Barcelona to have double digit lead in integer

http://www.eetimes.com/news/latest [...] =197700269 Is this enough? Is this enough for the enthusiast class?

I agree talk is cheap, but if these numbers are true, to what extent will Barcelona have to excel in integer, knowing that soi wont oc much better even at 65nm? Or will the posts of the future be : who needs oc when my Barcelona is faster already?

in bench we trust

But us enthusiasts love to oc...

It is a marketing piece. A double digit lead could be as little as 10%. There are no independent benchmarks of how well it runs applications.

And apparently, R600 is going to be more late.

john

Yeah, AMD's NUMA architecture works very well with SPECfp_rate. But as AMD has only given two rough numbers (this one plus a database related benchmark), there's not enough information to say other than the two architectures will be quite close to each other for the desktop. It'll probably end up being decided like it used to be, by CPU clock.

File this under "More Beast Sightings."

you were laughed at for calling the k10 a beast

looking more and more like u will laugh last

although the rabid INTELIOTS are eerily quiet...

File this under "More Beast Sightings."

you were laughed at for calling he k10 a beast

looking more and more like u will laugh last

I was basing it on real specs, not assumptions or hopes or fanboyism. I said I expect to see SPEC benches soon. They are getting closer with the R600 Demo having happened.

1 TFLOP with two Streams and probably two Barcelona's is like WOW!! Hell even 4 CPUs and two CPUs is like Wow. AMD has now won another race; TFLOPS IN A BOX.

The little engine that could is pulling a Beast. I am almost willing to bet that AMD will drop a box this year that will crack the top of the TPC-C. It's currently owned by 64P Itanium and Power.

Since we can see that 8xxx does have 4 HT links they will have a cache coherent 1-hop 32P box by Sept/Oct.

I'm not knowledgable/articulate enough to explain this but there is an article that I've posted and here is a snippet of it... explains SSE128 and I think it may delve a little bit into your question/comment

http://anandtech.com/cpuchipsets/s [...] i=2939&p=3

just to quote some from the article:


In general, the accuracy of a CPU's branch predictor determines how wide and how deep of a design you can make. The average number of instructions before the predictor mispredicts governs how many instructions you can have in flight, which in turn controls how many execution units you can realistically keep fed on a regular basis. The K8's branch predictor was quite good and very well optimized for its architecture, but there were some advancements Intel introduced in the Pentium M and Pentium 4 that AMD could stand to benefit from.

Barcelona adds a 512-entry indirect predictor which, believe it or not, predicts indirect branches. An indirect branch is one where the target of the branch is a location pointed to by an address in memory, in other words, a branch with multiple targets. Instead of branching directly to a label indicated by the branch instruction, an indirect branch sends the CPU to a memory location that contains the location of the instruction that it should branch to

And more:

Barcelona widens the execution units that handle SSE operations from 64-bits to 128-bits, so now 128-bit SSE operations don't have to be broken up into two 64-bit operations. This also means that you get more usable decode bandwidth since 128-bit SSE instructions now map to a single micro-op instead of two. The FP scheduler can now handle these 128-bit SSE operations as well.

It's the increase to SSE execution width that drove a number of other changes within the core. Since you effectively have more decode bandwidth when executing 128-bit SSE instructions AMD discovered a new bottleneck: instruction fetch bandwidth. These 128-bit SSE instructions tend to be quite large, and in order to maximize the number decoded in parallel the Barcelona core can now fetch 32-bytes per cycle, up from 16-bytes in K8. The 32B instruction fetch not only benefits SSE code but also seems to benefit integer code as well. Bigger instructions in general will see a performance boost here.

Now that you can fetch and decode more instructions, you need to be able to get more data to the execution core and thus AMD widened the interface between the L1 data cache and Barcelona's SSE registers. Barcelona can now perform two 128-bit SSE loads per cycle from the L1-D cache compared to two 64-bit loads per cycle in K8. AMD then widened the interface between the L2 cache and the memory controller so that now 128-bits can be transferred per cycle, once again to balance out all of the aforementioned changes

so do they have benchmarks out yet? not that I'm aware of..

it's one thing to know the answer, it's something else to know why the answer is...

It's not the destination that matters, it's the journey

Okay, so basically all these new K10 threads are responses to what is just currently unfounded marketing hype by AMD (regardless of whether or not it will be proven true in the future)? That's disappointing.

Yeah I didn't mean to suggest either you or the author was, I was just curious if there had been benchmarks or something because today there are seven new Barcelona threads on this page alone, which suggested there was something going on anyway.

