Is AMD going to be in trouble 2006?

[Zebo]

I don't mean financial, that seems to be a perpetual situation, I mean with us.

In the here and now, I fully expect AMD Veince Dual cores to dominate Intels prescott dually's this year. First off, Intel looses it's three primary advantages. Hyperthreading, which will be absent from duallys since two real live cores blows it away. SSE3. And finally clock speed because two presshots ain't going to run 3.5+ dual mode -- exponential heat increases with two cores will make it impossible, while AMD will easily clock to 2.5 absolutly destroying Intels sloppy netburst in everything. 05 is AMD.

However Dothan is in the shadow. I think, looking at the reviews, Dothan has higher IPC than A64. The only saving grace is A64's intergrated mem controller and moblie crippled dothan...And Dothan Runs cooler. Uses less power. Scales. Basically killer app.

All Intels got to do it tweak an already great chip a bit for desktop. Like adding 64bit instructions, quad pumped dual channel DDR2 it, dual core it etc and I think you'll have a chip laying the smack down on A64...lower power, faster, and maybe cheaper with mass production since it's smaller die now... and smaller yet again 65nm...prolly circa 06' we can see this.

So what's AMD got up it's sleeve? You think they'll be in trouble?

Here are a couple reviews of Dothans excellent performance on crippled old platform.

http://www.behardware.com/art/imprimer/546
http://techreport.com/reviews/2005q1/dfi-855gme-mgf/index.x?pg=1

 

[amol]

AMD will surprise us . . . just you watch

 

[Duvie]

Originally posted by: Zebo
I don't mean financial, that seems to be a perpetual situation, I mean with us.

In the here and now, I fully expect AMD Veince Dual cores to dominate Intels prescott dually's this year. First off, Intel looses it's three primary advantages. Hyperthreading, which will be absent from duallys since two real live cores blows it away. SSE3. And finally clock speed because two presshots ain't going to run 3.5+ dual mode -- exponential heat increases with two cores will make it impossible, while AMD will easily clock to 2.5 absolutly destroying Intels sloppy netburst in everything. 05 is AMD.

However Dothan is in the shadow. I think, looking at the reviews, Dothan has higher IPC than A64. The only saving grace is A64's intergrated mem controller and moblie crippled dothan...And Dothan Runs cooler. Uses less power. Scales. Basically killer app.

All Intels got to do it tweak an already great chip a bit for desktop. Like adding 64bit instructions, quad pumped dual channel DDR2 it, dual core it etc and I think you'll have a chip laying the smack down on A64...lower power, faster, and maybe cheaper with mass production since it's smaller die now... and smaller yet again 65nm...prolly circa 06' we can see this.

So what's AMD got up it's sleeve? You think they'll be in trouble?

Exactly what I am thinking though I think the dothan ones wont arrive until sometime into 06......

I agree starting with pbetter IPC and lowe rpower and heat is nice...sorta reminds you of how AMD got to where it is versus the presshit.....It will need a whole new core revision before Intel would and that is truly the shoe on the other foot...

 

[Zebo]

Other foot LOL, ya it will be funny seeing 3.2 Ghz A64 and "only" 2.8 Dothans..

 

[mustap]

have faith and u shall be deserved

 

[MisterChief]

Personally, I am very anxious to see how well Intel's dual core/64-bit CPU's will perform. Can you believe future Celerons will be 64-bit as well?

 

[mwmorph]

dont worry, amd always pulls through with something suprising and creative. i bet you they already have a secrect project combating the dolthan right now that will have even higher ipcs/clocks.

 

[Shenkoa]

Intel needs to realize that dothan is the only thing that will save its arse, untill then AMD will own Intel.

