For reasons we are not privy to, nvidia decided that the nvenc encoder
should apply aspect ratio compensation to 'DVD like' content, assuming that
the content is not BT.601 compliant, but needs to be BT.601 compliant. In
this context, that means that they make the following, questionable,
assumptions:
1) If the input dimensions are 720x480 or 720x576, assume the content has
an active area of 704x480 or 704x576.
2) Assume that whatever the input sample aspect ratio is, it does not account
for the difference between 'physical' and 'active' dimensions.
From these assumptions, they then conclude that they can 'help', by adjusting
the sample aspect ratio by a factor of 45/44. And indeed, if you wanted to
display only the 704 wide active area with the same aspect ratio as the full
720 wide image - this would be the correct adjustment factor, but what if you
don't? And more importantly, what if you're used to lavc not making this kind
of adjustment at encode time - because none of the other encoders do this!
And, what if you had already accounted for BT.601 and your input had the
correct attributes? Well, it's going to apply the compensation anyway!
So, if you take some content, and feed it through nvenc repeatedly, it
will keep scaling the aspect ratio every time, stretching your video out
more and more and more.
So, clearly, regardless of whether you want to apply bt.601 aspect ratio
adjustments or not, this is not the way to do it. With any other lavc
encoder, you would do it as part of defining your input parameters or do
the adjustment at playback time, and there's no reason by nvenc should
be any different.
This change adds some logic to undo the compensation that nvenc would
otherwise do.
nvidia engineers have told us that they will work to make this
compensation mechanism optional in a future release of the nvenc
SDK. At that point, we can adapt accordingly.
Signed-off-by: Philip Langdale <philipl@overt.org>
Reviewed-by: Timo Rothenpieler <timo@rothenpieler.org>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
As the nvEncodeApi.h header is now MIT licensed, this can be dropped.
The loaded CUDA and NVENC libraries are part of the nvidia driver, and
thus count as system libraries.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
The code needs only a few definitions from cuda.h, so define them
directly when CUDA is not enabled. CUDA is still required for accepting
HW frames as input.
Based on the code by Timo Rothenpieler <timo@rothenpieler.org>.
For some unknown reason enabling these causes proper CBR padding,
so as there are no known downsides just always enable them in CBR mode.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
Group them in a subsection of the external library section. That should
make them easier to find and understand how they fit in the scheme of
things.
Also, rewrite the description text in a similar way as in the previous
commit.
Add a more accurate description of what the switches actually do (i.e.
allow using the given library, not enabling the corresponding
codecs etc.).
Replace the library descriptions, in many cases boilerplate text without
useful information, with a short summary of what the library does.
There is no real advantage to listing some codecs or subsystems
separately simply because they are somehow "hw-accelerated", on the
contrary it makes them harder to find than in a plain alphabetically
ordered list.
Previously, we required the minimum number of bytes required for
the full box. Don't strictly require the astronomical body and additional
notes fields, but do require an altitude field (which currently isn't
parsed). This matches the initial length check at the start of the function
(which doesn't know about the variable length place field).
Signed-off-by: Martin Storsjö <martin@martin.st>
This was missed in e1eb0fc9601, when ff_interleaved_peek was
changed to include const during the evolution of the patch.
Signed-off-by: Martin Storsjö <martin@martin.st>
As long as caller only writes packets using av_interleaved_write_frame
with no manual flushing, this should allow us to always have accurate
durations at the end of fragments, since there should be at least
one queued packet in each stream (except for the stream where the
current packet is being written, but if the muxer itself does the
cutting of fragments, it also has info about the next packet for that
stream).
Signed-off-by: Martin Storsjö <martin@martin.st>
This allows callers with avio write callbacks to get the bytestream
positions that correspond to keyframes, suitable for live streaming.
In the simplest form, a caller could expect that a header is written
to the bytestream during the avformat_write_header, and the data
output to the avio context during e.g. av_write_frame corresponds
exactly to the current packet passed in.
When combined with av_interleaved_write_frame, and with muxers that
do buffering (most muxers that do some sort of fragmenting or
clustering), the mapping from input data to bytestream positions
is nontrivial.
This allows callers to get directly information about what part
of the bytestream is what, without having to resort to assumptions
about the muxer behaviour.
One keyframe/fragment/block can still be split into multiple (if
they are larger than the aviocontext buffer), which would call
the callback with e.g. AVIO_DATA_MARKER_SYNC_POINT, followed by
AVIO_DATA_MARKER_UNKNOWN for the second time it is called with
the following data.
Signed-off-by: Martin Storsjö <martin@martin.st>
Anton Khirnov
Philip Langdale
Timo Rothenpieler
Derek Buitenhuis
Diego Biurrun
Martin Storsjö
Michael Niedermayer
James Almer
Michael Niedermayer
Kyle Swanson
Paul B Mahol
Yong Lei