrive a Lucent io chip
- in the middle two legacy PCI slots
- at the right two (serial) PCI-X slots with broadcom controllers

[Pedro Estarque]

I don't know much about 32, 64, 128 bit but I thought that one of the differences between then was that because it processes information in 64-bit chunks, the 970 would process more information per cycle then a 32 bit ppc.
Isn't that the reason why altivec instructions are so much faster then normal ones?

[pgwalsh]

the 970 will process 32 bit and 64 bit. My understanding is that 32 bit apps will run throug 32 bit mode. It wont see any advantage of 64 bit. So most apps, unless altered, won't see any advantage. However, it' s good for server stuff since most - to my understanding - is done in 64

[Pedro Estarque]

yes I know, but there are people saying we will never need 128 bit addressing, because we will never need that much memory. But what about performance?I think a 128 bit chip with OS and applications optimized for it would be much faster then 32 bit just because it processes information in a bigger chunk per cycle.

[cr2sh]

Every thread on the 970 degenerates into this:

"32 vs 64, some apps will run...."

then it goes to

"Will osX be able to?"

and then the code-monkeys come out...

"Yes but it will need recompiled.. or blah."

But it's always fun to watch the progression. :p

[Frohickey]

Originally posted by beatle888
yes it will be in a server room, but the fact remains, its ugly, and so is their site.

Beauty is in the eye of the beholder.

[Frohickey]

Originally posted by backspinner
it looks like this (using transparant heatsink mode)

In transparent aluminum heatsink mode, you would be barely able to see the heatsink.

Geez... and I travelled millions of miles.... err thousands of miles to tell you this. :p

[Dont Hurt Me]

Originally posted by cr2sh
Every thread on the 970 degenerates into this:

"32 vs 64, some apps will run...."

then it goes to

"Will osX be able to?"

and then the code-monkeys come out...

"Yes but it will need recompiled.. or blah."

But it's always fun to watch the progression. :p No kidding the 970 will run 32 as well as 64 no emulation needed . 32 Native. yes it will be faster beacause of cpu speed, more instructions in and out per clock cycle, wider front side bus. Just thought i would say that for all those 32 -64 huh? people.

[beowolf]

To quote this page, http://www-3.ibm.com/chips/products/powerpc/newsletter/dec2002/newproductfocus2.html , on the sixth paragraph, "In addition to its support of new 64-bit solutions, the 970 retains full native support for 32-bit applications. This not only protects 32-bit software investments, but provides these 32-bit applications with the same high-performance levels that it extends to 64-bit uses. This native, nonemulated, 32-bit support is not limited to application code, which runs unmodified. 32-bit operating systems with minor updates can also take advantage of the PowerPC 970's outstanding performance." This is what arn was saying when he said the 970 would give 100% compatability, and, from the look's of this, something more since 32 bit will get the same high performance that 64 bit gets. Nice

One other thing of note, on this page http://news.zdnet.co.uk/story/0,,t269-s2131244,00.html?rtag=zdnetukhompage
in the 9th paragraph they say "Sources familiar with IBM and Apple's plans said that Apple will be a customer for the PowerPC 970 next year, but Apple has declined to comment." Is this of any important note coming from zdnet?

Phew, that was a long first post ever. Hi.

[beowolf]

On that zdnet page I linked to, it pointed out this, "A 1.8GHz PowerPC 970 dissipates 42 watts, according to IBM's figures, compared to 68.4 watts for a 2.8GHz Pentium 4 and 30 watts for a 1GHz G4e chip from Motorola. The more heat a chip dissipates, the more space and equipment is required for cooling." Does this mean a lower clocked 970 (1 GHz-1.3 or 1.4 GHz) could possibly run in a powerbook with a slightly updated design off the bat? Or maybe Stevo will try and go for a home run and update across the line to 970 (VERRRYY doubtful in my mind, but would be awesome none the less). Can anybody add to this thought (i.e. economic benifits of only selling puters with one chip to support?). Thanks

[Foxer]

Originally posted by law guy
What I find discouraging is that I just called the Apple store in Tysons Corner, VA and they tell me at least four (4) weeks until they begin to get 1.42 inventory. So, we won't even be seeing the 1.42 G4 machines until April

According to the Apple store, mine will be built next week. Hopefully his info is better than Tysons.

