Xbitlabs Xclusive preview on Athlon64.

[sirfergy]

I think a comparison to an HT P4 is perfectly valid since all new ones will have it.

 

[SexyK]

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: Sudheer Anne
duvie your opinion is very intel biased....... lets wait until we see some real final silicon reviews before placing any sort of judgement. The fact that it is doing so well at such a low clockspeed and early alpha silicon is pretty amazing in itself.

Come on, you have to be joking me? have you looked at the benchmarks? Obviously this isn't final silicon, we all know that, and we all understand that there are gains to be made. But this chip is getting absolutely BLOWN OUT in a majority of the benchmarks. I seriously doubt they will be able to make up all the ground on the P4 before release, but there's a chance I'll be wrong. As it stands now, calling this chip a 2800+ is a complete joke, and anyone that tries to justify the rating is pretty obviously biased. I hope as much as the next guy that they can release a "true" 3400+ on time (or... on time for this release date)... All anyone wants is a faster box, bottom line. But it's pretty insulting to have this being called a 2800+ the way it stands now

Its amazing how quickly people forget. Remember when the first scores showed up on the first Athlon. Remember this first review of the Athlon HERE. Well not to long after that the Athlon ruled. As a matter of fact Intel has just really regained control. The architecture of this cpu is radically different. It is technically superior to anything before it and anything that will released in the forseeable future which includes Intel. It might take a few bumps, but I do believe just like the first Athlon, given some time, this cpu will raise the bar.

Funny you say that since this processor basically is just like the first Athlon in most ways. It's rediculous to say that this is "technically superior" to everything that has been and will be when most of the core is a straight copy of the Athlon. Only time will tell, but based on the problems with SOI and the impending release of Prescott, IMHO Athlon64 has a long way to go to "raise the bar".

 

[dexvx]

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: Sudheer Anne
duvie your opinion is very intel biased....... lets wait until we see some real final silicon reviews before placing any sort of judgement. The fact that it is doing so well at such a low clockspeed and early alpha silicon is pretty amazing in itself.

Come on, you have to be joking me? have you looked at the benchmarks? Obviously this isn't final silicon, we all know that, and we all understand that there are gains to be made. But this chip is getting absolutely BLOWN OUT in a majority of the benchmarks. I seriously doubt they will be able to make up all the ground on the P4 before release, but there's a chance I'll be wrong. As it stands now, calling this chip a 2800+ is a complete joke, and anyone that tries to justify the rating is pretty obviously biased. I hope as much as the next guy that they can release a "true" 3400+ on time (or... on time for this release date)... All anyone wants is a faster box, bottom line. But it's pretty insulting to have this being called a 2800+ the way it stands now

Its amazing how quickly people forget. Remember when the first scores showed up on the first Athlon. Remember this first review of the Athlon HERE. Well not to long after that the Athlon ruled. As a matter of fact Intel has just really regained control. The architecture of this cpu is radically different. It is technically superior to anything before it and anything that will released in the forseeable future which includes Intel. It might take a few bumps, but I do believe just like the first Athlon, given some time, this cpu will raise the bar.

Classic Athlons were faster than Katmai's clock for clock. Coppermine was faster than the Classics, and Tbirds barely regained the crown.

 

[classy]

Originally posted by: SexyK

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: Sudheer Anne
duvie your opinion is very intel biased....... lets wait until we see some real final silicon reviews before placing any sort of judgement. The fact that it is doing so well at such a low clockspeed and early alpha silicon is pretty amazing in itself.

