A little bit of history about frame dropping:
I did a large number of experiments with frame dropping in old versions
of OBS1, and it's not an easy thing to deal with. I tried just about
everything from standard i-frame delay, to large buffers, to dumping
packets, to super-unnecessarily-complex things that just ended up
causing more problems than they was worth.
When I did my experiments, I found that the most ideal frame drop system
(in terms of reducing the amount of total data that needed to be
dropped) was in the 0.4xx days where I had a 3 second frame-drop buffer
where I could calculate the actual buffer size in bytes, and then
intellgently choose packets in that buffer to trim it down to a specific
size while minimizing the number of p-frames and i-frames dropped, and
preventing the actual impact of dropped frames on the stream. The
downside of it was that it required too much extra latency, and far too
many people complained about it, so it was removed in favor of the
current system.
The current system I just refer to just as 'packet dumping', which when
combined with low keyframe intervals (like most services use these
days), is the next-best method from my experience. Just dump the buffer
when you reach a threshold of buffering (which I prefer to measure with
time rather than in size), then wait for a new i-frame. Simple,
effective, and reduces the risk of consecutive buffering, while still
having fairly low impact on the stream output due to the low keyframe
interval of services.
By the way, audio will not (and should not ever) be dropped, lest you
end up with syncing issues (among other nasty things) specific to server
implementation.
- Fix an issue that could occur when using more than one video encoder.
Audio/video would not sync up correctly because they were expected to
be paired with a particular encoder. This simply adds a little
helper variable to encoder packets that specifies the system time in
microseconds. We then use that system time to sync
- Fix an issue with x264 with fractional FPS rates (29.97 and 59.94
particularly) where it would create ridiculously large stream
outputs. The problem was that you shouldn't set the timebase_*
variables in the x264 params manually, let x264 handle the default
values for it and leave them at 0.
- Make x264 use CFR output, because there's no reason to ever use VFR
in this case.
- Implement the RTMP output module. This time around, we just use a
simple FLV muxer, then just write to the stream with RTMP_Write.
Easy and effective.
- Fix the FLV muxer, the muxer now outputs proper FLV packets.
- Output API:
* When using encoders, automatically interleave encoded packets
before sending it to the output.
* Pair encoders and have them automatically wait for the other to
start to ensure sync.
* Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop'
because it was a bit confusing, and doing this makes a lot more
sense for outputs that need to stop suddenly (disconnections/etc).
- Encoder API:
* Remove some unnecessary encoder functions from the actual API and
make them internal. Most of the encoder functions are handled
automatically by outputs anyway, so there's no real need to expose
them and end up inadvertently confusing plugin writers.
* Have audio encoders wait for the video encoder to get a frame, then
start at the exact data point that the first video frame starts to
ensure the most accrate sync of video/audio possible.
* Add a required 'frame_size' callback for audio encoders that
returns the expected number of frames desired to encode with. This
way, the libobs encoder API can handle the circular buffering
internally automatically for the encoder modules, so encoder
writers don't have to do it themselves.
- Fix a few bugs in the serializer interface. It was passing the wrong
variable for the data in a few cases.
- If a source has video, make obs_source_update defer the actual update
callback until the tick function is called to prevent threading
issues.
Now that we have the priorties window in we can finally be able to
select windows for capture source such as window capture. Only took
about an hour or two to write.
Also, fixed some depednency issues on winmm.lib with obs-outputs
Improve the properties API so that it can actually respond somewhat to
user input. Maybe later this might be further improved or replaced with
something script-based.
When creating a property, you can now add a callback to that property
that notifies when the property has been changed in the user interface.
Return true if you want the properties to be refreshed, or false if not.
Though now that I think about it I doubt there would ever be a case
where you would have this callback and *not* refresh the properties.
Regardless, this allows functions to change the values of properties or
settings, or enable/disable/hide other property controls from view
dynamically.
- Add start/stop code to obs-output module
- Use a circular buffer for the buffered encoder packets instead of a
dynamic array
- Add pthreads.lib as a dependency to obs-output module on windows in
visual studio project files
- Fix an windows export bug for avc parsing functions on windows.
Also, rename those functions to be more consistent with each other.
- Make outputs use a single function for encoded data rather than
multiple functions
- Add the ability to make 'text' properties be passworded
- obs-outputs module: Add preliminary code to send out data, and add
an FLV muxer. This time we don't really need to build the packets
ourselves, we can just use the FLV muxer and send it directly to
RTMP_Write and it should automatically parse the entire stream for us
without us having to do much manual code at all. We'll see how it
goes.
- libobs: Add AVC NAL packet parsing code
- libobs/media-io: Add quick helper functions for audio/video to get
the width/height/fps/samplerate/etc rather than having to query the
info structures each time.
- libobs (obs-output.c): Change 'connect' signal to 'start' and 'stop'
signals. 'start' now specifies an error code rather than whether it
simply failed, that way the client can actually know *why* a failure
occurred. Added those error codes to obs-defs.h.
- libobs: Add a few functions to duplicate/free encoder packets
Note that this is a somewhat heavily modified custom version of librtmp.
