FFplay do not respect resize alghoritm set in the sourcecode

[ami_stuff]

issue 1804

​https://roundup.libav.org/file834/test.tif

When I run FFplay like this:

ffplay -x 504 -y 376 test.tif

the picture will be displayed incorrectly (looks like chessboard).

this is not reproducible with ffmpeg:

ffmpeg -i test.tif -s 504x376 out.bmp

---

C:\>ffmpeg -i test.tif
FFmpeg version SVN-r22379, Copyright (c) 2000-2010 the FFmpeg developers

built on Mar 9 2010 06:09:18 with gcc 4.4.2
configuration: --enable-memalign-hack --cross-prefix=i686-mingw32- --cc=ccache

-i686-mingw32-gcc --arch=i686 --target-os=mingw32 --enable-runtime-cpudetect --e
nable-avisynth --enable-gpl --enable-version3 --enable-bzlib --enable-libgsm --e
nable-libfaad --enable-pthreads --enable-libvorbis --enable-libtheora --enable-l
ibspeex --enable-libmp3lame --enable-libopenjpeg --enable-libxvid --enable-libsc
hroedinger --enable-libx264 --enable-libopencore_amrwb --enable-libopencore_amrn
b

libavutil 50.11. 0 / 50.11. 0
libavcodec 52.58. 0 / 52.58. 0
libavformat 52.55. 0 / 52.55. 0
libavdevice 52. 2. 0 / 52. 2. 0
libswscale 0.10. 0 / 0.10. 0

Input #0, image2, from 'test.tif':

Duration: 00:00:00.04, start: 0.000000, bitrate: N/A

Stream #0.0: Video: tiff, monob, 1024x768, 25 tbr, 25 tbn, 25 tbc

At least one output file must be specified

[Stefano Sabatini]

Replying to ami_stuff:

issue 1804

​https://roundup.libav.org/file834/test.tif

When I run FFplay like this:

ffplay -x 504 -y 376 test.tif

the picture will be displayed incorrectly (looks like chessboard).

this is not reproducible with ffmpeg:

Hi,

the problem depends on the brittle implementation of the -x/-y and filtering interaction logic.
Workaround: use -vf scale=504:376

I suggest to remove -x and -y (and the maintainance burden of keeping two distinct logic paths) in favor of -vf scale=w:h, and make docs and code points to -vf scale.

Would you consider acceptable such a solution?

[ami_stuff]

I'm ok with this change. There is also "-s XxY" option to remove/repleace.

BTW. Maybe it would be also possible to expand scale filter a bit to set the alghoritm at the command line, so this alghoritm would repleace the dafault algo set in the program's source code:

-vf scale=504:376:lanczos

[ami_stuff]

ok, there is an "-sws_flags" option for this, but I'm unable to make it to work with ffplay:

this works:

ffmpeg -i test.tif -vf scale=504:376 -sws_flags bilinear out.bmp

this doesn't work:

ffplay -i test.tif -vf scale=504:376 -sws_flags bilinear

and it seems there is another regression, because ffmpeg generates wrong resized pic with at least bilinear/fast_bilinear rescale algo selected (g5381823 11.04.2011)

[Michael Niedermayer]

Replying to ami_stuff:

ok, there is an "-sws_flags" option for this, but I'm unable to make it to work with ffplay:

this works:

ffmpeg -i test.tif -vf scale=504:376 -sws_flags bilinear out.bmp

this doesn't work:

ffplay -i test.tif -vf scale=504:376 -sws_flags bilinear

-vf scale=504:376:flags=2 can be used as a workaround

and it seems there is another regression, because ffmpeg generates wrong resized pic with at least bilinear/fast_bilinear rescale algo selected (g5381823 11.04.2011)

ffmpeg uses mono colospace as output did you mean another issue ?

[ami_stuff]

-vf scale=504:376:flags=2 can be used as a workaround

ah, ok

ffmpeg uses mono colospace as output did you mean another issue ?

IIRC the problem was releted to ticket #52, so it's fixed now

[Michael Niedermayer]

The remaining problem with sws_flags is no longer reproduceable