FX-8320 , 16 GB of ram, video in a 8 GB ramdrive.
Source is 1440x1080 h264 stream 2m 30s = 452,290,777 bytes , so about 24 mbps , using the following AVS script:
loadplugin("c:\Programs\MeGUI\tools\dgavcindex\DGAVCDecode.dll")
AVCSource("sequence.dga")
No video card acceleration, the Radeon 4850 is not supported well and it's AMD so I can't use the DGAVCIndex made for nVidia cards.
The recompression will be a bit faster with streams that have lower bitrate.
Also, the speed will be a bit higher with a better multithreaded h264 decoder. Using "DirectShowSource" instead of AVCSource to decode the h264 stream using another codec resulted in 26.8 fps for the CRF 28 test, about 3 fps more.
FX-8350 may be slightly faster but probably a few more frames. FX-8320 defaults to 3.5 Ghz and turbo boosts to 4 Ghz from, fx-8350 defaults to 4 ghz and turbo boosts to 4.2 ghz. Throughout the tests, the 8 cores stayed at about 88-93% cpu usage.
CRF 19
y:\>x264.exe --crf 19 -o seq.mp4 sequence.avs
avs [info]: 1440x1080p 0:0 @ 25/1 fps (cfr)
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX XOP FMA4 FMA3 SSEMisalign LZCNT BMI1
x264 [info]: profile High, level 4.0
aborted at input frame 1605, output frame 1550
encoded 1550 frames, 16.59 fps, 10689.29 kb/s
CRF 23
y:\>x264.exe --crf 23 -o seq.mp4 sequence.avs
avs [info]: 1440x1080p 0:0 @ 25/1 fps (cfr)
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX XOP FMA4 FMA3 SSE Misalign LZCNT BMI1
x264 [info]: profile High, level 4.0
[..]
aborted at input frame 1466, output frame 1411
encoded 1411 frames, 21.72 fps, 5333.93 kb/s
CRF 28
y:\>x264.exe --crf 28 -o seq.mp4 sequence.avs
avs [info]: 1440x1080p 0:0 @ 25/1 fps (cfr)
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX XOP FMA4 FMA3 SSE Misalign LZCNT BMI1
x264 [info]: profile High, level 4.0
[..]
aborted at input frame 1534, output frame 1479
encoded 1479 frames, 24.16 fps, 2685.54 kb/s