[shadowfax]

Originally posted by beowolf
This is what arn was saying when he said the 970 would give 100% compatability, and, from the look's of this, something more since 32 bit will get the same high performance that 64 bit gets. Nice
...
Phew, that was a long first post ever. Hi.

I don't know just what they mean by taking advantage of the 64 bit performance--it could just be some propagandized sort of thing exaggerating the fact that it will still run 32 bit apps &c. i'd be surprised if it did something like putting 2 32 bit instructions in a single operation to enhance performance. i suspect these will just be running 32 bit apps about as well as a 1.8-2.5 GHz G4 would. i could be off though :).

anyways, welcome to the forums; it's always good to start off relevant and backed by sources right up front!

[shadowfax]

Originally posted by beowolf
On that zdnet page I linked to, it pointed out this, "A 1.8GHz PowerPC 970 dissipates 42 watts, according to IBM's figures, compared to 68.4 watts for a 2.8GHz Pentium 4 and 30 watts for a 1GHz G4e chip from Motorola. The more heat a chip dissipates, the more space and equipment is required for cooling." Does this mean a lower clocked 970 (1 GHz-1.3 or 1.4 GHz) could possibly run in a powerbook with a slightly updated design off the bat? Or maybe Stevo will try and go for a home run and update across the line to 970 (VERRRYY doubtful in my mind, but would be awesome none the less). Can anybody add to this thought (i.e. economic benifits of only selling puters with one chip to support?). Thanks

i think when they pop the .09 micron design into the PPC 970 it will take considerably less power, which looks even better for powerbooks, provided the chips are cheap enough.

[Frobozz]

Originally posted by Pedro Estarque
I think a 128 bit chip with OS and applications optimized for it would be much faster then 32 bit just because it processes information in a bigger chunk per cycle.

Well, I'm no chip architect, but I believe that also means it'll waste more time and become less efficient if you can't fill up those chunks.

[beowolf]

I was letting my mind chew on this while feeding cows and had to wonder. A lot of people on here keep saying that they average joe isn't going to need a 64 bit computer. Well, from what I remember from senior economics in high school (just last year), a lot of getting somebody to buy something is both percieved need and want. Well, I'm pretty certain that the want is going to be there (hey, it's bigger, better, and faster! mindset) and I think that marketing should be able to make people PERCIEVE that they need it, whether they really do or not. Just my $.02

[shadowfax]

Originally posted by beowolf
I was letting my mind chew on this while feeding cows and had to wonder. A lot of people on here keep saying that they average joe isn't going to need a 64 bit computer. Well, from what I remember from senior economics in high school (just last year), a lot of getting somebody to buy something is both percieved need and want. Well, I'm pretty certain that the want is going to be there (hey, it's bigger, better, and faster! mindset) and I think that marketing should be able to make people PERCIEVE that they need it, whether they really do or not. Just my $.02

the want is going to be there, that's certainly given. whether or not it will be affordable enough to convince people to succumb to their desire is much more sketchy.

[eric_n_dfw]

Originally posted by beowolf
I was letting my mind chew on this while feeding cows and had to wonder. A lot of people on here keep saying that they average joe isn't going to need a 64 bit computer. Well, from what I remember from senior economics in high school (just last year), a lot of getting somebody to buy something is both percieved need and want. Well, I'm pretty certain that the want is going to be there (hey, it's bigger, better, and faster! mindset) and I think that marketing should be able to make people PERCIEVE that they need it, whether they really do or not. Just my $.02
What can kill the "want" is a lack of applications that show off the advantage of 64 bits. All it would take would be an x86 32Bit machine coming close to the same performance on video redering or something and the 64bit marketting bubble would pop.

