CONFIRMED IBM Power PC 970 - Page 4

[quote]Originally posted by moki:
<strong>
Wait until you see the benchmarks from this processor; I think you may be quite impressed.</strong><hr></blockquote>

Do indeed listen to the man.

Intel has been working hard for years to make the Itanium very fast, but IBM has been working harder for longer, as far as I'm concerned. IBM has been making supercomputers for a long time. They know how to push around big chunks of data fast.

And regardless of whether it's single core, I'm sure it can multiprocess quite well.

[quote]Originally posted by rogue27:
<strong>The processor speed has to be a multiplier of the bus speed, generally in 0.5 increments.

a 1.8Ghz 970 would be using a bus multiplier of 2 on a 900Mhz bus.

That is the fastest the chip will be. If you use a bus multiplier of 1.5, you end up with 1350Mhz. I don't know if Apple will even dare to ship a PowerMac at that clock speed by the time this chip is available.</strong><hr></blockquote>

Drawing conclusions from sketchy clues may lead you astray.*

This is a press release. It is written by marketing, not engineers.

While it says 900MHz, could not it be DDR 450MHz or QDR 225MHz?

There is no reason that multipliers HAVE to be in increments of 0.5. It's just nice for syncing data. 0.25 would probably be possible.

This could be the beginning of a new approach.

The multiplier of EVER chip is locked at 2x, and different bus speeds are used instead. IBM did say "up to" 900MHz in the press release.

1.8GHz CPU, 900MHz bus, 6.4GB/s, dual-channel DDR 400.

1.5GHz CPU, 750MHz Bus, 5.3GB/s, dual-channel DDR 333

It could be that the real bus is 450MHz, with multiplier increments of 0.25x and or different busses are used.

If the speculation that the PowerPC 970 uses ApplePI is true, then you can bet your arse that the solution will be a neat one.

----------------

The bus speed is 900MHz, yet data transfer is 6.4GB/s. Sounds like 1Hz in every 9 is used to address the data, and the other 8Hz is the actual data. This is how I got my 750MHz bus, which is 1 and 1/8th of 666MHz.

----------------

BRussell, as Programmer said, you're mistaken.

----------------

Barto

*Obscure quote. I doubt anyone will know it <img src="confused.gif" border="0">

Also, I doubt someone from IBM is going to be the most unbiased source on the performance of this chip.

Barto

[quote]Originally posted by moki:
<strong>If you assume that the Wintel world will be moving over to the Itanium, it might interest you to know that the current speed of the Itanium 2 -- which scores quite well on the benchmarks, btw -- is a palty 1ghz.

The people in the Wintel world who have been using ghz ratings as a measure of their penis size are going to be in for a rude awakening when the Itanium is pushed as mainstream. They very well may be forced to admit that there is something to this whole "mhz myth" after all.

One other thing to keep in mind is that the GP-UL is not just a stripped down, single-core POWER4 processor -- it isn't valid to assume that the GP-UL will simply be a scaled down POWER4 from earlier this year, with no enhancements whatsoever.

Wait until you see the benchmarks from this processor; I think you may be quite impressed.</strong><hr></blockquote>


I love it when you talk like this... you tease

Some notes. L2 cache size is 512KB and there is an included L3 cache controller. Support for L3 cache is either 4MB or 8MB. But you can elect to not include L3 in your designs.

Also to mention some things about bus speeds, the Power4 has a fixed bus speed ratio, 3:1. the 1.1GHz and the 1.3GHz each have different bus speeds: 366 and 433MHz respectively. there is no reason to suspect the GPUL will be any different, but with a lower ratio: 2:1. I imagine a processor to start out like this:

1.4GHz with 700MHz bus
1.6GHz with 800MHz bus
1.8GHz with 900MHz bus

This way a bus can better scale with a small increase in core speed. A 1.8GHz machine will have much better performance than a 1.4GHz despite there only being 400MHz of difference. Contrast this with the x86 world where a 2.2GHz P4 doesn't feel significantly faster than a 1.8GHz P4 if they use the same memory and bus speed (400MHz).

And remember RAM speed does not have to match bus speed or bandwidth. It would be perfectly alright for apple or IBM to use cheaper PC2700 DIMMs in their designs instead of more expensive 400MHz memory modules with dual channels and such.

The cat's outta the bag now. Now does everyone understand why Apple will be selling PPC 970 Towers sooner than a year from now? Sales are going tank, and remain tanked until any Mac lover can walk down to an Apple store or ChumpUSA and plonk down the bucks to walk home with a GPUL Tower.

