mpv/DOCS/man/en/vo.rst

725 lines
28 KiB
ReStructuredText

.. _video_outputs:
VIDEO OUTPUT DRIVERS
====================
Video output drivers are interfaces to different video output facilities. The
syntax is:
--vo=<driver1[:suboption1[=value]:...],driver2,...[,]>
Specify a priority list of video output drivers to be used.
If the list has a trailing ',' mpv will fall back on drivers not contained
in the list. Suboptions are optional and can mostly be omitted.
*NOTE*: See ``--vo=help`` for a list of compiled-in video output drivers.
*EXAMPLE*:
``--vo=opengl,xv,``
Try the OpenGL driver, then the Xv driver, then others.
Available video output drivers are:
xv (X11 only)
Uses the XVideo extension to enable hardware accelerated playback. This is
the most compatible VO on X, but may be low quality, and has issues with
OSD and subtitle display.
For information about what colorkey is used and how it is drawn run
mpv with ``-v`` option and look out for the lines tagged with ``[xv
common]`` at the beginning.
adaptor=<number>
Select a specific XVideo adaptor (check xvinfo results).
port=<number>
Select a specific XVideo port.
ck=<cur|use|set>
Select the source from which the colorkey is taken (default: cur).
cur
The default takes the colorkey currently set in Xv.
use
Use but do not set the colorkey from mpv (use the ``--colorkey``
option to change it).
set
Same as use but also sets the supplied colorkey.
ck-method=<man|bg|auto>
Sets the colorkey drawing method (default: man).
man
Draw the colorkey manually (reduces flicker in some cases).
bg
Set the colorkey as window background.
auto
Let Xv draw the colorkey.
x11 (X11 only)
Shared memory video output driver without hardware acceleration that works
whenever X11 is present.
*NOTE*: this is a fallback only, and shouldn't be normally used.
vdpau (X11 only)
Uses the VDPAU interface to display and optionally also decode video.
Hardware decoding is used with ``--hwdec=vdpau``.
sharpen=<-1-1>
For positive values, apply a sharpening algorithm to the video, for
negative values a blurring algorithm (default: 0).
denoise=<0-1>
Apply a noise reduction algorithm to the video (default: 0, no noise
reduction).
deint=<-4-4>
Select deinterlacing mode (default: -3). Positive values choose mode
and enable deinterlacing. Corresponding negative values select the
same deinterlacing mode, but do not enable deinterlacing on startup
(useful in configuration files to specify what mode will be enabled by
the "D" key). All modes respect ``--field-dominance``.
0
same as -3
1
Show only first field, similar to ``--vf=field``.
2
Bob deinterlacing, similar to ``--vf=tfields=1``.
3
motion adaptive temporal deinterlacing. May lead to A/V desync
with slow video hardware and/or high resolution.
4
motion adaptive temporal deinterlacing with edge-guided spatial
interpolation. Needs fast video hardware.
chroma-deint
Makes temporal deinterlacers operate both on luma and chroma (default).
Use no-chroma-deint to solely use luma and speed up advanced
deinterlacing. Useful with slow video memory.
pullup
Try to apply inverse telecine, needs motion adaptive temporal
deinterlacing.
hqscaling=<0-9>
0
Use default VDPAU scaling (default).
1-9
Apply high quality VDPAU scaling (needs capable hardware).
fps=<number>
Override autodetected display refresh rate value (the value is needed
for framedrop to allow video playback rates higher than display
refresh rate, and for vsync-aware frame timing adjustments). Default 0
means use autodetected value. A positive value is interpreted as a
refresh rate in Hz and overrides the autodetected value. A negative
value disables all timing adjustment and framedrop logic.
composite-detect
NVIDIA's current VDPAU implementation behaves somewhat differently
under a compositing window manager and does not give accurate frame
timing information. With this option enabled, the player tries to
detect whether a compositing window manager is active. If one is
detected, the player disables timing adjustments as if the user had
specified fps=-1 (as they would be based on incorrect input). This
means timing is somewhat less accurate than without compositing, but
with the composited mode behavior of the NVIDIA driver there is no
hard playback speed limit even without the disabled logic. Enabled by
default, use no-composite-detect to disable.
queuetime_windowed=<number> and queuetime_fs=<number>
Use VDPAU's presentation queue functionality to queue future video
frame changes at most this many milliseconds in advance (default: 50).
