1.6GHz is impressive, but i would like to know how the card reacts to overclocking memory, and memory alone.
they are not constrained by their memory in most cases. I haven't really had much of a chance to play with any of the board partner cards, but the reference rx 480 cards performance is limited by the maximum core clock you can achieve.
I'm not going to dig through all the posts, it's a long weekend and I'm with the future in-laws and posting from my phone, but if you look through the GTX 1080 thread you'll see people guessing clock speeds and me saying higher or lower. I indicated north of 3 GHz there
Anyway, 13:00 and lunchtime.
Alright, so do you still stand by your "cannot validate past 850Mhz, delayed to October" comment?
That you don't know that the clock depends on the architecture and can't be compared this way....
That is right, however, the Maxwell's result is a good 40+% boost, which translates quite well to performance. I'll keep watching the reviews and OC reports, and do all kind of tests on my 970 before I trade it for an AMD card.
Still not convinced that the 480 would be a significant step forward for me, but we shall see.
Polaris 10 is a beast. But it seems you guys got overhyped and expected too much.
The ref card is priced at $199 and is already the best p/p in the recent time. Don't force the 1070 performance on that tiny board with a blower cooler.
Still as I said before the 1070 performance can be achieved by pushing the card to the uncomfortable zone. It's when the AIBs are needed and of course you have to reach deeper into your pocket (The p/p still better than Pascal though )