Thursday, January 13th 2011

Bulldozer 50% Faster than Core i7 and Phenom II

Here, take some salt. AMD reportedly gave out performance figures in a presentation to its partners, performance figures seen by DonanimHaber. It is reported that an 8-core processor based on the "Bulldozer" high-performance CPU architecture is pitched by its makers to have 50% higher performance than existing processors such as the Core i7 950 (4 cores, 8 threads), and Phenom II X6 1100T (6 cores). Very little is known about the processor, including at what clock speed the processor was running at, much less what other components were driving the test machine.

Taking this information into account, the said Bulldozer based processor should synthetically even outperform Core i7 980X six-core, Intel's fastest desktop processor in the market. Built from ground-up, the Bulldozer architecture focuses on greater inter-core communication and reconfigured ALU/FPU to achieve higher instructions per clock cycle (IPC) compared to the previous generation K10.5, on which its current Phenom II series processors are based. The processor is backed by new 9-series core logic, and a new AM3+ socket. AMD is expected to unveil this platform a little later this year.Source: DonanimHaber
Add your own comment

424 Comments on Bulldozer 50% Faster than Core i7 and Phenom II

#1
TheMailMan78
Big Member
by: JF-AMD
Can you please give us a complete description of why a bulldozer core is not a "true" core? Please be thorough so that I only have to answer this once.
Get em' JF!
Posted on Reply
#3
pantherx12
Whilst this is all well and good, lets all assume it will be rubbish so regardless of what happens at launch we're all pleasantly surprised!
Posted on Reply
#4
DigitalUK
that presentation was interesting
http://www.youtube.com/watch?v=VIs1CxuUrpc

im probably wrong as bulldozer is abit hard to get my head round, but would that make from a windows point of view, windows would see it as 2 phyical cpus installed with 4 cores per cpu?
Posted on Reply
#5
Mussels
Moderprator
JF-AMD: i think i speak for all of us (except the trolls) when i say:

I F'ing love having you around here to clean up the FUD and bullcrap.
Posted on Reply
#6
MxPhenom 216
Corsair Fanboy
Oh man! If bulldozer is going to be that good and if some motherboards will allow SLI then i may just go Bulldozer instead of Ivy Bridge
Posted on Reply
#7
Tensa Zangetsu
The desktop parts will probably come out same time as Intel's Socket 2011 based processors and judging from Sandy Bridges' performance, I don't see them regaining the performance crown for AMD, unless they get these a few weeks/months to market before Intel.
Posted on Reply
#8
Imsochobo
by: Mussels
JF-AMD: i think i speak for all of us (except the trolls) when i say:

I F'ing love having you around here to clean up the FUD and bullcrap.
I 2nd that!
I can report that atleast I love it as marketing for the special piece of market :)
Posted on Reply
#9
1Kurgan1
The Knife in your Back
It's always nice to have a rep here to say whats really going on, or as much as he can.
Posted on Reply
#10
TheMailMan78
Big Member
JF-AMD thank you man for staying around. Really. Thank you.
Posted on Reply
#11
Over_Lord
News Editor
Considering AMD is quite firm about launching a 4 module 8 core processor bulldozer first, we can expect this claim to be based of that.


Now 1 BULLDOZER module is 218 million transistors. 4 Modules will be 872 million.

Compare the die sizes, if we assue similarity case with 32nm manufacturing process, the 955 million transistors on Sandy Bridge 4 core has a die space of 216mm^2

source - http://techreport.com/articles.x/20188

Put the calculations on Bulldozer 4 module 8 core on pen and paper and you arrive at roughly 189mm^2....


Now if 189mm^2 die from AMD manages to beat the bigger Sandy Bridge(although we shouldn't neglect the gpu taking a part up, but thats intel's decision to stick it into sandy in the 1st place, so we cant do nothing about it) 216mm^2 by 50%, then I must say it's only a WIN-WIN situation for AMD here.
Posted on Reply
#12
Imsochobo
by: thunderising
Considering AMD is quite firm about launching a 4 module 8 core processor bulldozer first, we can expect this claim to be based of that.


Now 1 BULLDOZER module is 218 million transistors. 4 Modules will be 872 million.

Compare the die sizes, if we assue similarity case with 32nm manufacturing process, the 955 million transistors on Sandy Bridge 4 core has a die space of 216mm^2

source - http://techreport.com/articles.x/20188

Put the calculations on Bulldozer 4 module 8 core on pen and paper and you arrive at roughly 189mm^2....
Now if 189mm^2 die from AMD manages to beat the bigger Sandy Bridge(although we shouldn't neglect the gpu taking a part up, but thats intel's decision to stick it into sandy in the 1st place, so we cant do nothing about it) 216mm^2 by 50%, then I must say it's only a WIN-WIN situation for AMD here.
Die comparison is not applicable here.
Far from it.
Addin chipset DIE size, intel have interigated for PCI-E and interlink between cpu and NB SB, how this is done when its interigated I dunno, but atleast its in the cpu.
Graphics...

Graphics is a feature that comes to amd bulldozer cpu's sometime in H2 2012 (Correct me if i'm wrong..) meaning they do not have 128-190 million transistors or whatever SB have just for its graphics.
depending on if its 2 or 3 series.. not very accurate numbers i come up with here, but its the point..
Bulldozer is most likely bigger if we look at your numbers and take a very rough estimate, but direct comparison isnt possible due to same components isnt included.
Posted on Reply
#14
alucasa
I am not entirely sure about this.