[Catfish_Man]

Originally posted by Shadowfax
I don't know just what they mean by taking advantage of the 64 bit performance--it could just be some propagandized sort of thing exaggerating the fact that it will still run 32 bit apps &c. i'd be surprised if it did something like putting 2 32 bit instructions in a single operation to enhance performance. i suspect these will just be running 32 bit apps about as well as a 1.8-2.5 GHz G4 would. i could be off though :).

anyways, welcome to the forums; it's always good to start off relevant and backed by sources right up front!

Nope, it should be about twice as fast (maybe a bit more) per clock than a G4 (for fp operations anyway). This is entirely unrelated to 64 bitness, which actually is a slight performance loss, not a gain, for most tasks (64 bit pointers take up more cache space). The extra power comes from a fast bus, out of order execution, big caches, and a second floating point unit.

In reply to the 128 bit posts. It wouldn't allow 4 32 bit chunks (that's what Altivec does) it would allow one 128 bit chunk. AFAIK, no one needs numbers THAT big (2^128).

[freemidnight]

Going to the average rumors here about speed increase compared to the real facts, this is still Blah Blah Blah! and for months and months to come!

This is becoming pathetic!:rolleyes:

[Catfish_Man]

Originally posted by freemidnight
Going to the average rumors here about speed increase compared to the real facts, this is still Blah Blah Blah! and for months and months to come!

This is becoming pathetic!:rolleyes:

I'm afraid I can't quite figure out what you're saying. Could you clarify it?

[shadowfax]

Originally posted by Catfish_Man
Nope, it should be about twice as fast (maybe a bit more) per clock than a G4 (for fp operations anyway). This is entirely unrelated to 64 bitness, which actually is a slight performance loss, not a gain, for most tasks (64 bit pointers take up more cache space). The extra power comes from a fast bus, out of order execution, big caches, and a second floating point unit.
right. the processor itself shouldn't be faster. if and when the G4 gets a DDR bus, and if you could put that bus at 900MHz and clock the processor at 2.5 GHz, i doubt there would be much of a speed difference. the speed difference isn't intrinsic within the processor, it's a change because of the bus and cache so on... the only internal advantage you mentioned was the 2nd FP unit.

[shadowfax]

Originally posted by eric_n_dfw
What can kill the "want" is a lack of applications that show off the advantage of 64 bits. All it would take would be an x86 32Bit machine coming close to the same performance on video redering or something and the 64bit marketting bubble would pop.

hehe, unless they cost the same (or so). if you could get a 64 bit processor that performed about as well as a 32 with 32 bit apps for about the same or slightly more money, wouldn't you? it's like insurance against the future. it there were to come out a 64 bit app you needed and you went for the 32 bit x86 just because you didn't think you needed the 64, wouldn't you feel like an idiot? that said, they really are going to need some good apps to run on the 64, as it's probably NOT going to be of comparable price to a cutthroat pentium.

[astrocity20]

well whenever we see new apple computers is when I'll believe. Hopefully they'll have something good for us whether that be these or soemthing else.

[freemidnight]

Originally posted by Catfish_Man
I'm afraid I can't quite figure out what you're saying. Could you clarify it?

Simple The announcement of new processors that are contender to the Mac platform are still in "announcement"

No g5, no 970 here. Nothing.

Yes there are alot of clues about new processors but we still talk and talk,

Hope it is not too hard to understand!:)

I want to have "real" and palpable thing.

I know that, from a rumor site, this is hard to understand from certain peoples but please read about the original subject of this post!

Yes I know that it could go to 2,5 Ghz, nice! but when? When for Apple? You would speculate fot one to two years about that!!! without having a single GHZ of that processor at hand!

Hope you understand my point.