See below for additional information.
output_surfaces=<2-15>
Allocate this many output surfaces to display video frames (default:
3). See below for additional information.
Using the VDPAU frame queueing functionality controlled by the queuetime
options makes mpv's frame flip timing less sensitive to system CPU
load and allows mpv to start decoding the next frame(s) slightly
earlier which can reduce jitter caused by individual slow-to-decode
frames. However the NVIDIA graphics drivers can make other window behavior
such as window moves choppy if VDPAU is using the blit queue (mainly
happens if you have the composite extension enabled) and this feature is
active. If this happens on your system and it bothers you then you can set
the queuetime value to 0 to disable this feature. The settings to use in
windowed and fullscreen mode are separate because there should be less
reason to disable this for fullscreen mode (as the driver issue shouldn't
affect the video itself).
You can queue more frames ahead by increasing the queuetime values and the
output_surfaces count (to ensure enough surfaces to buffer video for a
certain time ahead you need at least as many surfaces as the video has
frames during that time, plus two). This could help make video smoother in
some cases. The main downsides are increased video RAM requirements for
the surfaces and laggier display response to user commands (display
changes only become visible some time after they're queued). The graphics
driver implementation may also have limits on the length of maximum
queuing time or number of queued surfaces that work well or at all.
direct3d_shaders (Windows only)
Video output driver that uses the Direct3D interface.
prefer-stretchrect
Use IDirect3DDevice9::StretchRect over other methods if possible.
disable-stretchrect
Never render the video using IDirect3DDevice9::StretchRect.
disable-textures
Never render the video using D3D texture rendering. (Rendering with
textures + shader will still be allowed. Add disable-shaders to
completely disable video rendering with textures.)
disable-shaders
Never use shaders when rendering video.
only-8bit
Never render YUV video with more than 8 bits per component.
(Using this flag will force software conversion to 8 bit.)
disable-texture-align
Normally texture sizes are always aligned to 16. With this option
enabled, the video texture will always have exactly the same size as
the video itself.
Debug options. These might be incorrect, might be removed in the future, might
crash, might cause slow downs, etc. Contact the developers if you actually need
any of these for performance or proper operation.
force-power-of-2
Always force textures to power of 2, even if the device reports
non-power-of-2 texture sizes as supported.
texture-memory=N
Only affects operation with shaders/texturing enabled, and (E)OSD.
Values for N:
0
default, will often use an additional shadow texture + copy
1
use D3DPOOL_MANAGED
2
use D3DPOOL_DEFAULT
3
use D3DPOOL_SYSTEMMEM, but without shadow texture
swap-discard
Use D3DSWAPEFFECT_DISCARD, which might be faster.
Might be slower too, as it must (?) clear every frame.
exact-backbuffer
Always resize the backbuffer to window size.
direct3d (Windows only)
Same as ``direct3d_shaders``, but with the options ``disable-textures``
and ``disable-shaders`` forced.
corevideo (Mac OS X 10.6 and later)
Mac OS X CoreVideo video output driver. Uses the CoreVideo APIs to fill
PixelBuffers and generate OpenGL textures from them (useful as a fallback
for vo_opengl_).
.. _vo_opengl:
opengl
OpenGL video output driver. It supports extended scaling methods, dithering
and color management.
By default, it tries to use fast and fail-safe settings. Use the alias
``opengl-hq`` to use this driver with defaults set to high quality
rendering.
Requires at least OpenGL 2.1 and the GL_ARB_texture_rg extension. For older
drivers, ``opengl-old`` may work.
