AMD's Response to Intel Conroe: Energy Efficient Athlon 64 X2 CPU

[Ricochet J]

The first impressions left by the new Core 2 Duo processors from Intel that arrived to replace not the best Pentium 4/Pentium D processor families turned out to be highly positive. Since the Intel processors with Core microarchitecture have been officially launched already, we had a lot of opportunities to check out their strengths and attractive features. Core 2 Duo CPUs demonstrated unprecedented performance, relatively low heat dissipation and excellent overclocking potential. Everything indicated that a new favorite appeared in the processor market. It is true, all objective data proves that Athlon 64/Athlon 64 X2 processors that used to offer the best combination of consumer-friendly features cannot be a serious competitor to the new Intel CPUs on Core microarchitecture anymore. However, the actual situation is not quite like that.

In response to Intel’s extremely successful Core 2 Duo processors AMD reduced the prices on its Athlon 64 X2 CPUs dramatically. But it was only the first move. Today we are going to discuss the second move: the beginning of mass shipments of new energy efficient Athlon 64 X2 processors dissipating considerably less heat than their predecessors. Read our review for details now!

Source

Question is, will this be enough to keep up with Intel?

 

[Stelly]

I dont think so... and I'm just being honest... it needs to be a totally new chip

Stelly

 

[Durzel]

I suppose in theory more energy efficient chips could mean more overclockability from the X2s, but I think it would have to enable a radically higher overclock than currently available to remain even slightly competitive - and there may not be enough in the process to make that happen, regardless of heat/energy.

 

[Digital Punk]

I don't think its about 'keeping up' per se.
We know from a performance aspect that Intels latest are faster, but something like these would be good for shuttle's and small form PC's like media centre's, especially if you use a fanless heatsink.

Some businesses might be interested too, if they want to keep energy bills lower and also have a decent performing cpu.

 

[El Jimben]

Even if they do increase clockspeeds, AMD's chips are still a long way behind in terms of cache too.

 

[4p]

Yay, something ive been waiting for.

When would these be released then?

 

[UKTopGun]

Conroe celerons?

 

[InvaderGIR]

What everyone seems to overlook about AMD and Intel at the moment is that Intel's Conroe are 65nm which is why they are faster than AMD's (which are 90nm). Now when AMD launch their 65nm chips everyone will **** off Intel as AMD will have a very high performance new chip...it's the way the processor industry and people work.



InvG

 

[UKTopGun]

What everyone seems to overlook about AMD and Intel at the moment is that Intel's Conroe are 65nm which is why they are faster than AMD's (which are 90nm). Now when AMD launch their 65nm chips everyone will **** off Intel as AMD will have a very high performance new chip...it's the way the processor industry and people work.



InvG

That's over 3 months away though. 3months is far to long away considering a lot of the people on here will be on Conroe by then anyway! Thats the processor industry .

And there's nothing saying there going to be much better than conroe because in over 3months there will be a new conroe released!

AMD have got a tough job. They have to meet the "overclockability" (funny word ), the price (by quite a bit!), and performance!

 

[InvaderGIR]

AMD have got a tough job. They have to meet the "overclockability" (funny word ), the price (by quite a bit!), and performance!

Well AMD have been beating Intel on all of those points really, till the Conroe which has allowed overclocking, isn't highly priced and has performance.
AMD have time to sort this out, they also have time to make their CPUs the same/better/slightly worse than the Intel 65nm (you sure they will be 'Conroe'?) in 3 months time without the fuss from marketing and selling a chip on 65nm.

I admit most people have upgraded this summer because of Conroe, and that will hit AMD hard especially as most people seem to be upgrading in summer not winter anyways more nowadays due to going back to school or going to uni etc.
Due to how the market works it's hard to say what will happen in 3 months time, but I'd have thought that they should be level-pegging (roughly at least) then till the next advance in technology.

I'm probably wrong and Intel may well stay on top, if they do, then they do and AMD and ATi are both quite screwed.

InvG

 

[Gibbs]

to be honest i think this is just a stop gap for AMD, another product on te market that they can sell. Every chip they sell its 1 less Intel Chip. Thats how they have to spproach this. They cant lose to big a market share/profit.

A lot of people who follow hardware developemnt that AMD hasnt really got anything that will beat Core2 untill next year.

A low-power X2 only has 2 advantages, less power, less heat. They *mite* overclock better than current X2s with out much, but a water/phase cooled or even a very well air cooler could let you push you normal X2 to the same limits.

This is AMDs way of stiring the market for the average Joe. Average Joe will be, on, a new AMD chip for less money, but be better than the Core2 since its newer, fresher etc etc.

Its only the real informative people or who can be bothered to research that will know the difference.

AMD know they have a hill to clime. Intel know it to. Someones got to do something.

 

[Justintime]

What everyone seems to overlook about AMD and Intel at the moment is that Intel's Conroe are 65nm which is why they are faster than AMD's (which are 90nm). Now when AMD launch their 65nm chips everyone will **** off Intel as AMD will have a very high performance new chip...it's the way the processor industry and people work.