Announcement Jan or Feb, shipping in quantity by April.

And no, this is not pure speculation. I have it on good word that the PPC 970 is on an express freight train to your doors. IBM's new chip foundry is going to be running at full production by Jan 2003, so cushion that date with some lag time for Apple to slap those little CPU's onto their totally bad-ass PPC 970 muthaboards, and you've got yourself a date.

So, when does this damn IBM presentation finally start at the Microprocessor Forum
It's in SanJose, Oct 15, but I don't know the exact timing.

Does anyone know ?
How is media coverage ? Does anyone know which Mac media attend and or cover the session ?

I guess the speaker will keep repeating 'no comment' whenever the word 'Apple' is mentioned in the Q&A session anyway, but I hope we get some more details from the presentation.

I keep hitting refresh...

The conference starts at 8:30 AM PST, and IBM is the first speaker after the keynote. Probably get a brief report around 10:00 PST, and a full story after between the PC and Server sessions (or during the VIA presentation, *shrug*).

Here's excerpts from the pre-conference report filed by EE Times (accessed from <a href="http://www.siliconstrategies.com/story/OEG20021014S0059" target="_blank">http://www.siliconstrategies.com/story/OEG20021014S0059)</a>

[quote]
<strong>
IBM Corp. will detail at the Microprocessor Forum here its first 64-bit PowerPC microprocessor, a 1.8-GHz CPU that's expected to power next-generation Macintosh computers. Analysts expect Apple could roll out 64-bit Macs starting late next year, turning up the heat on Intel Corp. to extend its flagship Pentium processor to 64 bits. ...

Essentially a derivative of the company's Power4 microprocessor, IBM's PowerPC 970 adds 64-bit PowerPC compatibility, an implementation of the Altivec multimedia instruction-set extensions and a fast processor bus supporting up to 16-way symmetric multiprocessing.

The CPU is expected to spark debate over whether the desktop is ready to break through the 4-Gbyte addressing of 32-bit architectures. Whatever the answer, the 970 would give Apple Computer Inc. a chance to deliver high-performance 32-bit systems that could later be upgraded to full 64-bit computers. ...

Apple declined to comment last week, but a source close to the company said the computer maker indeed will use the 970. The device could also show up in communications gear and in entry-level servers from IBM and other OEMs.

"Apple would have to be crazy not to use this part," said Peter Glaskowsky, editor-in-chief of the Microprocessor Report, which hosts the annual Microprocessor Forum. "Its performance will be in the upper reaches of any CPU. Apple would be able to produce for the first time machines that not only have great performance but support full 64-bit addressing." ...

In terms of die size, a rough measure of cost, the PowerPC 970 measures 118 mm^2, against 131 mm^2 for the Northwood 2.X-GHz Pentium 4. Both the IBM and Intel parts are being made in 130-nanometer CMOS on 300-mm wafers.

Analyst Brookwood noted that the 970 represents a return to the Mac market for IBM after a two-year lull. ...

IBM's approach to implementing a 32/64-bit architecture appears straightforward. The 970 supports full 64-bit registers and addressing. When a flag bit is sent it triggers a 32-bit mode in which the high-order words on an arithmetic logic unit and on memory addresses are ignored. In either 64- or 32-bit mode, the processor issues up to eight instructions per clock cycle.

Thus, 32-bit PowerPC applications run unchanged on the 970. However, a 32-bit operating system would have to support new data structures in its memory management unit and new interrupt handlers.

"There's not too much more to it," said IBM's Sandon. "Certainly we found it a compelling idea that there is market segment asking for 64-bit systems and this allows them to still make use of 32-bit code." ...

Besides the 64-bit capability, the PowerPC 970 includes IBM's first support for what Motorola calls the Altivec instruction set and Apple refers to as its Velocity engine. IBM simply cites the capability of the 160 vector instructions in its SIMD engine to speed graphics and multimedia operations. ...

The 970 also sports a cache-coherent, 900-MHz processor bus capable of data rates up to 6.4 Gbytes/second. It will support symmetric-multiprocessing configurations of up to 16 CPUs. That capability could be valuable for Apple's new line of Xserve entry-level servers.