 

[Green Man]

Dothan has 2M L2, so I would be surprised to see it get a huge benefit from dual channel. I don't think there is much benefit to going with the higher latency of DDR2 until DDR2 speeds get up high enough to negate the effect. By then AMD will have DDR2 as well. All of the information I have seen has shown Yonah without iAMD64 as well. I don't see how Intel can plan to move it to the desk top no matter how well it performs in 32 bit when the buzzword is 64 bit. I think A64 will continue to hold an IPC as well as a clockspeed advantage over the Pentium M successor, but I hope intel can find a way to push AMD into at least releasing some faster clock speeds.

OTOH, if you buy a 3000+ now and overclock it to 3800+, you are sure to have a near top of the line processor for the next year and a half +.

 

[ponyo]

I hope not. Speaking as AMD shareholder.

 

[stratman]

I hope AMD doesn't ever get crushed in the performance race. Even though logistics have me typing from a dothan lappy, I'm an amd fanboi

But i gotta question related to all of this -- why are dothans so expensive right now -- are they more inherently more expensive than other procs to make, or is it just because intel is producing lower numbers of them?
And also, people are always talking about how the dothan is not a well-rounded proc, that it lags far behind the prescott and the A64 in some crucial areas, is there some way for intel's engineers to add to the core to improve it's performance in these areas (media encoding?) or is it a problem with the core?

These, along with Intel's reluctance to ditch netburst, are the only hinderances I can see for the dothan's rise to glory.

 

[Duvie]

L2 cache size probably...2mb of cache takes up considerable amount of the core and adds a bunch of transistors, not to mention is expensive....look at the p4EE with 2mb of l3 cache which doesn't even run at cpu clock speed likt the l2...I think it is SRAM and it is much more expensive then system ram we see today...

 

[Zebo]

Originally posted by: Naustica
I hope not. Speaking as AMD shareholder.

I'm selling soon. I own Intel too... go figure... I can't afford to sell until it hits $28 then I break even.

 

[DrMrLordX]

Dothan does not stand to gain much from increased memory bandwidth. It DOES stand to gain much from an increased ability to handle multi-threaded applications. That seems to be its "Achilles' Heel", so to speak. Put Dothan in dual and quad-core CPU setups(1 or 2 dual-core CPUs) with multi-threaded applications, and bingo, the problem is solved.

I'd like to see it happen, if only to force AMD to stay competative and silence the annoying Cell-fanatics floating around.

 

[Zebo]

You guys that think memory won't do much for dothan, fine. However there were some performance snips to the core they had to make for lower power in it's intended function, laptops. When they bring a derivative to desktop they are more free to add features, ramp etc which can only increase dothans already excellent performance since they are no longer restricted to 20-28W envelope.

Read about it here:
http://www.intel.com/technology/itj/200...ssue02/art03_pentiumm/p01_abstract.htm

 

[Avalon]

If it happens, it happens. I'm more partial to AMD, but only because they've given me the best performance for the price for a few years now. If Intel can do that with a Dothan in 2006, they'll be my new best friend. As long as CPU performance continues to improve steadily and there is competition, I'll be happy.

 

[stratman]

Originally posted by: Zebo

Originally posted by: CraigRT
competition is good! but this will only help AMD in the long run! :evil:

There will be no compitition in rendering or science apps. I mean the M just gets crushed...
http://www.gamepc.com/labs/vie...asp?id=gmso&page=7

http://www.gamepc.com/labs/vie...asp?id=gmso&page=9

Is there an easy way to fix this to make the dothan a well-rounded proc?

And could Intel throw on a on-die ddr controller?

 

[Sc4freak]

On Die? DDR? Intel? I think that if Intel ever includes an on-die memory comtroller it'll be for DDR2.

 

[DrMrLordX]

Originally posted by: Sc4freak
On Die? DDR? Intel? I think that if Intel ever includes an on-die memory comtroller it'll be for DDR2.

Intel won't go for on-die memory controllers if they continue to push BTX.

 

[Vette73]

Originally posted by: DrMrLordX

Originally posted by: Sc4freak
On Die? DDR? Intel? I think that if Intel ever includes an on-die memory comtroller it'll be for DDR2.