Haha yeah I was reading earlier about how AMD was complaining about one of Intel's presentation of benchmarks favouring Intel CPUs over AMD. I thought it was pretty funny considering AMD is spouting its own superiority claims, but at least Intel used benchmarks, even if they may have been biased.

Seeing as I haven't yet had time to read all the mentioned 'crap' I can't be sure, but he probably meant the whatever % benefit over Intel CPUs. He didn't say 'untrue', just 'not proven'.

Do I find that statement disturbing? No not really, I mean... if you had a product wouldn't you say it was the best so people bought it? Do I believe that? No I don't, even if I did, you still missed the point of the post which has nothing to do with AMD saying they 40% faster than Intel...

I don't know why AMD makes unfounded statements... every business does, get over it.

Umm... if it's not faster it's not faster, how do you make AMD pay? I don't know, perhaps look at the benchmarks first? And as you said there are none out and I doubt you would be stupid enough to buy off someone's claims so that situation really has no merit here. If a person does buy off those claims without seeing benchmarks and such then that's their problem.

I remeber these statements about the early Conroe.
Yes, it's all the news about the k10 is 100% BS. But most of the conroe claims turned out to be true.

Ok beer, I'm done explaining it to you guys.

The point of posting the article... to show you people who were interested what's new in the Barcelona core, how AMD plans to improve and such...

I understand exactly what you're thinking. There are no grounds for these kind of rumors to be spread... that's cool and all and I agree with you.. but you're looking at it from a performance perspective...

This article I posted doesn't say "Barcelona will outperform by this much X%" If you actually read it, it doesn't even hint at that... it's just stating what improvements are in the core... That's it... nothing more... nothing less... take it as you will this thread is retarded because it is getting reduced down from a person wanting to share some knowledge on upcoming technologies to "X CPU sucks and the company lies! omg!"

Oh yes.. COMPLETEY different articles... I posted a bit in here...

http://anandtech.com/cpuchipsets/s [...] i=2939&p=1

No performance, no guessing, no assuming, just what Barcelona has in it... that's all

Just curious, what does R600 have to do with the price of tea in China?

wes

Quite a bit, actually. When the R600 is released and bought, many Chinese gamers will marathon their favourite games for days straight. Thus, they will require caffeine in large doses. The sudden demand for tea will create shortages and prices will skyrocket. The tea suppliers will increase production to compensate, but then the novelty of the R600 will wear off and the gamers will suddenly need to catch up on all their lost sleep. Excess tea will be flooding the marketplace and prices will hit record lows. A few days later the demand for tea will slowly return to normal, but very slowly, because after the large quantities of tea consumed, most people will likely be sick of it. That, coupled with the tremendous quantities of unsold tea in stores, and the useless extra capacity of tea suppliers, could be disastrous for not only the tea industry but the whole Chinese economy.

Yea that's what I was going to say... lol

Perhaps you can remember back to this time last year when Intel was saying 90% faster than NetBurst without producing benches. Later they said it would outdistance Opteron also before the benches (unless I'm mistaken).

It's the same thing. SPEC wil turn up sooner rather than later so I guess we'll all fnd out. I tend to believe the specs and so does every other person who has looked at them. I guess it's unfortunate that it takes up to a year to release a new arch after tape out.

ohh ic.. so according to you, only INTEL can say "my new processor will be #$&#$&#% faster than the competition!"?

Are you kidding?
There is nothing new and nothing worth in the article. Just another bold claims from AMD's marketing machine. Millions of words are less worth than a single benchmark.

It's almost as if your life will end if these numbers are true. AMD is smart enough to know not to blatantly lie. Not only will your partners abandon you but Wall St will more than likely give you a beating.

Why can't you believe this? People say I'm biased but when the Core 2 specs were released I believed their numbers. As a matter of fact I mentioned that they may have an RHT implementation.

Actually I would say unsubstantiated news from either company is frowned upon. I guess we have all concluded at this point that the marketing machine of both companies tend to exaggerate to the positive side on this type of stuff.

Actually I would say unsubstantiated news from either company is frowned upon. I guess we have all concluded at this point that the marketing machine of both companies tend to exaggerate to the positive side on this type of stuff.

They may do that with concepts but never numbers. So they may say MCM or native is the better concept without "proof" but benchmark numbers have to be about exact.

I think thats the problem with the "enthusiast." You're (enthusiasts) like the guy in that movie "Fever Pitch" about the Boston Red Sox and their fans.