Commit f236aa47 authored by Diego Biurrun's avatar Diego Biurrun
Browse files

wording/spelling

Originally committed as revision 15333 to svn://svn.ffmpeg.org/ffmpeg/trunk
parent 8fd64adf
...@@ -26,7 +26,7 @@ Current (simplified) Architecture: ...@@ -26,7 +26,7 @@ Current (simplified) Architecture:
Swscale has 2 scaler paths. Each side must be capable of handling Swscale has 2 scaler paths. Each side must be capable of handling
slices, that is, consecutive non-overlapping rectangles of dimension slices, that is, consecutive non-overlapping rectangles of dimension
(0,slice_top) - (picture_width, slice_bottom) (0,slice_top) - (picture_width, slice_bottom).
special converter special converter
These generally are unscaled converters of common These generally are unscaled converters of common
...@@ -37,64 +37,63 @@ Main path ...@@ -37,64 +37,63 @@ Main path
The main path is used when no special converter can be used. The code The main path is used when no special converter can be used. The code
is designed as a destination line pull architecture. That is, for each is designed as a destination line pull architecture. That is, for each
output line the vertical scaler pulls lines from a ring buffer. When output line the vertical scaler pulls lines from a ring buffer. When
the ring buffer does not contain the wanted line then it is pulled from the ring buffer does not contain the wanted line, then it is pulled from
the input slice through the input converter and horizontal scaler, and the input slice through the input converter and horizontal scaler.
the result is also stored in the ring buffer to serve future vertical The result is also stored in the ring buffer to serve future vertical
scaler requests. scaler requests.
When no more output can be generated because lines from a future slice When no more output can be generated because lines from a future slice
would be needed, then all remaining lines in the current slice are would be needed, then all remaining lines in the current slice are
converted, horizontally scaled and put in the ring buffer. converted, horizontally scaled and put in the ring buffer.
[this is done for luma and chroma, each with possibly different numbers [This is done for luma and chroma, each with possibly different numbers
of lines per picture] of lines per picture.]
Input to YUV Converter Input to YUV Converter
When the input to the main path is not planar 8bit per component yuv or When the input to the main path is not planar 8 bits per component YUV or
8bit gray then it is converted to planar 8bit YUV. 2 sets of converters 8-bit gray then it is converted to planar 8-bit YUV. 2 sets of converters
exist for this currently, one performing horizontal downscaling by 2 exist for this currently: One performs horizontal downscaling by 2
before the conversion and the other leaving the full chroma resolution before the conversion, the other leaves the full chroma resolution
but being slightly slower. The scaler will try to preserve full chroma but is slightly slower. The scaler will try to preserve full chroma
here when the output uses it. It is possible to force full chroma with here when the output uses it. It is possible to force full chroma with
SWS_FULL_CHR_H_INP though even for cases where the scaler thinks it is SWS_FULL_CHR_H_INP even for cases where the scaler thinks it is useless.
useless.
Horizontal scaler Horizontal scaler
There are several horizontal scalers. A special case worth mentioning is There are several horizontal scalers. A special case worth mentioning is
the fast bilinear scaler that is made of runtime generated MMX2 code the fast bilinear scaler that is made of runtime-generated MMX2 code
using specially tuned pshufw instructions. using specially tuned pshufw instructions.
The remaining scalers are specially tuned for various filter lengths. The remaining scalers are specially-tuned for various filter lengths.
They scale 8bit unsigned planar data to 16bit signed planar data. They scale 8-bit unsigned planar data to 16-bit signed planar data.
Future >8bit per component inputs will need to add a new scaler here Future >8 bits per component inputs will need to add a new scaler here
that preserves the input precision. that preserves the input precision.
Vertical scaler and output converter Vertical scaler and output converter
There is a large number of combined vertical scalers+output converters There is a large number of combined vertical scalers + output converters.
Some are: Some are:
* unscaled output converters * unscaled output converters
* unscaled output converters that average 2 chroma lines * unscaled output converters that average 2 chroma lines
* bilinear converters (C, MMX and accurate MMX) * bilinear converters (C, MMX and accurate MMX)
* arbitrary filter length converters (C, MMX and accurate MMX) * arbitrary filter length converters (C, MMX and accurate MMX)
And And
* Plain C 8bit 4:2:2 YUV -> RGB converters using LUTs * Plain C 8-bit 4:2:2 YUV -> RGB converters using LUTs
* Plain C 17bit 4:4:4 YUV -> RGB converters using multiplies * Plain C 17-bit 4:4:4 YUV -> RGB converters using multiplies
* MMX 11bit 4:2:2 YUV -> RGB converters * MMX 11-bit 4:2:2 YUV -> RGB converters
* Plain C 16bit Y -> 16bit gray * Plain C 16-bit Y -> 16-bit gray
... ...
RGB with less than 8bit per component uses dither to improve the RGB with less than 8 bits per component uses dither to improve the
subjective quality and low frequency accuracy. subjective quality and low-frequency accuracy.
Filter coefficients: Filter coefficients:
-------------------- --------------------
There are several different scalers (bilinear, bicubic, lanczos, area, sinc, ...) There are several different scalers (bilinear, bicubic, lanczos, area,
Their coefficients are calculated in initFilter(). sinc, ...). Their coefficients are calculated in initFilter().
Horizontal filter coeffs have a 1.0 point at 1<<14, vertical ones at 1<<12. Horizontal filter coefficients have a 1.0 point at 1 << 14, vertical ones at
The 1.0 points have been chosen to maximize precision while leaving a 1 << 12. The 1.0 points have been chosen to maximize precision while leaving
little headroom for convolutional filters like sharpening filters and a little headroom for convolutional filters like sharpening filters and
minimizing SIMD instructions needed to apply them. minimizing SIMD instructions needed to apply them.
It would be trivial to use a different 1.0 point if some specific scaler It would be trivial to use a different 1.0 point if some specific scaler
would benefit from it. would benefit from it.
Also as already hinted at, initFilter() accepts an optional convolutional Also, as already hinted at, initFilter() accepts an optional convolutional
filter as input that can be used for contrast, saturation, blur, sharpening filter as input that can be used for contrast, saturation, blur, sharpening
shift, chroma vs. luma shift, ... shift, chroma vs. luma shift, ...
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment