-
Notifications
You must be signed in to change notification settings - Fork 0
/
CHANGELOG
119 lines (97 loc) · 10.8 KB
/
CHANGELOG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
[code]
[b]v2.0.3[/b]: Some more bug fixes, based on a bug report.
[size=8pt]*Fixed bug with split MLV files by piping the FRAMES output through "paste -s -d+ - | bc"[/size]
[size=8pt]*Fixed the color-ext LUTs, which were unfixed due to laziness.[/size]
[size=8pt]*Fixed mlv2badpixels.sh when mlv_dump wasn't on the PATH. Requred modding the script; a patch is in the main repo, and is distributed alongside binaries.[/size]
[size=8pt]*Added documentation that symbolic linking support is required.[/size]
[b]v2.0.2[/b]: Some fundamental & bug fixes, as well as now being tested to work with new compressed output.
[size=8pt]*Fixed some mlv_dump versions wouldn't detect # of frames.[/size]
[size=8pt]*Added set +f to ensure globbing is always enabled. Added a commented set -e to kill on error, which I intend to activate for 2.1.0.[/size]
[size=8pt]*Removed all for-loop globbing and replaced with 'find' subcommands. Should remove a class of 'set +f'-related bugs.[/size]
[size=8pt]*Added --relaxed to every mlv_dump invocation, so it won't freak out and develop nothing at the slightest touch of corruption.[/size]
[size=8pt]*Updated help page & PDF for clarity.[/size]
[size=8pt]*Fixed the LUTs - they were processing a too large (extended) range, at 0 benefit to color fidelity and to the detriment of output quality.[/size]
[size=8pt]*Clarified that convmlv's (really, dcraw's) XYZ output is a D65 space, not D50 as it's specced to.[/size]
[size=8pt]*We now depend specifically on @bouncyball's mlv_dump, as that's the only one I can guarantee works with compressed footage.[/size]
[size=8pt]*Get the required mlv_dump here: https://www.magiclantern.fm/forum/index.php?topic=18975.0 .[/size]
[b]v2.0.1[/b]: I'm back! Did some quick fixes. Also, there is now a bundled tarball available - no need to mess around with manual dependencies yourself anymore.
[size=8pt]*Updated PDF with all new features. Nothing is unexplained, now![/size]
[size=8pt]*Added some Mac compatibility.[/size]
[size=8pt]*DNG sequences now use symlinks instead of copying.[/size]
[size=8pt]*Fixed DNG sequence bug.[/size]
[b]v2.0.0[/b]: I think convmlv is the stablest it's ever been! Though I'm sure there's plenty more bugs to discover :).
[size=8pt]*Color Management: Real, 3D LUT-based gamma/gamut choices! See -g/-G. There is now minimal quality loss from sensor to image. Adds a dependency: color-core. [/size]
[size=8pt]*Extra Colorspaces: Add extra colorspace options with the optional dependency color-ext - it gives logspace, wide-gamut, etc. options. It's not yet complete - stay tuned![/size]
[size=8pt]*More Dependency Lists: Ubuntu, Fedora, and Homebrew lists have been added to the -K option, allowing easy install with those package managers.[/size]
[size=8pt]*Misc: Specify multiple LUTs in a chain, use a custom DCRAW path, updated + streamlined 'convmlv -h' page.[/size]
[size=8pt]*Speedups: ffmpeg filtering is faster for EXRs. movie creation is much faster if images were developed.[/size]
[size=8pt]*Error Checking: Though not complete, it's been drastically upscaled - there should be far fewer "weird errors"; it'll tell you what's wrong.[/size]
[size=8pt]*Bug Fixes: Reuse of DNGs, DPX colorspace bug, PNG compression format, "files left" display for DNG folders, ARG bug, filter application sequence, file-specific reset bugs.[/size]
[b]v1.9.4[/b]: [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Added Unsharp filter. Useful for counteracting all 6 denoisers ;).[/size]
[size=8pt]*Bug Fixes/Small Improvements: File-specific blocks/multiple files fixed. Filter application has been sanitized.[/size]
[b]v1.9.3[/b]: [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Deshake Support! Just specify -D, and auto-stabilization will be applied. You may wish to scale it later.[/size]
[size=8pt]*HQDN3D Denoising Support. A superior denoising method (used in Handbrake). See 'convmlv -h' for usage instructions..[/size]
[size=8pt]*removegrain Denoising Support. A more specialized denoising method, ripe for experimentation. See https://ffmpeg.org/ffmpeg-filters.html#removegrain .[/size]
[size=8pt]*Bug Fixes: Different formats should now look the same. They didn't before... I hate color spaces... Misc. Bug fixes too.[/size]
[b]v1.9.2[/b]: [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Atadenoise Support! It's a great denoising method, used with "-N <A>-<B>" See the documentation for how to configure A and B values.[/size]
[size=8pt]*General FFMPEG Filter Support: The framework for any ffmpeg -vf filters exists now! lut3d and atadenoise are currently implemented.[/size]
[size=8pt]*Bug Fixes: LUTs can now be applied to EXR's! Other bugs fixed too.[/size]
[b]v1.9.1[/b]: [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Fully Supported Config Files! Check the help page; all options for which this makes sense can be specified in global, local, or file-specific config files.[/size]
[size=8pt]*File-Specific Blocks: With this, just make one config file w/any special options per file, and you're done. Huge time saver in production situations![/size]
[size=8pt]*Create Darkframes: You no longer have to create your averaged .darkframe files yourself. 'convmlv -R df <MLV FILE>' creates a file named df.darkframe; -F develops using it.[/size]
[size=8pt]*DNG Improvements: It works now (that's nice :) ), but settings.txt is also created properly now. Plus, bug fixes abound![/size]
[b]v1.9.0[/b]: [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Sane Argument Parsing! You can now choose to use spaces![/size]
[size=8pt]*DNG: Multithreaded mlv_dump with progress bar![/size]
[size=8pt]*Config File: Can now use a global config file in $HOME/convmlv.conf with select options. More to come later![/size]
[size=8pt]*Misc: Auto threads on Linux, use s and e in frame range, better help, more error checking/bug fixes.[/size]
[b]v1.8.1[/b]: [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Specify Frame Range to develop! Note that this will still develop all DNGs; however, if you specify the dng_ folder of a previous run as the argument, it'll reuse those DNGs without redevelopment. This works for Dual ISO as well.[/size]
[size=8pt]* Camera WB is now default.[/size]
[b]v1.8.0[/b]:MASSIVE BUG FIX. [url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Dark Level Aware!!! This is a bug fix so big that [color=red]all previous versions are basically unusable [/color] (you'd sometimes find a magenta or green cast; fixable but very, very bad).[/size]
[size=8pt]*Color Space Output! Gone is the "gamma curve" of old; gamma and output color spaces are now synchronized.[/size]
[size=8pt]*Chroma Smoothing! Uses 2x2, 3x3, or 5x5; works perfectly with Dual ISO as well.[/size]
[size=8pt]*More features: Use badpixels file without generating one, print out footage attributes (simple stuff for now, and only works on MLV for now).
