Craig P. and anybody else who cares. I posted the wrong link like Craig had thought. At Ace's there was a thread that discussed the scores and evolved into compiler info here is that link http://www.aceshardware.com/forum?read=80039633 it starts to get specific about the compilers used here http://www.aceshardware.com/forum?read=80039637 sorry for the confusion.

Real World Technologies take on MPF 2002 Coverage - AMD Opteron
http://www.realworldtech.com/page.cfm?AID=RWT101602161848

Waah the AMD roadmap at THG lists Barton with a 266 FSB.

best two reasons for hammer?

keep competitive pressure on intel....which means inexpensive (relatively) P4's for you intel fans........who knows? i may even build a pentium system some day.

the day AMD tanks, all you intel fanboys are in for some sticker shock.

reason number two for hammer:
i just want one, because it's going to be fast even for 32bit code, and i'll get to play with 64bit linux (because i can)

"Most of the motherboard manufacturers have samples with a clock speed of 1.6 GHz, labeled unofficially as "XP 3200+."
"

Thats from Tomshardware

http://www.tomshardware.com/mainboard/02q4/021017/index.html

Microprocessor Forum presentation here:
http://www.amd.com/us-en/assets/content_type/DownloadableAssets/Fre...

[quote]Combine that with a release date very close to Itanium 3 and the recently announced lack of some of the reliability features high-end system vendors expect, and you've got what is starting to look like a budget web server processor and not much else. [/quote]

Yeah lots of 64-bit web servers out there. hoo boy that's the selling point of 64-bit computing! Because web servers need to address all those gigs of ram!

No, I won't use a sarcasm tag, I refuse.

Clawhammer will be available in Q2 2003 in the same sense that XP2400 is available today...

Oh, that makes sense. I guess they were lying to their stockholders yesterday.

Praise the lawd!!!

Hammer not delayed after all. From the Inq link:

[quote]Clarification AMD asks us to point out that Hammer schedules haven't slipped from its previous advice, as we originally suggested in this article. A spokesman from the company told us that desktop versions of Hammer are still planned to ship (for revenue) in Q1 2003 with systems on shelves at the turn of Q1 2003, not the second half of 2003 as we stated. [/quote]
Phew.

http://www.spec.org/osg/cpu2000/docs/runrules.html#toc_4.1

[q]4.1 Rules regarding availability date and systems not yet shipped

If a tester publishes results for a hardware or software configuration that has not yet shipped,

The component suppliers must have firm plans to make production versions of all components generally available, [b]within 3 month/b] of the first public release of the result (whether first published by the tester or by SPEC)[/q]

Don't forget those are only 32-bit benchmarks. They suggest another 20% improvement with the 64-bit compiles.

So, basically, the Sledgehammer is going to perform a little worse than the Itanium 3 in SPECint and a little better than Itanium 2 in SPECfp. Combine that with a release date very close to Itanium 3 and the recently announced lack of some of the reliability features high-end system vendors expect, and you've got what is starting to look like a budget web server processor and not much else. Crap.

Another good link is page 3 of this 3 page analysis here:
http://babel.altavista.com/urltrurl?lp=de_en&url=http%3A%2F%2Fwww.z...t.de%2Ftechexpert%2Fartikel%2Ftests%2Fcpu%2F2002...[sta.com]
Look at the graphs and see how the Opteron scales compared to the P4.

[q]
AFAIK Intel and AMD have specially optimized compilers to get high spec-scores.

AMD doesn't have _any_ compiler tech of its own, they used Intel's compilers for SPEC submissions in the recent past.
[q]

You´re right. >:´-(

The reason for the "estimated: spec is given here:
http://www.realworldtech.com/page.cfm?AID=RWT101602161848

