AMD 65nm and 45nm

[imported_Questar]

Originally posted by: Viditor

Originally posted by: myocardia
I don't see how moving to the 65nm process will be that much of an improvement for the Athlons, even if they make a much smoother transition than Intel did, which is unlikely, IMO. I believe that AMD thought that the C2D was going to along the lines of Intel's last two "launches", and not be what it's turned out to be, and so they just sat on their laurels, and raked in the profits.

Of course, I'm hoping that I'm wrong (how often do you hear anyone say that?), and AMD will at least catch C2D in performance, because the only time we, as consumers, benefit, is when both companies have similar performing products. And that applies to anything, not just processors.

Some questions...
1. Don't you think that delaying volume production for 6 months has allowed AMD to improve their 65nm process over Intel's?
2. As Intel was able to drastically reduce their power usage and increase clockspeed at the same time when they wen to 65nm, do you think AMD won't be able to do the same?
3. AMD's 90nm launch was FAR smoother than Intel's (after delaying for the same length of time)...is there a reason that you think this time will be different?

1. No. You are assuming that Intel will have made no improvements in the past year, which is false.

2. No. The power savings were due to architecture changes, not the shrink. I posted three months ago that AMD was having problems at 65nm, which has been proven true with current reports of of speed bins at 65nm.

3. A one year delay is smooth?

 

[Pabster]

Originally posted by: Regs
What scares me most is that with all these incremental improvements what hope can we get from the K8L? The same?

Nah, I think K8L will be a big improvement. The question is when? And if the reports of poor yields at AMD on 65nm are to be believed (I tend to believe them), it certainly raises some questions. And makes me (still) think any talk of AMD at 45nm is fanboys and pipe dreams from a realistic point of view.

Kentsfield is coming, and it is a beast. AMD's only real hope is K8L.

I'll now await the mass of AMD fans to stomp me down. Just remember, I predicted them being late to the game with 65nm, and I was right

 

[jiffylube1024]

Originally posted by: Viditor
Some questions...
1. Don't you think that delaying volume production for 6 months has allowed AMD to improve their 65nm process over Intel's?

Over Intel's? No way - Intel has had C2D in laptops for about 6 months themselves. But it does bode well for AMD having a smooth transition to 65nm.

2. As Intel was able to drastically reduce their power usage and increase clockspeed at the same time when they wen to 65nm, do you think AMD won't be able to do the same?

I think AMD should get a healthy drop in power usage. Clockspeed; I don't think AMD is going to get a monumental gain like Intel has. Probably 3 GHz and change for the first while.

3. AMD's 90nm launch was FAR smoother than Intel's (after delaying for the same length of time)...is there a reason that you think this time will be different?

I think AMD's 65nm transition should be smooth. I'm concerned about how the 45nm switch is going to take place...

Originally posted by: Soulkeeper
Intel is making a lot more money also by having 65nm out, i think this is one of the biggest factors as it is very difficult for AMD to compete on price/performance while using 90nm

Part of the reason Intel is making a lot more money is that they have a superior performing chip; part of the reason they have a superior performing chip is because they can pack an astounding 4MB of cache into the C2D and still use similar die space due to the much smaller process. So yeah, this will help AMD tremendously in competing with Intel by dropping die sizes so much.

--------------

Is it true that there will be no AMD 65nm for Socket 939? Because if so, that's a huge buzzkill. They would make a killing since X2/Opteron became the de-facto performance leader for a healthy period of time, and they did that on Socket 939.

I'd like to hear something official from AMD on this issue...

 

[oztrailrider]

I am just glad to see that there is strong competition between Intel and AMD. It is a win-win situation for us, as long as they both remain competitive and manage to keep themselves afloat in the marketplace anyway.

 

[Viditor]

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

I think you misunderstood my post...if you read back through the posts, I wasn't commenting so much on performance, only that a 65nm shrink (no matter WHO makes it) will reduce both power requirements and size while allowing for more clockspeed headroom.
Either that or you are assuming that it's the architectual changes that allow C2D to overclock (as opposed to the manufacturing process). If that's the case, then can I ask you what it is about the architecture you feel is responsible for this?

By the way, that 800Mhz you mentioned represents ~40% gain in clockspeed. A 40% gain in AMD's current peak clockspeed would get them up to near 4 GHz...and then they are scheduled to be shipping at 45nm in Q1-Q2 2008.

 

[Viditor]

Originally posted by: Questar

Originally posted by: Viditor

Originally posted by: myocardia
I don't see how moving to the 65nm process will be that much of an improvement for the Athlons, even if they make a much smoother transition than Intel did, which is unlikely, IMO. I believe that AMD thought that the C2D was going to along the lines of Intel's last two "launches", and not be what it's turned out to be, and so they just sat on their laurels, and raked in the profits.

