• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD ClawHammer Benchmarks!?!?!?!



<< If we take AMD?s info for granted, then its performance makes 1350 points, while the today?s eldest Athlon XP 2000+ gets only 730 points at the most. >>



Is this just a mock up or actual scores? There are chips on there that aren't even supposed to be made (yet) 😕

If they are actual performance numbers that would be kinda neat 😀 (eventhough they are from AMD)
 
I cant wait!. I think a GF4 and one of those ClawHammer will bridge the gaming performance gap between a 9x OS and XP!
 
If they are actual performance numbers that would be kinda neat (eventhough they are from AMD)

There were unconfirmed rumors (of course they were unconfirmed 😉) that AMD had .18 micron Hammer samples out and about (this was in August 2001).

I don't think it is unrealistic at all to say that AMD already has Hammer sampling, most likely with .13 micron and SOI (AMD's conference call notes released yesterday mentioned that they had been toying with SOI "since Q3 2001").

Since it looks as if this may be from AMD, I would venture to guess the score of 1350 points is based on an actual ClawHammer sample, and not guesses. Also, remember last year at the Intel Developer that a 3GHz P4 system (super cooled) was shown off. I bet those 3GHz P4 numbers are based off of AMD's own super cooling tests with an overclocked 2.2GHz Northwood. Although that is a total guess. 😉
 
?

All I see is projected performance charts...
No data on system config, test programs, software.

This is vapor, just like the GeForce3 integrated thread, HP's Merced projections, etc. etc.

Edit: I see the projection is for SpecINT, CPU 2000.
 


<< ?

All I see is projected performance charts...
No data on system config, test programs, software.

This is vapor, just like the GeForce3 integrated thread, HP's Merced projections, etc. etc.

Edit: I see the projection is for SpecINT, CPU 2000. >>



Says who. IBM's Power4 benchmarks are accurate just as all the other benchmarks are. The only benchmarks that are suspect are the 3GHz P4 and the ClawHammer numbers.

There's no reason to think that AMD doesn't have ClawHammer CPU's out and about and benchmarked. AMD has the ClawHammer set for the end of this year, and AMD would be definitely be crazy to think that they could release ClawHammer by the end of this year without already having working samples by now. The long gone Wingznut PEZ (and pm) have said that CPU's usually take 3-5 years to develop. I've seen Hammer articles dating back to 1996-97!)

Btw, there's a reason why exact configurations weren't shown for these tests. Duh! 😉
 
Going by Hector Ruiz, AMD presidents' statement in their conference call of Mid November, AMD was expecting to have engineering sample ClawHammer's in 1H 2002. As 2002 has just started now I think it unlikely they've managed to produce engineering sample processors already, unless their ahead of schedule... which would be rather unusual considering the latest information seems to be indicating that ClawHammer may slip from Q 2002 release to Q1 2003.

My bet is this is nothing more then AMD's software simulation of the ClawHammers expected performance, and it's well known that such simulations may not always very accurately reflect final performance.
This guess also seems to be backed up by the fact that AMD has previously disclosed software simulations to have put the ClawHammer at a release of 3400+ to attain a SpecINT 2000 score of 1300-1400... which is dead on what these supposed "benchmarks" put it at, as well as the same supposed PR rating. Either AMD's software simulations were shockingly accurate, or these results are nothing more then those very same simulations.
It sure sounds a hell of a lot like this is nothing more then a software simulation of the ClawHammer prior to actual silicon.
 


<< There's no reason to think that AMD doesn't have ClawHammer CPU's out and about and benchmarked >>



And if some people in Rumania [or insert any other country here] posted supposedly official slides about some Intel chip that had a 4400 GHz rating would you beleive it? I wouldn't. I didn't beleive any of the Pentium 4 hype two years ago (aka bullsh!t) and I see no reason at all that this ClawHammer info is valid.

There's no reason to think that AMD does have ClawHammer CPU's out and about and benchmarked.

I'll beleive it when the review sites post their benches.
 
Going by Hector Ruiz, AMD presidents' statement in their conference call of Mid November, AMD was expecting to have engineering sample ClawHammer's in 1H 2002.

I think Hector Ruiz was referring to final Hammer samples by 1H 2002. These benchmarks may not have been done with final samples of Hammer, but with unfinal Hammer samples that still need debugging before final samples are released.

Either way, we're both guessing on this info. There's probably not a lot of truth to this, but then again who can say for sure.

There's no reason to think that AMD does have ClawHammer CPU's out and about and benchmarked.

Yes there is. Going by what I've read and what some other more knowledgeable people on these forums have said, AMD should have had Hammer samples last year if they were to even be thinking about releasing ClawHammer to the public this year.
 
whatever happend to wingnut pez and pm? i went to vegas in late decemeber and since i have been back, i havent seen either posting. i assume both got fed up and left 🙁

pisser, was nice to have them both around, but dont blame them at all for leaving.
 


<< whatever happend to wingnut pez and pm? i went to vegas in late decemeber and since i have been back, i havent seen either posting. i assume both got fed up and left 🙁

pisser, was nice to have them both around, but dont blame them at all for leaving. >>



I know, it sucks. 🙁
 
The one number I'm most interested in is price. Sure it theoretically will be about double the speed (which is evident in those numbers). But what will AMD do with the price? A $2000 processor might not be worth it to many of you.
 
A $2000 processor might not be worth it to many of you.

The ClawHammer's die size is supposed to be 104mm^2 when it debuts on the .13 micron + SOI process later this year. That's 24mm^2 less than the 128mm^2 large .18 micron Athlon XP's that are selling right now for no more than $350.

Unless SOI is prohibitively expensive and the Hammer's architecture is exorbitantly complicated, I very much doubt we'll be seeing $2000+ ClawHammer processors.