Come on, you have to be joking me? have you looked at the benchmarks? Obviously this isn't final silicon, we all know that, and we all understand that there are gains to be made. But this chip is getting absolutely BLOWN OUT in a majority of the benchmarks. I seriously doubt they will be able to make up all the ground on the P4 before release, but there's a chance I'll be wrong. As it stands now, calling this chip a 2800+ is a complete joke, and anyone that tries to justify the rating is pretty obviously biased. I hope as much as the next guy that they can release a "true" 3400+ on time (or... on time for this release date)... All anyone wants is a faster box, bottom line. But it's pretty insulting to have this being called a 2800+ the way it stands now

Its amazing how quickly people forget. Remember when the first scores showed up on the first Athlon. Remember this first review of the Athlon HERE. Well not to long after that the Athlon ruled. As a matter of fact Intel has just really regained control. The architecture of this cpu is radically different. It is technically superior to anything before it and anything that will released in the forseeable future which includes Intel. It might take a few bumps, but I do believe just like the first Athlon, given some time, this cpu will raise the bar.

Funny you say that since this processor basically is just like the first Athlon in most ways. It's rediculous to say that this is "technically superior" to everything that has been and will be when most of the core is a straight copy of the Athlon. Only time will tell, but based on the problems with SOI and the impending release of Prescott, IMHO Athlon64 has a long way to go to "raise the bar".

What planet are you living on? The Athlon 64 is a huge leap over anything that has been introduced for desktop use ever. Period. The only 64 bit capable processors have been for server use. Its the only processor that has its own integrated memory controller. Prescott will be faster but will not introduce anything new to desktop usage outside of being faster. If it was up to Intel we would still using Pentium Pros.

 

[classy]

Classic Athlons were faster than Katmai's clock for clock. Coppermine was faster than the Classics, and Tbirds barely regained the crown.

When the first Athlon scores were released the K7 was slower than Katmai. As a matter of fact it even trailed the K63 in business apps. Maybe you didn't read many of the first previews.

 

[PliotronX]

I capped the pages with benchmarks and upped em:

Page 6 (95KB)
Page 7 (68KB)
Page 8 (65KB)
Page 9 (51KB)
Page 10 (62KB)

 

[dexvx]

Originally posted by: classy

Classic Athlons were faster than Katmai's clock for clock. Coppermine was faster than the Classics, and Tbirds barely regained the crown.

When the first Athlon scores were released the K7 was slower than Katmai. As a matter of fact it even trailed the K63 in business apps. Maybe you didn't read many of the first previews.

As a matter of fact I read that first preview back when it was released. Let me re-interate my point. The Final Classic Athlons were faster than Katmai's, but were beat soundly by Coppermines. Tbirds barely matched Coppermines clock for clock.

 

[SexyK]

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: Sudheer Anne
duvie your opinion is very intel biased....... lets wait until we see some real final silicon reviews before placing any sort of judgement. The fact that it is doing so well at such a low clockspeed and early alpha silicon is pretty amazing in itself.

Come on, you have to be joking me? have you looked at the benchmarks? Obviously this isn't final silicon, we all know that, and we all understand that there are gains to be made. But this chip is getting absolutely BLOWN OUT in a majority of the benchmarks. I seriously doubt they will be able to make up all the ground on the P4 before release, but there's a chance I'll be wrong. As it stands now, calling this chip a 2800+ is a complete joke, and anyone that tries to justify the rating is pretty obviously biased. I hope as much as the next guy that they can release a "true" 3400+ on time (or... on time for this release date)... All anyone wants is a faster box, bottom line. But it's pretty insulting to have this being called a 2800+ the way it stands now

Its amazing how quickly people forget. Remember when the first scores showed up on the first Athlon. Remember this first review of the Athlon HERE. Well not to long after that the Athlon ruled. As a matter of fact Intel has just really regained control. The architecture of this cpu is radically different. It is technically superior to anything before it and anything that will released in the forseeable future which includes Intel. It might take a few bumps, but I do believe just like the first Athlon, given some time, this cpu will raise the bar.

Funny you say that since this processor basically is just like the first Athlon in most ways. It's rediculous to say that this is "technically superior" to everything that has been and will be when most of the core is a straight copy of the Athlon. Only time will tell, but based on the problems with SOI and the impending release of Prescott, IMHO Athlon64 has a long way to go to "raise the bar".

