Linked by Thom Holwerda on Fri 14th Jul 2006 21:01 UTC, submitted by DigitalDame
AMD AMD executives said this week that they will promote its '4x4' enthusiast platform to counter the Core 2 Duo. AMD also plans to push a sort of 'performance number' into the market to redefine how consumers should think about power. The 4x4 platform will place two physical sockets on a motherboard, connected by AMD's Direct Connect architecture. Mounted on each socket will be an AMD Athlon 64 X2 processor, for a total of 4 cores. An eight-core '8X8' program will roll out in 2007.
Order by: Score:
Gadgetitus
by eggs on Fri 14th Jul 2006 21:13 UTC
eggs
Member since:
2006-01-23

I'm going to suffer a gadgetitus inflamation!

Reply Score: 2

Mod
by fretinator on Fri 14th Jul 2006 21:13 UTC
fretinator
Member since:
2005-07-06

I can't wait to see the '4x4' shift control on the side of the computer that switches the computer into '4x4' mode. It should include sound effects - Vrrroooooom, Vrooooooooooooom.

Reply Score: 4

4x4 'enthusiast platform?
by Ronald Vos on Fri 14th Jul 2006 21:19 UTC
Ronald Vos
Member since:
2005-07-06

I would assume the 4x4 platform would be for more than a handful of enthusiasts alone.

Reply Score: 1

RE: 4x4 'enthusiast platform?
by Wes Felter on Fri 14th Jul 2006 21:39 UTC in reply to "4x4 'enthusiast platform?"
Wes Felter Member since:
2005-11-15

It will probably be extremely expensive; think $1000 of processors and $1000 of graphics cards on a $250 motherboard.

Reply Score: 4

RE[2]: 4x4 'enthusiast platform?
by DittoBox on Fri 14th Jul 2006 22:14 UTC in reply to "RE: 4x4 'enthusiast platform?"
DittoBox Member since:
2005-07-08

..In a 300 dollar case, with a 350 dollar sound card, two or three 150 dollar hard drives, 40 dollar DVD burner, 200 dollar OS, 70 dollar mouse, 30 dollar mouse pad, 50 dollar keyboard, 1500 dollar 30" lcd panel...

I salute all 3 of you out there who can afford such a monster. Now, just remember that your powerbill will quadruple in price.

Reply Score: 2

Too much too soon?
by Pseudo Cyborg on Fri 14th Jul 2006 21:38 UTC
Pseudo Cyborg
Member since:
2005-07-09

Cores are the new Megahertz.

Reply Score: 4

RE: Too much too soon?
by atomicplayboy on Fri 14th Jul 2006 22:06 UTC in reply to "Too much too soon?"
atomicplayboy Member since:
2006-04-28

It does seem the best way right now to get large performance gains with little heat and energy consumption. This, however, doesn't seem the right strategy for AMD to be taking. I can see it offering a performance edge (I would think a pretty substantial one) for a very short time until Intel offers up a two socket platform of it's own. I also don't see why the enthusiast couldn't have done this before with an opteron board. It seems that AMD won't have a true edge again until there 4 core parts.

Reply Score: 5

RE[2]: Too much too soon?
by Pseudo Cyborg on Fri 14th Jul 2006 22:35 UTC in reply to "RE: Too much too soon?"
Pseudo Cyborg Member since:
2005-07-09

Exactly. It seems like a poor interim solution. Actually, not much of a solution as just marketing hype.

Reply Score: 5

RE[3]: Too much too soon?
by abraxas on Fri 14th Jul 2006 23:53 UTC in reply to "RE: Too much too soon?"
abraxas Member since:
2005-07-07

It does seem the best way right now to get large performance gains with little heat and energy consumption. This, however, doesn't seem the right strategy for AMD to be taking. I can see it offering a performance edge (I would think a pretty substantial one) for a very short time until Intel offers up a two socket platform of it's own.

Not exactly. Intel doesn't have DirectConnect, HyperTransport and an integrated memory contorller. AMD's solution will be a clear winner in terms of bandwith, and Intel won't be able to provide a similar system until they redesign their chipsets.