Some features are available with OpenGL 3 capable graphics drivers only
(or if the necessary extensions are available).
lscale=<filter>
Set the scaling filter. Possible choices:
bilinear
bicubic_fast
sharpen3
sharpen5
hanning
hamming
hermite
quadric
bicubic
kaiser
catmull_rom
mitchell
spline16
spline36
gaussian
sinc2
sinc3
sinc4
lanczos2
lanczos3
lanczos4
blackman2
blackman3
blackman4
bilinear
Bilinear hardware texture filtering (fastest, mid-quality).
This is the default.
lanczos2
Lanczos scaling with radius=2. Provides good quality and speed.
This is the default when using ``opengl-hq``.
lanczos3
Lanczos with radius=3.
bicubic_fast
Bicubic filter. Has a blurring effect on the image, even if no
scaling is done.
sharpen3
Unsharp masking (sharpening) with radius=3 and a default strength
of 0.5 (see ``lparam1``).
sharpen5
Unsharp masking (sharpening) with radius=5 and a default strength
of 0.5 (see ``lparam1``).
mitchell
Mitchell-Netravali. The ``b`` and ``c`` parameters can be set with
``lparam1`` and ``lparam2``. Both are set to 1/3 by default.
lparam1=<value>
Set filter parameters. Ignored if the filter is not tunable. These are
unset by default, and use the filter specific default if applicable.
lparam2=<value>
See ``lparam1``.
scaler-resizes-only
Disable the scaler if the video image is not resized. In that case,
``bilinear`` is used instead whatever is set with ``lscale``. Bilinear
will reproduce the source image perfectly if no scaling is performed.
Note that this option never affects ``cscale``, although the different
processing chain might do chroma scaling differently if ``lscale`` is
disabled.
stereo=<value>
Select a method for stereo display. You may have to use ``--aspect`` to
fix the aspect value. Experimental, do not expect too much from it.
0
Normal 2D display
1
Convert side by side input to full-color red-cyan stereo.
2
Convert side by side input to full-color green-magenta stereo.
3
Convert side by side input to quadbuffered stereo. Only supported
by very few OpenGL cards.
srgb
Enable gamma-correct scaling by working in linear light. This
makes use of sRGB textures and framebuffers.
This option forces the options 'indirect' and 'gamma'.
NOTE: for YUV colorspaces, gamma 1/0.45 (2.222) is assumed. RGB input
is always assumed to be in sRGB.
This option is not really useful, as gamma-correct scaling has not much
influence on typical video playback. Most visible effect comes from
slightly different gamma.
pbo
Enable use of PBOs. This is faster, but can sometimes lead to
sporadic and temporary image corruption.
dither-depth=<N|no|auto>
Set dither target depth to N. Default: no.
no
Disable any dithering done by mpv.
auto
Automatic selection. If output bit depth can't be detected,
8 bits per component are assumed.
8
Dither to 8 bit output.
Note that the depth of the connected video display device can not be
detected. Often, LCD panels will do dithering on their own, which
conflicts with vo_opengl's dithering, and leads to ugly output.
dither-size-fruit=<2-8>
Set the size of the dither matrix (default: 6). The actual size of
the matrix is ``(2^N) x (2^N)`` for an option value of ``N``, so a
value of 6 gives a size of 64x64. The matrix is generated at startup
time, and a large matrix can take rather long to compute (seconds).
Used in ``dither=fruit`` mode only.
dither=<fruit|ordered|no>
Select dithering algorithm (default: fruit).
temporal-dither
Enable temporal dithering. (Only active if dithering is enabled in
general.) This changes between 8 different dithering pattern on each
frame by changing the orientation of the tiled dithering matrix.
Unfortunately, this can lead to flicker on LCD displays, since these
have a high reaction time.
debug
Check for OpenGL errors, i.e. call glGetError(). Also request a
debug OpenGL context (which does nothing with current graphics drivers
as of this writing).
swapinterval=<n>
Interval in displayed frames between two buffer swaps.