What planet are you living on? The Athlon 64 is a huge leap over anything that has been introduced for desktop use ever. Period. The only 64 bit capable processors have been for server use. Its the only processor that has its own integrated memory controller. Prescott will be faster but will not introduce anything new to desktop usage outside of being faster. If it was up to Intel we would still using Pentium Pros.

What planet are you living on? The usefullness of 64-bit computing on the desktop is seriously in question. The integrated memory comtroller reduces latency, but doesn't enhance bandwidth - aside from the fact that it forces a processor upgrade if you want to use new memory technology. Other than those changes, this chip is basically an exact copy of the Athlon core - no new technology at all - and it's apparently not capable of scaling to higher clock speeds despite SOI and an increased pipeline length. Prescott will bring Hyperthreading 2 and SSE 3 to the table, along with the capability to hit somewhere around 5GHz. If you ask me, Hyperthreading will be a much bigger boon to desktop users than 64-bit computing will be, at least for the next 10 years or so. In regard to "if it were up to Intel we would still be using Pentium Pros." Well... that's just a good way to expose your ignorance. You probably wouldn't have a PC at all if it weren't for Intel, so to insinuate that they don't push the technology more than anyone else is a joke.

 

[classy]

Originally posted by: PliotronX
I capped the pages with benchmarks and upped em:

Page 6 (95KB)
Page 7 (68KB)
Page 8 (65KB)
Page 9 (51KB)
Page 10 (62KB)

For a cpu in the infant stage, that thing is showing some serious potential. The 2800+ and the Intels 2.8 are near the fastest cpus you can buy. I lay money right now when the smoke clears the Athlon 64 will be dominate. Expensive as hell probably, but dominate.

 

[classy]

Originally posted by: SexyK

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: classy

Originally posted by: SexyK

Originally posted by: Sudheer Anne
duvie your opinion is very intel biased....... lets wait until we see some real final silicon reviews before placing any sort of judgement. The fact that it is doing so well at such a low clockspeed and early alpha silicon is pretty amazing in itself.

Come on, you have to be joking me? have you looked at the benchmarks? Obviously this isn't final silicon, we all know that, and we all understand that there are gains to be made. But this chip is getting absolutely BLOWN OUT in a majority of the benchmarks. I seriously doubt they will be able to make up all the ground on the P4 before release, but there's a chance I'll be wrong. As it stands now, calling this chip a 2800+ is a complete joke, and anyone that tries to justify the rating is pretty obviously biased. I hope as much as the next guy that they can release a "true" 3400+ on time (or... on time for this release date)... All anyone wants is a faster box, bottom line. But it's pretty insulting to have this being called a 2800+ the way it stands now

Its amazing how quickly people forget. Remember when the first scores showed up on the first Athlon. Remember this first review of the Athlon HERE. Well not to long after that the Athlon ruled. As a matter of fact Intel has just really regained control. The architecture of this cpu is radically different. It is technically superior to anything before it and anything that will released in the forseeable future which includes Intel. It might take a few bumps, but I do believe just like the first Athlon, given some time, this cpu will raise the bar.

Funny you say that since this processor basically is just like the first Athlon in most ways. It's rediculous to say that this is "technically superior" to everything that has been and will be when most of the core is a straight copy of the Athlon. Only time will tell, but based on the problems with SOI and the impending release of Prescott, IMHO Athlon64 has a long way to go to "raise the bar".

What planet are you living on? The Athlon 64 is a huge leap over anything that has been introduced for desktop use ever. Period. The only 64 bit capable processors have been for server use. Its the only processor that has its own integrated memory controller. Prescott will be faster but will not introduce anything new to desktop usage outside of being faster. If it was up to Intel we would still using Pentium Pros.

