CONFIRMED IBM Power PC 970 - Page 5

Something I'd be interested to know is how conservative an early pace IBM has decided to set. I am inclined to think the chip will probably scale very well and be easily overclockable.

This is a hot little chip. Low voltage (the 7455 in the towers run at 1.8v), relatively compact, exceptionally low wattage for its class of processor, VMX, great spec numbers, and lots of bandwidth to the motherboard (one of the many things SPEC ignores) and to other CPUs.

It'll be moving to .09 micron relatively quickly, too, which means that if Apple can't get it in a PowerBook (or an iMac!) right off the bat they'll be able to before long.

In the meantime, Mot's offerings should scale up well, especially if they're moving an (evolved) G4 to RapidIO and the 0.09 micron fab they've bought into.

[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>

@ 2.8 GHz, the p4 has a Spec int of 976, and a Spec fp of 915. Its hardly "dusted" by the ppc970.

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

Granted, the ppc970 looks to be smaller, cheaper, cooler, 64bit and smp enabled... but at 1.8 Ghz, it doesn't look like it will be much faster than a 3ghz p4. I do hold hope, as Moki assured, that the ppc970 will scale rapidly...

I say this chip is a home run... I had been deluding myself into thinking I'd see a grand slam.

I just thought I'd share a couple pieces of information here.


The SPEC numbers for the P4 2.8GHz are as follows:

SPEC INT: 976
SPEC FP: 915


Now, we can deduce what a 3.8GHz P4's SPEC mark will be based on the above data:

SPEC INT: ~1324
SPEC FP: ~1241

This is, of course, providing that the P4's architecture stays the same and performance will linearly increase with the added MHz. Unfortunately, Intel will mostl likely have improved the P4's architecture by then so you may very well see SPEC numbers decently higher than this.


But since when has the SPEC mark been a good indicator of performace, right?


P.S.- I'm glad one of my computers still had this forum's cookie still stored--I can't login on any other computer <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />

EDIT: awww. Ompus beat me to the spec #s while I was trying to find a way to login

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

[quote]Originally posted by Nitzer:
<strong>

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? </strong><hr></blockquote>

The 1 GHz G4 sucks down 30 watts maximum, but 21 watts typical.

A 2 GHz P4 sucks down 52.4 watts, while a 2.53 GHz P4 sucks down 59.3 watts... @ 1.5 V

If you really want to compare the 970 to the P4 please have a lokk at this URL:
<a href="http://translate.google.com/translate?u=http%3A%2F%2Fwww.ix.de%2Fnewsticker%2F data%2Fciw-15.10.02-002%2F&langpair=de%7Cen&hl=de&ie=ISO-8859-1&prev=%2Flanguage_tools" target="_blank">http://translate.google.com/translate?u=http%3A%2F%2Fwww.ix.de%2Fnewsticker%2F data%2Fciw-15.10.02-002%2F&langpair=de%7Cen&hl=de&ie=ISO-8859-1&prev=%2Flanguage_tools</a>

This does not look so very impressive! <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />

<a href="http://www.theregister.co.uk/content/3/27621.html" target="_blank">Register Article</a>

Intersting fact from this article:

- "AMD told us today that when Opteron debuts in the first half of next year it will ratchet up a SPECint of 1202 and a SPECfp of 1170"

Now it also says that IBM is being conservative in it's1.4 to 1.8 ghz range but it is still saying second half while AMD will be out in the first half and with Higher specs...

So, in a year we will get about the performance of todays Pentium?

Disappointing

[quote]Originally posted by Krassy
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... <hr></blockquote>

Spec's not including VMX/altivec and bandwidth though, eh? Not sure.

-tink

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

How does the Spec-benchmarks relate to the fact that 970 is 8-way scalar? Is all 8 ways used when running the spec-test or can this processor do several spec-tests at once without penalty?