You'll tell me "Hey this is a rumor site"

Well this is it!:D :D :D

[beowolf]

ummmm, while you bring up a very valid and serious point, do you REALLY think that Apple would let this happen ANYTIME before Stevo announced them in new hardware? My thought on this is no. So what does that leave us with? blah blah blah blah. Why? Because Apple doesn't preannounce stuff like that. This policy is probably the single biggest reason Apple rumor sites exist, because things aren't announced ahead of time and ARE only rumor/speculation. As for how valid of rumor this is, well, I'm certain that this one will come true (either that or moto is doing one hell of a job of laying a false trail about the demise of the mythical G5...a bit less likely since where there is bread there are crumbs).

The question about it coming true is WHEN? Can this happen too soon? No. Can this happen too late? Yes. I've heard it said, and can't reference this and I apologize for it, that Steve claimed this would be the year Apple caught up with the wintel crowd in terms of speed. I hope this is true. I read today that zdnet said they had information saying that the PPC 970 would be used in Apple computers later this year. I hope this is true. I know (although I can't reference this) that Steve said this would basically be a breakout year for the mac. I hope this is true.

The bottom line is that I HOPE it is true and a fair number of people who read this thread do also (I think it has been viewed over 18k times). All Apple has ever really given us is hope and the promise that they are an amazing computer company.

Oh, and one more thing.....;)

[Stormcrow]

[QUOTE]

This is just it. When will these processors make it into a mac, if at all? Then, if they do, when will they make it into a mac I will buy? They need to make it into macs and in a hurry. Every time I use a mac, they are sooooo slow (the last time I used one was today, a dual 1.25 ghz, 1 gb of RAM). I'm not talking complicated tasks either. The simple act of launching an app seems like it takes forever. Editing a movie on iMovie seems never ending because each click seems like it takes 2 or 3 seconds to peform the task, like bringing up a clip, adding an effect, etc. Getting on the internet seems as slow as molasses, even with Safari. Even the encyclopedia is slow. Im not just perceiving this. There's no mhz myth making everything go in slow motion in my head. This is my actual experience.

Now, dont get me wrong. I'm not trying to start a flame war. I think Apple products are great. It has great software and great hardware, IMO. That's the sad part really. It would just be awesome if the speed of the chips matched the quality of the products. My main computer right now is a Gateway with a 2.2 ghz p4 with 512 mb of RAM. It is no where near the quality of a Mac with regards to both hardware and software, IMO. However, when I click or press a key to perform a task, the response is usually immediate. Like flipping on a light switch. Everything just seems so much snappier on my Gateway, from menus, to effects, to application performance.

BTW, I know some may say, who cares about having the fastest chip when you can do everything you need to do with a slower one? True, for now. I dont live in a time rift where now is forever. I want my machine to be useful for at least a few years. As time passes, applications (both mac and windows apps) get more complex and bulkier. I want the machine I get today to be able to handle those apps with good performance. I got my Gateway this time last year. The 2.2 ghz p4 had just come out. It was the fastest chip around. Now, 2.2 ghz is middle of the road in terms of performance for new computers (maybe on the slower side of middle), but still does everything I need it to do in terms of performance. I will tell you that I can notice a difference, though, when I work with a 3.06 ghz machine. The point is, I want a machine that can still perform with the passage of time.

I also want Apple to do well. We need Apple as a competitor. Plus I think Apple makes better products. I want a mac. I just might wait a little longer to get one in hopes that they get faster. No, Im not waiting for the PPC 970, I'm just waiting for macs to get faster, no matter what chips they use. Just my thoughts.


'

[Stormcrow]

Originally posted by beowolf
ummmm, while you bring up a very valid and serious point, do you REALLY think that Apple would let this happen ANYTIME before Stevo announced them in new hardware? My thought on this is no. So what does that leave us with? blah blah blah blah. Why? Because Apple doesn't preannounce stuff like that. This policy is probably the single biggest reason Apple rumor sites exist, because things aren't announced ahead of time and ARE only rumor/speculation. As for how valid of rumor this is, well, I'm certain that this one will come true (either that or moto is doing one hell of a job of laying a false trail about the demise of the mythical G5...a bit less likely since where there is bread there are crumbs).