What planet are you living on? The usefullness of 64-bit computing on the desktop is seriously in question. The integrated memory comtroller reduces latency, but doesn't enhance bandwidth - aside from the fact that it forces a processor upgrade if you want to use new memory technology. Other than those changes, this chip is basically an exact copy of the Athlon core - no new technology at all - and it's apparently not capable of scaling to higher clock speeds despite SOI and an increased pipeline length. Prescott will bring Hyperthreading 2 and SSE 3 to the table, along with the capability to hit somewhere around 5GHz. If you ask me, Hyperthreading will be a much bigger boon to desktop users than 64-bit computing will be, at least for the next 10 years or so. In regard to "if it were up to Intel we would still be using Pentium Pros." Well... that's just a good way to expose your ignorance. You probably wouldn't have a PC at all if it weren't for Intel, so to insinuate that they don't push the technology more than anyone else is a joke.

First of all many of the top game programmers have already said that the next level in graphics is with......................drum roll please..............64 bit apps. And AMD is the reason, the only reason Intel is working on pushing better desktop performance. I like both, use both, but I do believe that AMD is about to really break new ground.

 

[mechBgon]

From Firing Squad:

Evil Avatar: Can you give me any idea what kind of a performance increase people might see running the 64-bit version?

Tim Sweeney: In pure CPU performance, Athlon64 is about 15% faster than previous Athlon's of identical clock rate for 32-bit apps, and 30% for 64-bit apps, because it exposes double the number of CPU registers, enabling the compiler to generate more efficient code. As for UT2003 performance, we haven't benchmarked yet but expect it to be faster though less than 30% because the GPU is as significant a factor as CPU.

So I would be interested in seeing what the 64-bit OS and 64-bit-specific apps bring to this equation. Anyone have an opinion on whether media encoding apps would benefit from the full 64-bit treatment?

 

[SexyK]

First of all many of the top game programmers have already said that the next level in graphics is with......................drum roll please..............64 bit apps. And AMD is the reason, the only reason Intel is working on pushing better desktop performance. I like both, use both, but I do believe that AMD is about to really break new ground.

Well, if you want to believe that AMD is the only reason INtel is working on desktop performance, go ahead. I don't know why you think Intel would stop advancing technology if it weren't for AMD - considering they would never make any money since noone would ever have to upgrade

. 64-bit may make a difference in 10 years, but right now there's a lot holding back gaming other than the 32-bit barrier. By the time gaming needs 64-bits, Intel will have 64-bit chips in the desktop market, its as simple as that. We can bring back this thread in a few months and see where things stand then, but I'm not sold. I don't want AMD to fold, i dont want the chip to stink, but so far they are 20% below their goal clock speed, one year late delivering, and between 5% - 40% behind Intel's third fastest chip, with more on the way. It's going to to a lot to make Athlon 64 succede, we'll see if they have what it takes.

 

[classy]

Originally posted by: mechBgon
From Firing Squad:

Evil Avatar: Can you give me any idea what kind of a performance increase people might see running the 64-bit version?

Tim Sweeney: In pure CPU performance, Athlon64 is about 15% faster than previous Athlon's of identical clock rate for 32-bit apps, and 30% for 64-bit apps, because it exposes double the number of CPU registers, enabling the compiler to generate more efficient code. As for UT2003 performance, we haven't benchmarked yet but expect it to be faster though less than 30% because the GPU is as significant a factor as CPU.

So I would be interested in seeing what the 64-bit OS and 64-bit-specific apps bring to this equation. Anyone have an opinion on whether media encoding apps would benefit from the full 64-bit treatment?

The only problem is until we get a optimized and final silicon of Athlon 64 everything is just speculation. No one really knows how 64 bit will go over cause we don't have a desktop 64 bit capable processor and platform. Intel, the company of innovation (?), hasn't made one yet . But AMD has one on tap

 

[everman]

Very encouraging results, taking into consideration that this is not the final silicon. Now that there is a 64bit desktop cpu, software designers will have more incentive to create apps that can take advantage of it (not to mention 64bit windows and gaming ). Just being what it is, is a great step forward.

