AMD to Disclose Details About Bulldozer Micro-Architecture in August

[Mats]

No chip exists in a vacuum, every chip design is inevitably going to be compared to existing designs.

The point is that there's NOTHING to compare. We have zero benchmarks, not even Google-translated hieroglyphs.
Leave the comparing to when we actually have any info about it.

Until then, we're stuck with these kind of pointless speculations.

[OhNoes!]

The point is that there's NOTHING to compare. We have zero benchmarks, not even Google-translated hieroglyphs.
Leave the comparing to when we actually have any info about it.

Until then, we're stuck with these kind of pointless speculations.

Read posts 6, 7, and 9. Thank you.

[Hornet331]

What are you talking about? The fastest P4 released was 3.4 ghz (from wiki?), while core i7's under turbo (e.g. single core to single core) can approach that speed. You seem to have handily ignored things like transistor shrinkage as well as better design ...

guess you never heard of the 570J or the 670..

[Manicdan]

Read posts 6, 7, and 9. Thank you.

what point are you trying to prove? not with this exact post, but overall in this thread?

[OhNoes!]

what point are you trying to prove? not with this exact post, but overall in this thread?

I don't have to "prove a point," I can join the discussion, however. The direction of this thread was established with the second post, posts 6, 7, and 9 made it stick.

What point are you trying to prove in this thread, given the thread title?

[Mats]

Read posts 6, 7, and 9. Thank you.

Oh thank me.
I have already read them and nothing in them is new or actually tells me something about performance.

#6: "We have said many times that single threaded performance will be higher than current systems." That's old, and shows that we won't get any more specific info right now.

#7: Speculations

#9: Talks about the current X6, not Zambezi.

I'm not into this "you're a towel!" logic.

[Manicdan]

I don't have to "prove a point," I can join the discussion, however. The direction of this thread was established with the second post, posts 6, 7, and 9 made it stick.

What point are you trying to prove in this thread, given the thread title?

im trying to prove that there is no discussion about the original topic, and it seems your just looking for someone to trample with your "AMD is inferior" logic.

[OhNoes!]

im trying to prove that there is no discussion about the original topic, and it seems your just looking for someone to trample with your "AMD is inferior" logic.

You're the only one throwing words around. At least we had a "tech" discussion going before you started pointing fingers.

I have asked you at least twice what you'd rather discuss seeing the OP is about a date of disclosure. So you'd rather have us discuss possible dates? Or focus on what is an inherent weakness in AMD's previous micro-architecture, and which almost everyone agrees AMD has to improve on in order to catch up to Intel? It seems to me you're in this thread to point fingers rather than contribute to what was a relevant discussion to the impending battle between AMD and Intel's next gen.

[Sn0wm@n]

Again, people could you just relax for a second.Like Helmore said,very little is known about how Bulldozer module works ,let alone the IPC relative to Deneb or Bloomfield.Wait for August for more details and then we can discuss based on facts.

but jf-amd iluded that bulldozer would be faster clock per clock... can we take it as a given that bulldozer will have higher ipc????

[Manicdan]

You're the only one throwing words around. At least we had a "tech" discussion going before you started pointing fingers.

I have asked you at least twice what you'd rather discuss seeing the OP is about a date of disclosure. So you'd rather have us discuss possible dates? Or focus on what is an inherent weakness in AMD's previous micro-architecture, and which almost everyone agrees AMD has to improve on in order to catch up to Intel? It seems to me you're in this thread to point fingers rather than contribute to what was a relevant discussion to the impending battle between AMD and Intel's next gen.

keep going with your smear campaign, until you learn how a discussion works your only going to continue to blindly troll your way into your so called "discussions". you said yourself you dont have a point, which means your only responding to comments with the intent to start an argument. did you actually bring any useful information into this thread?

alot of people on here are very curious about how amd is going to unique handle multithreading, and might try and speculate on that. so please explain how the intel IPC of a cpu 4 years old, vs current phenoms, has anything to do with that? im here trying to see what ideas people have around CMT, but thanks to a few handful of people who constantly start brand wars, that quickly turned into the rest of the people trying to defend themselves.

so how exactly does one get the thread back on topic. since im here to mainly read, i shouldnt be posting much. but i think i have the right to ask people to stop derailing, dont ya think?

[informal]

but jf-amd iluded that bulldozer would be faster clock per clock... can we take it as a given that bulldozer will have higher ipc????

Yes we can. But until more details trickle out we should just stop this whole "IPC" story.