1 is equivalent to enable VSYNC, 0 to disable VSYNC.
no-scale-sep
When using a separable scale filter for luma, usually two filter
passes are done. This is often faster. However, it forces
conversion to RGB in an extra pass, so it can actually be slower
if used with fast filters on small screen resolutions. Using
this options will make rendering a single operation.
Note that chroma scalers are always done as 1-pass filters.
cscale=<n>
As lscale but for chroma (2x slower with little visible effect).
Note that with some scaling filters, upscaling is always done in
RGB. If chroma is not subsampled, this option is ignored, and the
luma scaler is used instead. Setting this option is often useless.
fancy-downscaling
When using convolution based filters, extend the filter size
when downscaling. Trades quality for reduced downscaling performance.
no-npot
Force use of power-of-2 texture sizes. For debugging only.
Borders will be distorted due to filtering.
glfinish
Call glFinish() before swapping buffers
sw
Continue even if a software renderer is detected.
backend=<sys>
auto
auto-select (default)
cocoa
Cocoa/OSX
win
Win32/WGL
x11
X11/GLX
wayland
Wayland/EGL
indirect
Do YUV conversion and scaling as separate passes. This will
first render the video into a video-sized RGB texture, and
draw the result on screen. The luma scaler is used to scale
the RGB image when rendering to screen. The chroma scaler
is used only on YUV conversion, and only if the video is
chroma-subsampled (usually the case).
This mechanism is disabled on RGB input.
Specifying this option directly is generally useful for debugging only.
fbo-format=<fmt>
Selects the internal format of textures used for FBOs. The format can
influence performance and quality of the video output. (FBOs are not
always used, and typically only when using extended scalers.)
fmt can be one of: rgb, rgba, rgb8, rgb10, rgb16, rgb16f, rgb32f,
rgba12, rgba16, rgba16f, rgba32f.
Default: rgb.
gamma
Always enable gamma control. (Disables delayed enabling.)
icc-profile=<file>
Load an ICC profile and use it to transform linear RGB to
screen output. Needs LittleCMS2 support compiled in.
icc-cache=<file>
Store and load the 3D LUT created from the ICC profile in
this file. This can be used to speed up loading, since
LittleCMS2 can take a while to create the 3D LUT.
Note that this file contains an uncompressed LUT. Its size depends on
the ``3dlut-size``, and can be very big.
icc-intent=<value>
0
perceptual
1
relative colorimetric
2
saturation
3
absolute colorimetric (default)
3dlut-size=<r>x<g>x<b>
Size of the 3D LUT generated from the ICC profile in each
dimension. Default is 128x256x64.
Sizes must be a power of two, and 256 at most.
alpha
Try to create a framebuffer with alpha component. This only makes sense
if the video contains alpha information (which is extremely rare). May
not be supported on all platforms. If alpha framebuffers are
unavailable, it silently falls back to a normal framebuffer. Note
that when using FBO indirections (such as with ``opengl-hq``), a FBO
format with alpha must be specified with the ``fbo-format`` option.
opengl-hq
Same as ``opengl``, but with default settings for high quality rendering.
This is equivalent to:
| --vo=opengl:lscale=lanczos2:dither-depth=auto:pbo:fbo-format=rgb16
Note that some cheaper LCDs do dithering that gravely interferes with
vo_opengl's dithering. Disabling dithering with ``dither-depth=no`` helps.
Unlike ``opengl``, ``opengl-hq`` makes use of FBOs by default. Sometimes you
can achieve better quality or performance by changing the fbo-format
sub-option to ``rgb16f``, ``rgb32f`` or ``rgb``. (Known problems include
Mesa/Intel not accepting ``rgb16``, Mesa sometimes not being compiled with
float texture support, and some OSX setups being very slow with ``rgb16``,
but fast with ``rgb32f``.)
opengl-old
OpenGL video output driver, old version. Video size must be smaller
than the maximum texture size of your OpenGL implementation. Intended to
work even with the most basic OpenGL implementations, but also makes use
of newer extensions, which allow support for more colorspaces.