Running AltiVec-optimized operation today still have no match on any processor. Witch _a_lot_ of more memory bandwidth how will it perform? c'ts article states 14.4 GFLOPS max @ 1.8 GHz.. A dual G4/1.25 pulls 18.3 GFLOPS today, so 970s VRF will be 55% more powerfull at 45% higher frequency. Not to impressive.

Ok, fishies, swim in the barrel...

first, Moki...
[quote]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.<hr></blockquote>

Yeah, and in 2007, when Itanium is projected to be more than 10% of Intel's unit shipments, maybe they'll have to deal with it. Saying ANYTHING about it now is rather crazy. I'm surprised you would bring this up. No consumers who "have been using ghz ratings as a measure of their penis size" are going to be buying Itaniums for a LONG time. This is a silly thing to bring up.

More from Moki:
[quote]Wait until you see the benchmarks from this processor; I think you may be quite impressed.<hr></blockquote>

Yup, after seeing those spec scores, I'm impressed enough to ... *yawn* ...

for KidRed
[quote]Actually, at 1.8ghz the 970 should wipe the floor with the 1.0ghz itanium2 <hr></blockquote>

You're right, it should. Obviously, based on the spec scores, however, it doesn't. a pitty.

for vr6
[quote]Is there anything dazzlingly new that we'll be able to do because of the new architecture? I'm thinking breakthroughs like GUI...<hr></blockquote>

Yes, it does look like the GUI will benefit - you'll be able to resize windows with only barely perceptible choppiness.


Ok, so I've gotten the trollishness out of my system. Seriously, it looks like a darn fine processors, especially if it comes out sometime in Q3 03. It won't take the performance crown for most things, IMO, but it won't be the slouching slow-poke that the current G4 is. I think I'd be happy having one of them in a machine of my own (if it ran windows, that is).

It will be interesting to see how well macs run OS X when they're not hindered by the G4s.

neye

I don't know the SSE or MMX or so specs or benchmarks of the P4. I Think it will be some lower than the ones of Altivec/970.
But I think this alone gives no good reason for beeing happy with the 970 or future macs

I wonder how well the PC970 will scale. For example, how long will it take to get a 33% increase? Not that it is a valid comparison, but does anyone know how long it has taken the P4 to scale this amount (if at all)?

Why did I pick 33%, since it makes a nice 2.4Ghz. At this speed would the approximate SPECInt be 937 / 1.8 * 2.4 = 1249 and SPECfp be 1051 / 1.8 * 2.4 = 1401?

[ 10-15-2002: Message edited by: lfrog ]

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

Perhaps it's time to post the SPECs for the G4?

C't tested a 1 GHz G4 some time ago:

SPECint: 306

SPECfp: 147-186

Has anyone seen this <a href="http://www.extremetech.com/article2/0,3973,635220,00.asp" target="_blank">ExtremTech article</a> yet?

[quote]Originally posted by JLL:
<strong>Perhaps it's time to post the SPECs for the G4?

C't tested a 1 GHz G4 some time ago:

SPECint: 306

SPECfp: 147-186</strong><hr></blockquote>

Yikes.

[quote]Yikes.<hr></blockquote>

nah, that was a 1Ghz machine for the spec scores. Yikes topped out at 400 Mhz I think

neye

[quote]Originally posted by disco:
<strong>Has anyone seen this <a href="http://www.extremetech.com/article2/0,3973,635220,00.asp" target="_blank">ExtremTech article</a> yet?</strong><hr></blockquote>

I did. I liked this part:

"If I were designing a processor for high end graphics desktops or servers this is what it would look like"

1.8 GHz, relatively low power consumption, a likely quick move to 0.09 microns, 6.4 GB/sec memory bandwidth, a full VMX implementation, full speed 32-bit mode, 64-bit support, 8/5-way superscalar, ~1000 int & fp specmarks, 96K L1 cache, 512K L2 cache, full SMP support up to 16-way (!!), all packed into only ~50 million transistors and you guys are still complaining??