The question about it coming true is WHEN? Can this happen too soon? No. Can this happen too late? Yes. I've heard it said, and can't reference this and I apologize for it, that Steve claimed this would be the year Apple caught up with the wintel crowd in terms of speed. I hope this is true. I read today that zdnet said they had information saying that the PPC 970 would be used in Apple computers later this year. I hope this is true. I know (although I can't reference this) that Steve said this would basically be a breakout year for the mac. I hope this is true.

The bottom line is that I HOPE it is true and a fair number of people who read this thread do also (I think it has been viewed over 18k times). All Apple has ever really given us is hope and the promise that they are an amazing computer company.

Oh, and one more thing.....;)

I hope this is true to, beowolf. But still, even later this year seems like a very long time. Additionally, if Apple uses the 970, I'm concerned that it will start with slower chips than it can get. I am concerned that Apple will make it's predictable 200 mhz jump, and of course, the upgrade will not be across the board. I'm afraid that we will see the 970 in one, maybe two top of the line powermacs. Everything else will probably get speed bumped g4s. I dont want it to happen this way, but I'm afraid that it will happen something like I just described. Could you imagine if Apple put a 2 ghz 970 in an iMac or Powerbook within the next few months? I thing a lot of people, including me, would be stumbling all over themselves to buy one. This wont happen even if Apple does use the chip because both computers were recently updated. For all its effort in keeping its new products secret, Apple is sure beginning to be very predictable. Just my thoughts.

[phampton81]

Every time I use a mac, they are sooooo slow (the last time I used one was today, a dual 1.25 ghz, 1 gb of RAM). I'm not talking complicated tasks either. The simple act of launching an app seems like it takes forever.
I dont mean to argue or say that you are flat out wrong, but I work day to day on a 700mhz g4 and each app lauchees just as fast as my AMD, and as far as newer macs go, with my experience they are blazing, my ONE experience with a 1.25 powermac was awesome. All apps launched nearly immediately and responsiveness was instantaneous. Now, do I think Macs could be faster.....yes, there is always room for improvement, but when you say that the newest macs are 'soooooo slow' then I just have to step in and say "NO".

[Stormcrow]

Originally posted by phampton81
I dont mean to argue or say that you are flat out wrong, but I work day to day on a 700mhz g4 and each app lauchees just as fast as my AMD, and as far as newer macs go, with my experience they are blazing, my ONE experience with a 1.25 powermac was awesome. All apps launched nearly immediately and responsiveness was instantaneous. Now, do I think Macs could be faster.....yes, there is always room for improvement, but when you say that the newest macs are 'soooooo slow' then I just have to step in and say "NO".

Ok. Fair enough. I acknowledge the fact that people will have different experiences, and I am actually glad to hear that your experience has been different. I WANT to be proven wrong. I have no axe to grind.

The only question I would ask is have you used a PC at 2 ghz or above to compare your experience? My thought here is that for those mac users who DO NOT use PCs at all or that much, they have no real point of comparison. So, a 700 mhz g4 will certainly be much faster that an equivalent g3 or slower g4. A 1.25 mhz g4 will be faster than a 700 mhz g4 and so on. To the mac user each increase will generally make everything you do faster, and the user will seem to blaze compared to slower macs. I think you get my point.

If you or someone else who reads this post uses PCs and can say, "I use PCs and my ___ mhz mac is just as snappy and fast as the ___ mhz PC that I use or have used, I say GREAT!! Let me know about it. Like I said before, I WANT to be proven wrong. Take it easy.

[MacBandit]

Originally posted by Stormcrow
Ok. Fair enough. I acknowledge the fact that people will have different experiences, and I am actually glad to hear that your experience has been different. I WANT to be proven wrong. I have no axe to grind.

