First REAL Barcelona Benchmarks

[Parrot]

Here are REAL Barcelona benchmarks:

CLICK ON IMAGE TO ENLARGE

 

[exarrkun]

if i see this on 5 independent sites with a good reputation
then i will trust it

just too much BS around on the web these days...

 

[San Pedro]

I don't count company propaganda as real benchmarks.

 

[ap90033]

Yeah its crap. If its so good give one to Toms Hardware and one to Anandtech and let them benchmark them...

 

[Synthetickiller]

It would be nice if some specifics were given other than processor and under what conditions the tests were administered.

This is silly. Its not even science. 😀

 

[skittle]

actually if you go by their charts, and assume they wanted to compare at the same frequency (it only makes them look better). it shows a 52.645% advantage in specfp, AND 50.133% advantage in perf/watt!!!!!!!

they simply chose a lower clocked AMD chip to fudge the graph... so the delta between the two processors doesnt look so... extreme (the amd chips would be hitting the very top of that chart).

Sounds like a bunch of BS to me.. give us some REAL benchmarks

 

[wolverinero79]

I wonder why they didn't use the 5355.

I wonder other things, but we'll get to those when we have real, non marketing benchmarks. :-D

I'm really surprised http://multicore.amd.com/us-en/AMD-Multi-Core/Products/Barcelona/Performance.asp x doesn't have even fake benchmarks 🙁

 

[jaywalker256]

is this just bothering me or did anyone else notice it...

why does the bar, show 69.5, when it actually goes well past 100 on the graph meter.... any special reason for this?

 

[Hatman]

That was released ages ago and I thought AMD already took it down.

 

[skittle]

the graph is scaled to show the percentage delta.

notice intel is at 100%, while the amd chip is at 125&130%

 

[NMDante]

A slide = benchmark?
Not only that, but 1 SPEC test = performance of the CPU for all other applications?
Right.

How about running more than 1 test, using independent parties before claiming anything? It's what...a few more weeks until release, and all we still see about Barcelona are slides? Great.

 

[Mephistopheles]

This is complete crap. It's useless.

This compares the highest-clocked Barcelona at launch that hasn't been launched yet to the 3rd highest clocked Clovertown Xeon (70% of the speed of the 5365) that is already on sale for a long time now.

It's really ridiculous. Anyone who seriously thinks this looks good is a fool.

Edit: Ooo, I got it now: "Facts are stubborn, but statistics are more pliable."!!!

 

[turpit]

 

[akahuddy]

Yeah its crap. If its so good give one to Toms Hardware and one to Anandtech and let them benchmark them...

Don't give one to Tom's... it'll be a year before they get it benched. Send it to Anand, XS, or my favorite, X C P U S! 😀

http://img341.imageshack.us/img341/4610/yeahright2hc4.jpg

Beautiful.

 

[croc]

Here are REAL Barcelona benchmarks:

CLICK ON IMAGE TO ENLARGE
http://www.iian.ibeam.com/events/th...0070726084721294707/default_large/Slide45.JPG

Uh, welcome back Parrot.... sort of.

Still spouting the same old AMD non-sense, I see. Please go back to wherever it was that you were hiding, until 1 of several things happens:
1. Intel goes bankrupt... One of your earlier 'facts' that was to happen in the near term. 2007, wasn't it?
2. Amd goes bankrupt. I'd like to think not, but could be closer than 1.
3. Real benchmarks get published by real testers using real cpu's that I can go out and buy. Oh wait... I CAN get the xeon tested against, and several better as well...

 

[Ycon]

So they admit Intel uses slightly less power at a higher clockspeed?

 

[akahuddy]

So they admit Intel uses slightly less power at a higher clockspeed?

Sigh... No. That was a joke by Turpi t.

 

[vsary6968]

can anyone can foreward to AMD some information? Why don't AMD continue using integrated L1 128 ?For the K7 they use L1 128 and now how come they use on K8 AND K10 64KB/64KB?

why don't AMD upgrade their L2 cache on Barcelona? they should upgrade for each core use 1mb x4 equal 4MB. But for the L3 can stay at 2mb. If they do some of this change for the Barcelclona, I think AMD Barcelona will have no competition. Even the Penryn or Hypertown come out, I don't think they can match wit h the Barcelona or Shanghai. Can anyone forward this information to

 

[vsary6968]

can anyone can foreward to AMD some information? Why don't AMD continue using integrated L1 128 ?For the K7 they use L1 128 and now how come they use on K8 AND K10 64KB/64KB?

why don't AMD upgrade their L2 cache on Barcelona? they should upgrade for each core use 1mb x4 equal 4MB. But for the L3 can stay at 2mb. If they do some of this change for the Barcelclona, I think AMD Barcelona will have no competition. Even the Penryn or Hypertown come out, I don't think they can match with the Barcelona or Shanghai. Can anyone forward this information to AMD?

this would be helpful for them to upgrade their L1 AND L2 cache. L1 use to be 128/128kb. Now is only 64kb/64kb . But for the L2 cache is losing by half from 1mb down to 512kb. Is their any specific reason to down grade their L1 and L2 cache?

 

[TechnologyCoordinator]

http://img341.imageshack.us/img341/4610/yeahright2hc4.jpg

Tuprit, that was mother bleeping hillarious!!! LMFAO!!!

 

[NMDante]

WTF? Why am I the first person point out that what Tuprit posted if mother bleeping hillarious? LMFAO!!!

No, I thought it was very clever and done quite nicely.

 

[turboflame]

Here are REAL Barcelona benchmarks:

CLICK ON IMAGE TO ENLARGE
http://www.iian.ibeam.com/events/th...0070726084721294707/default_large/Slide45.JPG

Parrot's back

woohoo?

Protip: "Benchmarks" by the first party are meaningless

 

[TechnologyCoordinator]

NEW BENCHMARKS!!!


Barcelona: 1,605,602
Conroe: 36,811

Barcelona clearly whoops Conroe.




(Benchmarks based off of population of Barcelona, Spain and Conroe, Texas)

 

[ches111]

Here are REAL Barcelona benchmarks:

CLICK ON IMAGE TO ENLARGE
http://www.iian.ibeam.com/events/th...0070726084721294707/default_large/Slide45.JPG

Dude you kill me.

First REAL benchmarks as released on a slide FROM the company that makes the product.

Brilliant! Simply Brilliant!!

Yeah I will wait for some "third party" results.

I bet you listen to the sales pitch every time someone is trying to sell you something.

Cmon,

Fess up! You own at least three time shares don't cha!!

 

[Spikke]

This thread is great. Is it an April fools joke? Oh wait, it's July. I was expecting a really real benchmark and all I got was something that looks like it came out of a company brochure.