The Fermi Thread - Part 3

[***Deimos***]

Kind of an eye opener thats for sure, too bad TSMC doesn't have a high-k metal gate process on 40nm, 28nm should be high-k metal gate or at least be an option with TSMC later this year.

I'm wondering how much of a role leakage may be contributing with the high transistor count.

TMSC 40nm is NOT the problem. Somehow magically AMD made a DUAL-GPU card with less power and lower idle power.

Last time I checked, 5970=4BT. GTX=3BT. 4>3 isn't it?

The problem is the design. AMD put a LOT of focus on maintaining 4870's very high performance/TDP. nVidia was focused on top performance irregardless of means.

Its exactly the same way P4 fell from grace at 3.8Ghz.

(this is why Intel doesn't just hope for 1%performance/1%power, but requires it and much higher)
Now, where are all those Dual-Fermi folks? or the Fermi laptop weirdos.. I wanna see them spin their way out.

Nvidia partners prolly going nerdrage right now for this massive fail, after waiting so long for this...

Who wants to be the first to offer 5 year warranty.. anybody? What about 3 year? Any takers? EVGA? BFG? They will provide SOME warranty (and house fire insurance), right?

Radeon HD 5870’s 27W idle board rating, achieved in part by clocking its GPU down to 157 MHz and its GDDR5 memory to 300 MHz. Nvidia goes even further, dropping clocks to 50 MHz core, 67 MHz memory (270 MT/s data rate), and 100 MHz for the shaders. Nvidia doesn’t cite its idle board power, but an educated guess would still put the GeForce GTX 480 around 60W at those frequencies

idle	core	mem	shader	power
5870	157	300	157	27
480	50	67	100	60!?