AMD expects Barcelona to be 40% faster than Clovertown!

[Fred_Pohl]

Tsk,tsk,tsk

Those are the final production revisions,NOT ES!
What do you think,that Sun will get quad core ES Barcelona in April and make an uber expensive supercomputer with a bunch of ES chips.You gotta be kidding...

April-June,final shipments of the so called RevB (final production rev.).
ES are probably wandering around in the wild as we speak.But we all know AMD,and this time the NDA curtain reached it's highest peak.Not surprising at all ,if we consider the fact that this chip is the basis for the next 2-3 years of AMD product lineup.

So then you either believe that AMD intends to skip the ES process with Barcelona and go straight into volume production based solely on internal testing or that everyone is scared silly of violating the NDA of little AMD but couldn't care less about violating Intel's NDAs?

Does this popular "AMD=TOP SECRET" fanboy myth really make sense to you? AMD hyped K8 for 3 years and the first ES BMs appeared 16 months before the launch yet you believe that K8L will go from vaporware rumors and task manager demos to hard product launch in ~3 months, Good luck with that.

[nn_step]

So then you either believe that AMD intends to skip the ES process with Barcelona and go straight into volume production based solely on internal testing or that everyone is scared silly of violating the NDA of little AMD but couldn't care less about violating Intel's NDAs?

Does this popular "AMD=TOP SECRET" fanboy myth really make sense to you? AMD hyped K8 for 3 years and the first ES BMs appeared 16 months before the launch yet you believe that K8L will go from vaporware rumors and task manager demos to hard product launch in ~3 months, Good luck with that.

actually AMD's current NDA includes ritual suicide and no I am not joking.

[informal]

So then you either believe that AMD intends to skip the ES process with Barcelona and go straight into volume production based solely on internal testing or that everyone is scared silly of violating the NDA of little AMD but couldn't care less about violating Intel's NDAs?

Does this popular "AMD=TOP SECRET" fanboy myth really make sense to you? AMD hyped K8 for 3 years and the first ES BMs appeared 16 months before the launch yet you believe that K8L will go from vaporware rumors and task manager demos to hard product launch in ~3 months, Good luck with that.

First of all,AMD is not a "lil" company anymore
They have become a large one and it seems that intel fans can't get use to it(and to the fact that AMD now owns one of the largest players in graphics-chip making)

Second,ES are in the wild and people (Charlie D. among them) have the "hard" perf. numbers already(at least in the "Cinebench" league).Charlie confirmed this a few days ago at Ace's.Why the ES don't have a "go" to be early pre-tested by an "enthusiast(s)" is the way AMD wants it to be.
They are floating around,the fact AMD is not "forced" to publish early ES results as they were in the days they had ZERO server and notebook marketshare and declining desktop one,is a no brainer actually(well it maybe is to you...).

[Fred_Pohl]

Thanks, Fred_Pohl, I'm glad to see somebody isolated the CPU to verify whether they cause the trouble or not.

I am very sad to see that AMD was willing to sacrifice even the biggest advantage they had over Netburst for this silly 4x4 platform

You're most welcome, uOpt. I'm glad to see that not everyone goes into instant denial mode when they don't like the facts.

IMO 4x4 will be much more appealing with 65nm CPU's but it's appeal will still be limited to a small niche in the desktop market.

[red]

They are floating around,the fact AMD is not "forced" to publish early ES results as they were in the days they had ZERO server and notebook marketshare and declining desktop one,is a no brainer actually(well it maybe is to you...).

So Intel published hard numbers, while AMD is saying 40% with no benches.

[[LCN]Knowledge]

More happy in my pants speculation I hope it's true though, not just because I like AMD but because I like to see competition and the leadership in the processor performance charts switching hands (just means both companies work harder to develop new technology).
It's AMD's turn for the spotlight again.

QFT.

I too am waiting for New AMD's CPU to be at least just as good as core 2 Duo (i hope they'll be better). I wanna see a constant prices war...

