2 |
|
|
3 |
|
|
4 |
*** build environment *** |
*** build environment *** |
5 |
|
|
6 |
- portable makefile [DONE] |
- portable makefile [DONE] |
7 |
more testing needed |
|
8 |
|
- QuickTime frontend |
9 |
|
- Java frontend |
10 |
|
|
11 |
|
|
12 |
*** api *** |
*** api *** |
13 |
- decore/encore2 wrappers; [DONE] |
- decore/encore2 wrappers; [DONE] |
|
|
|
14 |
- xvid_init() [DONE] |
- xvid_init() [DONE] |
15 |
handle cpu detection, yuv/dct lookup table init |
handle cpu detection, yuv/dct lookup table init [DONT] |
16 |
return api version number |
return api version number [DONE] |
17 |
return core build number [build number should be incremented automatically, how?] |
return core build number [HALTED; how?] |
|
|
|
18 |
|
|
19 |
*** common *** |
*** common *** |
20 |
|
|
21 |
- mpeg-4 quantizer (quant4) testing [DONE] |
- mpeg-4 quantizer (quant4) testing [DONE] |
22 |
- consder moving a modified mbprediction/mbdeprediction here [DONE] |
- consder moving a modified mbprediction/mbdeprediction here [DONE] |
|
|
|
23 |
- add [-2048,2048] to mmx dequant [DONE] |
- add [-2048,2048] to mmx dequant [DONE] |
|
|
|
|
|
|
24 |
- mmx quant4 [DONE] |
- mmx quant4 [DONE] |
25 |
- sse2: convert intel's idct/sad/comp apnotes to nasm [IN PROGRESS] |
- sse2 [DONE] |
26 |
|
|
27 |
|
- sse2 quant4 |
28 |
- mlib |
- mlib |
29 |
|
|
30 |
|
|
31 |
*** image *** |
*** image *** |
32 |
|
|
33 |
- sort out in/out & conversion func naming issues [DONE] |
- sort out in/out & conversion func naming issues [DONE] |
34 |
maybe, in_yuv, in_yuyv, in_uyuv, in_rgbXX |
- rgb16. [DONE] |
35 |
out_yuv, out_yuyv, out_uyuv, out_rgbXX |
|
36 |
|
|
37 |
- input flipping support |
- input flipping support |
38 |
output flipping support is not 'perfect'. |
output flipping support is not 'perfect'. |
41 |
- rgb24/32_mmx cause yellow dots to appear on left hand size |
- rgb24/32_mmx cause yellow dots to appear on left hand size |
42 |
(only visibile with high contrast colors) |
(only visibile with high contrast colors) |
43 |
|
|
44 |
- rgb16. [DONE] |
- investigate difference between basic-yuv and iso-yuv |
|
can be done generically using r,g,b masks |
|
|
or just code for the popular 16bit formats 555, 565 |
|
|
|
|
|
- diff between basic-yuv and iso-yuv |
|
45 |
|
|
46 |
|
|
47 |
*** decoder *** |
*** decoder *** |
49 |
- quant=1 encoder material appears "soft" [FIXED] |
- quant=1 encoder material appears "soft" [FIXED] |
50 |
- block based interpolation [DONE] |
- block based interpolation [DONE] |
51 |
|
|
|
- msmpeg-4 support... dont hold your breath [IN PROGRESS] |
|
|
best to write an entirely new 'decoder' than bolt stuff onto |
|
|
the existing one. |
|
|
|
|
|
- post processing [IN PROGRESS] |
|
|
|
|
52 |
- add basic mpeg4 stuff: |
- add basic mpeg4 stuff: |
53 |
- custom mpeg quant matrixes [DONE] |
- custom mpeg quant matrixes [DONE] |
54 |
- intra_dc_vlc_threshold [DONE] |
- intra_dc_vlc_threshold [DONE] |
55 |
- quarterpel [in progress] |
- quarterpel [in progress] |
56 |
- resync? |
- resync? |
57 |
|
|
58 |
|
- msmpeg-4 support [IN PROGRESS] |
59 |
|
- post processing [IN PROGRESS] |
60 |
- b-vops [IN PROGRESS] |
- b-vops [IN PROGRESS] |
|
- figure out what todo with time codes, and fatal error |
|
61 |
|
|
62 |
|
- proper handling of time codes and fatal/non-fatal errors |
63 |
|
|
64 |
|
|
65 |
*** encoder *** |
*** encoder *** |
66 |
|
|
67 |
- purple mb block (rare) [FIXED] |
- purple mb block (rare) [FIXED] |
68 |
- coeff range checks need to be verified against mpeg-4 spec [DONE/PARTIAL] |
- coeff range checks need to be verified against mpeg-4 spec [DONE] |
69 |
- dequants now clamp to [-2048,2047]. [DONE] |
- dequants now clamp to [-2048,2047]. [DONE] |
70 |
- fdct_mmx clamps to [-2048,2047], it doesnt have to (but prevents overflows in mmx quantization) |
- fdct_mmx clamps to [-2048,2047], it doesnt have to (but prevents overflows in mmx quantization) |
71 |
- quants _should_ clamp to [-127,127], but this introduces errors at low quant levels so we ignore it |
- quants _should_ clamp to [-127,127], but this introduces errors at low quant levels so we ignore it |
72 |
- idct _should_ clamp [-256,255], but its really unnessary overhead. |
- idct _should_ clamp [-256,255], but its really unnessary overhead. |
|
|
|
73 |
- pmvfast bug fixing [DONE] |
- pmvfast bug fixing [DONE] |
|
|
|
|
- quater pixel motion search [IN PROGRESS] |
|
|
|
|
|
- bidirectional coding [IN PROGRESS] |
|
|
|
|
|
- global motion compensation |
|
|
|
|
|
- redo coefficient thresholding like described in H26L |
|
|
|
|
74 |
- 'smearing bug' when motion occurs in straight vertical direction ( |
- 'smearing bug' when motion occurs in straight vertical direction ( |
75 |
http://www.videocoding.de/forum/viewtopic.php?topic=71&forum=2&6 ) |
http://www.videocoding.de/forum/viewtopic.php?topic=71&forum=2&6 ) |
76 |
coefficient thresholding problems? [FIXED/REMOVED] |
coefficient thresholding problems? [FIXED/REMOVED] |
|
|
|
77 |
- better luminance masking [DONE] |
- better luminance masking [DONE] |
78 |
|
|
|
- further improve adaptive quantisation, maybe implement JND maps |
|
79 |
|
|
80 |
- use N_VOP when a interframe is entirely composed of ublks |
- quater pixel motion search [IN PROGRESS] |
81 |
& frame dropping. eg. simple ublk threshold |
- bidirectional coding [IN PROGRESS] |
82 |
|
- virtualdub support [IN PROGRESS] |
83 |
|
|
84 |
|
- global motion compensation |
85 |
|
- redo coefficient thresholding like described in H26L |
86 |
|
- further improve adaptive quantisation, maybe implement JND maps |
87 |
|
- use N_VOP when a interframe is entirely composed of ublks & frame dropping. eg. simple ublk threshold |
88 |
|
|
89 |
|
|
90 |
*** gpl conversion *** |
*** gpl conversion *** |
91 |
|
|
92 |
mbdecoding.c [REPLACED] |
mbdecoding.c [REPLACED] |
|
|
|
93 |
encoder.c [REPLACED] |
encoder.c [REPLACED] |
94 |
mbmotionestcomp.c [REPLACED] |
mbmotionestcomp.c [REPLACED] |
95 |
mbcoding [REPLACED] |
mbcoding [REPLACED] |