The code performs very few checks, so if a feature does not work, this
might be because it is not supported by your card/OpenGL implementation
even if you do not get any error message. Use ``glxinfo`` or a similar
tool to display the supported OpenGL extensions.
(no-)ati-hack
ATI drivers may give a corrupted image when PBOs are used (when using
`force-pbo`). This option fixes this, at the expense of
using a bit more memory.
(no-)force-pbo
Always uses PBOs to transfer textures even if this involves an extra
copy. Currently this gives a little extra speed with NVidia drivers
and a lot more speed with ATI drivers. May need
the ati-hack suboption to work correctly.
(no-)scaled-osd
Changes the way the OSD behaves when the size of the window changes
(default: disabled). When enabled behaves more like the other video
output drivers, which is better for fixed-size fonts. Disabled looks
much better with FreeType fonts and uses the borders in fullscreen
mode. Does not work correctly with ass subtitles (see ``--ass``), you
can instead render them without OpenGL support via ``--vf=ass``.
rectangle=<0,1,2>
Select usage of rectangular textures which saves video RAM, but often
is slower (default: 0).
0
Use power-of-two textures (default).
1
Use the ``GL_ARB_texture_rectangle`` extension.
2
Use the ``GL_ARB_texture_non_power_of_two`` extension. In some
cases only supported in software and thus very slow.
swapinterval=<n>
Minimum interval between two buffer swaps, counted in displayed frames
(default: 1). 1 is equivalent to enabling VSYNC, 0 to disabling VSYNC.
Values below 0 will leave it at the system default. This limits the
framerate to (horizontal refresh rate / n). Requires
``GLX_SGI_swap_control`` support to work. With some (most/all?)
implementations this only works in fullscreen mode.
ycbcr
Use the ``GL_MESA_ycbcr_texture`` extension to convert YUV to RGB. In
most cases this is probably slower than doing software conversion to
RGB.
yuv=<n>
Select the type of YUV to RGB conversion. The default is
auto-detection deciding between values 0 and 2.
0
Use software conversion. Compatible with all OpenGL versions.
Provides brightness, contrast and saturation control.
1
Same as 2. This used to use nVidia-specific extensions, which
didn't provide any advantages over using fragment programs, except
possibly on very ancient graphic cards. It produced a gray-ish
output, which is why it has been removed.
2
Use a fragment program. Needs the ``GL_ARB_fragment_program``
extension and at least three texture units. Provides brightness,
contrast, saturation and hue control.
3
Use a fragment program using the POW instruction. Needs the
``GL_ARB_fragment_program`` extension and at least three texture
units. Provides brightness, contrast, saturation, hue and gamma
control. Gamma can also be set independently for red, green and
blue. Method 4 is usually faster.
4
Use a fragment program with additional lookup. Needs the
``GL_ARB_fragment_program`` extension and at least four texture
units. Provides brightness, contrast, saturation, hue and gamma
control. Gamma can also be set independently for red, green and
blue.
5
Use ATI-specific method (for older cards). This uses an
ATI-specific extension (``GL_ATI_fragment_shader`` - not
``GL_ARB_fragment_shader``!). At least three texture units are
needed. Provides saturation and hue control. This method is fast
but inexact.
6
Use a 3D texture to do conversion via lookup. Needs the
``GL_ARB_fragment_program extension`` and at least four texture
units. Extremely slow (software emulation) on some (all?) ATI
cards since it uses a texture with border pixels. Provides
brightness, contrast, saturation, hue and gamma control. Gamma can
also be set independently for red, green and blue. Speed depends
more on GPU memory bandwidth than other methods.
lscale=<n>
Select the scaling function to use for luminance scaling. Only valid
for yuv modes 2, 3, 4 and 6.
0
Use simple linear filtering (default).
1
Use bicubic B-spline filtering (better quality). Needs one
additional texture unit. Older cards will not be able to handle
this for chroma at least in fullscreen mode.
