Amd Officially Benches Bulldozer

[Apokalipse]

I disagree , logical is to compare threads vs threads.

But 1 core + HT is two threads... just like one BD module.

They both are designed to add only a small increase in die area for significantly higher multithreaded performance
Intel core has I think 5% more die area from HT for about 20% more performance
While a BD module is 12% larger than a single (full) core, and AMD claims 80% more performance.

[Pestilence]

...will Stores have them on Monday?

No. Im thinking newegg will have them thursday

[matari]

No point; with 2700k around the corner and price drop for the 2600k.

[qcmadness]

No point; with 2700k around the corner and price drop for the 2600k.

So we see a decline of MSRP of FX-8150P (from ~$300 to ~$270)

[duron]

No. Im thinking newegg will have them thursday

OK
i'll keep refreshing the Egg then

[Lanek]

No point; with 2700k around the corner and price drop for the 2600k.

If 2700K will be released ... you will not see a drop on the 2600K .. they will replace it...... Intel have launch 12 new CPU, who are for most the same cpu with a better clock for the same TDP.. look like they production is better and they can increase the clockspeed for the same TDP. We tend to forget this is not GPU, if Intel launch a series with a max TDP of 95W, they keep this TDP, if they can get 100mhz for the same TDP, so they release some new models ....

[accord99]

I think it's logical to compare one BD module (two cores) to one core + HT.
I think AMD have simply come up with a solution superior to hyperthreading.

So far, it looks like a BD module will have no throughput advantage when running two threads versus a comparably clocked SB core with HT. However, with one thread,SB will have a significant advantage. So it seems to me that the wide core + SMT strategy is superior to the AMD's dual narrow integer core CMT.

[Lanek]

we will need bench for see that ..... but don't forget, Bulldozer FX is a step and this the first step.. for see how this strategy is really good or not, personally i will wait Piledriver for have a real idea of what this strategy can bring on the table.

Then it is needed to see how this work on a server based system for start speak on how good or bad it is.....

If you base your comment on the Handbrake " results" ... it is really not enough... It is an opensource project, i have absolutely no idea on what instructions are used, called in it. And the result look to vary so much from a run to another, i don't know why reviewers use it as a test... if they use it, it's only cause many use it for encode mobile movies, it can be used as an information, but as a bench breaker, i really doubt of it's value.

[PerryR]

I am leaning towards Movieman knowing these new BD's are probably leading now. He seems pretty confident, and also this was just posted by xbit-labs spilling the beans. They obviously have the chips under testing and seem to confirm that AMD has the leadership with Bulldozer but for only a short while. So guess that is it. Bulldozer will be overall faster or at the minimum equal to what intel has out. Great news!!!

http://www.xbitlabs.com/articles/cpu...011-idf_5.html

Under the Final Remarks paragraph..

Anything to say about this Movieman?

P.S. they even mention that intel may push Sandy E up sooner... hmmmm

Still intrigued by this post. And I'm not too concerned about Sandy E, particularly if Bulldozer is priced to perform; or beats the current Core I5-I7 as that article seems to indicate.

[Apokalipse]

So far, it looks like a BD module will have no throughput advantage when running two threads versus a comparably clocked SB core with HT. However, with one thread,SB will have a significant advantage. So it seems to me that the wide core + SMT strategy is superior to the AMD's dual narrow integer core CMT.

Based on what benchmarks exactly?

[generics_user]

So far, it looks like a BD module will have no throughput advantage when running two threads versus a comparably clocked SB core with HT. However, with one thread,SB will have a significant advantage. So it seems to me that the wide core + SMT strategy is superior to the AMD's dual narrow integer core CMT.

in your crystal ball?

[Pestilence]

No point; with 2700k around the corner and price drop for the 2600k.

When the 2700K is released expect the 2600K to be discontinued

[accord99]

A lack of evidence hasn't stopped AMD supporters from claiming CMT is superior to Hyperthreading.

[Pestilence]

A lack of evidence hasn't stopped AMD supporters from claiming CMT is superior to Hyperthreading.

Thats because they're trying to be positive in every aspect.

Here's what i expect..

Benchmarks are released showing the 8150 beating up on the 2500K in multithreaded apps but losing in Gaming
Amd fans rejoice because they beat sandy bridge
Intel fans say you need 8 cores to beat 4
Amd fans say it doesn't matter we just want to beat intel


Mindless bickering ensues till SB-E arrives and then it starts all over again.

Intel fans say 3830K destroys Bulldozer in everything
Amd fans say you can't compare them because of the pricing
Intel fans say bow down and wash my car

To be Continued when IB gets released

[Apokalipse]

A lack of evidence hasn't stopped AMD supporters from claiming CMT is superior to Hyperthreading.

You don't think it's possible to make an educated guess based on the details of the architecture?

Also, AMD themselves claim 80% more multithreaded performance over a single thread.
And yes, we can't take that figure as gospel, but if it's even somewhat close to that it would be safe to say it's better than hyperthreading.

Thats because they're trying to be positive in every aspect.

Blanket statement much?

I wouldn't claim to know if BD can win against SB single threaded, in either IPC or frequency. I'd only guess it would be somewhere in the ballpark.

