IBM published Barcelona SpecCPU results.

[terrace215]

some manufacturers have been known to toss results into the ring early...won't name names, but...this is an OEM pissing match, remember.

dave

Have you ever known a Tier 1 OEM to submit pre-production scores to SPEC???

They have to say when the hardware will be available to customers... and pre-prod never would be.

[lapdog]

IBM submitted scores from a defective CPU!
I just can't believe it.

[dave_graham]

Have you ever known a Tier 1 OEM to submit pre-production scores to SPEC???

They have to say when the hardware will be available to customers... and pre-prod never would be.

*cough* it happens. i've known several companies to publish "machines" that suddenly have config changes (significant) at time of release.

cheers,

dave

[dave_graham]

IBM submitted scores from a defective CPU!
I just can't believe it.

didn't say that...was just questioning results.

dave

[lapdog]

guys, again, there are significant performance deltas between the release stepping of the core and the B0/B1 steppings.

without much more information than that, you just need to keep your panties in check.

dave

I guess I misunderstood; you certainly seem to suggest IBM used B0/B1 that had problems and have been fixed (were defective.)

[dave_graham]

I guess I misunderstood; you certainly seem to suggest IBM used B0/B1 that had problems and have been fixed (were defective.)

IF the identified CPUs were B1s, yes. don't know as that's not typically revealed.

dave

[kl0012]

[QUOTE=hollo;2421465]LS-DYNA, being a simulation program, will naturally be floating point heavy, like most simulations (eg molecular dynamics on supercomputers).
[QUOTE]

LS-DYNA is a heavy bandwidth benchmark and quad cores not the best solutions for this benchmark.
Here is the list of 8-core scores:
http://www.topcrunch.org/benchmark_r...count=8&type=5

[tam2]

Which is more important in stanford's folding@home?
SPECint/SPECfp????

-tam2-

[accord99]

Which is more important in stanford's folding@home?
SPECint/SPECfp????

Neither really give you a good hint:

http://www.techreport.com/articles.x/13176/7

[lapdog]

IF the identified CPUs were B1s, yes. don't know as that's not typically revealed.

dave

You have lost all credibility in my book.

[THE JEW (RaVeN)]

You have lost all credibility in my book.

You seem to give and take said credibility quite easily

Dave's input, whether here or on 2cpu or any other forum, is valued. People may disagree with it, hold it highly, or merely lump it in with other voices, but he's one of the people giving actual input (ie. not regurgitating FUD's tabloid articles). That, in itself, gives him "cred" in my book.

Personally, I'm waiting at least a month or two after the release to form an opinion. Launch day presents too small a sample most likely tainted by brown-nosing. The industry itself will let us know whether they're a good product or not.