| 1 |
2 |
kdv |
CHANGES
|
| 2 |
|
|
----------
|
| 3 |
|
|
December 8, 1998
|
| 4 |
|
|
- Rewrote X11 support to allow 8bpp, 15bpp, 16bpp, 24bpp, 32bpp
|
| 5 |
|
|
- 8bpp switched from dithered to grayscale since most people have
|
| 6 |
|
|
>8bpp these days: helps framerate for those with low-end machines
|
| 7 |
|
|
|
| 8 |
|
|
January 9, 1996 to July 17, 1996
|
| 9 |
|
|
- cleaned up some code which gave warnings.
|
| 10 |
|
|
- altered some code to be compatible with Sun CC compiler.
|
| 11 |
|
|
However, support in the makefile for non-ansi C compilers
|
| 12 |
|
|
has been dropped (this is a stupid thing to support).
|
| 13 |
|
|
|
| 14 |
|
|
December 20, 1995 to January 9, 1996:
|
| 15 |
|
|
|
| 16 |
|
|
verified on HHI #22, TCEH #23 bitstreams.
|
| 17 |
|
|
|
| 18 |
|
|
1. new arguments format. Got to be so many argument fields that
|
| 19 |
|
|
a new more consistent format was devised.
|
| 20 |
|
|
|
| 21 |
|
|
2. Frame_Store_Flag (-f) option now controls lower layer prediciton
|
| 22 |
|
|
picture format (field or framewise)
|
| 23 |
|
|
|
| 24 |
|
|
3. getpic() structural changes
|
| 25 |
|
|
|
| 26 |
|
|
|
| 27 |
|
|
Since December 18, 1995:
|
| 28 |
|
|
|
| 29 |
|
|
1. added special case for current B pictures subsframe.c which
|
| 30 |
|
|
loads entire reference frame for buffer substitution.
|
| 31 |
|
|
|
| 32 |
|
|
2. fixed -l omission (-lXext) in Makefile that drives Tristan nuts everytime.
|
| 33 |
|
|
|
| 34 |
|
|
|
| 35 |
|
|
Since December 14, 1995:
|
| 36 |
|
|
|
| 37 |
|
|
1. organized frame buffer substitution routines into subspic.c
|
| 38 |
|
|
2. added "big file" -b mode for Tristan ;-)
|
| 39 |
|
|
|
| 40 |
|
|
|
| 41 |
|
|
Since July 4, 1994:
|
| 42 |
|
|
|
| 43 |
|
|
1. Concatenated elementary sequences within same bitstream
|
| 44 |
|
|
|
| 45 |
|
|
Decode can now handle concatenated elementary video sequences of
|
| 46 |
|
|
arbitrary parameters.
|
| 47 |
|
|
|
| 48 |
|
|
2. TRACE and VERBOSE #ifdef flags
|
| 49 |
|
|
|
| 50 |
|
|
3. More disciplined naming convention
|
| 51 |
|
|
|
| 52 |
|
|
normative variables and bitstream elements defined in 13818 are
|
| 53 |
|
|
verbatim, lower case. Implementation specific routines and variables
|
| 54 |
|
|
are capitolized.
|
| 55 |
|
|
|
| 56 |
|
|
4. Spatial scalability corrections
|
| 57 |
|
|
- see Carsten's document (spatscal.doc)
|
| 58 |
|
|
|
| 59 |
|
|
5. D-pictures (picture_coding_type==D_TYPE)
|
| 60 |
|
|
|
| 61 |
|
|
Only two small changes were necessary to accomodate D-pictures:
|
| 62 |
|
|
|
| 63 |
|
|
a. in Decode_MPEG1_Intra_Block() added line which termines
|
| 64 |
|
|
subroutine after DC coefficient has been processed.
|
| 65 |
|
|
|
| 66 |
|
|
b. in picture_data(), added line which parses marker bit.
|
| 67 |
|
|
|
| 68 |
|
|
|
| 69 |
|
|
6. forced decoder to display frame progressively (regardless of whether
|
| 70 |
|
|
the picture is frame structured or field structured) when -f flag
|
| 71 |
|
|
is invoked in the command line arguements.
|
| 72 |
|
|
|
| 73 |
|
|
also: progressive_frame now decides whether a frame is to be displayed
|
| 74 |
|
|
as a frame picture to two field pictures, rather than the older convention
|
| 75 |
|
|
of testing progressive_sequence.
|
| 76 |
|
|
|
| 77 |
|
|
7. Adapted systems parser from Stefan's mpeg2play to mpeg2decode.
|
| 78 |
|
|
The major changes are:
|
| 79 |
|
|
|
| 80 |
|
|
mpeg2dec.c:
|
| 81 |
|
|
- fseek() called twice
|
| 82 |
|
|
|
| 83 |
|
|
gethdr.c, getpic.c:
|
| 84 |
|
|
instances of Flush_Bits(par,32) changed to Flush_Bits32(par)
|
| 85 |
|
|
|
| 86 |
|
|
gethdr.c
|
| 87 |
|
|
Get_Bits(par,32) changed to Get_32_Bits(par)
|
| 88 |
|
|
|
| 89 |
|
|
global.h
|
| 90 |
|
|
added rdmax, sysstream, and bfr[] to parameters struct.
|
| 91 |
|
|
|
| 92 |
|
|
8. Restructuring of getpic.c:
|
| 93 |
|
|
|
| 94 |
|
|
a. moved picture pointer rotation into Update_Picture_Buffers()
|
| 95 |
|
|
b. moved picture output logic into Output_Current_Frame() to
|
| 96 |
|
|
in anticipation of 3:2 pulldown
|
| 97 |
|
|
|
| 98 |
|
|
|
| 99 |
|
|
|