I do believe AMD's claim of 50% more speed, but I also do believe that it is from a selective benchmark.

I've gotten old enough not to fall for every hypes companies make.
Posted on Reply
#15
TheMailMan78
Big Member
by: alucasa
I am not entirely sure about this.

I do believe AMD's claim of 50% more speed, but I also do believe that it is from a selective benchmark.

I've gotten old enough not to fall for every hypes companies make.
Of course. But a 50% jump is nice no matter what bench. However I am willing to bet over all it will be around 20%. Just a guess......I'm waiting for Cadave to drop some knowledge of how wrong I am :laugh:
Posted on Reply
#16
CDdude55
Crazy 4 TPU!!!
by: Mussels
JF-AMD: i think i speak for all of us (except the trolls) when i say:

I F'ing love having you around here to clean up the FUD and bullcrap.
Agreed 100%.

I also doubt it will be that much faster, but hey, im always skeptical and if they can push that much past and i7's and Phenom II's then we've got ourselves an excellent platform coming :). Then again, if Bulldozer performs that well over current chips, then LGA 2011 Sandy Bridge CPU's must be monstrous, as Intel usually piggybacks with something that pushes past AMD performance wise. But really, this could be the chip that brings AMD back into the Athlon 64 days, when they were a force to be reckon with in the performance realm. Very exciting to see some good competition overall, let the price wars begin!! lol
Posted on Reply
#17
HTC
Personally, i won't like it if this 50% over I7 turns out to be true. My reasoning is simple: dominance = overpricing!

What i want is for AMD to put out a chip that is within 2% of I7 and Sandy's performance (lower or higher) but with the same or lower power consumption. If Bulldozer is that chip, then great for all consumers.

If Bulldozer turns out a bit better then this or if power consumption is a bit higher, it's OK, but not ideal.

If Bulldozer turns out to be quite a bit better then I7 or Sandy's, then we have a problem, in the sense that it will be AMD's turn to overprice and i would really hate that!
Posted on Reply
#19
cadaveca
My name is Dave
by: JF-AMD
Can you please give us a complete description of why a bulldozer core is not a "true" core? Please be thorough so that I only have to answer this once.
Can you blame people for being skeptical?

Last time was Phenom1 launch, where leslie was telling poeple that there would be 3ghz Phenoms. That didn't happen. TLB bug happened.

I'm just gonna shoot straight from the hip here....

Back then, we didn't have guys like you trying to quell the rumours. So I'm going to make a very specific request, that will work well for everyone:

Create a list of websites that post rumours pertaining to AMD that are not based in fact, and make it public. They want to throw mud at AMD, throw it back at them.

You guys need to be far more agressive. Your single voice here is lost in the echoes, so you need a choir to back you up.



I know I'm off topic here, but this news is fake, so I see no point in staying on topic.


I really want to be AMD's next CEO. this may seem a bit far-fetched to some, but I think my vision of what AMD should be is the only right one. But who knows, maybe I'm just crazy.

You guys need to increase spending on both R&D and marketing.


I don't mean to brag or anything like like that, but I've been pretty accurate when it comes to what AMD is doing for some time now....as you can see by Mailman's posts. It's taken some itme, but I've even won HIM over.

The time for "playing safe" is over. Bulldozer IS ready. And it's pretty good. But just good performance, at this point, isn't enough. Agressive pricing, while beneficial, doesn't add enough value, either.

I know that at this point, it really is impossible for you guys to pull up the release. Qualification testing needs to finish, and then the launch is as good as gold.


But if many of the points that I mentioned in past are not addressed, bulldozer will fall flat on it's face.


I WANT AMD to succeed. So does basically everyone else here, from what I can tell. And I know...it's definately more than possible...but that depends on how you measure success.

Don't let others set that metric...69XX cards are a failure to many, because expectations were set too high. We all know who set those expectations, and it defiantely wasn't AMD.
Posted on Reply
#20
Aleksander
Amd is expected to bring even more than 30% better performance
Check out anandtech!
Posted on Reply
#21
de.das.dude
Pro Indian Modder
wow. :D
intels gonna have a tough time beating this!'
even if they get that 22nm cpu thingy, they will never "feel" as fast as AMD.

i dont wanna sound like a fanboy, but intel HAD a few good CPUs
but now, i never used an intel CPU that satisfies me. however even i3/i5 users have said my computers are fast!
Posted on Reply
#22
Isenstaedt
I'm skeptical but at the same time I want this to be true so badly!
Posted on Reply
#23
Makaveli
While I want AMD to succeed. I'm going to have to call BS on this until I see some numbers.



:toast:
Posted on Reply
#24
Widjaja
As long as these processors beat Intels next release I'll be all smiles.

It would be like the little geek KO'ing the big douche bag.
Posted on Reply
#25
wolf
Performance Enthusiast
an 8 core 50% faster than a 3.06ghz 4c 8t chip.... this reeks of selective multithreaded benchmarking, but hey you have to promote the chips strengths somehow.

the i7 2600K is already par or faster than the i7 975 (3.33ghz) while consuming 40-45% less power, and intel have yet to release 6-8 and 12 core CPU's.

I can't help but think this round will go to intel again, but I sincerely hope AMD pull a rabbit out of their hat on this one, or at least price to compete like they have been doing, I love my 1090T.

come on clock for clock improvements over PII.
Posted on Reply
Add your own comment