avcodec/vc1: correct ff_vc1_dqscale
authorJerome Borsboom <jerome.borsboom@carpalis.nl>
Mon, 23 Apr 2018 18:59:28 +0000 (20:59 +0200)
committerPaul B Mahol <onemda@gmail.com>
Wed, 25 Apr 2018 20:07:20 +0000 (22:07 +0200)
According to VC-1 spec table 74, the last value in ff_vc1_dqscale should be
0x1041 instead of 0x1000.

Signed-off-by: Jerome Borsboom <jerome.borsboom@carpalis.nl>
libavcodec/vc1data.c

index 0df7d4d..19f1cad 100644 (file)
@@ -1090,7 +1090,7 @@ const int32_t ff_vc1_dqscale[63] = {
      0x1F08,  0x1E1E,  0x1D42,  0x1C72, 0x1BAD, 0x1AF3, 0x1A42, 0x199A,
      0x18FA,  0x1862,  0x17D0,  0x1746, 0x16C1, 0x1643, 0x15CA, 0x1555,
      0x14E6,  0x147B,  0x1414,  0x13B1, 0x1352, 0x12F7, 0x129E, 0x1249,
-     0x11F7,  0x11A8,  0x115B,  0x1111, 0x10C9, 0x1084, 0x1000
+     0x11F7,  0x11A8,  0x115B,  0x1111, 0x10C9, 0x1084, 0x1041
 };
 
 /* P Interlaced field picture MV predictor scaling values (Table 114) */