| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5876 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
| |
They are not a reliable indicator that the frame is truly progressive.
Also, fix mask mode. It generated nothing but black frames.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5875 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
out the existing one.
Based on an x264 patch by Anton Mitrofanov.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5870 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5864 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
|
| |
BDs can (and occasionally do) have different audio stream types in
different clips that compose a playlist. We try to pick the audios that
appear in the most clips, but this breaks down when there are only 2
clips to choose from. So in the case of only 2 clips, choose audio
from the longest clip.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5863 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
I have a poorly mastered BD that does not pad the adaptation field
of the PCR PID properly. Since these packets are only used to extract
the PCR, we can drop them early before the code that would generate
these error messages.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5862 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
| |
Don't close a list of buffers then also close the last item in the list :-\
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5861 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
| |
CC subtitles which are embedded in the video stream have an id
of 0 assigned. We should ignore these when setting up the PID
filter list since they do not use an independent PID.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5860 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
| |
unused variables in denoise.c
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5859 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
Based on http://git.libav.org/?p=libav.git;a=commit;h=0f583e6 by Loren Merritt.
Patch by BradleyS. Thanks!
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5858 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
Based on http://git.libav.org/?p=libav.git;a=commit;h=85e228c by Loren Merritt.
Patch by BradleyS. Thanks!
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5857 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
|
| |
Filters were leaking buffers when a job is cancelled.
decavcodec could leak when job cancelled.
decavcodec leaked audio extradata in BSInfo
encavcodec and encavcodecaudio leaked AVCodecContext
sync leaked subtitle_sanitizer_t data
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5853 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
Based on a patch by Maxym Dmytrychenko. Thanks!
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5851 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
| |
libavformat doesn't always check this, so the return code from
av_interleaved_write_frame doesn't indicate an error when there
is a failure to write to disk.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5850 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For some codecs, the samplerate is not known until we decode some data.
So postpone calculation of frame duration till after we have decoded
some data.
The effect of this error is that the first 2 audio frames could have the
same start time. This causes the "non monotonically increasing dts"
error in libavformat mp4 muxer.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5849 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
Part 1 of a broader OpenCL refactoring.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5844 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
Fixes "pts < dts" and "non-monotically increasing dts" errors in avformat
muxing.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5843 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5842 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5841 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
| |
the average bitrate to be consistent (we log the RF in CRF mode).
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5840 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
Patch by BradleyS. Thanks!
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5839 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
| |
The unused portions of the opencl scaling code were not removed before commit
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5837 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5834 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5833 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5832 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
| |
unavailable.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5831 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5830 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
Unlike the maximum limit, which is HandBrake-specific, this is not our bug, and we can't fix it, short of dropping support for all old Haswell drivers.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5813 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
All user-controlled settings except GOP-related ones now have an explicit default value (it's never "unknown").
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5812 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
It doesn't play well with CQP either.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5811 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
The option seems to have no effect in my testing, and is of limited usefulness, so let's leave it unknown (up to the implementation).
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5810 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5809 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
stuff found in testing.
Thanks to gmb for pointing it out, and Nina Kurina for pointing out the ExtBRC/AsyncDepth issue.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5808 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5807 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
|
|
| |
Modeled after x264's preset system.
MMBRC and LookAhead RC are now enabled by default, whereas Trellis is disabled by default.
Please note that Windows GUI support will come later, and that preset names are not definitive.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5806 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
| |
compilation without --enable-qsv.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5805 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5804 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5803 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
This is confirmed by marketing material and testing with the 3257 driver on Ivy Bridge hardware.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5802 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
| |
This seems to work around a weird and obscure issue on some systems.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5801 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5799 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
This makes libhb expect all strings passed to it to be in utf8 format.
The cli converts the converts from the current code page to utf8. libhb
converts back to the current code page when necessary for libraries that
expect it.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5798 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5797 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5793 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Patch originally by the Multicoreware Inc team, followed by improvements and fixes by Micheal Wootton from AMD Inc,
OpenCL:
This patch implements Bicubic Scaling in OpenCL.
Note that HandBrake currently uses Lanczos so the performance difference appears to be much more significant. We may offer an option of BiCubic in software later.
Bicubic scaling may appear a bit sharper than the equivalent Lanczos encode and may increase file size a bit. Quality may be better or worse depending on the scaling and content and personal preference towards sharpness.
When comparing performance with a custom HandBrake build that runs Software Bicubic to OpenCL Bicubic, performance increase is about 5~7% on average on a modern GPU.
Hardware Decode via DXVA:
We also have optional DXVA decoding which may come in useful for slower/lower end systems that have a capable GPU.
This is only available on input sources that use the libav decode path.
Most GPU hardware for decoding is designed for playback, so if you are running on a high end CPU, it will bottleneck the encode process.
Requires OpenCL 1.1 or later supporting GPU.
Front end changes and testing framework are not included in this patch. This will be resolved later.
Patch will be revised further before the UI is implemented.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5792 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
e.g. star trek into darkness
This reverts part of a patch that was originally written for dvdread
access. So we need to watch for regressions.
The patch was originally written for dvdread access and my not actually be
a bug in dvdnav. So I have reverted it in dvdnav only. I also tried to
reproduce the original bug in dvdread, but was unsuccessful. So I'm
wondering if the original bug also got "fixed" by subsequent changes.
The original bug report is here:
https://forum.handbrake.fr/viewtopic.php?f=4&t=9758&start=0
The original fix is here:
https://trac.handbrake.fr/changeset/2256
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5787 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
| |
code to update the ones that are used. Broke out types for each struct into their own file. Switched all references to hb_handle_s to IntPtr and deleted that non-working struct.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5786 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
| |
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5784 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
|
|
|
| |
Address the easy issues first.
Apologies if I broke HandBrakeInterop.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5783 b64f7644-9d1e-0410-96f1-a4d463321fa5
|
|
|
|
|
|
|
| |
When doing codec probes, the title hasn't been set yet.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@5776 b64f7644-9d1e-0410-96f1-a4d463321fa5
|