Unlike the original Power4, from which the CPU was derived, the 970 supports only one internal processor core. ...
</strong>
<hr></blockquote>

[ 10-15-2002: Message edited by: The Watchful Babbler ]</p>

I bet that MacCentral would give us the first peeks at the new info . . . they're pretty good about timelyness, and it seems as if they should be there.

Just keep your eyes open - I'm sure even (vomit) MOSR will have the info (although probably convoluted and all screwy) up soon.

It's my guess that IBM will put the full text of the paper on their website. It will be after the paper is presented at the forum, no doubt.

Babbler, thanks a lot.

- I guess this was already discussed, but anyway: The article you quote (I think it was the only one) mentions 'late next year' [for Apple], while other reports just say 'next year or '2nd half of next year'.
I do hope we get clearer news on availability...and maybe a hint that a first small production run could be available earlier (XServe ?! ) for Apple.

- The other marketing question is what will Apple do to keep PowerMac sales up between today and the release date ?
This story has been picked up by so many news outlets that everyone will have read about it.
I wonder how Apple will handle this. The usual Apple silence is probably not the best approach this time.
Will we see a processor/product roadmap for the first time at Apple MWSF ? Like, 'Yes, we will use the IBM 970, but it will be only available for XServe. Out in March 2003 and you can preorder starting today'. Oh well, I'm probably dreaming...
On the other hand, this is not an uncontrolled leak. Apple has had at least 2 months time to prepare for Oct 15 - but maybe they still underestimated the coverage this story would get ? I really wonder if they can get away with the 'no comment' this time.

Then there's also the Apple analyst Q&A tomorrow night (results) where they will hear the same question again...

Interesting day for the platform

[ 10-15-2002: Message edited by: jabba ]</p>

Speaking of FY02Q4, here's the Apple info on tomorrow's feed-the-animals call:

WHERE: Via conference call. The dial-in number for press is: 913-981-5524

WHEN: Wednesday, October 16, 2002, 2:00 p.m. (PDT)/5:00 p.m. (EDT)

REBROADCAST: The conference call will be available as a continuous rebroadcast beginning Wednesday October 16 at 5:00 p.m. (PDT)/8:00 p.m. (EDT) through Wednesday, October 23 at 5:00 p.m. (PDT)/8:00 p.m. (EDT). The dial in number for the rebroadcast is 719-457-0820. Please enter confirmation code: 738248

WEBCAST: Apple will provide live audio streaming of its FY 02 Fourth Quarter Results Conference Call utilizing Apple's popular QuickTime(R) 5, the industry-leading multimedia software for live and on-demand audio and video streaming. The live audio webcast will be accessible through Apple's web site at <a href="http://www.apple.com/quicktime/qtv/earningsq402/" target="_blank">http://www.apple.com/quicktime/qtv/earningsq402/</a> .
Viewers simply need to download a copy of QuickTime 5 in advance from <a href="http://www.apple.com/quicktime" target="_blank">www.apple.com/quicktime</a> to listen to the conference call.

Today being the big day, I crawled out of bed and headed over to apple.com for the latest official word from apple. There I found Apple's amazing new processor... ladies and gentlemen.. it's none other than Jimmy Carter! That's right kids, enough processing power to run a nation is now at your fingertips. Call now, supplies are extremely limited.

I'm suprised they've forgiven Apple for the sledgehammer thing.

Andrew

Hopefully the 4th quarter conference call will have some insight on the 970 chip much like last year how steve jobs said imacs would have superdrives this year when investors asked

[ 10-15-2002: Message edited by: O and A ]</p>

Here's a link about the 970 along with a picture of a powermac.

<a href="http://www.e-insite.net/electronicnews/index.asp?layout=article&articleId=CA251834" target="_blank">http://www.e-insite.net/electronicnews/index.asp?layout=article&articleId=CA251834</a>

In the article, an Intel spokesperson notes that there is not "enough details to compare the
upcoming PowerPC 970 to the Itanium 2." Is someone saying this thing has perfomance in the itanium range?
I'd be happy if the 970 can keep up with faster P4's.

[quote]Originally posted by sc_markt:
<strong>Here's a link about the 970 along with a picture of a powermac.

