So certain high-profile individuals were complaining that it was
difficult to configure recording settings for quality in OBS. So, I
decided to add a very easy-to-use auto-configuration for high quality
encoding -- including lossless encoding. This feature will
automatically configure ideal recording settings based upon a specified
quality level.
Recording quality presets added to simple output:
- Same as stream: Copies the encoded streaming data with no extra usage
hit.
- High quality: uses a higher CRF value (starting at 23) if using x264.
- Indistinguishable quality: uses a low CRF value (starting at 16) if
using x264.
- Lossless will spawn an FFmpeg output that uses huffyuv encoding. If a
user tries to select lossless, they will be warned both via a dialog
prompt and a warning message in the settings window to ensure they
understand that it requires tremendous amounts of free space. It will
always use the AVI file format.
Extra Notes:
- When High/Indistinguishable quality is set, it will allow you to
select the recording encoder. Currently, it just allows you to select
x264 (at either veryfast or ultrafast). Later on, it'll be useful to
be able to set up pre-configured presets for hardware encoders once
more are implemented and tested.
- I decided to allow the use of x264 at both veryfast or ultrafast
presets. The reasoning is two-fold:
1.) ultrafast is perfectly viable even for near indistinguishable
quality as long as it has the appropriate CRF value. It's nice if you
want to record but would like to or need to reduce the impact of
encoding on the CPU. It will automatically compensate for the preset at
the cost of larger file size.
2.) It was suggested to just always use ultrafast, but ultrafast
requires 2-4x as much disk space for the same CRF (most likely due to
x264 compensating for the preset). Providing veryfast is important if
you really want to reduce file size and/or reduce blocking at lower
quality levels.
- When a recording preset is used, a secondary audio encoder is also
spawned at 192 bitrate to ensure high quality audio. I chose 192
because that's the limit of the media foundation aac encoder on
windows, which I want to make sure is used if available due to its
high performance.
- The CRF calculation is based upon resolution, quality, and whether
it's set to ultrafast. First, quality sets the base CRF, 23 for
"good" quality, 16 for "very high" quality. If set to ultrafast,
it'll subtract 2 points from the CRF value to help compensate. Lower
resolutions will also lower the CRF value to help improve higher
details with a smaller pixel ratio.
Selecting any supported FFmpeg format where ff_format_desc_extensions
returns NULL would crash the std::string constructor, so we pass an
empty extension instead (rtsp is one candidate format that triggers
the crash, on my machine at least)
The user may not want their audio or their display to be captured when
creating a new scene collection. Make new scene collections default to
fully empty.
For the sake of consistency, always create a display capture source on
the very first run of the program, just to have something displayed.
(NOTE: The only exception here is on windows 7/vista, which isn't ideal
for display capture, so it'll continue to be left blank)
CBR is now always on by default for streaming, so there's no reason to
have a setting for this in particular. Still available in advanced
output settings of course, but simple output mode really should be kept
as simple as possible.
This is mostly just to remove the unnecessary clutter from the output
sections. The reconnect settings are generally rarely modified by users
as it is.
When stream delay is active, the "Start/Stop Streaming" button is
changed in to a menu button, which allows the user to select either the
option to stop the stream (which causes it to count down), or forcibly
stop the stream (which immediately stops the stream and cuts off all
delayed data).
If the user decides they want to start the stream again while in the
process of counting down, they can safely do so without having to wait
for it to stop, and it will schedule it to start up again with the same
delay after the stop.
On the status bar, it will now show whether delay is active, and its
duration. If the stream is in the process of stopping/starting, it will
count down to the stop/start.
If the option to preserve stream cutoff point on unexpected
disconnections/reconnections is enabled, it will update the current
delay duration accordingly.
I added stream delay options to advanced settings not just because I
feel it's an advanced option, but also to reduce clutter in the outputs
section and its sub-sections, which already have far too many options as
it is.
This reverts commit a508c17f0a7048a5592d91c0a6587bfb59c28e84.
I realized that this would become more of an annoyance for most people
rather than anything helpful. This has only happened only twice that I
am aware of in all the years that the program has been around.
Limits similar log entries (determined by a simple hash function that
sums the characters) to certain number of lines in a row. When a
different log entry occurs, it resets the repeat check and logs how many
times the last message was repeated.
