AMD expects Barcelona to be 40% faster than Clovertown!

[Fred_Pohl]

so where's the es benchmarks???

am2 benchmarks are out 3 months ahead of release
c2d benchmarks are out 5-6 months before release

so where's the k8l benchmarks? nda is a bad excuse. show the money shots!!

I think we all need to be more patient. The first K8 ES benchmarks appeared in June 2002 and K8 didn't launch until 16 months later in September 2003.

As for all the performance speculation, I am constantly amazed by how many people will believe anything when it's what they want to hear.

[Anonymous]

I think we all need to be more patient. The first K8 ES benchmarks appeared in June 2002 and K8 didn't launch until 16 months later in September 2003.

As for all the performance speculation, I am constantly amazed by how many people will believe anything when it's what they want to hear.

To some, computing is religion. You don't need facts, just faith

[[cTx]Philosophy]

To some, computing is religion. You don't need facts, just faith

I agree with this guy fully..

I hope AMD does what it has to do VS Intel, not because I want them back out front, but price to performance will drop consideraby for us all..

[turbox997]

Whether this will be TRUE or NOT...I hope INTEL is reading this and getting really scared/paranoid to get them to work harder for their next release. Complacency and arrogance can lead to a weak and dissappointing product.

[Fred_Pohl]

Whether this will be TRUE or NOT...I hope INTEL is reading this and getting really scared/paranoid to get them to work harder for their next release. Complacency and arrogance can lead to a weak and dissappointing product.

Amen to that. Just look at AM2, Brisbane and 4x4.

[speed bump]

so where's the es benchmarks???

am2 benchmarks are out 3 months ahead of release
c2d benchmarks are out 5-6 months before release

so where's the k8l benchmarks? nda is a bad excuse. show the money shots!!

We didn't see anything other than maybe 1 screenshot until about a week before AM2 came out. Its typical AMD you don't see anything until it comes out.

[safan80]

Anyone making any assumption about whether or not AMD is blowing smoke without seeing some actual results is a fanboy in denial.

they haven't publicly released a demo of anything with the purchase of ATI and with the delays of the R600 AMD needs to show they have new products not just die shrinks. How do they expect to keep investors when they haven't released a single new (R600 or any quad core processors) product? The RD600 was long in development before the buy out.

[Fr3ak]

true, but, C2D is being compared to a 3 year old AMD tech, quite funny that so many where shocked by the fact that C2D was faster then Athlon64. My point being months seem like an acceptable time span for comparison would'nt you agree?

Offcourse AMD's Stars CPU's will be faster then Intel's Core CPU's it is only natural progression of technology. Now 40% does seem like much but lets say thats the best case scenario against C2D and that actual average is 20% that means that Stars will be faster then Athlon64 chips by 40% (Because C2D's are 20% faster then Athlon64). Today you can actually find a benchmark where Opteron actually beats Xeon by a little bit, meaning if Xeon matches the Opteron performance in such test then the Barcelona based Opteron will actually beat Clovertown based Xeon CPU by 40%.

What is wrong comparing a C2D to a 3 year old architecture. K8 is that old, though AM2 just hit the market in may 06.

Furthermore I posted in belief that K8L wont be out before the end of Q3, which is false.

Funny how people react after I posted that.
I am not a Intel fanboy at all. I have owned 4 times as many AMD cpus as Intel cpus. I just tend to buy the best bang for the buck and I never believe in pre release benchmarks or announcements (as there are not even benchmarks in this case) of the company which is making that product.
I cant remember what kinda benchmarks Intel showed before C2D was released, because I didnt really care. I prefer to base my reasons to buy hardware on proven facts and not mystical proclamation.

[Syn.]

What is wrong comparing a C2D to a 3 year old architecture. K8 is that old, though AM2 just hit the market in may 06.

Furthermore I posted in belief that K8L wont be out before the end of Q3, which is false.

Funny how people react after I posted that.
I am not a Intel fanboy at all. I have owned 4 times as many AMD cpus as Intel cpus. I just tend to buy the best bang for the buck and I never believe in pre release benchmarks or announcements (as there are not even benchmarks in this case) of the company which is making that product.
I cant remember what kinda benchmarks Intel showed before C2D was released, because I didnt really care. I prefer to base my reasons to buy hardware on proven facts and not mystical proclamation.

you was commenting on how people are willing to compare C2D with Stars so i pointed out to you that the comparison is being already held about 2 products that are separeated years apart so you should not be suprised by the fact that products separeatd by months are compared.

AM2 is still K8, C2D is not NetBurst

[gOJDO]

The best of AMD will compete against the best of Intel: K8L vs Core2(be it 65nm or 45nm).

[rozzyroz]

personally, i think its ok to compare anything to anything. who cares if its not out yet, who cares if something is old tech. it may be a pissing match between intel and amd, but its not a pissing match between us. if one of us post some prereleased benchmarks of k8l, then its ok to compare those to any benchmark out there. its just a rough estimate to figure out how something will perform. if intel decided to throw up some benchmarks on the web of there next chip, then thats fair game to be compared to anything thats out there. if amd put some up, then that too is ok to compare to anything out there. who cares if this is going against that and this is old tech vs new tech? if we have the benchmarks, then lets see where that puppy stands!

[TSFroggy]

It's true that comparing Core2Duo to three-year old technology (A64) is a little unfair. AMD has the chance now to best Core2... but then what?

They are allocating their resources to creating K8L right now, slating a release in Q3/Q4. But what about Intel?

Intel is already working on a 45nm answer to Core2Duo - Nehelem - and they are already pumping out 45nm chips. What it comes down to is resources.

Intel has more money, more fabs, more R&D, and their die-shrink to 45nm has already been successful. AMD's K8L will be based off of 65nm technology still. What this means is, while AMD is working on going from 65nm to 45nm, Intel will (probably) almost be ready for the next die shrink (from 45nm to what? 22nm?)

I want AMD to keep Intel in a sweat. Competition is always good for the consumer. If AMD falls, Intel will be what's left, and then what improves? Intel would become lax.

Companies don't follow progression, they follow money.

I want some K8L numbers, not biased speculation.

[Ubermann]

I find EVERY post in here unreliable =)

[arisythila]

Brent..... I think everybody here has it wrong.... See Intel PWN's AMD. Especially when you overclock them.

wOOt for Intel. Common everybody say it with me.. "wOOt for Intel"

~Mike

[SlicerSV]

brent, it's been plenty of time now for AMD to have made a new architecture that'll completely blow Intel's away. AMD started announcing it's plans for K8L quite a while back, before C2D came out, as I recall.

the FACTS as they stand right now are that Intel costs a little bit more, but is a TON faster. the facts as they stood when it was K8 vs NetBurst were that Intel costed a LOT more, but performed a LOT worse. the facts when K8L is released? we can't know that yet, but i'm choosing to be optimistic and hope that AMD at least marginally beats Intel. granted, if it's only marginal that means a new architecture will be due a lot sooner rather than having a couple years. but we really don't mind that either!

[informal]

This one made my day
Peace

Yeah,it was me who forced that one out ^^ .I'm positively surprised by the event and i welcome it.

[onewingedangel]

Barcelona may be 40% faster than cloverton, but the 45nm generation will be clocked 40% higher, not to mention the extra cache etc. so any advantage will be gone within a short period of time.

[The Ghost]

Barcelona may be 40% faster than cloverton, but the 45nm generation will be clocked 40% higher, not to mention the extra cache etc. so any advantage will be gone within a short period of time.

well we know none of this for sure , but time will tell

[nn_step]

Barcelona may be 40% faster than cloverton, but the 45nm generation will be clocked 40% higher, not to mention the extra cache etc. so any advantage will be gone within a short period of time.

umm how many times does it have to be said 40% more speed isn't 40% more performance.

[onewingedangel]

It's not a truly linear increase, but performance does scale almost linearly with clocks assuming there aren't any IO bottlenecks. The extra cache (50% extra) should mean that the FSB is no more of a bottleneck than at present, so any bottlenecks would be those of the execution core itself, but compared to conroe they will still feature the same percentage of branch misses etc, they'll just get through the cycles quicker.

[josty2]

Barcelona may be 40% faster than cloverton, but the 45nm generation will be clocked 40% higher, not to mention the extra cache etc. so any advantage will be gone within a short period of time.

They will NEVER be able to scale that thing so hihg right away with 45nm, which is probably because of the power comsumption. The QX6700 is at 125W TDP and uses the full wattage of it, it just sucks if you look at the power it draws (so does AMD's FX's, but they're old) if they make 45nm parts with a voltage of 1.2 or 1.15 and clock it 40% higher, that would make for another 15-20% increase in TDP. I highly doubt they would be so stupid to realease a 150W TDP product... AMD though is stating that their Q-core will have a 95W TDP which is quite nice... let's just hope they keep each other going!!!

[Growly]

Whenever I see POS, I think "Point of Sale", which in this case doesn't make sense. For shame.

[onewingedangel]

They will NEVER be able to scale that thing so hihg right away with 45nm, which is probably because of the power comsumption. The QX6700 is at 125W TDP and uses the full wattage of it, it just sucks if you look at the power it draws (so does AMD's FX's, but they're old) if they make 45nm parts with a voltage of 1.2 or 1.15 and clock it 40% higher, that would make for another 15-20% increase in TDP. I highly doubt they would be so stupid to realease a 150W TDP product... AMD though is stating that their Q-core will have a 95W TDP which is quite nice... let's just hope they keep each other going!!!

a 3.66ghz part (333x11) should be doable within current TDP's, possibly even a 3.8ghz part(intel will be able to use half multi's with their 45nm chips, to help remove the huge difference in clockspeed throughout the range whilst maintaining the same ammount of sku's). With selective binning it would probably be possible on the 65nm process. A 4ghz part is not outside the realms of possibility for the 45nm process. It all depends on how tightly intel want to bin (look at amd's current 3ghz parts to see just how much you can push the binning process if need be).

Not to mention the potential overclocking (with the associated increase in TDP).

AMD saying that they will be 40% better than cloverton (in some applications) more than likely means they will remain semi-competitive, but certainly doesn't mean k8l will be the better product, in fact I'd argue the opposite. BUt until we have these chips in our hands and have some conclusive benches this is all idle speculation.

[uOpt]

Whenever I see POS, I think "Point of Sale", which in this case doesn't make sense. For shame.

Does journey consistency constraints ring a bell, too?

[onewingedangel]

I'm not saying that 45nm will clock 40% better than 65nm - there are still huge reserves in the 65nm process, but intel aren't tapping into them. Two of the main reasons are firstly not to fragment their product range by having core2's that range greatly far too greatly in power (which the half multi's will help - allows core clock differences of 166mhz when using a 1333fsb, so the market can be positioned more like merom with smaller jumps in performance when moving up the sku's) and secondly there is no need to strictly bin the products when they already handily outperform amd when they don't raise the clocks (which k8l will help with). Intel have announced they plan to clock the 45nm core2 in the 3.6-4ghz range, so they clearly have the intention of moving core clocks higher across the range.

The 45nm chips could be clocked 40% higher than the 65nm chips, in part because of the process, and in part because of greater competition between the two big cpu companies encouraging stricter performance binning.