I modified all the platform specific code that we were using for the
OBS1 to make it platform-independent.
I don't really like the code in this library, but it works well enough,
so I can't really fault anyone for it. It's just very.. unclean. Even
for a C library, quite unclean. Some parts are also a little less safe
than I'd prefer as well.
- Implement OBS encoder interface. It was previously incomplete, but
now is reaching some level of completion, though probably should
still be considered preliminary.
I had originally implemented it so that encoders only have a 'reset'
function to reset their parameters, but I felt that having both a
'start' and 'stop' function would be useful.
Encoders are now assigned to a specific video/audio media output each
rather than implicitely assigned to the main obs video/audio
contexts. This allows separate encoder contexts that aren't
necessarily assigned to the main video/audio context (which is useful
for things such as recording specific sources). Will probably have
to do this for regular obs outputs as well.
When creating an encoder, you must now explicitely state whether that
encoder is an audio or video encoder.
Audio and video can optionally be automatically converted depending
on what the encoder specifies.
When something 'attaches' to an encoder, the first attachment starts
the encoder, and the encoder automatically attaches to the media
output context associated with it. Subsequent attachments won't have
the same effect, they will just start receiving the same encoder data
when the next keyframe plays (along with SEI if any). When detaching
from the encoder, the last detachment will fully stop the encoder and
detach the encoder from the media output context associated with the
encoder.
SEI must actually be exported separately; because new encoder
attachments may not always be at the beginning of the stream, the
first keyframe they get must have that SEI data in it. If the
encoder has SEI data, it needs only add one small function to simply
query that SEI data, and then that data will be handled automatically
by libobs for all subsequent encoder attachments.
- Implement x264 encoder plugin, move x264 files to separate plugin to
separate necessary dependencies.
- Change video/audio frame output structures to not use const
qualifiers to prevent issues with non-const function usage elsewhere.
This was an issue when writing the x264 encoder, as the x264 encoder
expects non-const frame data.
Change stagesurf_map to return a non-const data type to prevent this
as well.
- Change full range parameter of video scaler to be an enum rather than
boolean
Add a fairly easy to use settings interface that can be passed to
plugins, and replaced the old character string system that was being
used before. The new data interface allows for an easier method of
getting/altering settings for plugins, and is built to be serializable
to/from JSON.
Also, removed another wxFormBuilder file that was no longer in use.
- Added some code for FFmpeg output that I'm still playing around with.
Right now I'm just trying to get it to output to file and try to
understand the FFmpeg/libav APIs. Hopefully in the future this plugin
can be used for any sort of output to FFmpeg.
- Fixed a cast warning in audio-io.c with size_t -> uint32_t
- Renamed the 'video_info' and 'audio_info' structures to
'video_conver_info' and 'audio_convert_info' to better represent their
actual purpose, and to avoid confusion with 'audio_output_info' and
'video_output_info' structures.
- Removed a few macros from obs-def.h that were at one point going to be
used but no longer going to be used (at least for now)
- First, I redid the output interface for libobs. I feel like it's
going in a pretty good direction in terms of design.
Right now, the design is so that outputs and encoders are separate.
One or more outputs can connect to a specific encoder to receive its
data, or the output can connect directly to raw data from libobs
output itself, if the output doesn't want to use a designated encoder.
Data is received via callbacks set when you connect to the encoder or
raw output. Multiple outputs can receive the data from a single
encoder context if need be (such as for streaming to multiple channels
at once, and/or recording with the same data).
When an encoder is first connected to, it will connect to raw output,
and start encoding. Additional connections will receive that same
data being encoded as well after that. When the last encoder has
disconnected, it will stop encoding. If for some reason the encoder
needs to stop, it will use the callback with NULL to signal that
encoding has stopped. Some of these things may be subject to change
in the future, though it feels pretty good with this design so far.
Will have to see how well it works out in practice versus theory.
- Second, Started adding preliminary RTMP/x264 output plugin code.
To speed things up, I might just make a direct raw->FFmpeg output to
create a quick output plugin that we can start using for testing all
the subsystems.
Completely revamped the entire media i/o data and handlers. The
original idea was to have a system that would have connecting media
inputs and outputs, but at a certain point I realized that this was an
unnecessary complexity for what we wanted to do. (Also, it reminded me
of directshow filters, and I HATE directshow with a passion, and
wouldn't wish it upon my greatest enemy)
Now, audio/video outputs are connected to directly, with better callback
handlers, and will eventually have the ability to automatically handle
conversions such as 4:4:4 to 4:2:0 when connecting to an input that uses
them. Doing this will allow the video/audio i/o handlers to also
prevent duplicate conversion, as well as make it easier/simple to use.
My true goal for this is to make output and encoder plugins as simple to
create as possible. I want to be able to be able to create an output
plugin with almost no real hassle of having to worry about image
conversions, media inputs/outputs, etc. A plugin developer shouldn't
have to handle that sort of stuff when he/she doesn't really need to.
Plugins will be able to simply create a callback via obs_video() and/or
obs_audio(), and they will automatically receive the audio/video data in
the formats requested via a simple callback, without needing to do
almost anything else at all.
jp9000
Palana
BtbN