Bulldozer 40% - 50% faster than Sandy Bridge?

[bobdozer]

So suggests donimhaber who got hold of AMD internal documents stating BD is 50% faster than an i7 950 in applications and gaming - if the gaming bit is true that would be amazing.

http://www.xbitlabs.com/news/cpu/display/20110114134306_AMD_s_Bulldozer_Microprocessors_Expected_tffer_50_Higher_Performance_than_Core_i7_Phenom_II_Chips.html

 

[jaguarskx]

I need to see some preliminary benchmarks.

 

[Terry1212]

The i7 950 isn't Sandy Bridge.

 

[fazers_on_stun]

JF-AMD already dispelled this rumor on another site - apparently this Turkish guy donimhaber doesn't have any access to AMD internal documents despite what his website claims..

 

[bobdozer]

Do you have a link to that?

 

[fazers_on_stun]

Do you have a link to that?

I'll have to look for it on my work PC since that is where I found it. But on Hardforum JF says this:

Conspiracy theories aside, we just don't release benchmarks prior to launch. And I personally know that I have never approved a "leaked" benchmark.

I can't speak for the other divisions, but I know that it is not policy to release benchmarks prior to launch. There are very specific business reasons why letting the cat out of the bag, regardless of what the performance is, can be a bad thing.

Obviously there are those who keep pushing becase they want to know, but from a business standpoint there is more reason not to release benchmarks prior. Did intel release any offical benchmarks on SB prior to launch? No. Were there "leaked" benchmarks? Probably.

We do occasionally do some demos prior to launch (like with the Zacate last September) but that was not a benchmark, it was a comparison, nothing was audited.

 

[fazers_on_stun]

OK, found the thread over on http://www.techpowerup.com/138328/Bulldozer-50-Faster-than-Core-i7-and-Phenom-II.html?cp=4

by JF-AMD (January 13th - 1:09 PM) - Reply
OK, first off, let me start by saying that we don't comment on speculation. If people want to speculate on this, have at it. This is not an AMD article and I have no idea who this guy is.

We are in the middle of quiet period so you would never see AMD making a performance statement. I have no idea about the validity of the article because, amongst other things, I don't speak turkish.

To date the only performance statement we have made is around the server throughput of Bulldozer.

IMO, this Donimhaber guy is just angling for his 15 minutes of fame..

 

[1965ohio]

Are they ever going to launch Bulldozer soon anyways? I been reading about this for ages. First some BS article similar to this on the Anandtech forums some months ago saying someone leaked an engineering sample and it is faster than all Intel current i7's... but hell, by the time they release it, Ivy bridge or something else will already be out to overtake it.... or even the Haswell architecture.

Every time Intel does a die shrink, they always incorporate more main board controller functions like the GPU and they double their bandwidth. By the time Bulldozer is released Intel will have everything integrated into the CPU, LOL... You will be able to buy a CPU that has RAM, SSD, GPU and all in one chip by the time they are done building the hype for Bulldozer and finally release it... haha

 

[jf-amd]

JF-AMD already dispelled this rumor on another site - apparently this Turkish guy donimhaber doesn't have any access to AMD internal documents despite what his website claims..

I never claimed that he didn't have access to documents (because I don't know). All I said was that this was not an AMD leak.

Are they ever going to launch Bulldozer soon anyways? I been reading about this for ages.

Client parts launch next quarter.

 

[Raidur]

If it was strong enough for AMD to claim that then BD would have near a 100% IPC increase from Phenom II.

:lol:

Right.

 

[fazers_on_stun]

I never claimed that he didn't have access to documents (because I don't know). All I said was that this was not an AMD leak.

OK, I posted that before going to see exactly what you did say about it. However your other statements seem to state you think there are no such documents, since it is against AMD policy and also your pre-reporting quiet period.

Anyway, I'm looking forward to the reviews once they come out.

 

[jf-amd]

OK, I posted that before going to see exactly what you did say about it. However your other statements seem to state you think there are no such documents, since it is against AMD policy and also your pre-reporting quiet period.

Anyway, I'm looking forward to the reviews once they come out.

Since it was on the client side, I have no idea what they are referring to, and quite frankly I'd rather not get into the middle of it, I have my own business to run.

I am not saying that this couldn't be true. I am just saying that it was not initiated by AMD. I am very strict on performance data. If there is any performance data that comes from AMD, it will be in my blog first. Outside of that, anything else is suspect.

 

[amdfangirl]

It's highly likely to be false given that how long it is before the product launch.

 

[bobdozer]

Why is that, anands preview of sandy bridge wasn't so far off the final result and that was months ago too.

There will be bulldozer chips out there now in the hands of people who shouldnt have them.

 

[amdfangirl]

Well Anand had the chip and he's reliable.

 

[fazers_on_stun]

^ +1. I seriously doubt this guy in Turkey had any documents, let alone an ES chip. Now if he were in Taiwan or some other mobo/OEM site, then I'd give it more credibility.

 

[amdfangirl]

Unlike tabloids like INQ and Fudzilla, I've never heard of the site, let alone seen if it's predictions are right.

 

[excelerater]

its the normal seesaw.......Intel steps up AMD steps above Intel
and then Intel steps above AMD.....

 

[bobdozer]

Donanimhaber is one of europe's largest websites and they often get hold of information before release. They were one of the first to get 6950 and 6970 graphics cards.

 

[Wisecracker]

It's highly likely to be false given that how long it is before the product launch.

Not necessarily so (and maybe John can help us with this).

AMD has teams of engineers pounding away on microcode to repair errata, improve efficiency of instructions, etc., to optimize, enhance or even add new functionality. With the new design this becomes even more critical for AMD as they ramp to production. No real reason to 'spill the beans' as they continue to optimize before production, right?

There is an inherent "double-whammy" in this round: new design AND new instructions. This is not necessarily a bad thing.

My understanding (always questionable ) is a primary result of the AMD/Intel legal wranglings is 'Chipzilla' must be more forthcoming on future instruction sets. In the past Intel would not divulge this information until a microprocessor was physically on the retail market -- keeping AMD months (if not years) behind the curve. What's the big deal you ask?

Instruction set optimizations are both backward- and forward-looking. We want 'old' software already implemented to perform better, and we want 'new' software to 'scale' with future instructions. Is this making sense?

The history of the 'Chip Wars' are full of examples -- the latest being SSE4. With Penryn/Nehalem Intel brought forth dozens of 'optimized' SIMD instructions: SSE4, SSE4.1 & SSE4.2. Being behind the curve, AMD brought forth their own 'unique' SSE4a instructions -- all three of them (being a 'Southern' American we would normally preface this with "Bless their heart" - LOL - just kidding, John!).

Guess what? Bulldozer will universally support almost all SSE4, SSE4.1 & SSE4.2 instructions, in addition to the new AVX instructions PLUS their own AMD "Instructions formally known as SSE5" (and now commonly referred to as XOP and FMA4).

And most importantly (the reason that AMD is so exceptionally tight-lipped about things these days): Randy Allen

Randy was doing his job a few years ago, and managed to stick both his feet into his mouth. He took a single unique instance of performance gains, and fairly or not, it was somehow implied as 'across the board' performance gains.

Bless his heart!

 

[jf-amd]

Instructions have been out with tools/compiler companies for a while. I think you are speculating a bit much on this.

There are obviously parts out there, but all I can say is what I know - this was not something that AMD engineered as a leak.

 

[Wisecracker]

Instructions have been out with tools/compiler companies for a while. I think you are speculating a bit much on this.

There are obviously parts out there, but all I can say is what I know - this was not something that AMD engineered as a leak.

Just to be clear --- Am I incorrect in speculating that BD will support SSE4, SSE4.1 & SSE4.2, along with portions of AVX and XOP and FMA4 ??

Thanks.

 

[Stardude82]

Just to be clear --- Am I incorrect in speculating that BD will support SSE4, SSE4.1 & SSE4.2, along with portions of AVX and XOP and FMA4 ??

Thanks.

SSE4 is a mess.. BD probably won't have the SSE4.1-2, but it will have SSE4a and the other extensions you listed.

EDIT: Well, maybe AMD got SSE4.1-2 licensing after all. Looks like they gave up on the SSE4a bit.

 

[bobdozer]

yeah theyre 50 percent faster thats why Dirk resigned

Dirk didn't resign.

 

[saint19]

Sweet, I love rumors