Reply Score: 4

RE[2]: Too much too soon?
by jcinacio on Sat 15th Jul 2006 01:11 UTC in reply to "RE: Too much too soon?"
jcinacio Member since:
2006-03-12

From the article: "The 4x4 platform will place two physical sockets on a motherboard, connected by AMD's Direct Connect architecture. Mounted on each socket will be an AMD dual-core processor (not necessarily an AMD Athlon64 X2, according to an AMD representative), for a total of four cores."

When i read 4x4, i really was expecting something more - either a surprise 2-way quad-core cpu's, or some kind of wonder technology...
Still, i have very mixed feelings about this: on it's own, this "strategy" is sure to be a fiasco, but i think it will be somewhat usefull to demo the capacity of AMD's Direct Connect architecture.

Will it be enough to regain the market that they for sure *will* loose to Intel until K8L? not even close.
I hope they can come up with something more meanwhile...

Reply Score: 1

RE[3]: Too much too soon?
by kaiwai on Sat 15th Jul 2006 06:06 UTC in reply to "RE[2]: Too much too soon?"
kaiwai Member since:
2005-07-06

Will it be enough to regain the market that they for sure *will* loose to Intel until K8L? not even close.
I hope they can come up with something more meanwhile...


Given that Sun is now shipping 16 way Opteron machines in U2 form factor, running Solaris, one has to ask, when are we going to start seeing mainstream computer assemblers offer the same level of scalable machines running Intels processors.

Intel had the opportunity to close the gap with the adoption of CSI, they chose to instead keep with the old crusty bus they have now; it might win them back some customers, but server customers have now opened up and realised that Opteron can deliver more for less.

Reply Score: 2

RE[2]: Too much too soon?
by suryad on Sat 15th Jul 2006 02:56 UTC in reply to "RE: Too much too soon?"
suryad Member since:
2005-07-09

Well whether it is a 4 x 4 or whatever AMD might call it the basica architecture is still K8 and not K8L right? So that just means they are slapping on more cores albeit on a 65 nm process so that should help clock speeds and power consumption and heat production somewhat. Still the architecture being the same, performance wont be different right? All you are doign is throwing more cores at the problem. The problem is that this is a bad idea because software is just not that multithreaded. There are few softwares like dvd rippers or encoders etc that will like having the extra cores but as an end user, it just does not help much for the price one will have to pay.

Intel did great because their cores seem to be single threaded as well as multithreaded champs. They run single threaded games a lot faster than their AMD counterpart and they are great at multitasking as well. But one thing surprises me is the size of the cache. Does anyone besides me think that is sorta on the large side?

Reply Score: 1

stopgap
by cutterjohn on Fri 14th Jul 2006 22:40 UTC
cutterjohn
Member since:
2006-01-28

Well, I guess that they had to come up with something, too bad it wasnt an announcement that K8L was ahead of schedule, and with some benches.

In any event, they're going to have to seriously reduce pricing...

Reply Score: 5

RE: stopgap
by suryad on Fri 14th Jul 2006 23:09 UTC in reply to "stopgap"
suryad Member since:
2005-07-09

And reducing prices is always good. That said AMD still has an advantage in the 64 bit arena. I wonder how the new Cores do in 64 bit.

Reply Score: 3

Computer just following car industry??
by somebody on Sat 15th Jul 2006 00:53 UTC
somebody
Member since:
2005-07-07

Turbo? - checked
Quadro? - checked
4x4? - checked

I bet my head, that the next one is to be CPU with ABS handled cooler spinning and TCS on bus.

After that we the poor souls will probably have hissing water cooled systems (except that instead of "tea ready" it will mean "system overheat" or roughly translated "cpu cooked") which come equiped together with car windshield wipers on our LCD. God help us!

Edited 2006-07-15 00:54

Reply Score: 2

megahertz -> cores + ?
by butters on Sat 15th Jul 2006 04:28 UTC
butters
Member since:
2005-07-08

Most tech savvy poeple now realize that clock frequeny isn't everything. They know that performance per watt in important, and the microprocessor companies are telling them that, to these ends, having more cores is better. What they might not know about is the sorry state of task-level multithreading in the software world. And even if they do, they don't have any idea how this can be fixed in a 5-year timeframe.