Intel won't go for on-die memory controllers if they continue to push BTX.

That and they make a PILE of money from their chipsets. Look at AMD's chipssets in the market and most offer about the same performance, so for a oem you would choose on price. That would hurt intels $$$ to much. But for AMD that only makes server chipsets, putting the mem controller on die adds performanmce and they don;t lose any money in the chipset war.

 

[jbh129]

Yonah (the dual core Dothan) will destroy its AMD competitors clock for clock. Its going to be updated with SSE3 instructions and some other improvements over the semi-ancient dothan/banias architecture. BTW, Dothans will benfit from higher memory speeds on the Alviso cipset even if dual-channel isnt more than a 5% gain

 

[Duvie]

Originally posted by: jbh129
Yonah (the dual core Dothan) will destroy its AMD competitors clock for clock. Its going to be updated with SSE3 instructions and some other improvements over the semi-ancient dothan/banias architecture. BTW, Dothans will benfit from higher memory speeds on the Alviso cipset even if dual-channel isnt more than a 5% gain

I think destroy is pushing it and I also think its arrival if late like everything else from Intel over the last several years will diminish any perceived leads it would have now....

I expect we wont see any of that until well into 2006 ad that is rather optimistic for Intel and pessimistic for AMD to think Intel will evolve AMD will not...

Intel is further behind then we think or they wouldn't be in a rush to throw out a dual core option using its obviously failed prescot architecture that is already consuming way to much power and producing way too much heat. The pipes wont change much and they are going to get rid of HT...That is the crutch that has at least kept it close or in the lead in a few remaining items today...SSE3 will be a moot point since AMD will have it as well....AMDs dual core will be more seamless since A64's were rumored to have been designed with this in mind....I think why they can put out a dual core with minimal redesign they have the next thing lieing in wait....

How about quad cores??? How about dual cores in dual cpu setups??? I still think we will go more and more into multiprocessors then just faster proecessors...I have been saying this for 3 years now, before any talk of dual core anything...

 

[AWhackWhiteBoy]

Originally posted by: stratman
Is there an easy way to fix this to make the dothan a well-rounded proc?

And could Intel throw on a on-die ddr controller?

the Banias/Dothan floating point unit is largely the same as the Pentium 3's from which the entire core was derived. so in this respect the Dothan is still very much inferior if a program leans on that aspect heavily. what your seeing is how a P3 would fair again an AMD64 in that benchmark.

 

[gobucks]

intel can't throw in an on-die memory controller unless it gives up on BTX, which i suppose is possible after they axe the P4, since the PM's lower heat output might make BTX unecessary. Unfortunately, though, that's why AMD can't take up the BTX form factor and why Intel is showing no interest currently in an integrated controller.

As for the AMD/Intel war in 2006, i admit it will be interesting. It will be a lot closer than the A64/P4 competition, and i suspect that AMD will dominate big time till desktop PMs take over. Still, I don't think Intel's gonna suddenly have a big lead, if they have any lead at all. They have a disadvantage in that the PM took a backseat to P4 for years, and as a result it is behind the technology curve. There is no good chipset for it right now, its bus is extremely slow, it was not built with dual core in mind, as AMD's K8 architecture was, its clockspeed is much lower than it could have been if it had been their primary focus, and it currently has no 64-bit functionality built into it. Those are a lot of hurdles that Intel needs to jump over before they can compete with AMD's best. As for clockspeed, sure, a 2.8GHz PM will own just about any processor, but since it's official highest speed is 2.0GHz, and usually intel boosts clockspeed by 100MHz, that's 8 more speed bumps, or 4 if Intel steps up the boosts to 200MHz. And that's just to keep up with AMD's current lineup, which won't be standing still.

 

[Jeff7181]

I wouldn't expect to see a major AMD core change before 2007... the Athlon-64 finally came out at the end of 2003 right? And it normally takes 3-5 years of development time to put a new core into production.