Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit bdc5cae

Browse files
author
nadeko
committed
CI autodeploy
1 parent ffc96c6 commit bdc5cae

File tree

1 file changed

+86
-3
lines changed

1 file changed

+86
-3
lines changed

‎manual/master/index.html

Lines changed: 86 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -6007,8 +6007,9 @@ <h2>ALSA audio output options</h2>
60076007
</div>
60086008
<div class="section" id="gpu-renderer-options">
60096009
<h2>GPU renderer options</h2>
6010-
<p>The following video options are currently all specific to <tt class="docutils literal"><span class="pre">--vo=gpu</span></tt> and
6011-
<tt class="docutils literal"><span class="pre">--vo=libmpv</span></tt> only, which are the only VOs that implement them.</p>
6010+
<p>The following video options are currently all specific to <tt class="docutils literal"><span class="pre">--vo=gpu</span></tt>,
6011+
<tt class="docutils literal"><span class="pre">--vo=libmpv</span></tt> and <tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>, which are the only VOs that implement
6012+
them.</p>
60126013
<dl class="docutils">
60136014
<dt><tt class="docutils literal"><span class="pre">--scale=&lt;filter&gt;</span></tt></dt>
60146015
<dd><p class="first">The filter function to use when upscaling video.</p>
@@ -6167,6 +6168,11 @@ <h2>GPU renderer options</h2>
61676168
<tt class="docutils literal">bilinear</tt> is used instead of whatever is set with <tt class="docutils literal"><span class="pre">--scale</span></tt>. Bilinear
61686169
will reproduce the source image perfectly if no scaling is performed.
61696170
Enabled by default. Note that this option never affects <tt class="docutils literal"><span class="pre">--cscale</span></tt>.</dd>
6171+
<dt><tt class="docutils literal"><span class="pre">--builtin-scalers</span></tt></dt>
6172+
<dd>Allow using faster built-in replacements for common scalers such as
6173+
<tt class="docutils literal">nearest</tt>, <tt class="docutils literal">bilinear</tt> or <tt class="docutils literal">bicubic</tt>. These have the disadvantage of
6174+
not being configurable, unlike normal scaler kernels. Defaults to
6175+
enabled. (This option only affects <tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>)</dd>
61706176
<dt><tt class="docutils literal"><span class="pre">--correct-downscaling</span></tt></dt>
61716177
<dd><p class="first">When using convolution based filters, extend the filter size when
61726178
downscaling. Increases quality, but reduces performance while downscaling.</p>
@@ -6234,6 +6240,14 @@ <h2>GPU renderer options</h2>
62346240
discontinuity every other minute.</p>
62356241
<p class="last">Set this to <tt class="docutils literal"><span class="pre">-1</span></tt> to disable this logic.</p>
62366242
</dd>
6243+
<dt><tt class="docutils literal"><span class="pre">--interpolation-preserve</span></tt></dt>
6244+
<dd>Preserve the previous frames' interpolated results even when renderer
6245+
parameters are changed - with the exception of options related to
6246+
cropping and video placement, which always invalidate the cache. Enabling
6247+
this option makes dynamic updates of renderer settings slightly smoother at
6248+
the cost of slightly higher latency in response to such changes. Defaults
6249+
to on. (Only affects <tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>, note that <tt class="docutils literal"><span class="pre">-vo=gpu</span></tt> always
6250+
invalidates interpolated frames)</dd>
62376251
<dt><tt class="docutils literal"><span class="pre">--opengl-pbo</span></tt></dt>
62386252
<dd>Enable use of PBOs. On some drivers this can be faster, especially if the
62396253
source video size is huge (e.g. so called &quot;4K&quot; video). On other drivers it
@@ -6725,7 +6739,7 @@ <h2>GPU renderer options</h2>
67256739
values will sharpen the image (but add more ringing and aliasing). Negative
67266740
values will blur the image. If your GPU is powerful enough, consider
67276741
alternatives like the <tt class="docutils literal">ewa_lanczossharp</tt> scale filter, or the
6728-
<tt class="docutils literal"><span class="pre">--scale-blur</span></tt> option.</dd>
6742+
<tt class="docutils literal"><span class="pre">--scale-blur</span></tt> option. (Only for <tt class="docutils literal"><span class="pre">--vo=gpu</span></tt>)</dd>
