Bulldozer Information Thread.

[CDdude55]

You think this would perform equal to nehalem performance?

Possibly, Nehalem is almost around Sandy Bridge performance at times and i do believe that AMD's Bulldozer chips will probably hit around SB performance but i still believe that the overall performance crown will remain at Intel with SB-E.

Then again, Bulldozer is expected to be more affordable then SB-E and more likely competitively priced toward the 1155 SB chips, and if performance is decent enough for your average gamer/enthusiast who doesn't work at Goldman Sachs, then BD might be one popular line of chips.

 

[repman244]

Official Benchmarks:

translated:

In the infamous hotel room two blocks away from IDF, AMD is our first official benchmark results of the AMD FX "Bulldozer" processor shown. Since both tests are presumably carefully chosen for the new platform to put in a good light, we can based on these numbers no conclusions. Yet we want the results do not remember.

First they showed a comparison between a new unspecified AMD FX processor and an unspecified Intel processor i5 Sandy Bridge, with the help of the program Handbrake a video of 5 minutes is converted to H.264 video in SD resolution. The AMD FX processor with eight cores perform this function with an average of 223 frames per second, the i5 with four cores came in at 188 fps. Both systems will be comparable in price according to AMD, which it wants to show that AMD a better price / performance offering. That may be the case, but who has a more negative view would conclude that AMD is looking to double the number of cores needs to less than 20% better performance available.

In the second demonstration showed up one game in three Dirt 2560x1600 resolution, running on two Radeon HD 6790 cards in Crossfire. In one system, the cards are combined with an Intel Core i7 980X, in the second system with an AMD FX processor. The Intel machine could produce an average of 80.9 fps, the AMD machine averaged 82.8 fps. The much cheaper AMD FX processor is faster than the Core i7 980X - if you want to show - though we should really stick through it: Dirt 3 to this resolution and the chosen card is very GPU thus limited. The extra 2 fps of the AMD-101 system can be explained ways, all of which are not necessarily traceable to CPU performance.

For real benchmarks, we'll have to wait for the AMD FX processors actually available. Fortunately, that probably will not take too long.

However revealed also a working laptop with AMD processor Trinity, the next generation AMD APU based on Bulldozer cores and a next generation GPU. Trinity middle of next year will reach the market.

http://nl.hardware.info/nieuws/24619/eerste-officiele-benchmarks-amd-fx-processor


The benchmarks don't tell very much, especially the gaming one since it's probably more GPU limited than CPU.

 

[xenocide]

How exactly are those official? There is no data that you can actually see. No video of the games running, no charts depicting max\min FPS, just some guys saying they saw something. How is that any different from the random Chinese sites that post Cinebench SS' and such?

 

[repman244]

How exactly are those official? There is no data that you can actually see. No video of the games running, no charts depicting max\min FPS, just some guys saying they saw something. How is that any different from the random Chinese sites that post Cinebench SS' and such?

Well the title on the site says so (I only wrote what I saw ), but I agree you can't draw any conclusions from this.

 

[xenocide]

Well the title on the site says so (I only wrote what I saw ), but I agree you can't draw any conclusions from this.

I mean if that's true, that sounds awesome, and means Bulldozer might actually be worth the wait. One could easily point out that Dirt 3 test is a little wonky because of the strain on the GPU though. :/

 

[heky]

If this is true, the 8 core FX with a higher clock(stock) is only a little faster than a 4 core i5 2500K with a lower clock(stock). Weak, seriouslly.

 

[Dent1]

If this is true, the 8 core FX with a higher clock(stock) is only a little faster than a 4 core i5 2500K with a lower clock(stock). Weak, seriouslly.

Read the article again. The 8 Core FX beat out a $600 an Intel Core i7 980X.

 

[heky]

Ehm, nope, the article says an unspecified i5 sandy bridge and an unspecified AMD processor for the first test(video conversion) and in the second test(gaming), there is the i7 980x, but we all know why they chose that processor, becouse sandy bridge is faster in games than i7 980. So i have to dissapoint you, BD just isnt that great as it appears, judgeing by the article.

 

[ensabrenoir]

If this is true, the 8 core FX with a higher clock(stock) is only a little faster than a 4 core i5 2500K with a lower clock(stock). Weak, seriouslly.