The only question I would ask is have you used a PC at 2 ghz or above to compare your experience? My thought here is that for those mac users who DO NOT use PCs at all or that much, they have no real point of comparison. So, a 700 mhz g4 will certainly be much faster that an equivalent g3 or slower g4. A 1.25 mhz g4 will be faster than a 700 mhz g4 and so on. To the mac user each increase will generally make everything you do faster, and the user will seem to blaze compared to slower macs. I think you get my point.

If you or someone else who reads this post uses PCs and can say, "I use PCs and my ___ mhz mac is just as snappy and fast as the ___ mhz PC that I use or have used, I say GREAT!! Let me know about it. Like I said before, I WANT to be proven wrong. Take it easy.

My question is simple. How can an app open faster then instantaneous?

[Stormcrow]

Originally posted by MacBandit
My question is simple. How can an app open faster then instantaneous?

I guess the meaning of "instantaneous" could be debated, but the short answer to your question is that it cant open any faster. I guess some of this would depend on how many other apps you have open and what other processes you're doing at the same time. That is where I have really noticed a speed difference.

Anywho, I am glad your experience has been better than mine. But, if I am going to purchase a mac, I have to go by my own experience, not by what others say. Additionally, the main thing I'm trying to illustrate is that I believe I represent a number of Windows users who would break down the doors to buy a mac, if they had faster processors. You also have to remember that many Windows users experiences with a mac are at CompUSA, and the macs at all the CompUSA stores are as slow as Christmas. Happily, I have used macs more extensively than just a few minutes at CompUSA, and I know they aren't that slow. However, many windows users have not used macs more than the few minutes they tinker with them at CompUSA. The perception of slowness that this few minutes inculcates and confirms in the mind of Windows users is hard to overcome.

But I digress. IMO, there is no denying that if Apple put a 1.5-2.0 ghz 970 chip (or higher) in their powerbooks, the same in their iMacs, and 2.0-2.5 ghz (or higher) in their powermacs, they would get a flood of customers. I know that there are many permutations of how this move would eat sales, and how they just updated everything, blah, blah, blah... But, such a move would be surprising and noticeable. I think it would have at least two very good effects. First, people would think that apple has some fast machines, check them out and some would actually buy the machines. Second, it would tell people that Apple is moving fast to give customers what they want--Apple wants to give us the best and fastest products it can as soon as it can. No more disappointing 200 mhz increases every now and again. I guess a third, related effect is that it would throw everyone off. People would not expect it. Apple would be able to bolster its mystique. It would be good stuff.

[Stormcrow]

Originally posted by Stormcrow
You also have to remember that many Windows users experiences with a mac are at CompUSA, and the macs at all the CompUSA stores are as slow as Christmas.

I meant to say that the macs at all the CompUSA stores I have been to are as slow as Christmas.

[Dont Hurt Me]

If apple really wants to turn heads and sell computers then stormcrows post about a 970 imac and across the line is correct. If tomorrow they announced we are now selling 970 cpu's in every model they would be flooded with buyers. I would be scrambling as fast as i could to get my 970 imac. But if all we get are motobumps well then ill take my time and get a motobump upgrade for my powermac and apple looses another sale. If they are to go from 3% to 5% this year the only answer i see is the 970's everywhere except for emac or ibook.

[os10geek]

And how is something slower than Christmas? And is my birthday any faster?:D

[DavPeanut]

People keep saying that they would not update to 970s accross the board because everything was updated recently. That's true now, but it won't be true in July.

[nuckinfutz]

If you think 970's are going to hit the iMac soon you're severely mistaken.

I would be scrambling as fast as i could to get my 970 imac. But if all we get are motobumps well then ill take my time and get a motobump upgrade for my powermac and apple looses another sale. If they are to go from 3% to 5% this year the only answer i see is the 970's everywhere except for emac or ibook.

