[Cin] another 10 bit encoding patch

Andrew Randrianasulu randrianasulu at gmail.com
Thu Nov 14 18:38:18 CET 2024


чт, 14 нояб. 2024 г., 20:27 Terje J. Hanssen <terjejhanssen at gmail.com>:

> Thank you for the patch.
>
> A few clarification questions in between before I start:
>
> Den 14.11.2024 14:21, skrev Andrew Randrianasulu:
>
> > you need to reset your git tree,
> >
> > something like
> >
> > git reset --hard 2037bf63cdf25462ca606f4a28636e12df77c2a0
>
> I did use "git reset --hard" without number at the end for the previous
> patch
>
> What number here, should I find mine somewhere in my git tree?
>


you can fig them in 'git log' screen, you should have two more patches on
top of "ORIGIN" line

you can do

git pull -r

first, it will rebase two patches on top of what currently in git repo.

then you can use 'git log' to look up those 'commit hashes'. In theory they
should be unique, so one I already provided should work *after rebase* (git
pull -r). Without rebase commits and objects from remote repo do not exist
yet in your local repo



> >
> > before applying this patch
> >
> > then run cin like
> >
> > CIN_10BIT_ENC=p010le bin/cin
>
> No export as before before bin/cin ?
>


you can do export on separate line, but then if you want to undo its effect
you should unset this variable, or close and reopen terminal window.

I think same-line invocation is easier

>
> > or replace p010le with whatever equivalents y210le (?)
>
> So need to restart Cingg again before using next vaapi format?
>

for this patch - yes.


>
> > sorry, I just hanged my X session so can't copy/paste exact output,
> > but if you launch cingg with
> >
> > error=debug in bin/ffmpeg/encode.opts it will print list of accepted
> > by vaapi library formats at encoding attempt.
>
> Something I have seen and wondered about:
>
> Is "debug" and "verbose" the same or different ffmpeg output(s)?
>

I think they different, but how much depend on ffmpeg's code that runs at
those moments. For example if you run vaapi encode than all carefully added
logging functions in qsv encoder will be of no use ....



>
> >
> > I think this is still more like prototype to see if we can get it
> > working, I hope to wire it to existing profile reading mechanism.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cinelerra-gg.org/pipermail/cin/attachments/20241114/29e3c2df/attachment.htm>


More information about the Cin mailing list