AMD's 65nm is perfect!

[rettihSlluB]

read it http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~103048,00.html

Maybe that was the reason they were delaying 65nm till next year. :lol:

 

[sledgehammer70]

I think AMD is going in the right direction..... as Intel thinks by going lower watts and adding a ton more cores will help them sell more CPU's.... I still think AMD's new CPUs will over dominate Intels new line-up. I think AMD is letting Intel release there news and get going on all there new chips before they bush wack them again! with there new CPU's all's I have to say is FX 60 and AMD Athlon X2 5000+

 

[slvr_phoenix]

Just because they've worked out a new process technology doesn't mean that they're tooled up for it at the FAB level. (Or for that matter have even finished the core redesigns yet.) :?

 

[RichPLS]

on paper everything is perfect. If only on paper perfection is required, Intel P4's accell beyond 5 ghz

 

[sledgehammer70]

on paper everything is perfect. If only on paper perfection is required, Intel P4's accell beyond 5 ghz

5Ghz with diminishing results!

 

[slvr_phoenix]

Actually, if Intel could get Scotty to speed up, it'd be with increasing results.

Intel threw in a nasty cache latency so that it'd scale high. But for other reasons it didn't scale high. So the cache latency is really holding it back at low speeds. If they could get to higher speeds, that horrid cache latency wouldn't be as noticable, as it'd finally be hitting the sweet spot where Intel intended it to be, instead of dragging down the low end like it is now.

 

[sledgehammer70]

if they went 5Ghz they would need a turbine Jet cooler shipped with every CPU.

"Hey guys I got my new 5 GHz CPU for $40,000 ... man oh man does it scream"

 

[mpjesse]

Intel still has the advantage. They've already migrated over to 65nm in Oregon and I believe they've started in Dresden.

Intel is on the attack right now.

-mpjesse

 

[sledgehammer70]

You have to understand, just because they have swithced to the 65nm or even the 45nm doesn't mean they will have the better CPU.

 

[slvr_phoenix]

No, but their new process is looking very good right from the beginning.

In fact, Intel was looking up and up until they switched to the 90nm process. So maybe their suckiness was only for that process.

 

[MeTaLrOcKeR]

No, but their new process is looking very good right from the beginning.

In fact, Intel was looking up and up until they switched to the 90nm process. So maybe their suckiness was only for that process.

You say that now...wait till we see real world results...

Maybe they'll be as worthless as the Xbox 360...who knows?

 

[Atolsammeek]

Remember what happen with the Presscott everyone on intel side went Presscott going to kick butt. And it turn out to be a heater and I almost went for a Prescott laptop vers a Northwood Laptop.

 

[Action_Man]

Worthless as the xbox 360? WTF.

I can't wait till they hit the 65nm process, cheap dual cores for all.

 

[endyen]

In fact, Intel was looking up and up until they switched to the 90nm process. So maybe their suckiness was only for that process.

That is funny. A prescott on any process is still a big leak. Well maybe not FD SOI, but even then, @ 65nanos, it would still leak badly. Just too many interconnects, with too many dissimilar charges next to each other.
As for Amd and 65 nanos, It was my understanding that fab 39 was designed and built for 300mm wafers and 65 nanos. I don't remember hearing that that had changed. I think it qualified @ 65 nanos during the summer, though Amd has been running aditional tests. The above press release may explain some of that.
I have also heard that Amd has an automation program that allows them to transition seamlessly between 90 and 65 nanos. This also suggests that Amd is already 65 nano capable. Since they have offered the tech to SMC, or one of the other chip giants, I'd guess it must work.

 

[sonoran63]

Intel still has the advantage. They've already migrated over to 65nm in Oregon and I believe they've started in Dresden.

No flame - just a correction - Intel has no FAB (or any other presence that I see listed) in Dresden. That's that other company... And we've got at least 3 (or is it 4) FABs ramped, or ramping, on the 65nm 300mm process.

* Not speaking for Intel Corp *

 

[MrAnderson]

I think they are both going in the right direction for each of them.

AMD doesnt have the fab abilities of Intel so in order to get their chips out they need to partner with other companies while they build up a new fab or two.

Intel is a fab powerhouse unfortunately this is a problem with declining sales of cpu's both by market and by AMD taking a chunk. I think Intel should sell manufacturing to other companies since they can supply more than their own chips.

 

[julius]

if amd is already 65nm capable, whats stopping them from switching over? i think theres some finishing touches they need to complete. In any case, 65nm is not going to give intel any advantages, except within its own product line. amd doesnt really need 65nm for right now, as its fine with 90nm, and will be fine when presler comes out, although it would be cheaper to make 65nm. amd is saving up its 65nm recipe for conroe, m2 65nm will likely be at least 3ghz dc, scary thought for amd, but thats most likely what theyll need to compete with conroe.

 

[endyen]

if amd is already 65nm capable, whats stopping them from switching over?

Aside from what you mentioned, fab 39 is not thier production facility yet. That task still rests with it's next store neighbour.

 

[K8MAN]

No, but their new process is looking very good right from the beginning.

In fact, Intel was looking up and up until they switched to the 90nm process. So maybe their suckiness was only for that process.

I remeber when prescott was supposed to be king shit and Spud @ the time had listed an entire page of improvements out of his head(he had really high hopes) and then a few days before release the pipline increase info was leaked and then the final product only for reviewers to find that it was actually slower overall and especially bad in games. :lol:
At least all of the Itanium cores have been a shining success......wait a second............ :lol:
*Sits back waiting for endyen to post with further Pro-AMD propoganda while Xeon writes a 10 page essay to post about how he doesnt care about computers anymore*

 

[julius]

well, i meant more that why doesnt amd just open fab36 witth 65nm production, there must be a reason.