+1

I believe this is bulldozer's performance level. (i5). I believe its a good chip, just a few years late to the party. If this came out before sandy Bridge, Amd would've ruled. Now no matter when they release it....Intel will just pull the release trigger and

CURVE-STOMP

Sandy bridge E is on the scene

 

[TheMailMan78]

According to that article they are using a AMD cpu thats in the same price range as an i5 Sandy. The AMD processor is just under 20% faster then the i5 Sandy for the same price.

In the second test its about two frames faster then a 980 and costs half. So yeah Bulldozer looks to be a beast no matter what.

On a side note at that resolution its all GPU based. So the gaming tests are no good. But it does show that BD wont be a bottle neck in crossfire/SLI rigs unlike the Phenom.

 

[Dent1]

Ehm, nope, the article says an unspecified i5 sandy bridge and an unspecified AMD processor for the first test(video conversion) and in the second test(gaming), there is the i7 980x, but we all know why they chose that processor, becouse sandy bridge is faster in games than i7 980. So i have to dissapoint you, BD just isnt that great as it appears, judgeing by the article.

The i7 980x would murder the i5 Sandy Bridge in a proper multi-threaded application i.e. one which takes advantage of its 6 cores and 12 threads.

And before you say, "well the Dirt isnt a game that takes advantage of multi-threading".

The question is, is Dirt 3 multi threaded and can it take advantage of 8 cores. If it CANT take advantage of 8 cores then the Bulldozer is almost just as handicapped as the i5 and i7 980 yet still is faster core per core. Imagine the performance boost when a game that can utilise those 8 cores are used.

The real question is can dirt utilise 8 cores?

 

[TheMailMan78]

+1

I believe this is bulldozer's performance level. (i5). I believe its a good chip, just a few years late to the party. If this came out before sandy Bridge, Amd would've ruled. Now no matter when they release it....Intel will just pull the release trigger and

CURVE-STOMP

Sandy bridge E is on the scene

The only thing that will be "CURVE-STOMP" is the price.

 

[linoliveira]

The only thing that will be "CURVE-STOMP" is the price.

+1
I won't give a leg an arm and a kidney to buy SB-E considering its peformance over SB.

 

[Crap Daddy]

These test are made exactly as they were not supposed to be. It should have been the 6 core Intel vs. the 8 core AMD for handbrake (encoding) and the i5 (unspecified model - might be 2400, 2300?) vs FX in gaming. I'm sure the results would have been totally different. Wonder why they didn't do it that way?

 

[TheMailMan78]

These test are made exactly as they were not supposed to be. It should have been the 6 core Intel vs. the 8 core AMD for handbrake (encoding) and the i5 (unspecified model - might be 2400, 2300?) vs FX in gaming. I'm sure the results would have been totally different. Wonder why they didn't do it that way?

Dude just look at the resolution they used for the gaming. Its obvious these tests were designed to not show anything. All fluff.

The only thing this shows is that BD scales multi-GPU's well. Which is a good sign.

 

[ensabrenoir]

The only thing that will be "CURVE-STOMP" is the price.

Maybe... ok probably. But you gotta admit for its performance level sb was surprisingly cheap for intel. A $300 bruiser that goes toe to toe with its 6 core $1000 beast. Sb-e will more expensive but not ultra expensive.

 

[TheMailMan78]

Maybe... ok probably. But you gotta admit for its performance level sb was surprisingly cheap for intel. A $300 bruiser that goes toe to toe with its 6 core $1000 beast. Sb-e will more expensive but not ultra expensive.

I agree. But the early benches also show that its performance over the current SB are not ultra good ether.

Wild guess is Bulldozer will be on par or nipping at sandys toes and Sandy-E will be maybe 20% faster.

At the end of the day we all win man. Competition is good. The stiffer the better!............thats what she said.

 

[heky]

The i7 980x would murder the i5 Sandy Bridge in a proper multi-threaded application i.e. one which takes advantage of its 6 cores and 12 threads.

And before you say, "well the Dirt isnt a game that takes advantage of multi-threading".

The question is, is Dirt 3 multi threaded and can it take advantage of 8 cores. If it CANT take advantage of 8 cores then the Bulldozer is almost just as handicapped as the i5 and i7 980 yet still is faster core per core. Imagine the performance boost when a game that can utilise those 8 cores are used.

The real question is can dirt utilise 8 cores?

Are you for real, 1core vs 1core @same clock, sandy bridge obliterates the i7 980. Wtf. And even if DIRT cant take advantage of 8 buldozer cores, sandy bridge still beats it, becouse it beats the i7 980 in gaming. i am not saying it sucks, but for so many years of developing its just weak, imo.

 

[cadaveca]

OK boys, there's you're performance, now let's line up like some nice school girls, empty our wallets, and bend over.

Because everyone likes a good ass kicking.

Oh look, they used some apps I use in my reviews. GO figure. And since i bench those apps, I understand the workload they produce.

 

[repman244]

Ah the good old number of cores vs number of cores comparison. Like any of you buy a cpu by comparing that.
And since I see many of you are interested in gaming, you should be looking for the 4 core FX numbers and not the 8 core version (and as we know the 4 core version has the same L3 cache of an 8 core version and very aggressive frequency, which games like).

These numbers aren't really telling much except maybe for the first one (video encoding) but we have no idea which intel chip model was used, second was IMO highly GPU limited and again doesn't say anything.

If we go with the rumored price of ~$260 (which is, mind you for the top model!) which beats a similar priced counterpart in multithread (if we trust these numbers), that's a win. As to single thread performance, well that is a tough one since aggressive turbo could make up for the lower IPC compared to say SB CPU.

All in all I think we should all still wait for more numbers and better comparisons, these don't tell us much.

 

[TheMailMan78]

Ah the good old number of cores vs number of cores comparison. Like any of you buy a cpu by comparing that.
And since I see many of you are interested in gaming, you should be looking for the 4 core FX numbers and not the 8 core version (and as we know the 4 core version has the same L3 cache of an 8 core version and very aggressive frequency, which games like).

These numbers aren't really telling much except maybe for the first one (video encoding) but we have no ideas which intel chip model was used, second was IMO highly GPU limited and again doesn't say anything.

If we go with the rumored price of ~$260 (which is, mind you for the top model!) which beats a similar priced counterpart in multithread (if we trust these numbers), that's a win. As to single thread performance, well that is a tough one since aggressive turbo could make up for the lower IPC compared to say SB CPU.

All in all I think we should all still wait for more numbers and better comparisons, these don't tell us much.

Logic has no place on a tech forum when it comes to rumors man. I should report you!

 

[repman244]

Logic has no place on a tech forum when it comes to rumors man. I should report you!

Sorry! What I meant to say was it will be on par with a Pentium 4, maybe a little better in multithread.

 

[cadaveca]

These numbers aren't really telling much except maybe for the first one (video encoding) but we have no ideas which intel chip model was used, second was IMO highly GPU limited and again doesn't say anything.

Actually, I can play Dirt3 just fine with a single 6950 on three monitors. It's NOT 60 FPS, but it's more than playable. It's not realyl ug limited, but it IS such a res that it's possible a few things are at paly here. This bench is partially useless. With Crossfire added, they should be just breaking above 60 FPS. 2560x1600 is not a killer res like it used to bem adn that app is NOT really GPU-limited.


Encode numbers are 8 AMD BD cores VS. 4 cores of Intel, and I do not see BD doubling i5 performance.

Of course, the starting CODEC for the video file matters too.

 

[mastrdrver]

If this is true, the 8 core FX with a higher clock(stock) is only a little faster than a 4 core i5 2500K with a lower clock(stock). Weak, seriouslly.

These test are made exactly as they were not supposed to be. It should have been the 6 core Intel vs. the 8 core AMD for handbrake (encoding) and the i5 (unspecified model - might be 2400, 2300?) vs FX in gaming. I'm sure the results would have been totally different. Wonder why they didn't do it that way?

Encode numbers are 8 AMD BD cores VS. 4 cores of Intel, and I do not see BD doubling i5 performance.

Of course, the starting CODEC for the video file matters too.

Who reports FPS on Handbrake? I've never seen a benchmark site use FPS on Handbrake as the measure, just time.

Like you said though Cadaveca, without the CODEC (and I'd add settings) it is impossible to know if that was good or bad result for BD compared to the i5.

 

[cadaveca]

Who reports FPS on Handbrake? I've never seen a benchmark site use FPS on Handbrake as the measure, just time.

Since you clearly do not read my reviews:

UH, yeah, that's why I say those benches are for me, because guess what, I report FPS, as I found the time thing a bit suspect: