More Bulldozer Benchmarks (Supposedly from AMD)

[inf64]

Yes. But when will the FMA4 and XOP recompile happen, and to what degree in the server environment?

And we need the most important part, good power measurements. Really looking forward to Johans review.

It can happen really soon. It's a matter of compiler switch. All of the compilers (except ICC) will have the FMA, XOP and AVX targets for Bulldozer.
Note that on desktop ,Zambezi may need and update to windows task scheduler. Remember that JF-AMD said the final silicon production started in mid or late August ,so whatever was shown to the press on the August 31st was not this. This fits good with October launch as you need ~6-8 weeks from production to launch.

 

[dbigers]

IF THESE ARE REPRESENTATIVE OF FINAL PERFORMANCE

Only score I am interested in is the Cinebench score. I have a small rendering farm that needs updating and I run Cinebench and Lightwave daily. If these are real, I am not too impressed. The Cinebench score at stock is barely better than the 1100T. An "8 core" at 3.6 ghz only gets a hair better than the X6?

Like I said, if these are real. That is a big IF and until reviewers get them in their hands and test them properly I will hold off on my decision. If these are correct though, the 2600k looks like it will be going in a few boxes for me.

Also, the gaming comparison is pure PR at its best. No one buys a 980 for gaming. They might use it for that, but that is probably in addition to its main purpose. Like I said, PR at its best.

 

[bunnyfubbles]

Totally pathetic. And I don't mean the slides. I mean "if the slides are real" the people still hoping they are fake just because they are in bed with Intel.

yeah, the fact that we've been inundated with fake slides and benchmarks (some very recently from the very source website) has nothing to do with it anyone's skepticism

also, these slides really aren't very flattering, an intel fanboy should be ecstatic, they really only suggest - at least for desktop/consumer use - that AMD has what might be, at best, a competitive product up to the i5 2500K level. But considering there are plenty of scenarios that aren't benched, such as single-threaded performance and power consumption, there's still a chance that Bulldozer could be in pretty big trouble in the grand scheme of things if it turns out to be a dog in such aspects despite these favorable slides.

My past 4 systems have all been Intel and haven't gone AMD since my Opteron 165 (i.e. since Core 2). If anything I'm hoping these are fake because I have been hoping for higher multithreaded performance to make Bulldozer a clear winner against my 2600K at least for my encoding work (of which, if these slides are true, it appears my 2600K wins more than it loses in that regard), particularly ever since discovering the nerfing of X79 for s2011 giving me little reason to invest in what now appears to be a very short lived and expensive platform, I'd be better off grabbing an i7 980 for my old s1366 rig and waiting for Ivy Bridge.

 

[Skott]

I still want to see independent reviews from non bias legit reviewers of retail chips. I think its the only way to finally settle all the rumors and so called leaked info.

 

[IlllI]

i cant wait until these are finally released. not b/c i'm anxious to get one, but just so this whole saga can finally be put to rest.

 

[aigomorla]

Am i the only one thats CRACKING UP at these AMD marketing ideas?

No seriously.. did they hier a bunch of ADD people to add bling and brilliant colors ?

can we agree this is really really funny... comparing it to a 980X and increasing the cost of when u can compare it to a 2600K or a 2500K.

Can we even say Bulldozer is in the same class as the 980X?
And what happened to the 990X?

What me to show u guys benchmarks which would smash bulldozer silly on a 990X?
And yes i know real bulldozer scores arent out there, but my 990X scores are nothing to laugh about, and im sure they will still slaughter bulldozer.

Seriously... here im trying to exclude all the unfair CPU's intel has against AMD in other threads, and AMD themselves is commiting seppuku.

 

[OVerLoRDI]

dude, it's been long-established that both the x-axis and the y-axis goes to 11 for all things bulldozer.

def true

Who knows if these are real. They look encouraging compared to all the doom and gloom we have heard so far, but again are they real? ugh...

I think I'm going to just avoid this sub forum until october 12th.

 

[veri745]

the fact that these charts have no units on their x/y axis makes me rage with the force of a thousand suns

The only chart I see in the entire slide deck that doesn't have the axes labeled is the TruCrypt comparison between PhII X6 (labeled as 1X) and FX (labeled as 3.7X faster)

 

[aigomorla]

u know what i see why those people left amd now.

with advertising like this, i would even jump ship.

they obviously have no idea how to present a believable scale.

 

[NostaSeronx]

u know what i see why those people left amd now.

with advertising like this, i would even jump ship.

they obviously have no idea how to present a believable scale.

LOL, The one who shall not be named has better numbers than these scores lol!!!!

:awe:

 

[bryanW1995]

Soo... AMD slides claiming they can beat the 2500k, and almost match the 2600k.

AMD made huge improvements over the Phenom II if right... the benchmarks most impressive are not the gameing ones, but rather the software ones (those where the ones AMD was doing worst in compaired to Intel).

Not bad and if their prices are right... ~245$ for the top model...
thats actually gonna give Intel a run for its money (until SB-E is out).

I think these #s are real, because the slides look "real" (like the AMD ones).
*IF* they are Im extremly happy, this means AMD is nearly at 2600k levels in all area's.

Why will it do anything at all to intel? They'll just introduce the 2700/2800/2900k to replace the 2600k's pricepoint and drop 2600k down to take 2500k's spot. They've been planning to do so for months, they're just waiting for BD to finally arrive.

 

[classy]

I don't know if these benchmarks are real or not, but this notion that the 2600K is some god cpu in gaming over a 980X is a laugh. They are similar in gaming, very close and the 980X is faster than the 2600K in multi-threaded apps. So all these cracks about AMD picking the 980X for comparison is a joke. The 980X is arguably the 2nd fastest desktop cpu on the planet.

http://techreport.com/articles.x/20486/1

http://www.anandtech.com/bench/Product/142?vs=287

 

[Rubycon]

And yes i know real bulldozer scores arent out there, but my 990X scores are nothing to laugh about, and im sure they will still slaughter bulldozer.

X5690 x 2 at 4.5GHz. Cinebench. To the track, now! :biggrin:

 

[Imouto]

All these slides are fake. They're damn awful with terrible typesetting and it's impossible that they've been made by any professional in the matter. Even lines are broken in some of these images.

Stop looking at the scores and think a bit about the container ffs.

 

[T_Yamamoto]

Just stop listening to rumors and just wait for bd to comeout! Lol

 

[Piano Man]

I want gaming benchmarks vs. i5 2500k and i7 2600k at stock and then both OC'd to max with air.

gogogogogo

 

[bunnyfubbles]

Why will it do anything at all to intel? They'll just introduce the 2700/2800/2900k to replace the 2600k's pricepoint and drop 2600k down to take 2500k's spot. They've been planning to do so for months, they're just waiting for BD to finally arrive.

I think that will only be if Bulldozer is actually competitive with the 2600K, the most recent word is that the 2700K is just going to be an even higher end product and the 2600K will remain in its same price bracket. It would make sense; no need to shift pricing if there's no competition and chips continue to sell well

 

[wanderer27]

Am I the only one that noticed that the Intel System has twice (at least) the RAM that the AMD System does?

I guess they're trying to do a $ comparison, but seems to me the extra RAM would help just a smidge.


What the hey, they're doubtless fake slides anyway . . .


.

 

[micaturbo]

I was thinking the same...I'm guessing the whole dual vs triple channel would mean the 980x rig would cost a bit more in the RAM dept. That definitely looks like double the price, though.

 

[RobertR1]

These marketing guys are sure earning their money! They're having to pick n choose which Intel processor to compete against on each slide.

 

[dac7nco]

I'm beginning to have second thoughts on the FX chips from "joke" to "possibility". I just talked with a NewTek rep who strongly suggested I prepare to replace my 12U of Lynfield Xeons with AM3+.

Daimon

 

[JFAMD]

If benchmarks are from AMD, they will be on AMD.com.

 

[HurleyBird]

Besides AMD crippling performance on the Intel rig, I don't see how those eyefinity benchmarks could possibly be explained.

That being said, as an eyefinity owner, I'm intrigued. I couldn't care less if I lost 20% performance on everything else if I gained that 20% back in gaming. That being said, it's a bit too weird to believe.

 

[RussianSensation]

Whenever Intel, AMD, Nvidia release their own slides, they always present their benchmarks in the best possible light (no offense to any of these firms, but who is going to present slides and show worse performance for their product than their competitor???).

Remember this?

Mind you, this was published by a high-end hardware website - Techspot, similar to TechPowerUp, not by Fudzilla/Semiaccurate. Really, GTX580 SLI is 60-80% faster than HD5870 CF?

What about this?

TechPowerUp published HD6990 deck:
http://www.techpowerup.com/141317/AMD-Radeon-HD-6990-Press-Deck-Leaked.html

HD6990 is 67% faster than GTX580 on Average (at 2560x1600), according to AMD.

Yet, the same website found HD6990 to be only 38% faster at 1920x1200 and 53% faster at 2560x1600.

 

[Silverforce11]

They forgot to put Hawx2 demo in there.. the green bars would be OFF THE CHARTS!!

Edit: 6990 OC, 1600p. It's about right.