[b]v1.7.1[/b]:[url=https://bitbucket.org/so-rose/convmlv/src]https://bitbucket.org/so-rose/convmlv/src[/url]
[size=8pt]*Use any DNG sequence as input now; just specify the folder.[/size]
[size=8pt]*Darkframe Subtraction! Simply specify the darkly recorded MLV with -F; it'll automatically be averaged and used.[/size]
[b]v1.7.0[/b]:[url=https://bitbucket.org/so-rose/convmlv]https://bitbucket.org/so-rose/convmlv[/url]
[size=8pt]*Multithreading! Image sequences and Dual ISO processing are multithreaded - from 4x to 8x speedup in my 4-core (8 thread) testing![/size]
[size=8pt]*Dual ISO! Specify -u to process the file with cr2hdr, which is now a dependency. ***Badpixel removal might be a little strange, but it shouldn't affect you otherwise.[/size]
[size=8pt]*Format Support! Use -f to choose TIFF, EXR, DPX, or PNG (feel free to request more). This includes automatic compression for all of them with -c.[/size]
[b]v1.6.1[/b]:EXRs and Bug Fixes.
[size=8pt]*EXR output support (now default)! This includes PIZ compression. Unfortunately LUTs cannot be applied (ffmpeg doesn't support exr output).[/size]
[b]v1.6.0[/b]:Piping and compression.
[size=8pt]*Speedup! Proxies are now created simultaneous to the image/movie. In the future, images and movies will be made simultaneously as well.[/size]
[size=8pt]*Lossless TIFF compression! I chose ZIP, due to its superior effect on 16-bit files (78% of the original size in one test).[/size]
[size=8pt]*-A now denotes the amount of images used to average the WB, regardless of the length of the sequence.[/size]
[b]v1.5.2[/b]: Speedups, fixes, and can specify own .badpixels file!
[size=8pt]*-H2 looks good, but blew highlights in some files :( -H0 is now default again![/size]
[size=8pt]*Four color interpolation (-f) is now available, for fixing weirdness in AHD and VNG modes.[/size]
[size=8pt]*You can now specify your own .badpixels file, along with the generated one (they're pasted together), for bad pixels that aren't focus pixels. Use -a<path>.[/size]
[size=8pt]*LUT application to images has been sped up & streamlined immensely![/size]
[size=8pt]*It will no longer ask to overwrite the output directory, only the individual file directory.[/size]
[b]v1.5.1[/b]: FPS patch.
[size=8pt]*FPS is now read from the MLV or RAW file. -f option has thus been removed.[/size]
[b]v1.5.0[/b]: UI and Speed, plus cleanup (and probably new bugs :) ).
[size=8pt]*AWB speedup - it's now fast, and can be made faster by raising the number in the new -A[int] option.[/size]
[size=8pt]*Redid UI - You can choose to make any combination of image sequence, movie, and proxies for each. Other options changed names, and the help page was reorganized.[/size]
[size=8pt]*Faster Movie Creation - There's no longer any need to create TIFFs, then JPGs, before getting to the MOV/H.264, which translates to having a video file faster.[/size]
[b]v1.4.0[/b]: AWB and .RAW!
[size=8pt]*Auto White Balance returns! Requires new Python dependencies to invoke a companion script, balance.py. A little slow, but not unusably so.[/size]
[size=8pt]*Process .RAW files. Next step is Dual ISO processing :) .[/size]
[size=8pt]*Bug fixes + cleanups. Who knew things could go wrong...[/size]
[b]v1.3.0[/b]: Pixels & Bug Fixes.
[size=8pt]*Badpixels removal! Put dfort's script (get it [url=https://bitbucket.org/daniel_fort/ml-focus-pixels/src]here[/url]) in the same directory, or specify the path with -B.[/size]
[size=8pt]*Cleaned up some pretty serious bugs, as well as organizing the output better. Should be neater than ever :) ![/size]
[b]v1.2.0[/b]:Denoising!
[size=8pt]*Wavelet denoising[/size]
[size=8pt]*Undefaulted auto white balance, put warning - auto WB is WIP.[/size]
[b]v1.1.0[/b]: Apply LUTs (.cube). A bit slow, but most certainly works!
[b]v1.0.0[/b]: Initial Release.
[/code]