2
Use cubic filtering in horizontal, linear filtering in vertical
direction. Works on a few more cards than method 1.
3
Same as 1 but does not use a lookup texture. Might be faster on
some cards.
4
Use experimental unsharp masking with 3x3 support and a default
strength of 0.5 (see `filter-strength`).
5
Use experimental unsharp masking with 5x5 support and a default
strength of 0.5 (see `filter-strength`).
cscale=<n>
Select the scaling function to use for chrominance scaling. For
details see `lscale`.
filter-strength=<value>
Set the effect strength for the `lscale`/`cscale` filters that support
it.
stereo=<value>
Select a method for stereo display. You may have to use ``--aspect`` to
fix the aspect value. Experimental, do not expect too much from it.
0
Normal 2D display
1
Convert side by side input to full-color red-cyan stereo.
2
Convert side by side input to full-color green-magenta stereo.
3
Convert side by side input to quadbuffered stereo. Only supported
by very few OpenGL cards.
The following options are only useful if writing your own fragment
programs.
customprog=<filename>
Load a custom fragment program from <filename>. See
``TOOLS/edgedect.fp`` for an example.
customtex=<filename>
Load a custom "gamma ramp" texture from <filename>. This can be used
in combination with yuv=4 or with the customprog option.
(no-)customtlin
If enabled (default) use ``GL_LINEAR`` interpolation, otherwise use
``GL_NEAREST`` for customtex texture.
(no-)customtrect
If enabled, use texture_rectangle for customtex texture. Default is
disabled.
(no-)mipmapgen
If enabled, mipmaps for the video are automatically generated. This
should be useful together with the customprog and the TXB instruction
to implement blur filters with a large radius. For most OpenGL
implementations this is very slow for any non-RGB formats. Default is
disabled.
Normally there is no reason to use the following options, they mostly
exist for testing purposes.
(no-)glfinish
Call ``glFinish()`` before swapping buffers. Slower but in some cases
more correct output (default: disabled).
(no-)manyfmts
Enables support for more (RGB and BGR) color formats (default:
enabled). Needs OpenGL version >= 1.2.
slice-height=<0-...>
Number of lines copied to texture in one piece (default: 0). 0 for
whole image.
sw
Continue even if a software renderer is detected.
backend=<sys>
auto
auto-select (default)
cocoa
Cocoa/OSX
win
Win32/WGL
x11
X11/GLX
wayland
Wayland/EGL
sdl
SDL 2.0+ Render video output driver, depending on system with or without
hardware acceleration. Should work everywhere where SDL 2.0 builds. For
tuning, refer to your copy of the file SDL_hints.h.
sw
Continue even if a software renderer is detected.
switch-mode
Instruct SDL to switch the monitor video mode when going fullscreen.
null
Produces no video output. Useful for benchmarking.
caca
Color ASCII art video output driver that works on a text console.
image
Output each frame into an image file in the current directory. Each file
takes the frame number padded with leading zeros as name.
format=<format>
Select the image file format.
jpg
JPEG files, extension .jpg. (Default.)
jpeg
JPEG files, extension .jpeg.
png
PNG files.
ppm
Portable bitmap format.
pgm
Portable graymap format.
pgmyuv
Portable graymap format, using the YV12 pixel format.
tga
Truevision TGA.
png-compression=<0-9>
PNG compression factor (speed vs. file size tradeoff) (default: 7)
jpeg-quality=<0-100>
JPEG quality factor (default: 90)
[no-]jpeg-progressive
Specify standard or progressive JPEG (default: no).
[no-]jpeg-baseline
Specify use of JPEG baseline or not (default: yes).
jpeg-optimize=<0-100>
JPEG optimization factor (default: 100)
jpeg-smooth=<0-100>
smooth factor (default: 0)
jpeg-dpi=<1->
JPEG DPI (default: 72)
outdir=<dirname>
Specify the directory to save the image files to (default: ``./``).