This is thinking like a Consumer. You know what Apple is thinking. They're thinking that they already KNOW you want a 970 in an iMac but they're going to put it in the Powermacs only. Do they care that you won't buy....no really. There is a finite amount of 970 CPU's that they will have and they're looking to ensure that they profit as much as they can from these chips. The G4 still has life and it's a know quantity. I'd take a 1.2Ghz G4 imac for $1299 in a heartbeat.

I think I'm reading too many expectations of Apple that will only cause you to be pissed off when reality hits. Even PC users have their reality. You don't buy Xeon based PC's for the price of a Consumer PC.

Folks the PPC 970 IS a Xeon Class processor. Thoughts of it in an iMac are far too premature.

[Dont Hurt Me]

I hear you nuckinfutz on the 1.2 imac i have thought of it also. But I dont expect to see a 1.2 ghz imac for $1299, maybe a 1 giger 15" this fall. The 970 was actually made to be a consumer entry of the power4. Just turns out that it can outperform the power 4 due to many factors and is a lot cheaper to make. But if you are going to mac from 3%market to 5% you are going to have to do something very big. Just using the 970 in the powermac line alone to me wont make that happen. It will take more

[nuckinfutz]

Ahhhh what the hell do I know :(

I'm hoping Apple gets aggressive DHM.

Hell my "hopes" tell me to think 970 everywhere but my sensibilitis won't let me get too giddy.

Here's hoping for the best which is 970's wherever they fit!

[wms121]

someone asked me (this is my first MacRumors post..thankyou..thankyou)...could IBM make the 970/POWER4 series...the "Book F" ppc processor..and insert it into the Motorola
architecture for the current 8540/8560 series?

Hmm..maybe..but the bus speed might be prohibitive. And...

the 8540/8560 is set up for Rapid IO busses..NOT the Infiniband
bus fabric switch which definitely Apple may need post 2005 for its server series...any MOto engineers out there who want to defend the current "7550,7540..and 7560 64-bit series" for Rapid IO..please speak up now..or wait till Steve formally strangles you.

<--embedded developer in Texas

(check this out...next stupid question I was asked.."Could IBM
produce a 'teraFlop chip' out of the 970 series..?" ) Hmm..possibly
IF a) they did a second dual-mode version i.e. 4 code cores per
cross buss
b) they also planned to upgrade to 256 bit floating
point...with 1024 bit internal busses (like TransMeta did)
c) this new chip clocked in above 5.5 GHz


~2.307 TeraFlops ( 14.4GFlops x 4 x 16 x 2.5 ) in 32/64 bit apps

yes..please check my math..this may be too slow..

[DavPeanut]

Who needs a teraflop? Ok, don't answer that. Pixar would need ten of them. Anyway, what kind of prices would these machines go for? The processors are not too expensive, I know, but that about everything else? Would they have SATA and PCI-X and that kind of stuff?

[whooleytoo]

Just a few points:

- Will Apple use this processor? 99.9% certainly yes! :-) The fact that IBM have gone to such effort to add a SIMD unit, plus the fact that it's labelled Altivec virtually guarantee it was created with Apple's needs in mind. (Plus, the fact that Apple's alternatives aren't really compelling!)

- Does 64 bit mean faster? Quite apart from the extended memory abilities, it should also give a performance boost, though mostly in very compute-intensive work. (As IBM states in the pdf above..)

- Why do Motorola "suck"? Dunno, but it might have something to do with the major rift between them and Apple when Apple pulled the cloning licenses after Motorola had invested a lot in it's Mac clone development!

Mike.

[phampton81]

plus the fact that it's labelled Altivec virtually guarantee it was created with Apple's needs in mind.
I do agree that the use of the term Altivec says a lot, and that if IBM was not planning on selling this CPU to Apple then there would be no reason to even use that term. So with that in mind, I am confident we will see this proc. in macs as soon as Apple (and IBM of course) is ready.

