65nm AMD CPUs this December

ooOOOooo shiny! :cool:

Can't wait to see the OC results for some of these...however those comments were right, Intel will be well on the way with their 45nm chips before the 65nm AMDs gain any real strength probably. It's crazy to think just how small these chips are getting though...pretty soon they will be down to like 5nm and one chip you will have like 4 individual dies on the same socket chip.
 
kill4killin said:
It's crazy to think just how small these chips are getting though...pretty soon they will be down to like 5nm and one chip you will have like 4 individual dies on the same socket chip.
Click to expand...

"system on a chip" is coming. CPU/MCH/ICH all on 1 chip. :D
 
I had thought that the 65nm chips were supposed to have come out a while ago...
 
kill4killin said:
pretty soon they will be down to like 5nm and one chip you will have like 4 individual dies on the same socket chip.
Click to expand...

This is a stretch. 5 nm is too close to having each transistor the size of a single molecule chain. They are still working on molecular transistors that consistantly work and are consistantly produceable.

When you get to the atomic level, you can't go any smaller (actually there are smaller particles when you smash atoms in a particle acclerator, but look at the size of the accelerator in the Los Almos labs and you'll see why subatomic is not going to be practicle for another century, that's even if they can figure out how to keep smashing and get consistant states for logic purposes).

This is why Mr Moore recently stated that his own theory is now dead or soon to be. http://www.techworld.com/opsys/news/index.cfm?NewsID=3477

Gordon Moore said:
In terms of size [of transistor] you can see that we're approaching the size of atoms which is a fundamental barrier, but it'll be two or three generations before we get that far - but that's as far out as we've ever been able to see. We have another 10 to 20 years before we reach a fundamental limit. By then they'll be able to make bigger chips and have transistor budgets in the billions.
Click to expand...

I think when he says generations, I think that's in terms of processor generations (K9, K10, K11 and then bam, or Core 3, Core 4, Core 5 and then bam). Instead you'll either see huge chips or they'll break them up into multiple cells that you install into a system (picture 10 CPU cells on a single MB with the chips/sockets still fiting on a ATX MB with some massive high speed serialized bus interconnect).
 
Interesting indeed. I hope they can reach 3.0-3.5 more easily :)
 
Actually, I was more interested to see that all of the chips are 2x512kb cache instead of 2x1mb in some of them. They raised the clock speeds 100mhz to compensate.
 
conroe prices...thats all i have to say ;)

will these perform better than the current am2? (not overclocked)
 
jcll2002 said:
conroe prices...thats all i have to say ;)

will these perform better than the current am2? (not overclocked)
Click to expand...

No, since they are based on the same architecture. One of these clocked at the same speed as an AM2 chip should perform nearly identical to it. The switch is designed to 1). save money because there is less silicon being used, and 2). to get higher clockspeeds.
 
You must log in or register to reply here.
Back
Top