67296743
<dt><tt class="docutils literal"><span class="pre">--opengl-glfinish</span></tt></dt>
67306744
<dd>Call <tt class="docutils literal">glFinish()</tt> before swapping buffers (default: disabled). Slower,
67316745
but might improve results when doing framedropping. Can completely ruin
@@ -7062,6 +7076,29 @@ <h2>GPU renderer options</h2>
70627076
boost), and slightly increase the boost up until 1.2 for 256 lux.</p>
70637077
<p class="last">NOTE: Only implemented on macOS.</p>
70647078
</dd>
7079+
<dt><tt class="docutils literal"><span class="pre">--image-lut=&lt;file&gt;</span></tt></dt>
7080+
<dd>Specifies a custom LUT file (in Adobe .cube format) to apply to the colors
7081+
during image decoding. The exact interpretation of the LUT depends on
7082+
the value of <tt class="docutils literal"><span class="pre">--image-lut-type</span></tt>. (Only for <tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>)</dd>
7083+
<dt><tt class="docutils literal"><span class="pre">--image-lut-type=&lt;value&gt;</span></tt></dt>
7084+
<dd><p class="first">Controls the interpretation of color values fed to and from the LUT
7085+
specified as <tt class="docutils literal"><span class="pre">--image-lut</span></tt>. Valid values are:</p>
7086+
<dl class="last docutils">
7087+
<dt>auto</dt>
7088+
<dd>Chooses the interpretation of the LUT automatically from tagged
7089+
metadata, and otherwise falls back to <tt class="docutils literal">native</tt>. (Default)</dd>
7090+
<dt>native</dt>
7091+
<dd>Applied to the raw image contents in its native colorspace, before
7092+
decoding to RGB. For example, for a HDR10 image, this would be fed
7093+
PQ-encoded YCbCr values in the range 0.0 - 1.0.</dd>
7094+
<dt>normalized</dt>
7095+
<dd>Applied to the normalized RGB image contents, after decoding from
7096+
its native color encoding, but before linearization.</dd>
7097+
<dt>conversion</dt>
7098+
<dd>Fully replaces the color decoding. A LUT of this type should ingest the
7099+
image's native colorspace and output normalized non-linear RGB.</dd>
7100+
</dl>
7101+
</dd>
70657102
<dt><tt class="docutils literal"><span class="pre">--target-prim=&lt;value&gt;</span></tt></dt>
70667103
<dd><p class="first">Specifies the primaries of the display. Video colors will be adapted to
70677104
this colorspace when ICC color management is not being used. Valid values
@@ -7177,6 +7214,11 @@ <h2>GPU renderer options</h2>
71777214
to <tt class="docutils literal">mobius</tt> or even <tt class="docutils literal">clip</tt>.</p>
71787215
</div>
71797216
</dd>
7217+
<dt><tt class="docutils literal"><span class="pre">--target-lut=&lt;file&gt;</span></tt></dt>
7218+
<dd>Specifies a custom LUT file (in Adobe .cube format) to apply to the colors
7219+
before display on-screen. This LUT is fed values in normalized RGB, after
7220+
encoding into the target colorspace, so after the application of
7221+
<tt class="docutils literal"><span class="pre">--target-trc</span></tt>. (Only for <tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>)</dd>
71807222
<dt><tt class="docutils literal"><span class="pre">--tone-mapping=&lt;value&gt;</span></tt></dt>
71817223
<dd><p class="first">Specifies the algorithm used for tone-mapping images onto the target
71827224
display. This is relevant for both HDR-&gt;SDR conversion as well as gamut
@@ -7258,6 +7300,13 @@ <h2>GPU renderer options</h2>
72587300
probably also perform horribly on some drivers, so enable at your own risk.
72597301
The special value <tt class="docutils literal">auto</tt> (default) will enable HDR peak computation
72607302
automatically if compute shaders and SSBOs are supported.</dd>
7303+
<dt><tt class="docutils literal"><span class="pre">--allow-delayed-peak-detect</span></tt></dt>
7304+
<dd>When using <tt class="docutils literal"><span class="pre">--hdr-compute-peak</span></tt>, allow delaying the detected peak by a
7305+
frame when beneficial for performance. In particular, this is required to
7306+
avoid an unnecessary FBO indirection when no advanced rendering is required
7307+
otherwise. Has no effect if there already is an indirect pass, such as when
7308+
advanced scaling is enabled. Defaults to on. (Only affects
7309+
<tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>, note that <tt class="docutils literal"><span class="pre">--vo=gpu</span></tt> always delays the peak.)</dd>
