use endianless format names
add but comment out a few types; these are crashing ffmpeg, need to find out why prefer BGRA; faster for nVidia cards
This commit is contained in:
@@ -22,7 +22,7 @@ namespace avcodec
|
||||
static struct AVPixelFormat_t
|
||||
{
|
||||
int bpp;
|
||||
int masks[4];
|
||||
uint32_t masks[4];
|
||||
avcodec::PixelFormat pf;
|
||||
bool bHighColor;
|
||||
bool bByteSwapOnLittleEndian;
|
||||
@@ -40,31 +40,50 @@ static struct AVPixelFormat_t
|
||||
PixelFormatYCbCr_YUYV422,
|
||||
},
|
||||
{
|
||||
/* This format is really BGRA, and is affected by endianness, unlike PIX_FMT_RGB24
|
||||
* and PIX_FMT_BGR24. */
|
||||
32,
|
||||
{ 0x0000FF00,
|
||||
0x00FF0000,
|
||||
0xFF000000,
|
||||
0x000000FF },
|
||||
avcodec::PIX_FMT_BGR32_1,
|
||||
avcodec::PIX_FMT_BGRA,
|
||||
true,
|
||||
true,
|
||||
false,
|
||||
PixelFormatYCbCr_Invalid,
|
||||
},
|
||||
{
|
||||
/* This format is really ARGB, and is affected by endianness, unlike PIX_FMT_RGB24
|
||||
* and PIX_FMT_BGR24. */
|
||||
32,
|
||||
{ 0x00FF0000,
|
||||
0x0000FF00,
|
||||
0x000000FF,
|
||||
0xFF000000 },
|
||||
avcodec::PIX_FMT_RGB32,
|
||||
avcodec::PIX_FMT_ARGB,
|
||||
true,
|
||||
true,
|
||||
false,
|
||||
PixelFormatYCbCr_Invalid,
|
||||
},
|
||||
/*
|
||||
{
|
||||
32,
|
||||
{ 0x000000FF,
|
||||
0x0000FF00,
|
||||
0x00FF0000,
|
||||
0xFF000000 },
|
||||
avcodec::PIX_FMT_ABGR,
|
||||
true,
|
||||
true,
|
||||
PixelFormatYCbCr_Invalid,
|
||||
},
|
||||
{
|
||||
32,
|
||||
{ 0xFF000000,
|
||||
0x00FF0000,
|
||||
0x0000FF00,
|
||||
0x000000FF },
|
||||
avcodec::PIX_FMT_RGBA,
|
||||
true,
|
||||
true,
|
||||
PixelFormatYCbCr_Invalid,
|
||||
}, */
|
||||
{
|
||||
24,
|
||||
{ 0xFF0000,
|
||||
|
||||
Reference in New Issue
Block a user