Exactly what I was looking for. Since we guestimate 40/28= 42% performance gain from direct shrink, the difference from "about 30%" underscores the expenditure on functionality tradeoff on the new architecture.
Originally Posted by SSPrncVegeta
Updated first post with more news.
Though it's still a gross estimate at this point. It will be interesting to see if the plain VLIW shrink is nearer the 42%.
Wooot 7970 is awesome!
(Oh and now I'm poor again)