Of course, I'm hoping that I'm wrong (how often do you hear anyone say that?), and AMD will at least catch C2D in performance, because the only time we, as consumers, benefit, is when both companies have similar performing products. And that applies to anything, not just processors.

Some questions...
1. Don't you think that delaying volume production for 6 months has allowed AMD to improve their 65nm process over Intel's?
2. As Intel was able to drastically reduce their power usage and increase clockspeed at the same time when they wen to 65nm, do you think AMD won't be able to do the same?
3. AMD's 90nm launch was FAR smoother than Intel's (after delaying for the same length of time)...is there a reason that you think this time will be different?

1. No. You are assuming that Intel will have made no improvements in the past year, which is false.

Nothing like that at all...I'm assuming that AMD will have a better yield ratio (production dice/candidate dice) than Intel did in January (that's when they launched 65nm...)

2. No. The power savings were due to architecture changes, not the shrink. I posted three months ago that AMD was having problems at 65nm, which has been proven true with current reports of of speed bins at 65nm.

Ummm...what architecture changes did this. Why did the Smithfield (90nm) run so much hotter than Pressler (65nm with ~same architecture)?
As to your post on the 65nm bins, it was shown to be BS by several sources...(it was originally from a piece by Charlie Demerjian at The Inq, he has since retracted the statement...)
Charlie's retraction
The problem was an early bios...

3. A one year delay is smooth?

Some real fuzzy math there...Intel shipped in January, and AMD shipped in August (I always thought that was 8 months...).
Also, Intel had such a shortage in product that Dell pulled all of their Prescot systems until Intel could get some quantity shipped in April/May...so it's more like a 4 month delay.
If you go back over the old CCs from the 2 companies, AMD hit their 50% 90nm shipping target (meaning that half of their shipped chips were 90nm) about 2-3 months before Intel did...

 

[Viditor]

Originally posted by: Pabster

Originally posted by: Regs
What scares me most is that with all these incremental improvements what hope can we get from the K8L? The same?

Nah, I think K8L will be a big improvement. The question is when? And if the reports of poor yields at AMD on 65nm are to be believed (I tend to believe them), it certainly raises some questions. And makes me (still) think any talk of AMD at 45nm is fanboys and pipe dreams from a realistic point of view.

Kentsfield is coming, and it is a beast. AMD's only real hope is K8L.

I'll now await the mass of AMD fans to stomp me down. Just remember, I predicted them being late to the game with 65nm, and I was right

Ummm...what reports of poor yields? I have read exactly the opposite!

"The company has made a number of improvements to yields from its silicon. One reason being that the company supplying its masks now has a factory down the road in Germany. The wafers are still sent to Singapore to be cut and packaged, however.

Another good reason for increased yields is its adoption of Front Opening Shipping Boxes (FOSBs). These keep the wafers almost hermetically sealed for the vast majority of the production process ? only exposing them to its cleanroom standard of 100 particles per cubic metre of air on very rare occasions"

As to AMD being "late to the game on 65nm", they are shipping earlier than expected, so I'm not sure what you are referring to...

 

[Nyati13]

Originally posted by: jiffylube1024

Over Intel's? No way - Intel has had C2D in laptops for about 6 months themselves. But it does bode well for AMD having a smooth transition to 65nm.

...

I know you're refering to Intel having 65nm in production at that time, but it was Core Duo, not Core 2 Duo in those laptops. Merom isn't shipping yet, so C2D in a laptop 6 months ago is impossible.

 

[Henny]

Originally posted by: Viditor

Originally posted by: Pabster

Originally posted by: Regs
What scares me most is that with all these incremental improvements what hope can we get from the K8L? The same?

Nah, I think K8L will be a big improvement. The question is when? And if the reports of poor yields at AMD on 65nm are to be believed (I tend to believe them), it certainly raises some questions. And makes me (still) think any talk of AMD at 45nm is fanboys and pipe dreams from a realistic point of view.

Kentsfield is coming, and it is a beast. AMD's only real hope is K8L.

I'll now await the mass of AMD fans to stomp me down. Just remember, I predicted them being late to the game with 65nm, and I was right

Ummm...what reports of poor yields? I have read exactly the opposite!

"The company has made a number of improvements to yields from its silicon. One reason being that the company supplying its masks now has a factory down the road in Germany. The wafers are still sent to Singapore to be cut and packaged, however.