This means PowerPC suddenly has a visible future, and will be fully competitive performance-wise within a year from today. At 40W and with likely good yields (due to the relatively low transistor count and IBM's experience with the POWER4), Apple will be able to deliver dual (or quad, or more) processor machines at the top end of its line -- machines which can go up against the best of x86 and more than hold their own... not even accounting for the advantages of SIMD and OS X. This is also just the first of a new line of PowerPCs, and it represents a new core that has legs just like the PentiumIV... except that at the same clock rate this thing will clobber the existing G4, unlike the PentiumIII -&gt; IV transition where Intel took a significant loss in efficiency per clock.

This processor can probably move down through the entire Apple lineup much faster than I was expecting. The iMac might be able to handle the 0.13 micron version, and the 0.09 version ought to be fully portable. Future chips will be able to go to higher rates and multi-core quite quickly.

I don't think it is this particular chip that worries Intel so much as the new future of the PowerPC. The Intel comment about "it must have bottlenecks somewhere" is laughable -- IBM has always been much better than Intel when it comes to avoiding bottlenecks. This chip isn't suddenly going to make x86 curl up and die, but it does mean that PowerPC is back in the game in a big way... and realistically that is all we can hope for. There isn't some magic technology that will suddenly render the biggest chip maker in the world incapable of competing.

[quote]Originally posted by moki:
<strong>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>

</strong><hr></blockquote>

So we're looking at about 15-25% better performance than top flight intel at about 60% the power consumption, well a little more than that, but still much better efficiency. I don't expect the P4 to scale linearly, but Intel is not averse to making modifications (either to the core or the compiler, so who knows.) Now the fastest Hammer and Itanic might be faster still, and a bit earlier in arriving. The real differentiator will be price performance. If IBM is supplying these things at IA32 price levels (high-end, natch) then they have a winner because all the early evidence suggests that Itanic will cost a hell-of-a-lot and while Hammer may be cheaper than Itanic, it's going to cost a fair bit too.

[RAMPANT UNCHARACTERISTICALLY OPTIMISTIC MODE ENABLED:]

42 watts is some seriously good power consumption for the 1.8Ghz part. If they slow it down to the 1.2-1.3 range that should take them under 30watts. X86 builders have been overcoming similar/greater obstacles in their notebook designs for quite some time. While GPuL's numbers suggest it has a shot at reaching the mobile market in current .13u form, and it looks certain to do it at .09, nothing about either Hammer or Itanium suggests you'll be using one to fry your testicles any time soon. And that's a very good thing, as to me at least, it suggests that at some point 12-18 months from now, Apple will return to the early Lombard/Pismo days of having the only true POWER notebook on the market... Matsu licks lips... yummy...

[/RAMPANT OPTIMISM SWITCHED OFF]

Also, if previous price and efficiency arguments hold any water, then GPuL also looks a lot better for desktop duals than either the P4Xeon 3.8-4.x-whatever it's at by then, or the Hammer, or the Itanium...

Could be good times ahead...

[/OKay, RAMPANT OPTIMIST MODE REALLY TURNED OFF NOW]

[ 10-16-2002: Message edited by: Matsu ]</p>

[quote]Originally posted by Carbonide:
<strong>So, in a year we will get about the performance of todays Pentium?

Disappointing </strong><hr></blockquote>

Yea... I'm so bummed too...

FOLKS!!!!!!!!!!!

And when I say FOLKS I really mean doom sayers and trolls... Not that they will pay attention to this but...

Does the G4 1.25Ghz dual run X at what you would consider an 'okay' level? Most people who own them I'm betting would say YES. I'm willing to bet most people here who are pissed about the 'bad news' don't even have a dual 1.25 in the first place (or the money to buy one - but that's not the point just an observation) but lets continue anyway.

Fact: The Dual 1.25Ghz G4 is a pretty good machine to begin with.
Fact: The MOT 1.25 doesn't even rank on the SPEC charts.
Fact: The IBM 1.8Ghz gets ALMOST to the TOP of the charts!
Fact: The IBM 1.8 is starting out at 1.8 on a .130 process
Fact: The IBM 1.8 is going to be moved to the .90 process by year end.
Fact: The smaller the process the faster the CPU will clock and the cooler it will run

All these numbers freaks who are bummed just because we don't TOTALLY WIPE THE FLOOR on some silly SPEC rating are majoring in minors.