Yeah u know i wanna build a rig with a $100-150 Quad Core CPU

The same goes for ATI/AMD... Nvidia prices are too high.

[nn_step]

So K8L is year and a half away!

no but interesting theory

[Fred_Pohl]

First of all,AMD is not a "lil" company anymore
They have become a large one and it seems that intel fans can't get use to it(and to the fact that AMD now owns one of the largest players in graphics-chip making)

Second,ES are in the wild and people (Charlie D. among them) have the "hard" perf. numbers already(at least in the "Cinebench" league).Charlie confirmed this a few days ago at Ace's.Why the ES don't have a "go" to be early pre-tested by an "enthusiast(s)" is the way AMD wants it to be.
They are floating around,the fact AMD is not "forced" to publish early ES results as they were in the days they had ZERO server and notebook marketshare and declining desktop one,is a no brainer actually(well it maybe is to you...).

How could I forget? Since acquiring ATI (on credit), AMD is now so large that they are almost 20% of Intel's size. With their clearly superior products and net income (loss) of -$168M in 2006 to Intel's paltry $5B (profit) it is virtually assured that AMD will continue to grow larger while Intel continues to shrink.

I also see your point about AMD not needing any positive press from K8L benchmarks. Considering how competitive their current lineup is and how well their stock is doing for their investors, AMD can well afford to keep K8L's revolutionary performance top secret as long as they wish to. So while Intel is hurriedly demonstrating 45nm Penryns running all manner of real world apps and letting the press actually use them, AMD can afford to sit back and ride the glory wave of their December K8L task manager demo right up to the May hard launch. After all, we all know that AMD could have demonstrated K8L running any app 40% faster than C2 back in December but they simply chose not to. Likewise, final silicon K8L ES chips have been floating about since December but no one would dare to risk the wrath of such a huge corporation by breaking their NDA. Clearly AMD doesn't need to prove to anyone how great K8L is because we already know it is, based on press releases and rumors. I'm sure this all makes perfect sense to you.

[Fred_Pohl]

So K8L is year and a half away!

Congratulations on so completely missing the point!

[Canibuz]

You guys really act such as if ES are candy and are as common as the morning newspaper and that the big companies are like Oh well just send these to anybody or somebody who will post performance. I hate the whole oh if no es's have been benched yet then they are trying to hide bad performance. NDA's are there for a reason maybe maybe not they are sending out ESs knowing that they will be leaked benchs BUT KEY WORD BUT signing an NDA means you know that you will be sued if you are proved to be the source of information leaking. With those consequences not EVERYONE would be willing to take the chance. So If i don't see any Rev H (which it should be called as AMD is calling it) till the day or day after I wouldn't be surprised because thats how its SUPPOSED to go, Im not going to say they are not going to come out sooner Im just saying stop crying because we haven't seen them yet. And ES's are chips that are being TESTED they can be a performance drop or increase on the final product incase no body remembered this.

[duploxxx]

its appearant that loe doesn't understand much engineering.

after a0 silicon comes back, you will have chances to de/activate some of the dfm redundancies to save your silicon and do some minor adjustments in the stepping.

another tape out would be called revB and so on.

april-june shipments will be a joke. more like sampling. i dont expect k8l to enter volume production for desktops before september. amd doesnt even have enough 65nm capacity yet to support both server and desktops.

exiting 2007, amd should still sell more 90nm am2 chips than 65nm am2 and 65nm k8l combined.

wrong:

The only major problem with the conversion that really sticks out is that the capacity will be somewhere between 15,000wspm and 20,000wspm. In the current scale of 300mm fabs this is rather small and pushes the breakeven point higher and the available operating margins lower. However, it is more than a stop-gap and the extra capacity, should demand continue to grow will provide sufficient capacity for AMD through 2008

[Poodle]

I'm glad that there is so much debate about this. What I make of all this is that both companies are running at Warp 8 to get the perfomance crown. Both are pushing their technologies as far as they can go. So I wouldn't be suprised if it turns out to be a hard to claim an overall winner when you take all data into account.

