OpenCores
URL https://opencores.org/ocsvn/mpeg2fpga/mpeg2fpga/trunk

Subversion Repositories mpeg2fpga

[/] [mpeg2fpga/] [trunk/] [tools/] [mpeg2dec/] [CHANGES] - Rev 2

Compare with Previous | Blame | View Log

CHANGES
----------
December 8, 1998
- Rewrote X11 support to allow 8bpp, 15bpp, 16bpp, 24bpp, 32bpp
- 8bpp switched from dithered to grayscale since most people have
  >8bpp these days: helps framerate for those with low-end machines

January 9, 1996 to July 17, 1996
- cleaned up some code which gave warnings.
- altered some code to be compatible with Sun CC compiler.
  However, support in the makefile for non-ansi C compilers
  has been dropped (this is a stupid thing to support).

December 20, 1995 to January 9, 1996:

verified on HHI #22, TCEH #23 bitstreams.

 1. new arguments format.  Got to be so many argument fields that
    a new more consistent format was devised.

 2. Frame_Store_Flag (-f) option now controls lower layer prediciton
    picture format (field or framewise)

 3. getpic() structural changes


Since December 18, 1995:

1. added special case for current B pictures subsframe.c which
   loads entire reference frame for buffer substitution.

2. fixed -l omission (-lXext) in Makefile that drives Tristan nuts everytime.


Since December 14, 1995:

 1. organized frame buffer substitution routines into subspic.c
 2. added "big file" -b mode for Tristan ;-)


Since July 4, 1994:

1. Concatenated elementary sequences within same bitstream

   Decode can now handle concatenated elementary video sequences of 
   arbitrary parameters.
  
2. TRACE and VERBOSE #ifdef flags

3. More disciplined naming convention

   normative variables and bitstream elements defined in 13818 are 
   verbatim, lower case.  Implementation specific routines and variables
   are capitolized.

4. Spatial scalability corrections  
   - see Carsten's document (spatscal.doc)

5. D-pictures (picture_coding_type==D_TYPE)

   Only two small changes were necessary to accomodate D-pictures: 

    a. in Decode_MPEG1_Intra_Block() added line which termines
       subroutine after DC coefficient has been processed.

    b. in picture_data(), added line which parses marker bit.


 6. forced decoder to display frame progressively (regardless of whether
    the picture is frame structured or field structured) when -f flag
    is invoked in the command line arguements.

    also: progressive_frame now decides whether a frame is to be displayed
     as a frame picture to two field pictures, rather than the older convention
     of testing progressive_sequence.

 7. Adapted systems parser from Stefan's mpeg2play to mpeg2decode.
    The major changes are:  

    mpeg2dec.c: 
     - fseek() called twice 

    gethdr.c, getpic.c:
     instances of Flush_Bits(par,32) changed to Flush_Bits32(par)

    gethdr.c 
     Get_Bits(par,32) changed to Get_32_Bits(par)

    global.h
      added rdmax, sysstream, and bfr[] to parameters struct.

 8. Restructuring of getpic.c:

      a. moved picture pointer rotation into Update_Picture_Buffers()
      b. moved picture output logic into Output_Current_Frame() to 
         in anticipation of 3:2 pulldown



Compare with Previous | Blame | View Log

powered by: WebSVN 2.1.0

© copyright 1999-2024 OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.