Some questions I'd love answers to:

How often do you guys really run apps that the SPEC marks TRULY represent?

Can you list what apps you run and to what parts of the SPEC tests they are represented by?

For gosh sakes... This CPU is INDEED a HUGE leap for OS X users.

The IBM CPU will give us stuff we already have:

- TRUE SMP SUPPORT - Unlike the P4
- 128 Bit VMX - Unline the P4

The IBM CPU will also give us things we've never had:

- 64 Bit computing - Unlike the P4
- 1.8Ghz clock
- Super fast 900Mhz bus - Unlike the P4

(plus a bunch of stuff that I'm sure a CPU geek could better tell you about than I ever could)

Okay so what does the P4 have on us?

- Mhz at a price of heat
- Better SPEC scores

If you wanna trade in 64bit + AltiVec + 900Mhz bus + 6.4Gbs of bandwidth + TRUE SMP support for and extra Ghz or so and a top spot in the SPEC lists well go ahead...

Oh but one more thing... You'll have to give up X while your at it.

Sorry but folks PLEASE don't listen to the whiners of the world.. It isn't worth it.

Dave

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

What kinda RAM is gonna feed that FSB ???

DDR-II ? Even then doesn't it maxx out around 4.2GB/s. Maybe some kinda dual channel interleaved DDR-II solution?

Any answers?

[quote]Originally posted by Matsu:
<strong>What kinda RAM is gonna feed that FSB ???

DDR-II ? Even then doesn't it maxx out around 4.2GB/s. Maybe some kinda dual channel interleaved DDR-II solution?

Any answers?</strong><hr></blockquote>

Yea... I forgot didn't they call a stop to all advances in memory technology? Sorry for the dig but it was aching to be said.

Fact is I'd rather have a CPU that support more than what TODAYS memory can do since as we all know nothing stands still. Heck we are going from a cpu that maxes out at 166Mhz to a cpu that maxes out at 900Mhz this IS a GOOD THING(tm) trust me!

Again I didn't mean to take that swipe but all we need are folks going around using THAT as a reason to think the CPU is sub-standard.

Dave

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

I am wondering the SPEC score for the MIP processors used by SGI's IRIX-based boxes.

I didn't mean it as a criticism, I was honestly wondering what the next step will be? Absolutely better to have a higher FSB limit than a lower one, I don't see how anyone could use it as a knock against the CPU.

matsu~

Read through the old posts--actually, at 1.2 gig it only uses 19 WATTS (emphasis mine).

I can smell portables showing up a hell of a lot sooner!

[quote]Originally posted by Programmer:
<strong>1.8 GHz, relatively low power consumption, a likely quick move to 0.09 microns, 6.4 GB/sec memory bandwidth, a full VMX implementation, full speed 32-bit mode, 64-bit support, 8/5-way superscalar, ~1000 int & fp specmarks, 96K L1 cache, 512K L2 cache, full SMP support up to 16-way (!!), all packed into only ~50 million transistors and you guys are still complaining??</strong><hr></blockquote>

I couldn't agree more.

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

[quote]Originally posted by Matsu:
<strong>I didn't mean it as a criticism, I was honestly wondering what the next step will be? Absolutely better to have a higher FSB limit than a lower one, I don't see how anyone could use it as a knock against the CPU.</strong><hr></blockquote>

The next step is DDR II, but that's not what the 6.4GB/s of bandwidth is really for.

Think Apple Processor Interconnect. There are a number of ways Apple could lay this out, but imagine communication between processors at that speed! Heck, Apple could fit RAM and AGP bandwidth on one bus and have room to grow.

This is workstation grade stuff. This is very cool. And it's the kind of thing that SPEC doesn't even touch.