To quote:
" ....these numbers must be classified as "estimated" to satisfy the strict disclosure requirements of the SPEC score submission protocol, Fred Weber claims that the numbers were indeed obtained on real hardware as it exists in AMD's laboratory. The AMD Opteron, operating at a frequency of 2.0 GHz, and coupled with registered PC2700 DDR SDRAM memory achieves a SPEC int 2000 score of 1202, and a SPEC fp 2000 score of 1170. These numbers were obtained by using Intel compilers and the Opteron processor operating in 32 bit mode. Fred Weber claims that with compilers that can properly optimize for the AMD-64 bit architecture and take advantage of the additional architectural register available, upwards of 20% performance may be obtained for the same processor operating at the same frequency in 64 bit mode. The additional caveat that the 20% performance delta would be application dependent, and subject to the memory access patterns of the workload. For a workload that is CPU bound and has a heavy register pressure, the speedup of 20% may be observed. For a workload that is not similarly bound, and accesses memory frequently, the benefit of the additional architectural register is greatly lessened."

The reason they are "estimated" is that SPEC does not let you publish "official" scores unless the whole system is shipping. The Opteron is obviously not shipping, but that doesn't make the scores any less accurate, just harder to verify

I see the best niche for these processors is in the supercomputer areas where there are hot-swappable board arrays. The FPU crunching power and cost/performance ratios make these a good processor to use, but the power consumption may be a bit of a problem. Can you say Nuclear?

Otherwise they will more than likely be relegated to the ranks of the chosen few... the tinkerer (i.e. TechReport junkies) who wants to pump out folding units in record time and get good gaming performance. With AMDs history of not delivering on promised parts, I am sure some vendors are more than annoyed and will stay away until there is something really compelling to use them.

-LS

Why is it estimated? uh, because it's not a final product yet. What a silly question!

I work in one of those places where the "REAL" money is :-)

We're mostly Sun boxes in various configs so the lure of 64 bit computing via Hammer isn't eactly tempting. It is kinda tough to see where Hammer would fit in a big corporate data center. We have wintel boxes for file/print/exchange/etc, but our real growth is on the unix side. So far, every attempt to use Linux has resulted in utter failure for a myriad of reasons so it's Sun or nothing for now. I suspect many other companies are in similar situations with IBM and HP unix.

Still, I want for hammer to succeed. I can se it being a big hit for small business servers and workstations so I keep my fingers crossed. AMD only has the fab space to supply ~10% of the CPU market so it doesn't need to be a huge success for AMD to make a lot of money.

Originally Posted by davidwright
I am getting one of those AMD Bart Simpson things, when they come out.

The "REAL" money is in places most of us have never seen... Think the render farm for LOTR, think nVidia's corp server farm, or Google's, or what have you. That $5000 desktop, or even $10,000 workstation is not really that worthwhile for the CPU mfr. The $2000 server chip, on the other hand....

I want AMD to be waiting for a better economy in which to release Hammer, but I am worried that is not the reason for this delay.

[quote]With losses mounting and AMD's one potential savior delayed, I fear we may not see Intel forced into a round of price cuts any time soo /quote]

Does this mean the only reason you are eager to see AMD release Barton and Hammer is to see Intel prices drop?

the 2GHz estimates are impressive but it's an opteron not a clawhammer being quoted. opteron means 1MB cache and not one but two channels ddr. the claw has one channel and est half that cache. so if the memory controller isn't supa dupa and cache is critical then barton may indeed be faster than a clawhammer.

btw ace's thread is worthless. the real money is in premium desktop not server so why put it in server first.

[q]Read this thread over at Ace's http://www.aceshardware.com/forum?read=80039892 They seem to have all the reasons for the "estimated" score with alot less speculation.[/q]I see a lot of speculation about what's happening with the processors and no discussion whatsoever of the "estimated" score. Did you post the right link?

So will Barton keep us satisfied until then?

Will AMD stay close to the top of performance with Barton?

I am still waiting. I guess this p4 will have to do until Hammer graces my desktop.

I am waiting AMD! You got my money... if it ever comes out.

Don't forget all the results released so far are 32 bit and there is speculation over at Silicon Investor that a true 64-bit compile will give another 20% performance increase. Gotta go work I have a job, yay!

Read this thread over at Ace's http://www.aceshardware.com/forum?read=80039892 They seem to have all the reasons for the "estimated" score with alot less speculation.