Railven, just when you start to sound reasonable you revert to lower attacks than the people you are arguing with. "Only AMD people" "AMD users think".
"GTX 780 is still a viable card". Relative to what? Relative to the competition? Outside of the mining inflation period, Nvidia marketed and sold the 780 as a superior product to the 290(non X). I know you have read the threads here so there's no need to remind you of the relative drop in performance.
I think AtenRa is on the wrong path though. Again, at max OC vs max OC how can the Fury X possibly surpass the 980 Ti except in a few outlier games? The 980 Ti will gain 25% today when both are at max OC. What evidence should we use to show a 25% AMD performance gain? Kepler has dropped about one performance tier, around 15% or so, compared to GCN in 2012-2013. Unless you think architecturally that Maxwell will age much and noticeably worse than Kepler did, and somehow make up for the VRAM gap?
In most cases where GCN outlived Kepler and has even made gains on Maxwell it is when the AMD card has a better memory setup and equal or more ROPs. 290 series has these advantages over the 780/970, for example. It should have been common foresight 1-2 years ago that the 290 series would age better relative to these cards, even if the exact amount was a mystery. The 980 Ti has 50% more ROPs than the Fury X. It also has 50% more memory. It also typically gets 30%+ higher clocks than a Fury X at max OC. It would take outrageous IPC improvements on AMD's end.
It's not going to happen. Please, if you want AMD market share to rise this is not the argument to make. The mass market who are buying 960s-970s and intending to keep them for 2-3 years are the ones you need to convince would be better met with a 380-390. It is there that we see repeats of 7970 vs 680 and 290 vs 780 potentially reliving.
Facebook
Twitter