[Pestilence]

You don't think it's possible to make an educated guess based on the details of the architecture?
Also, AMD themselves claim 80% more multithreaded performance over a single thread.
And yes, we can't take that figure as gospel, but if it's even somewhat close to that it would be safe to say it's better than hyperthreading.

Didn't the same thing happen with Thuban? So much hype then mediocre results?

[Apokalipse]

Didn't the same thing happen with Thuban? So much hype then mediocre results?

What hype are you referring to?
Personally I didn't expect it to be anything more than Deneb with 2 more cores, with single threaded performance staying the same.

[Pestilence]

Blanket statement much?

I wouldn't claim to know if BD can win against SB single threaded, in either IPC or frequency. I'd only guess it would be somewhere in the ballpark.

More like an educated guess. I believe BD will be good and if bloomfield was still top dog would surpass it but SB is a whole nother animal.

[Movieman]

I am leaning towards Movieman knowing these new BD's are probably leading now. He seems pretty confident, and also this was just posted by xbit-labs spilling the beans. They obviously have the chips under testing and seem to confirm that AMD has the leadership with Bulldozer but for only a short while. So guess that is it. Bulldozer will be overall faster or at the minimum equal to what intel has out. Great news!!!

http://www.xbitlabs.com/articles/cpu...011-idf_5.html

Under the Final Remarks paragraph..

Anything to say about this Movieman?

P.S. they even mention that intel may push Sandy E up sooner... hmmmm

Hard for me to comment when I haven't even seen a SB-e chip yet let alone worked with one.

[Apokalipse]

More like an educated guess.

More like what's an educated guess? That AMD fans are trying to be positive in every aspect? Or do you mean something else?

I'm an AMD fan. Yet I'm trying to be realistic.

I believe BD will be good and if bloomfield was still top dog would surpass it but SB is a whole nother animal.

Yes, I'd say it's probably better than bloomfield. Against SB, that's still unknown.

[Movieman]

Just a thought and probably one that will get laughed at but how about for once
WE ALL TAKE A STEP BACK and look at these things for what they are.
We have two companies that compete for our and the rest of the worlds business.
They both bring out at times excellent products and once in a while real dogs.
There's a ton of rumors going around on the BD chips and I can honestly tell you
that many of them are SO wrong it's not even funny.
I was there in Austin and saw with my own eyes a lot of what these can do BUT
that is different from setting up your own machine and then KNOWING what it will do.
Same with Intel, they all want to show you their products in the best light possible and that they should be doing.
After all, this is business right guys?
I wish I could say more but when you give your word to someone to keep your big mouth closed till X date then you have to do so.
Bottom line is patience my friends, it won't be that long till we all have GOOD hard numbers in front of us and then we can decide with information instead of speculation.

[El Mano]

Just a thought and probably one that will get laughed at but how about for once
WE ALL TAKE A STEP BACK and look at these things for what they are.
We have two companies that compete for our and the rest of the worlds business.
They both bring out at times excellent products and once in a while real dogs.
There's a ton of rumors going around on the BD chips and I can honestly tell you
that many of them are SO wrong it's not even funny.
I was there in Austin and saw with my own eyes a lot of what these can do BUT
that is different from setting up your own machine and then KNOWING what it will do.
Same with Intel, they all want to show you their products in the best light possible and that they should be doing.
After all, this is business right guys?
I wish I could say more but when you give your word to someone to keep your big mouth closed till X date then you have to do so.
Bottom line is patience my friends, it won't be that long till we all have GOOD hard numbers in front of us and then we can decide with information instead of speculation.

/thread

[Loque]

according to my scientific research, to get the real benchmark numbers from manufacturers* you have to reduce 15%~ from their inflated results.

*this includes AMD, Intel & nvidia.

[Lanek]

Thats because they're trying to be positive in every aspect.

Here's what i expect..

Benchmarks are released showing the 8150 beating up on the 2500K in multithreaded apps but losing in Gaming
Amd fans rejoice because they beat sandy bridge
Intel fans say you need 8 cores to beat 4
Amd fans say it doesn't matter we just want to beat intel


Mindless bickering ensues till SB-E arrives and then it starts all over again.

Intel fans say 3830K destroys Bulldozer in everything
Amd fans say you can't compare them because of the pricing
Intel fans say bow down and wash my car

To be Continued when IB gets released

I have make 15 sites in the last 4 hours for find a semblant of leaked numbers ( French, UK, Australian, japan, and US sites ), and i can't still find them ......... so i will made like all, i will wait the reviews.

I agree with your words, it#s exactly how act the peoples. whatever is the side, Nvidia vs AMD; Intel vs AMD, MAC vs PC, Samsung vs Apple .... but right now, you can't apply it, as you as me, and many other here, just don't know anything on how BD will perform....


I need say, Nvidia and AMD thoses 2 last years, have surely make us write miles and miles of text, They just keep us in dark till the release ..... but we all speak aboujt them ..

[hurleybird]

A lack of evidence hasn't stopped AMD supporters from claiming CMT is superior to Hyperthreading.

CMT probably is better than hyperthreading, but that alone isn't going to make bulldozer faster than Sandy Bridge. Core 2 Duo didn't have an IMC and still took the performance crown from AMD after all.

I wouldn't be surprised to see CMT in haswell, or even a combination of CMT and HT.