Another good reason for increased yields is its adoption of Front Opening Shipping Boxes (FOSBs). These keep the wafers almost hermetically sealed for the vast majority of the production process ? only exposing them to its cleanroom standard of 100 particles per cubic metre of air on very rare occasions"

As to AMD being "late to the game on 65nm", they are shipping earlier than expected, so I'm not sure what you are referring to...

Improvements to yield??? There is no yield since they aren't in production!! Yield improvements come with a prodution learning curve. It's foolish to think they can do much until they go to production.

Yes they are late to the game. The game is being played with Intel.

 

[Polish3d]

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

No way, it's more than that. It's more like +30% average at a minmum. At the LOW end its usually 20% faster per clock. I'm sick of people lowballing this kind of thing. Hell, in some games its ~50% faster per clock, and in SuperPi it's 60% faster

 

[Viditor]

Originally posted by: Henny

Originally posted by: Viditor

Originally posted by: Pabster

Originally posted by: Regs
What scares me most is that with all these incremental improvements what hope can we get from the K8L? The same?

Nah, I think K8L will be a big improvement. The question is when? And if the reports of poor yields at AMD on 65nm are to be believed (I tend to believe them), it certainly raises some questions. And makes me (still) think any talk of AMD at 45nm is fanboys and pipe dreams from a realistic point of view.

Kentsfield is coming, and it is a beast. AMD's only real hope is K8L.

I'll now await the mass of AMD fans to stomp me down. Just remember, I predicted them being late to the game with 65nm, and I was right

Ummm...what reports of poor yields? I have read exactly the opposite!

"The company has made a number of improvements to yields from its silicon. One reason being that the company supplying its masks now has a factory down the road in Germany. The wafers are still sent to Singapore to be cut and packaged, however.

Another good reason for increased yields is its adoption of Front Opening Shipping Boxes (FOSBs). These keep the wafers almost hermetically sealed for the vast majority of the production process ? only exposing them to its cleanroom standard of 100 particles per cubic metre of air on very rare occasions"

As to AMD being "late to the game on 65nm", they are shipping earlier than expected, so I'm not sure what you are referring to...

Improvements to yield??? There is no yield since they aren't in production!! Yield improvements come with a prodution learning curve. It's foolish to think they can do much until they go to production.

Yes they are late to the game. The game is being played with Intel.

Of course they've been in production since October of 2005...just not VOLUME production.
Also, AMD have a hardware/software system that Intel does not called APM 3.0...
It analyzes and adjusts the manufacturing process 24/7, all the way down to doping levels on individual dice. AMD have stated for months that the 65nm launch is to be at mature yields...
I'm not sure what you mean by "late to the game"...they're producing a little ahead of when they wanted to, and they continue to gain market share. That seems to be the only game worth winning in my book.

 

[Viditor]

Originally posted by: Frackal

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

No way, it's more than that. It's more like +30% average at a minmum. At the LOW end its usually 20% faster per clock. I'm sick of people lowballing this kind of thing. Hell, in some games its ~50% faster per clock, and in SuperPi it's 60% faster

Frackal, I think you're confusing performance with clockspeed. We aren't discussing performance in this...

 

[customcoms]

Some of you aren't even paying attention to what is going on with AM2 b/c of C2D; people are hitting 200-400mhz MORE with an AM2 processor equivlant to its S939 counterpart and a good motherboard (like the DFI Infinity and Lanparty which have just been released). Now, anything is possible with the switch to 65nm; the switch from 130nm to 90nm went far more smoothly for AMD than Intels switch to 65nm, but the latter is a harder transisition. I seriously doubt a huge performance increase from this switch (maybe on the order of 5% like the winchester and venice), but maybe higher clockspeeds will be attainable (like I said, AM2 is already oc'ing better). The extras silicon in revG is probably for 4x4 systems, and some tweaks like the venice SSE3 instruction set addition.

However, big picture, this isn't going to be able to regain the performance crown. It will probably keep AMD highly competitive on the low end for the time being (until we get more and cheaper C2D's and their mobo's). At the moment, the biggest cost appears to be ram and psu prices, at least in my book (ram is getting expensive, and some psu's prices doubled in the last month.

Just for reference, I was planning on building a C2D system as a "budget" rig for my family but that might change with cheaper/better boards and cheaper processors available on AM2.

 

[Polish3d]

Originally posted by: Viditor

Originally posted by: Frackal

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

No way, it's more than that. It's more like +30% average at a minmum. At the LOW end its usually 20% faster per clock. I'm sick of people lowballing this kind of thing. Hell, in some games its ~50% faster per clock, and in SuperPi it's 60% faster

Frackal, I think you're confusing performance with clockspeed. We aren't discussing performance in this...