InvG

sorry, no. I'm no Intel fanboy, been using AMD mainly up to Conroe.. but clock for clock the conroe destroys K8, and nothing AMD has now even overclocked to the moon competes. As for shrinking to 65nm, AMDs past track record has shown its not the best when changing processes, the initial release may not even clock that much more than the current speeds and then they'll finally tweak and tweak and get some clockspeed, though it might be late in the game and unless they bring out something that can clock in the 4Ghz+ region at launch i think Intels safe. K8L might also not hold the key for regaining the crown..i think a new core from scratch is AMDs only salvation.

 

[Sparky]

sorry, no. I'm no Intel fanboy, been using AMD mainly up to Conroe.. but clock for clock the conroe destroys K8, and nothing AMD has now even overclocked to the moon competes. As for shrinking to 65nm, AMDs past track record has shown its not the best when changing processes, the initial release may not even clock that much more than the current speeds and then they'll finally tweak and tweak and get some clockspeed, though it might be late in the game and unless they bring out something that can clock in the 4Ghz+ region at launch i think Intels safe. K8L might also not hold the key for regaining the crown..i think a new core from scratch is AMDs only salvation.

So 130nm to 90nm didnt see a big increase in performance in your opinion?

 

[OC_A64]

The die shrink to 65nm will be the first step, still not going to bring AMD close to Intel though... It'll help but Conroe will reign supreme.
K8L should be where AMD claw back their market share, should see a nice performance boost over the current gen 939's and the rev G A64's due out soon.
Also dont forget AMD has a major advantage over Intel, and thats the use of HTT over the "normal" FSB, along with the integrated memory controller. Should stand them in good stead as the number of cores are ramped up. Ok it might hurt via the cold bug, but lets face it AMD/Intel arent really designing chips for overclockers are they...?

Just cos' a large number of overclockers/enthusiasts are moving to conroe, doesnt mean AMD are loosing out on sales either. We make up a minute amount of the market. I don't think AMD will be hit as badly as many people make out by core, especially as many people know how good A64's are and how "bad" Intel has been in the past. Price drops along with a mature platform and a superior range of boards will keep AMD in the game for a little while yet. By the time sales of Core start to ramp up, AMD should have K8L pretty much ready to roll - then the fun begins

 

[manoz]

The die shrink to 65nm will be the first step, still not going to bring AMD close to Intel though... It'll help but Conroe will reign supreme.
K8L should be where AMD claw back their market share, should see a nice performance boost over the current gen 939's and the rev G A64's due out soon.
Also dont forget AMD has a major advantage over Intel, and thats the use of HTT over the "normal" FSB, along with the integrated memory controller. Should stand them in good stead as the number of cores are ramped up. Ok it might hurt via the cold bug, but lets face it AMD/Intel arent really designing chips for overclockers are they...?

Just cos' a large number of overclockers/enthusiasts are moving to conroe, doesnt mean AMD are loosing out on sales either. We make up a minute amount of the market. I don't think AMD will be hit as badly as many people make out by core, especially as many people know how good A64's are and how "bad" Intel has been in the past. Price drops along with a mature platform and a superior range of boards will keep AMD in the game for a little while yet. By the time sales of Core start to ramp up, AMD should have K8L pretty much ready to roll - then the fun begins

When are the K8Ls out then?, I'm saving up at the mo, might wait for them.

 

[Gashman]

not for a fair while yet, not even by end of year i think

 

[Justintime]

So 130nm to 90nm didnt see a big increase in performance in your opinion?

They have, but not to the extent that they originally predicted, also it took a longish time to get where they are now, like i said, it usually takes AMD much longer to tweak a process to ramp some decent speed than Intel. Its only in the later days the process tweaks are really working and its definetly not enough to counteract intel, making matters worse is they have yet to adopt 65nm, which might be another nightmare to get right.. heard talk of a jump to 45nm instead of a transition, but that'd be a huge accomplishment that might not be that feasible. I'd love to eat crow on this.. but knowing AMDs past..

 

[manoz]

not for a fair while yet, not even by end of year i think

Ho-hum, I'd rather not wait that long . Oh well, I'll weigh my options when I actually need the system.

EDIT: When the new K8Ls come out, would we expect to see a price drop on opterons?

 

[james.miller]

So 130nm to 90nm didnt see a big increase in performance in your opinion?

not massively, no. the 130nm a64's clocked well. all 90nm really did was reduce heat. it didnt chance performance to any real extent, and if thats anything to go by, the move to 65nm wont either.

 

[Justintime]

not massively, no. the 130nm a64's clocked well. all 90nm really did was reduce heat. it didnt chance performance to any real extent, and if thats anything to go by, the move to 65nm wont either.

Exactly, the clockspeed jump was'nt that great. The best i've seen AMD in good form was with the XP cpus, using that .18nm process all the way up to 1.73Ghz when even Intel could'nt get thier P6 much past 1Ghz on .18 (1.1 if you count the Celeron Coppermine).