[Dont Hurt Me]

Still no one has answered my market share question. can a 970 alone in the powermac line this summer move 3% to 5%? i dont see it. I think they would need more. The question becomes where will they use it and when. powermacs this summer only just wont make up to 5% marketshare in my opinion. to make up for lower and lower marketshare i think they would have to get aggressive about this.

[MrMacMan]

Um... I'm sorry if this was posted but I want to look at the IBM press release not someone telling me what it was.

I clicked on the link when arn posted it but I didn't download it.

If anyone has it please post it.

[whooleytoo]

Originally posted by whooley
Just a few points:


One more point :-)

- Do we need a faster Mac? Yes. YES. YESSS!! I want to play all my games at higher resolutions, with more complex graphics, at a higher framerate.

I want realtime iMovie effects, not 10 to 20 second delays. I want to be able to rip a CD in seconds, not minutes. I want more CPU cycles to devote to worthy projects like Folding@Home.
I want some kind of parity between Mac and PC speeds such that porting from PC to Mac (especially for games) isn't a tough choice.

But mostly, I want a faster Mac for all the applications that haven't been written yet because it wasn't thought to be possible on a destkop machine!

Mike.

[Dont Hurt Me]

Originally posted by whooley
One more point :-)

- Do we need a faster Mac? Yes. YES. YESSS!! I want to play all my games at higher resolutions, with more complex graphics, at a higher framerate.

I want realtime iMovie effects, not 10 to 20 second delays. I want to be able to rip a CD in seconds, not minutes. I want more CPU cycles to devote to worthy projects like Folding@Home.
I want some kind of parity between Mac and PC speeds such that porting from PC to Mac (especially for games) isn't a tough choice.

But mostly, I want a faster Mac for all the applications that haven't been written yet because it wasn't thought to be possible on a destkop machine!

Mike. There are a lot of us that feel this way,over 22 thousand views now tell me apple is it time to replace the g4? I think its a big YES!

[beowolf]

I know one thing for sure though, if Apple were to update across the board like was mentioned here, I think that EVERYBODY would be surprised, even if it were a rumor most wouldn't believe it because it would seem too good and impossible to be true. Although how they would keep up with demand would be amazing...IBM's new fishkill, ny plant might go a long ways towards taking care of that problem. Maybe if they used this chip in enough puters the price would come down noticibly (esp since it is expected to be easier to make than the G4 and have higher yields)?

[beowolf]

I recall reading a little while back somebody saying that the bus speed of 900mhz was not based on 1/2 the processor speed. To correct them (with a reference none the less!) this page http://www.arstechnica.com/wankerdesk/3q02/powerpc.html
has this

"One of the most important and least-discussed features of the PowerPC 970 is its 900MHz DDR frontside bus. This bus physically runs at 450MHz, but it's double-pumped. Its architecture is interesting in that the bus actually consists of two, 32-bit unidirectional point-to-point links. David Wang described it in a post to comp.arch as follows:

It's two 32 bit links: one from CPU to "companion chip" [the northbridge], and one back from that chip to the CPU. Each link runs at 900 MHz (1.8 GHz CPU core. the interface link runs at integer fraction of the CPU core, in this case 1/2)

So 4 bytes to, 4 bytes from, at 900 MHz that's 3.6 GB/s raw BW each way. The link multiplexes command and address info over the same pins, so it's some sort of packet based protocol. The math gets you 7.2 GB/s of raw bandwidth, but after subtracting out command and address overhead, raw peak data bandwidth is supposed to be about 6.4 GB of that 7.2 GB/s.

This high-bandwidth link to the northbridge is one of the elements that's going to make this chip as a media machine; it's exactly what Apple's current bandwidth-starved G4 systems lack, and it's going to be a major selling point for systems based on the PPC 970. When coupled with the right memory in an SMP configuration, the 970 should do quite well in bandwidth-intensive applications."

So it appears that the bus speed is 1/2 the processor speed (which makes me wonder about motherboard design issues with the bus speed being different for each processor. Does anybody know something about this?).