 

[NFS4]

http://www.theinformationminister.com/press.php?ID=612220000

 

[Idoxash]

... I thought this cpu was going to do much better then this ... Next thing you know VIA c3 be out doing AMD. This is sad for me since I like AMD.

--Idoxash

 

[ElFenix]

Originally posted by: PliotronX
I capped the pages with benchmarks and upped em:

Page 6 (95KB)
Page 7 (68KB)
Page 8 (65KB)
Page 9 (51KB)
Page 10 (62KB)

from that page 6 its pretty clear to me that the athlon 64 really needs a dual channel memory controller. the one it has is pretty damn uber, but an uber one isn't as good as two good ones in this case. the fact that it attains anywhere near DDR's theoretical max of 3200 read and 1600 write is pretty amazing.

they also need some type of HT...

and of course more Hz would help.

 

[mechBgon]

Originally posted by: ElFenix

from that page 6 its pretty clear to me that the athlon 64 really needs a dual channel memory controller. the one it has is pretty damn uber, but an uber one isn't as good as two good ones in this case. the fact that it attains anywhere near DDR's theoretical max of 3200 read and 1600 write is pretty amazing.

they also need some type of HT...

and of course more Hz would help.

I guess that describes Opterons, which do have a DC memory controller (and can be set up for true SMP as well). XBit Labs was hinting at a third-party Opteron chipset being revealed on the 22nd, and from their hints I think they mean nVidia (or perhaps Serverworks). If it has an AGP slot... mmmmm, yummy! Not only that, but Opterons will be available sooner, rather than later. LOL, I'm such an optomist

 

[RaynorWolfcastle]

Wow, I hope the Sandra SSE2 benchmarks aren't correctly showing what the final silicon will be able to do because the P4 wrecks the Athlon64 in the floating point SSE2 ops (though the Athlon64 is within striking distance on the integer SSE2 ops). Otherwise the scores are good, but nothing like what AMD built us up to believe it would be like. Judging by AMD's earlier comments I thought the Athlon64 would thoroughly massacre the P4 in everything from the get go.

I'd be curious to see how many benchmarks are seriously effected by the use of a 1 MB L2 cache as Prescott will be evening that playing field very shortly after the Athlon64's launch. Also, judging by the Athlon64's performance in 3DS Studio and Lightwave AMD engineers still have to work heavily on that SSE2 unit to bring it up to par with Intel's. The Opteron's memory latency on the other hand is exceptional, too bad AMD is choking the chip with single channel DDR though.


All in all I hope that they can significantly tweak this CPU before it's released because otherwise Prescott will be very bad news for AMD. I'm a little disappointed in the benches as I expected AMD to clearly be ahead of the P4 in almost all tasks.


OT: Does anyone else see how this shows that AMD's model number scheme is fundamentally flawed? (despite the fact that they claim it is a Tbird comparison)

 

[mechBgon]

Recall that the Hammers can also use external memory controllers, so you might be able to get the bandwidth that way, too.

 

[imgod2u]

Originally posted by: RaynorWolfcastle
Wow, I hope the Sandra SSE2 benchmarks aren't correctly showing what the final silicon will be able to do because the P4 wrecks the Athlon64 in the floating point SSE2 ops (though the Athlon64 is within striking distance on the integer SSE2 ops). Otherwise the scores are good, but nothing like what AMD built us up to believe it would be like. Judging by AMD's earlier comments I thought the Athlon64 would thoroughly massacre the P4 in everything from the get go.