EDIT: Here's the link to ClawHammer's die size.
 
The die size may have a $50 effect on the processor price. The real effect on the price is on AMD's opinion on how to maximize long term profits. I just picked $2000 out as a randomly high value (although I made sure it was $1000 less than the other 64 bit processors - since AMD likes to be under the competition). We know two things: (1) the Hammer is initially targetted more for businesses and not home users and (2) AMD seriously needs to raise the average selling price of its processors. Thus I think the Hammer will cost significantly more than the $300 for their 2000+XP. I have no clue what the price will be - thus my statement that I'm highly interested in price. My best guess is $500-$750.

I honestly don't think the initial Hammer release will sell many processors. I could be very wrong. But how many home consumers have a 64 bit version of Windows? (The unix community isn't large enough to make a significant effect on sales). How much does an upgrade to the 64 bit Windows version cost? How many home users have programs that will take advantage of the 64 bit processing? I don't know, but the fewer people that are estimated to buy it, the higher I think AMD's initial price will be.

Edit: To make it very clear, I'll state it again. I'd love to be wrong.
 
Mna......I don't want to dampen spirts because i'm as eager for the "Hammers" as anybody, but, I knew I had read this somewhere............

  • Hammer/SSE2 in 2003? (AMD)
    By Brian Neal
    Tuesday, January 15, 2002 7:38 PM EST
    A big thanks goes to wlaote for helping out on the translation of this article from German site tecchannel.de regarding support for SSE2 in AMD's upcoming Hammer line of processors. While it has been known that x86-64 processors from AMD will support SSE2 since the x86-64 specification was released back in 2000, there have been some questions lately as to when we might see it.

    AMD's Jan G&uuml;tter has stated that the company expects to support SSE2 as soon as it achieves broad-based support as a standard, something they expect to see happen in 2003:


    Not least therefore did AMD decide to built Intels SSE2 into their own CPUs, but with deliberation [care/carefully?]. AMD-spokesman Jan G&uuml;tter said to tecchannel.de: "AMD plans indeed, to support SSE2, as soon as it succeed [will be accepted] as a standard. We expect that for 2003." With that G&uuml;tter confirmed again, what was already casually known at the presentation of the hammer-architecture: The battle of the multimedia-extentions is over, the winner is SSE2. But the new version of the Athlon-core in 0,13 micron (codename thoroughbred) comes to the market without SSE2.
    However, the article raises the question as to whether or not AMD is licensed to release an SSE2-capable processor before 2003:

    Already at the time AMD anounced to support SSE with the palomino-core a high Intel-manager foamed at the edge [brink?]: "Everything, what they do, is to copy us!" In those days AMDs decision was made public several days after the expiration of a patent exchange agreement. After inquiry Jan G&uuml;tter denied to explain how his company come to a licence for SSE2. But the AMD-spokesman sees no problems in this respect. It's possible, that Intel still exclusively reserved SSE2 for the year 2002, and that therefore AMD can only come up with corresponding processor [=with SSE2] in 2003.
    At this point, it's only speculation, but if it's true that SSE2 is a Pentium 4 exclusive for 2002, then we may not see an x86-64 processor from AMD until 2003. The current AMD roadmap places ClawHammer's debut in the last quarter of this year.
 
My best guess is $500-$750.

I would guess that too Dullard. Maybe more like $800. We'll see of course.

ToBeMe:

I caught that news a couple days ago. It basically says that Intel could have a less than positive effect on Hammer's release date (Claw and Sledge), pushing it into 2003. But even if it is 2003, it probably won't be more than 1 year away since AMD has set up ClawHammer to debut late this year.

Trust me, I don't expect Hammer to come out any earlier than 2002/2003.
 
I thought I had also read that "Clawhammer" was to be AMD's answer to Itanium................if so, and from what has been written about it's capabilities...........$800 may be a REAL conservative estimate..............🙁
 


<< I thought I had also read that "Clawhammer" was to be AMD's answer to Itanium................if so, and from what has been written about it's capabilities...........$800 may be a REAL conservative estimate..............🙁 >>


Isn't it the sledgehammer the answer to Itanium? The clawhammer is AMD's card up the sleeves, since Intel doesn't yet have a single processor version of a 64 bit processor. $2000-$3000 is where the sledgehammer may reside. Who knows, I have no crystal ball into AMD's pricing strategy.
 


<< I thought I had also read that "Clawhammer" was to be AMD's answer to Itanium................if so, and from what has been written about it's capabilities...........$800 may be a REAL conservative estimate..............🙁 >>



Actually, I believe ClawHammer will be the desktop rival of the Pentium 4, whereas SledgeHammer will be the rival of Itanium/McKinley. However, I wouldn't be surprised if ClawHammer debuted for the desktop market at much higher prices than what the high-end AXP's are selling for currently.

Isn't it the sledgehammer the answer to Itanium? The clawhammer is AMD's card up the sleeves, since Intel doesn't yet have a single processor version of a 64 bit processor. $2000-$3000 is where the sledgehammer may reside. Who knows, I have no crystal ball into AMD's pricing strategy.

EDIT: Exactly, Intel doesn't have a 64-bit desktop processor planned for the near future. I wouldn't be surprised in the least if SledgeHammer was in that price range dullard.
 
The benchmarks are believeable to me. AMD has never thrown out any bullsh*t statements it cant back up, and there is little reason to think they would start doing such a thing now. Assuming those figures came from AMD directly, I wouldent doubt them for a secound.
 
Hammer engineering samples have been out for months now. I'm told these latest figures are compiled from .13 micron cores with SOI, running at a clock of 2.0GHz. Indeed, the future looks very, very bright. 😀
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top