[x265] [PATCH x265] Fix warnings and Win32 build error for ssimDistortion primitive

# HG changeset patch # User Akil Ayyappan # Date 1551966566 -19800 # Thu Mar 07 19:19:26 2019 +0530 # Node ID bf2f1880f5ba6f264cdff9711f548801edd1960b # Parent 0eccd62725b6a24ae27d52189c4a624dffdd7a07 Fix warnings and Win32 build errors for ssimDistortion primitive diff -r 0eccd62725b6 -r

[x265] [PATCH] Backed out changeset: fef63866bb60

# HG changeset patch # User Pradeep Ramachandran # Date 1552022473 -19800 # Fri Mar 08 10:51:13 2019 +0530 # Node ID 636258ebc7a90e0a35466e9b605ab335b9ce2194 # Parent 0eccd62725b6a24ae27d52189c4a624dffdd7a07 Backed out changeset: fef63866bb60 diff -r 0eccd62725b6 -r 636258ebc7a9

Re: [x265] Less FPS

Thanks Alex! Can you share your command that you are using? On Thu, Feb 28, 2019 at 8:06 PM Alex Giladi wrote: > Sunil, all, > From our experiments a few months ago, gains from AVX-512 were noticeable > (>5%) at veryslow preset and 4K resolution. > Best, > Alex. > > On Thu, Feb 28, 2019, 06:06

[x265] (NASM for Win32) pixel-a.asm: error: symbol `r7d' undefined

E:\MABS\msys64\mingw32\bin\nasm.exe -f win32 -DPREFIX -DARCH_X86_64=0 -I E:/MABS/build/x265-hg/source/common/../common/x86/ -DHAVE_ALIGNED_STACK=1 -DHIGH_BIT_DEPTH=0 -DBIT_DEPTH=8 -DX265_NS=x265 -o common/CMakeFiles/common.dir/x86/pixel-a.asm.obj

Re: [x265] [PATCH x265] Add AVX2 assembly code for normFactor primitive.

Hi Chen, Thanks for the feedback. Will do the possible changes. On Thu, Mar 7, 2019 at 4:37 PM Niranjankumar Balasubramanian < niran...@multicorewareinc.com> wrote: > Hi Chen, > Thanks for your suggestions. Your feedback is noted. > > On Thu, Mar 7, 2019 at 3:41 PM chen wrote: > >> Just say it

Re: [x265] [PATCH x265] Add AVX2 assembly code for normFactor primitive.

Hi Chen, Thanks for your suggestions. Your feedback is noted. On Thu, Mar 7, 2019 at 3:41 PM chen wrote: > Just say it works. > > First at all, > The expect algorithm is square of (x >> shift) > It is 8 bits (I assume we talk with 8bpp, the 16bpp are similar) multiple > of 8-bits and result is

Re: [x265] [PATCH x265] Add AVX2 assembly code for normFactor primitive.

Just say it works. First at all, The expect algorithm is square of (x >> shift) It is 8 bits (I assume we talk with 8bpp, the 16bpp are similar) multiple of 8-bits and result is 16 bits. The function works on CU-level, the blockSize is up to 64 only, or call 6-bits. So, we can decide the

Re: [x265] [PATCH x265] Add AVX2 assembly code for ssimDistortion primitive.

# HG changeset patch # User Akil Ayyappan # Date 1551251102 -19800 # Wed Feb 27 12:35:02 2019 +0530 # Node ID d12a4caf7963fd47d646040689ad5f02754ad879 # Parent cb3e172a5f51c6a4bf8adb7953fe53277f5a1979 x86: ssimDistortion primitive This patch adds AVX2 assembly for this primitive. Pushed

Re: [x265] [PATCH x265] Add AVX2 assembly code for normFactor primitive.

# HG changeset patch # User Akil Ayyappan # Date 1551693998 -19800 # Mon Mar 04 15:36:38 2019 +0530 # Node ID 19f27e0c8a6f8250af5e2f7c7984dc3b57bea7e4 # Parent d12a4caf7963fd47d646040689ad5f02754ad879 x86: normFactor primitive This patch adds AVX2 assembly for this primitive. Pushed to