GK104 / Tahiti die size comparison

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
There's no indication that their new architecture is less efficient per mm2 or per w than VLIW, infact, its more efficient.

So the caveat that somehow they redesigned for GPGPU means its inefficient. In terms of overall gaming load, VLIW was very inefficient, not using 25-33% of its shaders.

If the game is coded for VLIW architectures, the GCN has Less performance efficiency per transistor count (2.64B Cayman vs 2.8B Pitcairn) and operational core frequency (880MHz vs 1000MHz) than Cayman.

http://www.techpowerup.com/reviews/AMD/HD_7850_HD_7870/1.html

avp_1920_1200.gif


dragonage2_1920_1200.gif


shogun2_1920_1200.gif


Also in some older DX-9 games like COD

cod4_1920_1200.gif
 

Homeles

Platinum Member
Dec 9, 2011
2,580
0
0
To be fair, it's a bit difficult to claim an architecture is less efficient when the drivers are (likely) not mature yet.
 
Feb 19, 2009
10,457
10
76
If the game is coded for VLIW architectures, the GCN has Less performance efficiency per transistor count (2.64B Cayman vs 2.8B Pitcairn) and operational core frequency (880MHz vs 1000MHz) than Cayman.

That's true, VLIW can be very good if the games are specifically coded for it, otherwise its very inefficient and performance is bad.

This is why GCN cards are the future since they do well in every game like NV's architecture, no need for devs to try harder to optimize their engines.