[Mats]

Or focus on what is an inherent weakness in AMD's previous micro-architecture, and which almost everyone agrees AMD has to improve on in order to catch up to Intel?

So what is there left to discuss?
You can't go offtopic just because the topic is boring. Get your own thread.

[-Sweeper_]

but jf-amd iluded that bulldozer would be faster clock per clock... can we take it as a given that bulldozer will have higher ipc????

higher single threaded performance can also be achieved with higher clocks (or a more agressive turbo mode), so, I wouldnt take it as a given

[Helmore]

but jf-amd iluded that bulldozer would be faster clock per clock... can we take it as a given that bulldozer will have higher ipc????

AMD has never said that Bulldozer will be faster clock for clock, they have only indicated that single threaded performance will be higher. I'm only repeating what has already been mentioned by informal and Sweeper though.

[cegras]

guess you never heard of the 570J or the 670..

But i7s can turbo to 3.6 ghz (870). Surely 200 mhz isn't what separates the pentium from the core i7.

[informal]

When BD design was in its infancy,back in 2005-2007,scheduled to be released @ 45nm with SSE5(instead of AVX) ,AMD targeted it to be the highest performing x86 compute core in both single and multi thread scenarios. Then they decided to improve the design and release it @ 32nm with additional performance and scalability options(all due to competitive reasons and 45nm node ramp combined with good Shanghai design results).So ,if it was to be a highest performing compute core in 2008,they certainly worked on it to improve on that baseline with the BD ver2 we will be seeing next year. My guess is both higher IPC per core and per Mhz AND better power gating meaning very aggressive turbo mode on the core level(inside the module).There could be asynchronous clocking too and many other things that can all beef up the performance of Bulldozer when compared to what we have today(both intel and AMD).

[Sn0wm@n]

Yes we can. But until more details trickle out we should just stop this whole "IPC" story.

agreed!!!!!!

AMD has never said that Bulldozer will be faster clock for clock, they have only indicated that single threaded performance will be higher. I'm only repeating what has already been mentioned by informal and Sweeper though.

better single thread performance is a good indication of higher ipc dont you think.... and dont tell me that they could have taken a 2ghz cpu against a 3ghz cpu to make those claims... we should all drop that subject and wait till september for the real numbers....

[Helmore]

agreed!!!!!!

better single thread performance is a good indication of higher ipc dont you think.... and dont tell me that they could have taken a 2ghz cpu against a 3ghz cpu to make those claims... we should all drop that subject and wait till september for the real numbers....

I don't expect any performance numbers in 2010 though. I don't see much reason to completely drop the subject, all we're doing is some harmless speculation and there's little wrong with that as long as we're civil.

[JF-AMD]

but jf-amd iluded that bulldozer would be faster clock per clock... can we take it as a given that bulldozer will have higher ipc????

All I have said is that bulldozer will be faster than current products. I have not made any clock speed statements.

The statement that I made was that Interlagos would have 33% more cores and will be 50%+ faster than Magny Cours. If you are more than 50% faster with 33% more cores, then your "per core" performance is faster. That is the only statement that we will make on performance.

[MrMojoZ]

You're the only one throwing words around. At least we had a "tech" discussion going before you started pointing fingers.

Tech discussion = jumping into an AMD thread to talk about how much better Intel is, real or imagined. Gotcha.

[OhNoes!]

Tech discussion = jumping into an AMD thread to talk about how much better Intel is, real or imagined. Gotcha.

Sorry if that bothers you, but my post was actually pointing out the obvious. Another member posted links of benchmarks. I wonder where you've been the last 2 years? Hopefully not avoiding the obvious?

[-Sweeper_]

looks like they wont talk about clock speeds

150+% of the performance with 133% the number of cores

thats 12,7+% more performance per core, something easily achievable with higher clocks (magny cours works at a modest 2.3ghz)

[Mad Pistol]

I think AMD is on the right track. They've posted Quarterly profits I think for the first time in years this past quarter (correct me if I'm wrong on that. they have made a profit though) so that shows that they're doing something right. Obviously, it's chump change compared to Intel's earnings, but they are doing fine. AMD/ATI's R&D budget is much smaller than Intel's or Nvidia's, so that means they're being smart about what they're doing. The Radeon cards this generation have been fantastic. Who's to say that Bulldozer won't be the same way?

[Sn0wm@n]

im with mad pistol ... amd has changed in the past couple of years ... pleave give em credit that bulldozer is a beast .....

[radaja]

to ALL WHO ARE bickering here,remember it takes two to tango and two sides to ruin a thread.