I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

 

[88NovaTwincam]

How does the 32 vs. 64 bit OS manisfestation come into play?

I won't consider vistadoze until the wonderous MS SP2 arrives..

 

[Bobthelost]

Originally posted by: 88NovaTwincam
How does the 32 vs. 64 bit OS manisfestation come into play?

I don't think it does really, unless i'm being dense and missing the reference.

 

[Viditor]

Originally posted by: Bobthelost

Originally posted by: 88NovaTwincam
How does the 32 vs. 64 bit OS manisfestation come into play?

I don't think it does really, unless i'm being dense and missing the reference.

My understanding is that AMD still performs better than Intel on 64bit with 4+ GB of Ram, but not enough to overcome C2Ds basic core advantage...(~5-10% better, but that is just a rumour at this point).

 

[harpoon84]

Originally posted by: Frackal

Originally posted by: Viditor

Originally posted by: Frackal

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

No way, it's more than that. It's more like +30% average at a minmum. At the LOW end its usually 20% faster per clock. I'm sick of people lowballing this kind of thing. Hell, in some games its ~50% faster per clock, and in SuperPi it's 60% faster

Frackal, I think you're confusing performance with clockspeed. We aren't discussing performance in this...

I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

That's gaming only, and at 640x480... how about you dig up more benchmarks hmmm?

I think overall 1.2x is quite accurate. Anandtech concluded the E6600 is equal to or slightly faster than an FX-62 overall, which runs at 2.8GHz. Using my 1.2x calculations, the E6600 equates to the equivalent of a 2.88GHz X2... which is pretty close.

Perhaps it is closer to 1.25x, I don't know, it's just a rough calculation man, why get so worked up over it?

Hey, if you think 1.2x is bad, wait till you see AMD fanboys who claim C2D is only 5 - 10% faster... you'd probably smash your computer in if you read those messages.

 

[Viditor]

Originally posted by: harpoon84

Originally posted by: Frackal

Originally posted by: Viditor

Originally posted by: Frackal

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

No way, it's more than that. It's more like +30% average at a minmum. At the LOW end its usually 20% faster per clock. I'm sick of people lowballing this kind of thing. Hell, in some games its ~50% faster per clock, and in SuperPi it's 60% faster

Frackal, I think you're confusing performance with clockspeed. We aren't discussing performance in this...

I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

That's gaming only, and at 640x480... how about you dig up more benchmarks hmmm?

I think overall 1.2x is quite accurate. Anandtech concluded the E6600 is equal to or slightly faster than an FX-62 overall, which runs at 2.8GHz. Using my 1.2x calculations, the E6600 equates to the equivalent of a 2.88GHz X2... which is pretty close.

Perhaps it is closer to 1.25x, I don't know, it's just a rough calculation man, why get so worked up over it?

Hey, if you think 1.2x is bad, wait till you see AMD fanboys who claim C2D is only 5 - 10% faster... you'd probably smash your computer in if you read those messages.

I tend to agree with you harpoon...it's also only with comparing the 6800 vs the FX62.
Although the C2D is a wonderful core, it's not a Panacea...
It is faster at all levels, but only at the very top end (which almost nobody buys) do you find the higher numbers. At the mid and low range, they really are very close (e.g. the e6300 vs the X2 4200 which sell for almost the exact same price). As Anand points out in his blog:

"Just like the E6400 vs. X2 4600+ battle, it's tough to declare a true winner here. The E6300 pulls ahead in all of the gaming tests, while the 4200+ manages a few wins in the encoding and 3D rendering benchmarks. General application benchmarks generally favor the E6300 over the 4200+ but there are some cases where there's effectively a tie or the 4200+ pulls ahead"
Anand's blog

 

[Furen]

Originally posted by: Frackal
I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

What are you talking about, Core 2 is barely faster than the K8. Just look at the Sciencemark score: 1608 for the X6800, 1578 for the FX-62... [/sarcasm]

Personally, I think that overall, the X6800 is between 20-25% faster than the FX-62, and that's with a higher clock speed. That's not to say that I wouldn't take a Conroe over a K8 in most situations (I still favor the K8 in the low-end) but gaming is not the only use for a PC in my case. And let's not even get started on SuperPi.

Back to the original topic: AMD is behind quite a bit, in my opinion. Yes, I know SOI makes AMD's process something like half a step superior to Intel's equivalent production node, and yes, I know that AMD's production ramp will be faster than Intel's (it seriously took 'till like March for 65nm P4s to actually be available) but, even being optimistic, AMD is at least something like 7 months behind. I really hope the Inquirer's rumors about AMD's first native quad core being based on plain K8 instead of K8L are wrong, or AMD will need massive clock speeds to be able to compete.

 

[zsdersw]

It is faster at all levels, but only at the very top end (which almost nobody buys) do you find the higher numbers. At the mid and low range, they really are very close (e.g. the e6300 vs the X2 4200 which sell for almost the exact same price).

Just you wait (and not long of a wait) for the E4200 and the Celerons based off of Conroe/Allendale.

 

[TanisHalfElven]

Originally posted by: Hard Ball

Originally posted by: Furen

Originally posted by: dexvx
Not much info, but since K8L is 4 issue, similar to Core2, its IPC should be as good as or better than Core2.

I dont think K8L is actually 4-issue... not that it makes a huge difference, since I doubt even Conroe can hit more than 2 ops per clock.

They are not, integer performance would be quite comparable clock for clock with K8, moderately improved through larger BTB, better OoO execution, out of order loads, and a few other smaller improvements.

Both FP and Vector will will be double issue; as opposed to Conroe only being 128bit in vector datapath. So you can expect quite a bit of improvement in terms of FP, at least through FADD and FMUL units. FMISC is not clearly explained at this point; but many speculate that FP ld/st will also be of comparable width, in order to take full advantage of the other execution units in the FP datapath.

there isn't a line in this post i fully understood

 

[imported_inspire]

Originally posted by: Frackal

Originally posted by: Viditor

Originally posted by: Frackal

Originally posted by: harpoon84
Viditor - You can't compare 65nm Conroe to 65nm K8 - different architectures, different clockspeed ceilings.
Look how much 130nm to 90nm gained for AMD over 3 years - a mere 800MHz.

I think 65nm K8 is just a dress rehearsal for the real 'next big thing' - K8L.

Even a 3GHz K8 will not be competitive with the top end C2Ds. As a rough rule, C2D x 1.2 = eqv K8

E6600 = 2.88GHz K8
E6700 = 3.2GHz K8
E6800 = 3.51GHz K8

A 65nm K8 vs Conroe is about as exciting as a 65nm Presler P-D vs AMD X2 - ie. not very.

Bring on K8L!

No way, it's more than that. It's more like +30% average at a minmum. At the LOW end its usually 20% faster per clock. I'm sick of people lowballing this kind of thing. Hell, in some games its ~50% faster per clock, and in SuperPi it's 60% faster

Frackal, I think you're confusing performance with clockspeed. We aren't discussing performance in this...

I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

Actually, no. 20% is a good estimate if you take all the individual benchmarks done in that AT review and look at the relative difference in performance between the FX-62 and the X6800 and weight them evenly to compute the average. I don't know how you come up with "Not one bench is even at 20%" - in the General Performance PC Worldbench 5 bench on Nero, the FX-62 actually posted a gain over the X6800 (2.3%).

I have a spreadsheet I made that incorporates all the benches from the AT review from a couple months ago when harpoon84 and I got into a debate about this exact same issue. I uploaded it somewhere, but it's been so long that I don't have the link anymore...

 

[jiffylube1024]

Originally posted by: Frackal
I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

You've also gotta factor all the things the X6800 has going for it: ginormous cache, high FSB and a tad more clockspeed.

While I'd agree that the 4MB L2 Conroes are in the order of ~20%+ faster than similarly clocked AMD's with dual 1MB cache, it will be interesting to see if 65nm AMD can close this gap or at least narrow it significantly.

 

[coldpower27]

Originally posted by: jiffylube1024

Originally posted by: Frackal
I don't think so. He's saying a C2D performs at an average of Clockspeed x1.2 that of K8, and I'm saying that's a very lowball estimate. It is at least +30% faster per clock vs the X2's.

For instance, from AT's Conroe review:

CPU Bound gaming performance, x6800 vs. FX 62:


Advantage (Intel) 22.8% 55.0% 43.9% 30.9%


Not one bench is even at 20%. The average across the board shows that it is nearly 40% faster, and the x6800 is only slightly higher clocked than the FX-62

You've also gotta factor all the things the X6800 has going for it: ginormous cache, high FSB and a tad more clockspeed.

While I'd agree that the 4MB L2 Conroes are in the order of ~20%+ faster than similarly clocked AMD's with dual 1MB cache, it will be interesting to see if 65nm AMD can close this gap or at least narrow it significantly.

That is assuming AMD will be releasing higher end SKU's from the get go on 65nm which i haven't seen any indications to be the case, it's all about the 3800+, 4200+, 4600+ SKU's that will be shifting to 65nm.