72617310
<dt><tt class="docutils literal"><span class="pre">--hdr-peak-decay-rate=&lt;1.0..1000.0&gt;</span></tt></dt>
72627311
<dd>The decay rate used for the HDR peak detection algorithm (default: 100.0).
72637312
This is only relevant when <tt class="docutils literal"><span class="pre">--hdr-compute-peak</span></tt> is enabled. Higher values
@@ -7367,6 +7416,30 @@ <h2>GPU renderer options</h2>
73677416
content. The default of <tt class="docutils literal">no</tt> means to use the profile values. The special
73687417
value <tt class="docutils literal">inf</tt> causes the BT.1886 curve to be treated as a pure power gamma
73697418
2.4 function.</dd>
7419+
<dt><tt class="docutils literal"><span class="pre">--lut=&lt;file&gt;</span></tt></dt>
7420+
<dd>Specifies a custom LUT (in Adobe .cube format) to apply to the colors
7421+
as part of color conversion. The exact interpretation depends on the value
7422+
of <tt class="docutils literal"><span class="pre">--lut-type</span></tt>. (Only for <tt class="docutils literal"><span class="pre">--vo=gpu-next</span></tt>)</dd>
7423+
<dt><tt class="docutils literal"><span class="pre">--lut-type=&lt;value&gt;</span></tt></dt>
7424+
<dd><p class="first">Controls the interpretation of color values fed to and from the LUT
7425+
specified as <tt class="docutils literal"><span class="pre">--lut</span></tt>. Valid values are:</p>
7426+
<dl class="last docutils">
7427+
<dt>auto</dt>
7428+
<dd>Chooses the interpretation of the LUT automatically from tagged
7429+
metadata, and otherwise falls back to <tt class="docutils literal">native</tt>. (Default)</dd>
7430+
<dt>native</dt>
7431+
<dd>Applied to raw image contents in its native RGB colorspace (non-linear
7432+
light), before conversion to the output color space.</dd>
7433+
<dt>normalized</dt>
7434+
<dd>Applied to the normalized RGB image contents, in linear light, before
7435+
conversion to the output color space.</dd>
7436+
<dt>conversion</dt>
7437+
<dd>Fully replaces the conversion from the image color space to the output
7438+
color space. If such a LUT is present, it has the highest priority, and
7439+
overrides any ICC profiles, as well as options related to tone mapping
7440+
and output colorimetry (<tt class="docutils literal"><span class="pre">--target-prim</span></tt>, <tt class="docutils literal"><span class="pre">--target-trc</span></tt> etc.).</dd>
7441+
</dl>
7442+
</dd>
73707443
<dt><tt class="docutils literal"><span class="pre">--blend-subtitles=&lt;yes|video|no&gt;</span></tt></dt>
73717444
<dd><p class="first">Blend subtitles directly onto upscaled video frames, before interpolation
73727445
and/or color management (default: no). Enabling this causes subtitles to be
@@ -8275,6 +8348,16 @@ <h1>VIDEO OUTPUT DRIVERS</h1>
82758348
with <tt class="docutils literal">rgb32f</tt>. If you have problems, you can also try enabling the
82768349
<tt class="docutils literal"><span class="pre">--gpu-dumb-mode=yes</span></tt> option.</p>
82778350
</dd>
8351+
<dt><tt class="docutils literal"><span class="pre">gpu-next</span></tt></dt>
8352+
<dd><p class="first">Experimental video renderer based on <tt class="docutils literal">libplacebo</tt>. This supports almost
8353+
the same set of features as <tt class="docutils literal"><span class="pre">--vo=gpu</span></tt>. See <a class="reference internal" href="#gpu-renderer-options">GPU renderer options</a> for a
8354+
list.</p>
8355+
<p>Currently, this only supports <tt class="docutils literal"><span class="pre">--gpu-api=vulkan</span></tt>, and no hardware
8356+
decoding. Unlike <tt class="docutils literal"><span class="pre">--vo=gpu</span></tt>, the FBO formats are not tunable, but you can
8357+
still set <tt class="docutils literal"><span class="pre">--gpu-dumb-mode=yes</span></tt> to forcibly disable their use.</p>
8358+
<p class="last">Should generally be faster and higher quality, but some features may still
8359+
be missing or misbehave. Expect (and report!) bugs.</p>
8360+
</dd>
82788361
<dt><tt class="docutils literal">sdl</tt></dt>
82798362
<dd><p class="first">SDL 2.0+ Render video output driver, depending on system with or without
82808363
hardware acceleration. Should work on all platforms supported by SDL 2.0.

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /