Fudzilla: Bulldozer performance figures are in

[NostaSeronx]

The decision to bring Opterons into the fold was driven by customer interest in the new Interlagos Bulldozer processor that started shipping this summer.

But it was the latter chip that brought the customers running. According to Fultheim, starting at the end of 2010, he started getting inquiries about support for Interlagos. In particular a couple of "very significant customers" approached ScaleMP and AMD and basically told them if you build it, they will come. ScaleMP now has about 10 customers in the pipeline for the AMD product and expects more coming, thanks to some impressive attributes of the revamped Opteron.

Based on the new Bulldozer architecture, Interlagos should deliver very competitive performance. Coming in both 12-core and 16-core versions, the integer and floating point performance is expected to stack up quite well against the 8-core Sandy Bridge Xeon EP. A 16-core Interlagos can retire 64 floating operations per cycle, compared to 32 FP ops for an 8-core Sandy Bridge. Note that while the 8-core Xeon can also hit 64 FP operations with the new AVX floating point instructions, not all applications will be able to take advantage of AVX -- at least not right away.

They must be really profitable for vSMP to just go exclusive intel to AMD and Intel

http://www.hpcwire.com/hpcwire/2011...ron_support_to_its_virtual_smp_portfolio.html

 

[Abwx]

A guy at Hardware.fr provided this link with what seems
to be the AMD official numbers...

http://www.donanimhaber.com/islemci/galerileri/AMD-Bulldozer-FX-resmi-test-sonuclari.htm

 

[Crap Daddy]

It's from donanimhaber. here's the link:

http://www.donanimhaber.com/islemci...lemciler-teknik-ozellikler-test-sonuclari.htm

Check the photo gallery.

 

[Abwx]

Perfs should be better once there s some optimisations made...:biggrin:

 

[psolord]

A note to reviewers and Anandtech.

F1 2011 is a very good cpu test regarding gaming.

A friend's i7 2600k@5Ghz scored 60% more than my i7-860@4Ghz. We both used GTX 570 @850Mhz, 1280X720, DX11 ultra noAA.

I'd like to see the same on BD when the time comes. Thanks.

 

[grimpr]

A note to reviewers and Anandtech.

F1 2011 is a very good cpu test regarding gaming.

A friend's i7 2600k@5Ghz scored 60% more than my i7-860@4Ghz. We both used GTX 570 @850Mhz, 1280X720, DX11 ultra noAA.

I'd like to see the same on BD when the time comes. Thanks.

Well and heavy threaded perhaps, got a taskman shot of it running on all cores?

 

[IndyColtsFan]

Sigh.....I am really upset with AMD for this botched release. I have a brand new i7-2600K and an Asus P8Z68-V Pro sitting unopened in their original shipping boxes, debating whether I should return them or not for Bulldozer. I don't know why they can't just lift the NDA and let some benchmarks happen; if the release is October 12, they're already manufacturing the chips and have maybe have even started shipping some, I'd guess.

I've got until 10/18 to return these for a full refund but I hate waiting 3 more weeks when they could delay again.

 

[classy]

Sigh.....I am really upset with AMD for this botched release. I have a brand new i7-2600K and an Asus P8Z68-V Pro sitting unopened in their original shipping boxes, debating whether I should return them or not for Bulldozer. I don't know why they can't just lift the NDA and let some benchmarks happen; if the release is October 12, they're already manufacturing the chips and have maybe have even started shipping some, I'd guess.

I've got until 10/18 to return these for a full refund but I hate waiting 3 more weeks when they could delay again.

I would wait. Probably next week some real leaks will happen. But man you are really only talking about saving what $75-125. Which is a nice little savings, but not something I would lose sleep over. And it is the fastest in most things.

 

[psolord]

Well and heavy threaded perhaps, got a taskman shot of it running on all cores?

This is a video link of the benchmark I uploaded, showing what task manager registered (at low update speed) right after F1 2011 benchmark finished, on my 860@4Ghz+570@850Mhz.

I got the same result at 720P, NoAA, Ultra DX11, hence uber cpu limit and the cpu usage would be the same as the above video anyway. This game just likes Sandy.

I got 72fps in both this video and 720p Noaa, while the 2600k friend got 115fps!

 

[IndyColtsFan]

I would wait. Probably next week some real leaks will happen. But man you are really only talking about saving what $75-125. Which is a nice little savings, but not something I would lose sleep over. And it is the fastest in most things.

I'm not worried about saving money (the board, CPU, and the drive I ordered were basically free to me since I bought them with credit card rewards). I don't upgrade often and want to make sure I get the best CPU for the next 3 to 4 years. I tend to think that over that timeframe, a faster 4 core CPU will be more desirable than a slower 8 core, but I could be wrong.

I was really wanting to purchase a BD but AMD's continuous delays and the fact that their behavior doesn't paint a good picture made me go ahead and purchase the 2600K and board. The numbers posted a few posts above look impressive but they're likely from AMD and cherry picked so who knows how much of an impact they would actually have on real-world performance.

 

[classy]

I'm not worried about saving money (the board, CPU, and the drive I ordered were basically free to me since I bought them with credit card rewards). I don't upgrade often and want to make sure I get the best CPU for the next 3 to 4 years. I tend to think that over that timeframe, a faster 4 core CPU will be more desirable than a slower 8 core, but I could be wrong.