 

[endyen]

It's easier and safer for them to just follow the plan. At this point the plan does not even include shifting fab 36 to 65 nanos. It may be used for 90 nano parts, perhaps even chipsets, for the near future.

 

[ltcommander_data]

slvr_phoenix may have blaimed the 90nm process, but the failure of the P4 was just due to the Prescott architecture, specifically the pipeline increase. A lot of people have negative things to say about the 90nm process, but that has always been in context with Prescott. There really isn't anything wrong with the 90nm process when compared to the 130nm or any other Intel process. Just looking at the Pentium M, the 90nm process allowed Dothan with a 400MHz FSB to increase clock speed from 1.7GHz to 2.1GHz, doubled the cache, while still decreasing the TDP from 24.5W to 21W compared to Banias. The power life for Dothan was virtually identical to that of Banias despite the higher clock speeds, and the increase in power consuming cache. Intel's 90nm process compares quite favourably to AMD's considering a Dothan with a 533MHz FSB has the similar battery life to the Turion 64, despite the Turion's SOI advantage. Granted Dothan may have a few more power optimizations than Turion but the 90nm process certainly isn't working to its disadvantage.

It's funny you mentioned how Intel processors are bad for games. Extremetech actually did some research to discover the reason behind that.

http://www.extremetech.com/article2/0,1697,1895945,00.asp

It seems that game developers, in a rush to get the game out of the door, usually fail to optimize the code for the latest instruction sets. It seems that many games are only optimized for the Pentium III generation which means only SSE support. This means the code penalizes the Pentium IV by not taking into account its higher latencies or its support for SSE2 or SSE3. Now people may feel that optimizing code for the Pentium IV would penalize AMD, but that may not be the case.

"This is unlikely to penalize AMD specifically, though unrolling loops and other P4-specific operations might possibly penalize the Athlon 64, but it's hard to know without actually trying it. But using SSE/SSE2 shouldn't adversely affect AMD. Even Fred Weber, AMD's former chief technology officer, acknowledged that SIMD was the way to go with floating point as we move into the future."

It seems that AMD has no problems with game developers optimizing code for the Pentium IV generation as AMD processors likewise support SSE, SSE2, and SSE3.

What's even more interesting is that in many cases, game developers don't even activate support for SSE as even AMD recommends. They only use FPU code which runs slower on the Pentium IV.

If game developers spent a bit more time to optimize their code for SSE, SSE2, and SSE3 as AMD's Weber suggests, Intel's processors would see better performance in games. It probably won't be enough to dethrone AMD at the very top, but it offers free performance improvements to everyone by making full use of the processor whether AMD or Intel.

 

[K8MAN]

http://www.extremetech.com/article2/0,1697,1895945,00.asp

It seems that game developers, in a rush to get the game out of the door, usually fail to optimize the code for the latest instruction sets. It seems that many games are only optimized for the Pentium III generation which means only SSE support. This means the code penalizes the Pentium IV by not taking into account its higher latencies or its support for SSE2 or SSE3. Now people may feel that optimizing code for the Pentium IV would penalize AMD, but that may not be the case.

"This is unlikely to penalize AMD specifically, though unrolling loops and other P4-specific operations might possibly penalize the Athlon 64, but it's hard to know without actually trying it. But using SSE/SSE2 shouldn't adversely affect AMD. Even Fred Weber, AMD's former chief technology officer, acknowledged that SIMD was the way to go with floating point as we move into the future."

It seems that AMD has no problems with game developers optimizing code for the Pentium IV generation as AMD processors likewise support SSE, SSE2, and SSE3.

What's even more interesting is that in many cases, game developers don't even activate support for SSE as even AMD recommends. They only use FPU code which runs slower on the Pentium IV.

If game developers spent a bit more time to optimize their code for SSE, SSE2, and SSE3 as AMD's Weber suggests, Intel's processors would see better performance in games. It probably won't be enough to dethrone AMD at the very top, but it offers free performance improvements to everyone by making full use of the processor whether AMD or Intel.

Yes possibilities are endless but unfortunatly money and time is not and game developers need to get this stuff out the door on time and budget. Most serious gamers are AMD user's anyway so why bother.

 

[ltcommander_data]

I know there are always time and money limitations but sometimes they are just taken too seriously. When developers are too concious of deadlines rushed games like Battlefield series arise where the game is is buggy and the patches only make it worse. Besides, in the case of SSE optimizations all a developer needs to do is to click a checkmark before pressing compile. I can understand a game developer being hesitant with the latest instruction set like SSE3 breaking his code, however unlikely, but older intruction sets like SSE or even SSE2 have long been ingrained in compilers.

"In discussions with game developers over the past few years, I've learned that they tend to be pretty wary of automatic optimizations generated by simple use of compiler switches. Sometimes a large software build will break when certain automatic optimizations are turned on. Some of this is likely institutional memory, as compilers have improved over the years."

There really isn't any reason why developers shouldn't at least activate the original SSE instruction set which has been around since 1999. Even AMD processors would benefit from that.

 

[endyen]

It's not the compiling they worry about. Sure , adding SSE2 would take a few extra hours to compile, but no big deal. The problem arises out of debugging. The more switches you turn on, the harder it is to find the bug. Why take the chance, when the in game results are so small?