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

pegasus1 · 9 Aug 2006 at 21:02

james.miller said:
can't see it happening, and that's the honest truth. To be any threat at all will require a redesign.

James, surely AMD must have seen Conroe coming and have something up their sleeve, i would be very surprised if they were cought napping by intel.
Some said the same when NV brought out the 7800 series only to see ATI launch the 1800 series. Not being a fanboy but as AMD have been fairly inovative over the years id be surprised if they didnt have a new CPU round the corner.

4p · 9 Aug 2006 at 21:02

I have an idea.

Intel and AMD should work together to make a beasty chip.

Not that it will ever happen, let us pray.

Iraklis F.C. · 9 Aug 2006 at 21:49

these chips will make great HTPC's

Dolph · 9 Aug 2006 at 23:07

Kamakazie! said:
It would surely also run hotter/consume more power too though?
This means it probably wouldn't OC so well on air. With money no object cooling it should still get up to the same sorta speeds mind.

It would generally require/be able to handle more voltage, however power dissapation is normally pretty similar, because you have a larger surface area on a larger die.

Dolph · 9 Aug 2006 at 23:20

Justintime said:
Die shrink did'nt solve it? Explain the Tualatin then They weren't really any different then coppermine apart from being .13nm cpus and some had 512kb L2. The situation is reveresed, AMD now need high clockspeed with thier current chips to perform similar to conroes of less clockspeed.

And a significant amount of data prefetch logic taken from the P4.

The real problem with the Coppermine P3 wasn't that it wouldn't go faster than 1ghz or so, it was really that there wasn't much benefit in doing so, due to the architecture limitations, and the systems would give issues due to limits in the design, that were largely eliminated via additional logic in the Tualutin. The die shrink made space on the die for additional logic to stop the processor starving at high speeds...

Useful reading link http://www.realworldtech.com/page.cfm?AID=RWT061202231808

Justintime · 9 Aug 2006 at 23:27

Good ol days, twas a shame with the 1.13Ghz Coppermine

- a reminder of when you try to sell overclocked cpus

Too bad about the Tualatin though, i know it had a few hundred MHz left in it as all of my Cel Ts ran 1.7Ghz+ with minimal vcore bumps and my 1.4Ghz 512 was limited by bus speed at 1627Mhz. The die shrink did add the capability to clock higher imo, the .18 P6 process just could'nt hack that 1- 1.1Ghz wall , saying that i haven't seen much 1.1 Coppermine celerons in my time either.

Dolph · 9 Aug 2006 at 23:54

Justintime said:
Good ol days, twas a shame with the 1.13Ghz Coppermine - a reminder of when you try to sell overclocked cpus Too bad about the Tualatin though, i know it had a few hundred MHz left in it as all of my Cel Ts ran 1.7Ghz+ with minimal vcore bumps and my 1.4Ghz 512 was limited by bus speed at 1627Mhz. The die shrink did add the capability to clock higher imo, the .18 P6 process just could'nt hack that 1- 1.1Ghz wall , saying that i haven't seen much 1.1 Coppermine celerons in my time either.

I still have a dual P3 board that I bought to fill with two Tualatins (but never did due to the price of the damn things) here. Great little chips they were though.

IIRC I used to be able to get over 1.1ghz with my coppermine celeryon (I had one of the 633mhz 66mhz fsb chips.... they clocked really well as they weren't bus limited)

, but from what I remember, much above 1ghz you had to put loads of voltage through them and the benchmark scores hardly improved at all so it wasn't worth the effort. The days of 100% overclocks.... How we miss them....

Justintime · 10 Aug 2006 at 00:04

You can prolly pick up 2 Tualatin 1.4s for cheapish you know where. Still a bit on the dear side though! Amazed what old tech can still command! I sold a Barton 3000+ for 58 quid u know where 2 days ago, pure profit as i got it free from a clients dead packard bell

Digital Punk · 10 Aug 2006 at 08:08

pegasus1 said:
Not being a fanboy but as AMD have been fairly inovative over the years id be surprised if they didnt have a new CPU round the corner.

Of course they do. Conroe is barely out of the womb and people are writing off AMD already.
It took Intel 3 years to catch up and pass the A64, even with their huge R+D budgets.
It wont take AMD anywhere near as long to catch up to Conroe.

Dolph · 10 Aug 2006 at 10:20

Digital Punk said:
Of course they do. Conroe is barely out of the womb and people are writing off AMD already.
It took Intel 3 years to catch up and pass the A64, even with their huge R+D budgets.
It wont take AMD anywhere near as long to catch up to Conroe.

Supposition and assumption. There's no evidence to suggest that. AMD have a new chip roadmapped late this year, early next (the K8L) but there's no guarantees it will take the performance crown back.

Likewise you seem to suggest that Intel couldn't have caught up faster than they did, yet Netburst was mostly competative up to the end, at least in some benchmarks. It required a high clockspeed (and ultimately it was being unable to ramp that clockspeed higher and get benefits from it that led to the need to get a new architecture), and in some cases was a bit power hungry, but it kept up reasonably well.

AMD are nothing special, I'm afraid. They got the drop on Intel because Intel gambled on something revolutionary, while they followed Intel's previous path (because Intel refused to let them in on how the netburst CPU's would work). Intel's gamble failed and AMD pulled ahead. AMD, while they did design good chips, did not do anything clever or impressive to get ahead.

