using x264vfw as ffmpeg?

[marcorocchini]

Hi to,

in this moment I try to use in the SpeedRazor for cats HD 3.11.095.2014-rev.15.32547854IIbeta8547859999 the x264VFW plugin to export in mpeg4



in the "extra command line" I have to add options so that encoding of x264VFW is similar to the ffmpeg commandline:

ffmpeg.exe -y -i input.avi -c:v libx264 -profile:v high422 -level:v 3 -g 33 -bf 1 -crf 18 -pix_fmt yuv422p -flags +ildct+ilme -top 1 -x264opts tff=1  -aspect 16:9 out.avi

please note that I need to export in a mpeg4 in avi, if possibile, this time. Not a .mp4 file

An I wonder: if the fields "preset tuning profile level" are not congruent with the "extra comman line" option what happens?

So I how I modify the extracommandline to match the ffmpeg commanline? thanks

[poisondeathray]

extra commandline options override the GUI

--keyint 33 --bframes 1 --tff --crf 18


(you can use CRF 18 in the drop down menu as well, instead of CQP , select CRF quality based)

SAR x:y is for Sample Aspect Ratio (you can think of it as the pixel aspect ratio , or the w:h of the pixels). -aspect 16:9 in ffmpeg is Display Aspect Ratio (DAR)

Display Aspect Ratio = Frame Aspect Ratio * Sample Aspect Ratio

e.g
for 1920x1080

16/9 = 1920/1080 * 1/1

So you would enter 1:1 (ie square pixels)

[marcorocchini]

very useful, and for the SD (720x576) case how have to configure?

[poisondeathray]

very useful, and for the SD (720x576) case how have to configure?

Depends if you're using ITU or non-ITU aspect ratio interpretation

For 16:9, mathematically correct, is non-ITU 64:45

16/9 = 720/576 * 64/45



But ITU ratio is 16:11, because they base their calculations on the inner 702 width (rounded to 704)

[marcorocchini]

mmm should the same aspect ratio of this attached mp4 transcoded with ffmpeg (using -aspect 16:9)

[poisondeathray]

ffmpeg uses non-ITU rules. Broadcast, NLE's usually use ITU rules, but these days most places don't care (unless you're submitting to BBC, they are more strict). It's just that you should be aware of slightly different AR interpretations