WTF LOL...
From http://au.gamespot.com/forums/topic/29451500/xbox-one--7790-confirmed-by-xbox-one-architec.
You can't properly add memory bandwdith.
X1 can add memory bandwidth since it has two memory pools (think of L-shaped multi-memory controller setups) while PS4 has a single memory pool.
-----
JIT compression was from vgleaks and has similar pattern when ATI added 3DC+ texture compression to Xbox 360.
LOL at your PRT statement. Using AMD PRT is not automatic you clown. LOL. Game ports from the PC doesn't support DX11.2 LOL.
GPUs must have "Move Engines" if it has to copy large blocks of data efficiently. GPU has been doing that since 1985 era Amiga's custom chipset.
Your Pitcairn vs Bonaire codename junk is just noise to actual CU design being the same across GCN. Games like BF4 or Crysis 2 ((without factoring 7770's tessellation bottleneck with NVIDIA Fermi bias patch) scales with GCN's CU count. I have shown you CU bounded games and thier results.
With mostly CU bounded games (e.g. BF4, Crysis 2), the prototype 7850 with 12 CUs would be rendering sub-1080p if the target frame rate is 60. If the target frame rate is ~30, you get results like NFS Rivials at 1080p for X1 (with slightly less shader/CU based effects) and PS4. I have shown you the results with CU count gap with BF4 and Crysis 2.
If NFS Rivials targeted 60 fps, you most likely end up like BF4's X1 (720p) and PS4 (900p) results.
Btw, KZSF multiplayer has frame rates hitting high 30s to 50 and one should see why BF4 went for 900p for PS4.
With SHAPE, your the clown who compares GFLOPS across different processor designs. You have FMOD+Turn10's PR stating they can do more with SHAPE when compared to Xbox 360 in relation to audio.
Your mind is only limited to Xbox 360/X1/Lem vs PS3/PS4/Cows.
Log in to comment