<a href="http://www.e-insite.net/electronicnews/index.asp?layout=article&articleId=CA251834" target="_blank">http://www.e-insite.net/electronicnews/index.asp?layout=article&articleId=CA251834</a>

In the article, an Intel spokesperson notes that there is not "enough details to compare the
upcoming PowerPC 970 to the Itanium 2." Is someone saying this thing has perfomance in the itanium range?
I'd be happy if the 970 can keep up with faster P4's.</strong><hr></blockquote>
Thanks for the link, the Intel representative seems to be scared by the few specifications published on the PPC 970 compared to the itanium : 900 mhz front bus versus 400 mhz and 1,8 ghz compared to 1 ghz . I laugh when i eard him saying : there must be some bottlenecks somewhere : small cache ...
It looks like good news, an awfully efficient bus reduce the need of a very large on die cache. Can't wait to see this new beast

That intel rep is funny. Intel should be concerned if this chip is everything it has been made out to be.

info is as follows with more data forthcoming:


IBM POWER4 based processor. - PPC 970

Target process - 0.13um SOI 8 layer Cu interconnect
Target frequency 1.4 ~ 1.8 GHz
Target sample date 2Q03
Target ship date 2H03
Power 42W @ 1.8 GHz 1.3V (low power mode @ 1.1 V)

Est. SPEC INT 937 @ 1.8 GHz
Est. SPEC FP 1051 @ 1.8 GHz

Chip features: Elastic unidirectional point-to-point interconnects between CPU and "companion chip" (i.e. memory controller/northbridge) Elastic link may run "UP TO" 900 MHz, offering 6.4 GB/s of memory bandwidth.

POWER4 internals. Max of 8 inst fetch per cycle, 8 inst issue per cycle, and 5 (4 + branch) inst dispatch per cycle.

32 KB L1 Dcache, 64 KB L1 Icache.
512K L2.

32 64 bit GPR (general purpose register)
32 64 bit FPR (float)
32 128 bit VRF (vector)

nice keep it coming!

[quote]Originally posted by sc_markt:
<strong>Here's a link about the 970 along with a picture of a powermac.

<a href="http://www.e-insite.net/electronicnews/index.asp?layout=article&articleId=CA251834" target="_blank">http://www.e-insite.net/electronicnews/index.asp?layout=article&articleId=CA251834</a>

In the article, an Intel spokesperson notes that there is not "enough details to compare the
upcoming PowerPC 970 to the Itanium 2." Is someone saying this thing has perfomance in the itanium range?
I'd be happy if the 970 can keep up with faster P4's.</strong><hr></blockquote>

Actually, at 1.8ghz the 970 should wipe the floor with the 1.0ghz itanium2

[quote]Originally posted by Producer:
<strong>Target sample date 2Q03
Target ship date 2H03</strong><hr></blockquote>So we're looking at by MWSF 2004 for sure, MWNY 2003 at the earliest.

Shouldn't the talk be over by now? Let's get that info rollin' in!

I wonder where theRegister got their info from? I don't see it anywhere else. Could it be that they actually got the scoop before anyone else?

There's a separate VPU for Altivec/VMX is very good news...more to come

[quote]Originally posted by Producer:
<strong>There's a separate VPU for Altivec/VMX is very good news...more to come</strong><hr></blockquote>

how about just give the link:

<a href="http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=8300945231&m=347094333 5&p=9" target="_blank">http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=8300945231&m=347094333 5&p=9</a>

actually the info is from here:

<a href="http://www.realworldtech.com/forums/index.cfm?action=forum&roomID=11" target="_blank">RealWorldTechnologies Forum</a>

Opps...

[ 10-15-2002: Message edited by: sc_markt ]</p>

Are the ppc 970's numbers impressive?

Next year's PPC 970 @ 1.8 GHz:

Est. SPEC INT 937
Est. SPEC FP 1051

Today's p4 @ 2.53:

SPEC INT 893
SPEC FP 878

I2 @ 1GHz:

SPEC INT: 807
SPEC FP: 1356

What does a 64 bit chip mean for the OS and applications and for the product line-up? (I have no idea, but here are my questions)

1. Do we need a 64 bit aware OSX?
1a. What will run faster on it, what won't be affected?
1b. How long will it take to optimize OS X for 64 bits?

2. Do applications need to be 64 bit aware?
2a&b. same as above.

3. Is there anything dazzlingly new that we'll be able to do because of the new architecture? I'm thinking breakthroughs like GUI, image editing, movie editing, etc, have been in the past for Apple. Is there a new application that will be enabled by these.

4. Does it make sense for Apple to have the professional products as dual chip 970s and consumer products a single chip 970s or are we going to wait 4 years to get a 970 equipped iMac?

Discuss.

Looks pretty tasty to me -- and given the number of transistors, the move to a .9 process, this puppy has legs as well...