Speaking of things SPEC doesn't touch, you know that one little thing that allows a G4 to completely smoke a P4? AltiVec? Well, this chip's got it on board too. How badly will it smoke the competition once that's taken into account? Keep in mind that there is currently more than one effort to optimize PPC Linux for AltiVec. Apple is not alone here. Can you imagine what top-flight SPEC marks, unheard-of (for PCs) bandwidth, unheard-of (for PCs) MP capabilities, full-blown AltiVec, all crammed into a small, cool processor is going to do? Especially since the GPU will be conscripted to take a lot of the work away from them in OS X?

Even if something bests it by 10% or so in SPEC marks by the time it comes out, it'll still own in overall, real-world power and in power/initial cost and, because it's relatively small and cool, in power/watt. And it'll be in PowerBooks! If not immediately, then before long.

Oh, and it won't introduce a whole new ISA (like Itanic) nor (apparently) require hard resets to go from 32 to 64 bit mode (like Opteron).

This is the best goddamn news to hit the PowerPC world since... I don't know, the PowerPC. And, astonishingly, people are still whining. Get a grip, people. This rocks. IBM surpassed my expectations by a comfortable margin: I expected something like this level of performance, but I thought the tradeoff would be a big, hot, expensive monster. Not something that could ship in a notebook right off the bat!

[ 10-15-2002: Message edited by: Amorph ]

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

Programmer, moki & DaveGee...thank you. You all consistantly help keep my AI ulcer in remission.

-G

Has AltiVec finally been officially confirmed or is that still just an educated guess?

[quote]The IBM CPU will also give us things we've never had:

- 64 Bit computing - Unlike the P4<hr></blockquote>

here's something i need some explaining on

We've got a 970

can you mix and match 64-bit/32-bit apps on a 64-bit OSX?

Can you mix and match 64/32 on a 32-bit OSX?

Wouldn't everything have to be the same all across?


and also, would performance be different running 32bit apps/OS compared to 64bit apps/os?

I've read previously that 64bit apps/os would be slightly slower in some respects because of larger chunks goin through, wasting a bit more space here and there.

Can we mix and match this stuff?

If not - then the GPUL970IBMthingamajib won't give us 64-bit

Apple and software developers will have to.

And a reminder - what exactly needs to be done for 32but apps to go 64bit?
Just a recompile?
Or recompile and tweaking?
(meaning, software companies would wait until a larger .5 or .0 release to implement it)

regardless - this is kinda exciting.
although I thought the IBMGPUL970thingamajib would end up wiping the floor with the Itantium 2 - but hey - the Itantium 2 needs 64-bit windows and apps. WinXP-64 is out - a lot of apps aint.
(not to mention the cost!)

Carry on.

(and can one of the admins puh-leeeease fix that ridiculously long URL above? It's ruining my enjoyment of this page! Thanks.)

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

GPUL's VMX == AltiVec, this is confirmed. It has 2 VMX units.

Don't worry about 32-bit vs 64-bit compatibility, the OS that runs on these machines will run both kinds of software natively and at full speed. If you don't have any word-size assumptions in your code then you can just recompile into 64-bit mode, but why would you unless you are taking advantage of the 64-bitness of it? Otherwise just recompile the 32-bit app with a GPUL-aware compiler and there will likely be a bit of a performance improvement.

19 watts @ 1.2 GHz! WOW!

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

That 6.4GB sounds suspiciously like another standard we've been hearing about.

I want a 64bit notebook, I just want it, Apple, you might finally earn a sale, imagine that...

I could be in a switch add, an anti-switch add of sorts,

"You know, it's fvcking fast! Forget all the other bullsiht, it's fast." Do do do... Apple music playing in background... "BTW, it's fast!"

"My name is Matsu, and I used to dream of burning Cupertino to the ground..."

IBM will use a 0.13-micron SOI process with 8 levels of copper to manufacture the chip, which should require a 576 pin package; Sandon did not disclose the die size. IBM expects the chip will output between 19 watts and 42 watts of power, depending on the whether a 1.2-GHz (1.1 volts operating voltage) or 1.8-GHz (1.3 volt) clock speed is used.

from <a href="http://www.extremetech.com/article2/0,3973,635220,00.asp" target="_blank">http://www.extremetech.com/</a>

edit: whoops, redundant. sorry.

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

[quote]Originally posted by Amorph:
<strong>
It'll be moving to .09 micron relatively quickly, too, which means that if Apple can't get it in a PowerBook (or an iMac!) right off the bat they'll be able to before long.
</strong><hr></blockquote>

<a href="http://www.geek.com/procspec/apple/g4.htm" target="_blank">http://www.geek.com/procspec/apple/g4.htm</a>

Given this spec, the 667, which I believe is the one in the PBG4 667, uses 19watts.

In an ideal world I'd like the 970 to completely scrap the compatibility factor and lose a couple-million transistors in the process, but I'll take what I can get.

So, while I don't guess that Apple will immediately launch a bad-ass PB970, it looks as though it may be an option. Secondly, I don't know how much is to gain by reducing the fab size. How much lower can the voltage go before it's below the threshold for Si? I think the bottom limit &gt; 0.7V.

. . . So if 90nm can move to 1V, or maybe even 0.9V I guess that will help a bit.

This time around it better be andonized. . .

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

[quote]Originally posted by Programmer:
<strong>GPUL's VMX == AltiVec, this is confirmed. It has 2 VMX units.</strong><hr></blockquote>

Really? 2 subunits like the 7400 (VPU + VALU) or two full units or how do I have to understand this? Where can I read about it?

G4@1GHz: 10.5M RC5 keys/s
GPUL@1.8GHz: 18M RC5 keys/s

Hmm... ok.

This is a far more telling quote from the same article:

[quote]
Perhaps more importantly, the front-side bus can transfer up to 7.2 Gbytes per second, roughly four times the bandwidth of the current Pentium 4 front-side bus, according to MDR's Halfhill.
<hr></blockquote>

I don't know where he gets 7.2, but even at 6.4 that means it'll be doing considerably better than the PIV's FSB... and then consider that in most applications AltiVec is currently held back almost entirely by memory bandwidth. If Apple builds a top-of-the-line memory system to feed this beast I think we're going to see some very impressive media performance -- something that SPEC doesn't even attempt to benchmark.

[quote]Originally posted by Splinemodel:
<strong>In an ideal world I'd like the 970 to completely scrap the compatibility factor and lose a couple-million transistors in the process, but I'll take what I can get.
</strong><hr></blockquote>

Which "compatibility factor" are you refering to?!

[quote]<strong>
So, while I don't guess that Apple will immediately launch a bad-ass PB970, it looks as though it may be an option. Secondly, I don't know how much is to gain by reducing the fab size. How much lower can the voltage go before it's below the threshold for Si? I think the bottom limit &gt; 0.7V.

. . . So if 90nm can move to 1V, or maybe even 0.9V I guess that will help a bit.
</strong><hr></blockquote>

I don't know about the voltage reduction, but the smaller process means a smaller die. A smaller die means more parts per wafer which means better yields and thus cheaper parts. It should also allow higher clock rates.

[quote]Originally posted by 123:
<strong>

Really? 2 subunits like the 7400 (VPU + VALU) or two full units or how do I have to understand this? Where can I read about it?

G4@1GHz: 10.5M RC5 keys/s
GPUL@1.8GHz: 18M RC5 keys/s

Hmm... ok.</strong><hr></blockquote>

Linear scaling by clock rate isn't surprising. In fact, its a relief. I was concerned that we'd lose some of AltiVec's magic efficiency with the introduction of IBM's VMX equivalent. RC5 isn't very bandwidth intensive, I believe, so you're not seeing the benefits of the 6.4x improvement in bandwidth.

[quote]Originally posted by Programmer:
<strong>GPUL's VMX == AltiVec, this is confirmed. It has 2 VMX units.</strong><hr></blockquote>

So aside from being "8/5-way superscalar", it has TWO AltiVec units? So theoretically double the AltiVec power of a current G4?

All in all, this is a super-nice chip.