It's clear that application vendors are not going to parallelize their code unless their competitors do it first, a catch 22 that results in nothing doing. Further, there are severe limits to how effectively programmers can parallelize legacy code. I conclude that if efficient threading is going to exist any time soon, it's going to have to happen through a combination of compiler/vm and CPU technologies.

The way I envision this is as follows: AMD (for example) adds zero-cycle instructions to their ISA that enable compilers to pass hints to the processor that identify chunks of code that can be parallized if there are available hardware threads. Implementing heuristics for optimal threading would be far easier to do at the compiler level than in hardware, although algorithms that make decisions based on dynamic runtime state must still be implemented in hardware (or in a hypervisor!).

Code written for virtual machine platforms could be even more effectively threaded through intelligent algorithms implemented in the virtual machine. I don't quite understand why Sun and Microsoft aren't devoting massive research dollars to developing this technology for Java and .NET, since this wouldn't require any special hardware support at all. They could put the "less than native speed" arguments to bed for good if they started beating the pants off natively compiled code on multicore processors.

Say what you want about the Gigahertz race, but back in those days, the AMD and Intel were offering complete solutions to the the performance problem. Without efficiently threaded applications, we are missing a piece of the performance puzzle that must fit in the empty space next to multicore architecture.

Reply Score: 4

The best encounter
by hraq on Sat 15th Jul 2006 05:59 UTC
hraq
Member since:
2005-07-06

AMD should reduced the prices of their X2 line immediately and hurry up with quad core ( X4) line before intel even on their 90nm process if not 65nm.

No other solution will save AMD from Intel's slaughter.

Reply Score: 1

RE: The best encounter
by kaiwai on Sat 15th Jul 2006 06:10 UTC in reply to "The best encounter"
kaiwai Member since:
2005-07-06

AMD should reduced the prices of their X2 line immediately and hurry up with quad core ( X4) line before intel even on their 90nm process if not 65nm.

No other solution will save AMD from Intel's slaughter.


And price reductions are exactly what almost killed AMD; AMD should stick to the high margin, business of developing workstation and server class CPU's.

Reply Score: 1

4x4 won't necessarily be so expensive
by joshv on Sat 15th Jul 2006 13:24 UTC
joshv
Member since:
2006-03-18

Socket AM2 CPU prices are going to come down, they have to in order to compete. A retail dual socket motherboard might have a $100 premium on the outside. An extra AM2 dual core CPU, $300 or less (though granted, not if you want the latest 'FX'). I think many enthusiasts would pay an extra $400 for '4x4'.

Granted this is only a stopgap measure. AMD has to work on their CPU architecture. I am sure the 65nm transition (and the extra cache that will bring) will help things, but AMD better have more than that in its bag of tricks in order to stay competitive.

Reply Score: 1

...
by suryad on Sat 15th Jul 2006 17:32 UTC
suryad
Member since:
2005-07-09

The way I see it is like this. Intel introduced the ace up their sleeve. AMD is going to do the same. But not this time but next year. It is just a smart move on Intel's part that they chose a quite time to release CPUs. But I am not going to underestimate AMD. There could be a chance that in the next few months, AMD will have a surprse announcement that the K8L is ready to go...in 90 nm!!! For those who choose to wait can wait for the 65 nm counterparts coming out next year. The reason why I think this could be quite plausible is because I have been hearing and reading over the Net for a long time that AMD has been working on a brand new architecture supposed to be released around 2008 since the first time I saw Athlon 64 bit processors based on the K8 get released. That was to be the K10 or something. But K8L seems to be like a stepping stone to K10. I am quite convinced that in all this time, AMD probably has finalized the design for K8L by now and probably even decided to toss in a few nifty features. I predict an earlier than expected K8L release to keep down the mighty Conroe.

This sort of competition is great as long as nobody ends up killing each other! ;P

Reply Score: 1

v chortle
by cutterjohn on Sun 16th Jul 2006 00:29 UTC