Well, the SSE2 performance is understandable. While Intel skimped quite a bit on x87 support and many integer-intensive tasks are very branch-heavy (hence, the hyperpipelined design would hold it back), SSE/SSE2 performance on the P4, even on a clock-normalized scale, was focused on quite a bit. The K8 has the same SSE/SSE2 execution resources as the P4 does and since the data-to-instruction ratio is very high in such operations, the P4's limited scheduling and issuing bottlenecks aren't a concern. It's mostly data-limited and the P4's caching and memory subsystem should be quite sufficient at supplying the data neccessary for SSE/SSE2 operations.

I'd be curious to see how many benchmarks are seriously effected by the use of a 1 MB L2 cache as Prescott will be evening that playing field very shortly after the Athlon64's launch. Also, judging by the Athlon64's performance in 3DS Studio and Lightwave AMD engineers still have to work heavily on that SSE2 unit to bring it up to par with Intel's. The Opteron's memory latency on the other hand is exceptional, too bad AMD is choking the chip with single channel DDR though.

The SSE2 unit, by all means, are on par with the P4's. The difference is, the P4 has a significantly higher clockrate. You can't really do much more than 1 SSE2 instruction per clock, the parallel execution resources neccessary would be tremendous and more than anything available of any x86 processor I've ever seen.

All in all I hope that they can significantly tweak this CPU before it's released because otherwise Prescott will be very bad news for AMD. I'm a little disappointed in the benches as I expected AMD to clearly be ahead of the P4 in almost all tasks.

Well, expectations lead to disappointment because your expectations are almost always going to be unreasonably high. I think the chip did pretty well for a 1.6 GHz part that's suppose to be scalable up to 2.4 GHz and beyond.

OT: Does anyone else see how this shows that AMD's model number scheme is fundamentally flawed? (despite the fact that they claim it is a Tbird comparison)

A lot of people did and still do to this day. There have been many a threads about it. But there will always be fanboys eating out of the hands of marketing of their favorite pet company.

 

[Electrode]

Seems interesting, but I'm dissappointed that there were no 64-bit tests done. I guess they've never heard of Linux (which has been x86-64 compatable for quite some time now)...

 

[PliotronX]

Originally posted by: mechBgon
Recall that the Hammers can also use external memory controllers, so you might be able to get the bandwidth that way, too.

My take on this is that it creates redundancy and ends up costing the consumers more, because the IMC has the advantages of reduced latency and lessened cost of motherboards due to absence of a MC, but when we enthusiasts go for a DC controller because AMD castrated the Opteron for consumers, we still pay for the IMC. AMD doesn't have to give us a DC IMC, but I'd rather they give us the option of cheaper A64's w/o the IMC than the IMC alone.

 

[Duvie]

Lets look at the benchmark results more closely....Lets also throw out all the MARKS (IE pcmark, sysmark, and 3dmark) since they tend to favor P4s and often are terrible system comparisons....



P4 2.8 (800fsb) w/ HT - vs - Atlon64 2800+

Business Winstone: Athlon64 (+4.1%)
Lame MP3 encoder: P4 (+22.0%) note: P4 2.53 w/o HT was faster....
Winrar Data Comp.: Athlon64 (+16.2%)
Mpeg4 (Divx 5.02): P4 (+24.2%)
WMP 9 encoder: P4 (+35.5%)

RTCW: P4 (+7.2%)
UT2003 dm-antalus: Athlon64 (+14.7%)
UT2003 flyby: Athlon64 (+7.2%)

Sciencemark (mol.): P4 (+30.7%)
Sciencemark (Pri.): P4 (+16.0%)
Sciencemark (Cip.): Athlon64 (+2.6%)
3DSMax 5: P4 (+54.4%)
Lightwave 7.5 (Ray)4 (+17.0%) note: P4 2.53 w/o HT was faster....
Lightwave 7.5 (Sun)4 (+30.3%)
Cinema 4D: P4 (+48.2%)


Hmmm.....Didn't do as well on the SSE2 programs as I would have thought....With someof these percentage leads I don't think any final silicon is going to make up these numbers. If you look at this a 2.0ghz Athlon64 is going to be needed to make this even against a 2.8 p4 w/ HT. What will it take to equal a 3.4ghz P4 prescott w/ 1mb L2 cache???


Now lets see if the 2800+ PR rating is deserved

Athlon XP 2800+ - vs - Atlon64 2800+

Business Winstone: XP (+8.5%)
Lame MP3 encoder: XP (+7.6%)
Winrar Data Comp.: Athlon64 (+27.9%)
Mpeg4 (Divx 5.02): XP (+4.8%)
WMP 9 encoder: XP (+12.7%)

RTCW: Athlon64 (+8.4%)
UT2003 dm-antalus: Athlon64 (+3.9%)
UT2003 flyby: Athlon64 (+6.6%)

Sciencemark (mol.): XP (+30.9%)
Sciencemark (Pri.): XP (+22.4%)
Sciencemark (Cip.): XP (+17.6%)
3DSMax 5: XP (+23.3%)
Lightwave 7.5 (Ray):Athlon64 (+17.0%)
Lightwave 7.5 (Sun):XP (+24.9%)
Cinema 4D: XP (+14.9%)


I think the answer to that is NO!!!! The leads by the XP were in the average of 15-20%. This chip with these test should be called no better then say between a 2500-2600+...

 

[Lord Evermore]

You forget that AMD bases its ratings on a specific set of tests which THEY think is a valid way of comparing them, and compares them based on performance "officially" compared to a Thunderbird core Athlon, and naturally the tests that the Athlon64 blows by a T-bird are going to probably be used for the final number more than others.

Unfortunately the article is either off the server now, or the server is overloaded.

AMD's rep is going to take a hammering if they end up with a processor that isn't as fast as the existing Barton at least in EVERYTHING and that it be marketed with the proper ratings. At a minimum it needs that, and they can say that the true power can be seen when in 64-bit mode. Obviously it's faster on a clock-for-clock basis by at least a small amount, and will be better when it's finalized, so they just need to not try to over-rate the PR numbers and they're golden even without 64-bit mode. If a 1.6GHz A64 beats an XP 2600+ at all things, market it as a 2600+ level CPU, don't be stupid and try to tell us it's a 2800+ when it's so easy to prove otherwise. The enthusiast world is still their biggest market share I think, and the enthusiast world is much less swayed by marketing numbers.

Hell, they don't even NEED to worry about clock-for-clock comparisons. Intel has already proven that having a processor that runs at a higher speed but performs less well than the last product isn't a hindrance, and AMD's got it even better because of the PR ratings. They're basing their selling prices and marketing on a performance rating number, not on the final clock speed. If it takes a 2GHz Athlon64 to get a valid, accurate PR rating of 2800+, so what? When you buy it, you're still getting 2800+ performance no matter what the clock speed is. That's the whole point of generation changes, the clock speed isn't comparable just like comparing 3DMark 2001 scores to 3DMark 2003 scores isn't valid. Sure it makes it easier for them to "ramp up PR numbers" by having a wide gap with a low clock speed and high PR, but it doesn't help their reputation at all to try and stretch the truth.

Would any of us really be upset if AMD's next chip ended up being comparable to a P4's performance on a direct clock-for-clock basis? If a 2.5GHz Athlon64 ran at the same performance level as a 2.5GHz P4? (Don't worry about the bus speeds in this case, just assume that AMD's highest bus speed processor is compared to the highest bus speed P4, since anytime one company increases it, the other can do the same.) I think that'd be pretty nice. I WISH for AMD to stop trying to get higher PR numbers out of lower clock speed processors. Eventually they're going to be having like a 4GHz core that they call an 8000+ level processor.

Of course one hopes that AMD is just going to astound us all at the actual launch. With a top-end 2GHz part that performs across the board at 3GHz performance ratings in 32-bit mode and even higher in 64-bit mode with 64-bit apps and OS.