After looking at all the "data" at hand atm, I think that Penryn will most likely be the chose of the overclocker like C2D is 2day. However the K8L seems to have some nice feats when it comes to power saving and independant IMC/core clocking. So servers builders, silent seekers and low end builders might never leave the AMD plattform. 4X4 is a bad idea today but in three months or so when Agena and the cooler 690 chipset are both on the market, it will probably be the best multi core/gpu platform you can get. Eight cores and all that mem bw + CF should be good when we see more multi core tweaked games on the market.

[red]

http://www.overclockers.com/tips01099/

"We expect across a wide variety of workloads for Barcelona to outperform Clovertown by 40 percent."

Note the careful and very specific language used here. Not "in general." Not "most of the time," or "for most apps," or "on average." They win by 40% in a "wide variety of workloads." That's not even a "wide variety of apps," for all we know, it could be just one app!

Heh.

[red]

That was the exact point of the article, that AMD did not clearly state 40% overall better than Clovertown, which a lot of articles are suggesting. That is all. I hope you are wrong too, but you most likely won't be

[situman]

At this point nothing is certain. Everyone can speak what they think, but it doesn't mean it is true. BTW that site happens to like intel a lot more than amd, you can see it in almost each of their articles.
I personally think 40% overall is too much as well, I hope I am wrong thou

They are not as biased as you think. They do give "pats on the back" to AMD when its warranted. You just feel that way...well cause AMD isn't really upfront about things most of the time. He gave props to AMD for finally admitting that they need to get bigger in order to survive, even though their admittance was forced upon them by an analyst asking a direct question. This article regarding Barcelona being 40% faster without anything specific is just AMD blowing smoke again, so he tells it like it is. No I am not trolling.

[oldblue]

AMD did not clearly state 40% overall better than Clovertown, which a lot of articles are suggesting.

The title of this thread certainly suggests it. That's why I agree with uOpt that the thread title should be changed.

Perhaps something like "AMD: Barcelona to outperform Clovertown by 40% in wide variety of workloads."

[mascaras]

AMD Quad Core Barcelona Re-Appears






AMD showed off their native quad core Barcelona processor again and this time round during Windows Vista launch in Japan. The previous appearance is back in November last year where AMD Chief Sales & Marketing Executive Henri Richard did a comparison between Barcelona and Clovertown.







http://www.vr-zone.com/index.php?i=4553

cheers

[7he ]-[0rr0r]

Barcelona cant be this good.

Cities names = mainstream (normal cpus)

Star names = especial CPUs (high end)


expect stunning new from me in minutes

Give us the news I'm hoping coming from you we have something interesting the time that has past now, while it can still be measured in minutes, would be many minutes.
maybe you can end the flame wars.

[Enoc]

ES are in the wild and people (Charlie D. among them) have the "hard" perf. numbers already(at least in the "Cinebench" league).Charlie confirmed this a few days ago at Ace's.

what's the rendering score?

[alayashu]

what's the rendering score?

probably somewhere near 70-90% faster clock per clock than K8 and Conroe.

[informal]

And WHERE did that number come from? Any hard (or even SOFT) evidence to back that up?

He hasn't posted any numbers since he said he doesn't want to piss some ppl off .But when he gets "a go" ,he will,I am sure.

[Enoc]

He hasn't posted any numbers since he said he doesn't want to piss some ppl off .But when he gets "a go" ,he will,I am sure.

nice... , i hope is the other camp he does not want to piss off...

[informal]

Just a tiny clarification: the guy in question(the one i meant) is Charlie Demerjian(from inquirer).
He said at Aces's he has the "hard" numbers on Barcelona,but won't post them since it could cause some problems to his sources.