Due to certain design changes for delay, it's better to simply determine
whether outputs are active via booleans rather than an activeRefs
variable, which could get decremented more than once if say, the signal
for stopping the stream gets called more than once for whatever reason
(which may happen in the case of delay due to the way delay works)
Replaces all the json/config loading/saving functions with safe
variants to reduce the chance of potential file corruption as much as
possible.
Also does a minor refactor of json writing by using
obs_data_save_json_safe for writing json files instead of manually using
obs_data_get_json and os_quick_write_utf8 each time.
API Changed:
---------------------------
From:
- bool obs_startup(const char *locale, profiler_name_store_t *store);
To:
- bool obs_startup(const char *locale, const char *module_config_path,
profiler_name_store_t *store);
Summary:
---------------------------
This allows plugin modules to store plugin-specific configuration data
(rather than only allowing objects to store configuration data). This
will be useful for things like caching data, for example looking up and
storing ingests from remote (rather than storing locally), or caching
font data (so it doesn't have to build a font cache each time), among
other things.
Also adds a module-specific directory for the UI
If a user was using FFmpeg output before pathc 0.12.0, they had to type
in the full file name to the FFmpeg output URL/Path box, which isn't
exactly compatible with the new settings.
This changes each profile's config file so that the FFmpeg output
detects whether files are used, and then extracts the file's directory
and extension and sets them accordingly to make it compatible with the
new FFmpeg file output handling.
This changes the way the advanced output section's FFmpeg output
settings work by allowing the user to select whether they want to output
to a file or output to a URL, and makes it so file names are
automatically generated like other recording outputs.
If they choose to output to a file, it'll only require an output
directory similarly to how other recording outputs work. They can
select a directory to output to rather than being required to type in a
full path and filename; the filename is automatically generated. The
extension is also automatically retrieved from libff depending on the
format selected.
Otherwise if they have Output to URL selected, it'll show a simple edit
box where they can type in the target URL.
Apparently using QPointer container for the QObject being given to the
connect function makes it so it can't find the actual function to use.
I'm guessing this is incomplete functionality or a bug that existed in
GCC 4.8. Doesn't happen in 4.9+.
When settings the base resolution, try to find the closest output
resolution to the old output resolution, and use that for its value. If
the aspect ratio is about the same, then don't modify the value. If the
aspect ratio is significantly different, then find the closest
approximation while keeping with the new aspect ratio.
This particular issue has been an annoyance for quite some time.
Due to all the threads in libobs it wouldn't be safe to make that
parameter reconfigurable after libobs is initialized without adding
even more synchronization. On the other hand, adding a function to set
the name store before calling obs_startup would solve the problem of
passing a name store into libobs, but it can lead to more complicated
semantics for obs_get_profiler_name_store (e.g., should it always return
the current name store even if libobs isn't initialized until someone
calls set_name_store(NULL)? should obs_shutdown call
set_name_store(NULL)? Passing it as obs_startup parameter avoids
these (and hopefully other) potential misunderstandings
Fixes a minor bug introduced by the windowless graphics context merge.
When setting a new base resolution, it would not recalculate the preview
size, and the preview would display the wrong size until the main window
was resized by the user. This patch makes it so that it calls the
recalculation function when the base resolution is changed to prevent
that from happening.
Now that we aren't dependent upon a window for our context, we can
safely move the obs context creation/destruction to obs-app.cpp, and use
the OBSContext helper class to automatically shut down obs.
(Non-compiling commit: windowless-context branch)
Now that all obs_display related code has been moved to OBSQTDisplay,
we can prune a whole bunch of boilerplate code that had to be repeated
for the displays of each window.
Affects:
- Properties
- Filters
- Interact
- Projector
(Non-compiling commit: windowless-context branch)
Gets rid of all functions/data related to setting up the main preview to
be associated with obs_reset_video and instead uses OBSQTDisplay for the
main window display (thus associating it with an obs_display object).
(Non-compiling commit: windowless-context branch)
This makes it so that OBSQTDisplay now uses/controls an obs_display
object directly (rather than having the owner have to associate an
OBSDisplay with it). It was separated before because the main window
for the basic UI would was using the "main preview" stuff before the
windowless context and had to be handled differently, so you couldn't
just associate an obs_display object with OBSQTDisplay, meaning that all
"secondary" previews such as properties/filters/etc had to handle the
obs_display alone, which caused a lot of needlessly duplicated code.
Also adds a DisplayCreated signal to allow owners to be able to add
callbacks and such on creation of the actual obs_display context.
