5.1 to stereo downmix

[pm1315]

svnpenn,

After trying your script and the one provided by Robert Collier in the Doom forum, on a couple of files,
I think yours is the one I prefer, as it actually bumps the center channel volume up -- I want to hear
the dialogue more clearly. I had to boost the overall volume a little bit on one of my files, and after
using ffmpeg's 'volumedetect' to examine the levels, I used the following script on the source:

-filter:a "pan=stereo|FL < FL + 1.414FC + .5BL + .5SL|FR < FR + 1.414FC + .5BR + .5SR, volume=1.25"

Thanks for your work.

[pandy]

Sorry for asking but what is wrong with -af aformat=sample_fmts=fltp:channel_layouts=downmix

[svnpenn]

Sorry for asking but what is wrong with -af aformat=sample_fmts=fltp:channel_layouts=downmix

Did you even read the thread

[svnpenn]

I have updated the original post with a shorter command!

[rogerdpack]

Jokes aside, have you actually tried submitting a patch to FFmpeg? That is how contributions are usually made to the project. I don't understand your negative attitude towards patch submissions.

I do submit patches, when I feel like it.

article.gmane.org/gmane.comp.version-control.git/244180

In this case I dont feel like it, so submit it yourself if you care so much.

If anybody does have a patch I'd be happy to submit it upstream for them.

[svnpenn]

If anybody does have a patch I'd be happy to submit it upstream for them.

If they would be interested in merging it I could make a patch

stackoverflow.com/a/24007019

Changes need to be made in

libswresample/swresample.c

and

libswresample/rematrix.c

[rogerdpack]

If anybody does have a patch I'd be happy to submit it upstream for them.

If they would be interested in merging it I could make a patch

stackoverflow.com/a/24007019

Changes need to be made in

libswresample/swresample.c

and

libswresample/rematrix.c

OK especially if you have some type of input audio file that demonstrates the problem [and the patch fixes it] that would help.
Thanks!
-roger-

[svnpenn]

OK especially if you have some type of input audio file that demonstrates the
problem [and the patch fixes it] that would help.

Is a GitHub pull request the preferred way to do this?

[rogerdpack]

OK especially if you have some type of input audio file that demonstrates the
problem [and the patch fixes it] that would help.

Is a GitHub pull request the preferred way to do this?

that would work, basically I need a .patch file that's it.

[svnpenn]

that would work, basically I need a .patch file that's it.

I dont understand what you are getting at here. Why do *you* need a patch?
Unless you have commit rights or want to take credit for the patch I dont see
the point.