GP-UL Est. SPEC INT 937 @ 1.8 GHz
GP-UL Est. SPEC FP 1051 @ 1.8 GHz

from: the presentation at the MPF

Intel P4 SPEC INT 833 @ 2.4GHz
Intel P4 SPEC FP 812 @ 2.4GHz

Intel P4 Xeon SPEC INT 921 @ 2.8GHz
Intel P4 Xeon SPEC FP 878 @ 2.8GHz

from: <a href="http://www.spec.org/osg/cpu2000/results/res2002q3/" target="_blank">http://www.spec.org/osg/cpu2000/results/res2002q3/</a>

[ 10-15-2002: Message edited by: moki ]</p>

did that article say 900 MHz system bus???? Wow... very impressive.

[quote]Originally posted by Producer:
<strong>Power 42W @ 1.8 GHz 1.3V (low power mode @ 1.1 V)</strong><hr></blockquote>

In other words, one of these 1.8 GHz chips uses twice as much power as one 1 GHz 7455. I was hoping for more impressive power requirements, but eh, this a pretty fair trade-off if the performance is there. Not much better than Intel's P4 family though...

yes but i wonder what a 1.4GHz 1.1v chip dissipates.

[quote]Originally posted by Eugene:
<strong>

In other words, one of these 1.8 GHz chips uses twice as much power as one 1 GHz 7455. I was hoping for more impressive power requirements, but eh, this a pretty fair trade-off if the performance is there. Not much better than Intel's P4 family though...</strong><hr></blockquote>

According to the charts on geek.com. The P4 2.8GHz sucks down 68.4 watts. The 7455 at 1GHz sucks down 30. Doesn't seem so bad, perfectly acceptable for towers.

Now, when can I buy one?

So links with some more specs to compare the 970 to:

<a href="http://www.theinquirer.net/?article=4225" target="_blank">http://www.theinquirer.net/?article=4225</a>

<a href="http://www.theregister.co.uk/content/archive/22328.html" target="_blank">http://www.theregister.co.uk/content/archive/22328.html</a>

[quote]Originally posted by Outsider:
<strong>yes but i wonder what a 1.4GHz 1.1v chip dissipates.</strong><hr></blockquote>

19 Watts at 1.2 GHz 1.1v. According to the translation of C't article in the Ars forums.

"At 1.8 GHz, the CPU runs at 1.3 volts, and uses 42 watts. At 1.1 volts, it uses only 19 watts, and runs at 1.2 GHz"

How many watts is ok for a laptop? How many watts is the current powerbook G4 800?

Also this is 19 watts at 1.2 ghz on a .13 manufacting process...it should move pretty quickly to .09 so how much should .09 decrease power requirements and/or increase clock?

I think, Eugene, that it convincingly dusts a Pentium 4, especially in fpu, and its just, estimated behind an Itanic 2 on fpu. Impressive.

Not bad for a 'stripped down' processor.

And the Xeons and Itanics cost way more.

The 970 is cited for desktop use. And it's 64 bit and it's got altivec 'vmx'. As a Lightwave user...I'm about to break out the champers...

Moki has ruled out a Moto 'G5'. S'funny in one of the articles I read that a Motorola 'G5' rippost hasn't been ruled out and further that said part will be developed further. I don't see it though. I agree with many on this board.

...just see a Moto G4 die shrink on better memory structure this Jan'.

Yet...very strange those old Register links...was there a 8500 desktop that was canned? Did Moto/Apple decide to go with a G4 chip on better bandwidth because it was cheaper or because they'd already made the decision to go with IBM? Or maybe the links were pure fluff stories ala MOSR?

At least the 970 is real.

It remains to see what Apple says if quizzed during their conference call...

Lemon Bon Bon

[ 10-15-2002: Message edited by: Lemon Bon Bon ]</p>

[quote]Originally posted by Lemon Bon Bon:
<strong>I think, Eugene, that it convincingly dusts a Pentium 4, especially in fpu, and its just, estimated behind an Itanic 2 on fpu. Impressive.
</strong><hr></blockquote>

hmmm... i hoped for 1300 spec .... i think intel will be on 1100 in the usual win-pc next year. so if a 1.8gig 970 is in this range too that's nothing to get really exciting about isn't it? it's just to be as fast as the wintels....

but one thing is for sure: i prefere my 400mhz g4 over a 2.8ghz pentium 'cause it runs os x ;-)