Digital Punk · 10 Aug 2006 at 10:44

Dolph said:
Supposition and assumption. There's no evidence to suggest that. AMD have a new chip roadmapped late this year, early next (the K8L) but there's no guarantees it will take the performance crown back.

You're assuming I said they would take the performance crown back, which I didn't.

Likewise you seem to suggest that Intel couldn't have caught up faster than they did, yet Netburst was mostly competative up to the end, at least in some benchmarks. It required a high clockspeed (and ultimately it was being unable to ramp that clockspeed higher and get benefits from it that led to the need to get a new architecture), and in some cases was a bit power hungry, but it kept up reasonably well.

In some benchmarks, the current A64's keeps up reasonably well to Conroe.
Intel were only able to keep competative because of their huge marketing power, massive discounts and other bribes that companies were offered not to use AMD.

AMD are nothing special, I'm afraid. They got the drop on Intel because Intel gambled on something revolutionary, while they followed Intel's previous path (because Intel refused to let them in on how the netburst CPU's would work). Intel's gamble failed and AMD pulled ahead. AMD, while they did design good chips, did not do anything clever or impressive to get ahead.

Nice patronising lesson in cpu's there.
If I didn't work in IT and already know that stuff, I might have been impressed.

Dolph · 10 Aug 2006 at 12:21

Digital Punk said:
You're assuming I said they would take the performance crown back, which I didn't.

You certainly implied it, much as you implied that the natural position is AMD on top, which is again a supposition.

In some benchmarks, the current A64's keeps up reasonably well to Conroe.
Intel were only able to keep competative because of their huge marketing power, massive discounts and other bribes that companies were offered not to use AMD.

And in some benchmarks, the various netburst P4's could happily keep up with the A64 and FX series chips. Hence such a statement doesn't really prove anything

Incidentally, while you were ranting about Intel keeping competative you forgot to mention the most relevant factor, Intel could actually supply the market, AMD could not

Your pro-AMD attitude is doing you no favours when it comes to making your point.

Nice patronising lesson in cpu's there.
If I didn't work in IT and already know that stuff, I might have been impressed.

If you knew the stuff already, why did you forget it when you posted something that doesn't match up with it.

pegasus1 · 10 Aug 2006 at 15:02

So to calm things down a little, what kind on innovations would AMD have to come up with to get its CPU competative?

Justintime · 10 Aug 2006 at 15:05

Prolly a super great SSE4 engine and better L2 cache effiency and design + a few hundred MHz more in the closer to 4Ghz range.

Digital Punk · 10 Aug 2006 at 15:10

Dolph said:
You certainly implied it, much as you implied that the natural position is AMD on top, which is again a supposition.

Rubbish.
I certainly did not suggest they would take back the performance crown, thats merely your assumption. I actually said 'Amd wouldn't take three years to catch up to Conroe'. If they do, then they'll be out of business.
Again, you assume I think AMD's natural position is on top. I don't. They aren't.

And in some benchmarks, the various netburst P4's could happily keep up with the A64 and FX series chips. Hence such a statement doesn't really prove anything

You've already said that. And why did you make a point of saying it in you're previous post if it doesn't prove anything.

Incidentally, while you were ranting about Intel keeping competative you forgot to mention the most relevant factor, Intel could actually supply the market, AMD could not Your pro-AMD attitude is doing you no favours when it comes to making your point.

LMFAO. Ranting about Intel! Pro AMD! Your pompous and patronising tone are doing you no favours when it comes to making your points tbh.

Justintime · 10 Aug 2006 at 15:11

Digital Punk said:
Your pompous and patronising tone are doing you no favours when it comes to making your points tbh.

Unfortunately i must agree with that statement.

NathanE · 10 Aug 2006 at 18:40

I still remember the " :rolleyes:

." Digital Punk gave me when I first suggested (about Nov 2005 I think?) that Conroe would knock AMD off the top spot

In some benchmarks, the current A64's keeps up reasonably well to Conroe.

Unless I'm missing something, "the P4's kept up with the A64's in some benchmarks" is also a true statement... which seems to negate most of the impact this statement was intended to have.

James, surely AMD must have seen Conroe coming and have something up their sleeve, i would be very surprised if they were cought napping by intel.

You would have thought so. But back then AMD were, like pretty much everyone in this very forum - bar the few that knew Conroe really was something special, were blinded by the success of the A64. Remember all those "duels" AMD kept offering up to Intel - effectively gloating in a very public way. They were too loved up with themselves, prancing around like a clown, to realise the sleeping giant's eyes were opening.

K8L will almost certainly "match" Conroe on performance. But it's still the aging K8 architecture at its core whose scalability is running out fast. Intel was initially planning to launch Conroe at much higher clock speeds but realised there was no need to and so decided that they'd save those higher clocks for a rainy day (such as the K8L launch day

)

dante6491 · 10 Aug 2006 at 19:55

but higher clockspeeds are already available now, just ahve to overclock them.

NathanE · 10 Aug 2006 at 20:21

dante6491 said:
but higher clockspeeds are already available now, just ahve to overclock them.

Exactly

locutus12 · 10 Aug 2006 at 21:28

i think a lot of people are forgetting that 95% of the CPU`s sold will be going desktop producers and sold off the shelf, then sold to people who`s only experience of over clocking something is winding up the alarm clock for the morning.

conroe`s overclocking ability will have very little impact on AMD or intel`s market share for that matter. what will have an impact is price, and for business it will be energy consumption + price, and on that front they are both pretty damn good.

overclocking shouldnt be discussed when talking about the things that will make these CPU`s popular to most of the market.