I was really wanting to purchase a BD but AMD's continuous delays and the fact that their behavior doesn't paint a good picture made me go ahead and purchase the 2600K and board. The numbers posted a few posts above look impressive but they're likely from AMD and cherry picked so who knows how much of an impact they would actually have on real-world performance.

If thats the case just use the 2600K. There is always something faster around the corner. You can't go wrong with the 2600K, you just can't.

 

[Accord99]

The numbers posted a few posts above look impressive but they're likely from AMD and cherry picked so who knows how much of an impact they would actually have on real-world performance.

How is it impressive? Basically it looks to be the equivalent of a 8 core Sandy Bridge running at 2 GHz that turbos up to maybe 2.5 GHz with 4 threads or less.

 

[IndyColtsFan]

How is it impressive? Basically it looks to be the equivalent of a 8 core Sandy Bridge running at 2 GHz that turbos up to maybe 2.5 GHz with 4 threads or less.

Maybe you're right. Hasn't this guy posted fakes before?

 

[Imouto]

Those slides/promos are 100% fake. Maybe results from an ES but not from AMD since the design is awful, text is misplaced and there's several errors in typesetting and other mistakes.

 

[RussianSensation]

Those slides/promos are 100% fake. Maybe results from an ES but not from AMD since the design is awful, text is misplaced and there's several errors in typesetting and other mistakes.

Looking at their wPrime32M scores:

2600k is about 13% or so faster than the 2500k.

Real test scores:
http://www.bit-tech.net/hardware/cpus/2011/01/03/intel-sandy-bridge-review/7

2600k = 7.222 sec
2500k = 10.70 sec

That's 32.5% quicker. How is "AMD's" 2600k 20% slower than the 2500k in this benchmark?

 

[Olikan]

Those slides/promos are 100% fake. Maybe results from an ES but not from AMD since the design is awful, text is misplaced and there's several errors in typesetting and other mistakes.

yeah.
it's actually the first time that i see slides/promos, without any details about the chip arquitecture.
we may be wrong, but is indeed odd

 

[BlueBlazer]

Are these slides made up? Looking at the Cinebench R11.5 numbers........

Exactly same as here >> Cinebench score of 5.95 @ stock and 7.8 @ 4.8GHz with a FX8150 mentioned on XS forums. :hmm:

 

[Ryun]

They look pretty legit. The only thing that bothers me/looks fishy is that if you look at the Cinebench screen shot in the background the benchmark hasn't even been run. Further, they're not even using an AMD system if you look at the system info. It's a Xeon running Max OS X.

Also, and correct me if I'm wrong, but isn't 5.95 for BD a great score since it only has 4 FP units?

Edit:

Oh that's why... they took the image from the Cinebench site? http://www.maxon.net/uploads/pics/cinebench_screen_09.jpg

 

[BlueBlazer]

They look pretty legit. The only thing that bothers me/looks fishy is that if you look at the Cinebench screen shot in the background the benchmark hasn't even been run. Further, they're not even using an AMD system if you look at the system info. It's a Xeon running Max OS X.

Also, and correct me if I'm wrong, but isn't 5.95 for BD a great score since it only has 4 FP units?

Edit:

Oh that's why... they took the image from the Cinebench site? http://www.maxon.net/uploads/pics/cinebench_screen_09.jpg

That's about as fast as AMD Phenom II X6 1100T (which scores around 5.9x). And it does look fishy, why not show the actual Cinebench run but instead use the screen from Cinebench website? :hmm:

 

[Ryun]

That's about as fast as AMD Phenom II X6 1100T (which scores around 5.9x). And it does look fishy, why not show the actual Cinebench run but instead use the screen from Cinebench website? :hmm:

But the X6 has 6 FP units right? So wouldn't that mean BDs FP units' performance increased greatly?

But I digress, we've been getting a lot of info lately so launch really can't be too far behind.

 

[AtenRa]

so launch really can't be too far behind.

12 October it seams

 

[Abwx]

Are these slides made up? Looking at the Cinebench R11.5 numbers........



Exactly same as here >> Cinebench score of 5.95 @ stock and 7.8 @ 4.8GHz with a FX8150 mentioned on XS forums. :hmm:

CB is still not optmised to take advantage of BD s FPUs,
once it s implemented , we can expect way better perfs
in this application...

 

[Idontcare]

CB is still not optmised to take advantage of BD s FPUs,
once it s implemented , we can expect way better perfs
in this application...

Wasn't CB a bit of a johnny-on-the-spot about rushing out an optimized version for Intel's i7 chips or some such when they debuted? (going by a vague recollection here)

Seems reasonable to expect them to pull together a similar 11th hour rev of CB on the eve of zambezi launch which will be optimized for BD as well.

 

[Abwx]

Wasn't CB a bit of a johnny-on-the-spot about rushing out an optimized version for Intel's i7 chips or some such when they debuted? (going by a vague recollection here)

Seems reasonable to expect them to pull together a similar 11th hour rev of CB on the eve of zambezi launch which will be optimized for BD as well.

AMD has made a good job with its FlexFpu wich can execute
64 Single Precision FLOPs/cycle whatever the code while
SB can reach this throughput only with AVX , hence with
little parts of a code , 10% at best...

According to this slide , BD should do well if FP heavy apps like CB are
optimised accordingly.

 

[inf64]

That slide is very vague. "128 bit FP" is what exactly? FMA4 optimized ? Or legacy SIMD? C11.5 leaked numbers show us that in legacy SIMD "8C" Bulldozer is not faster than X6 that works at lower clock.