+- Selur's Little Message Board (https://forum.selur.net)
+-- Forum: Talk, Talk, Talk (https://forum.selur.net/forum-5.html)
+--- Forum: Small Talk (https://forum.selur.net/forum-7.html)
+--- Thread: Esxi running macos For selur (/thread-1495.html)
Pages:
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
RE: Esxi running macos For selur - shijan - 10.10.2020
Deinterlace->VIVTVC (Vapoursynth) -> VFM -> TDeintMod ERROR:
Failed to evaluate the script:
Python exception: positional argument follows keyword argument (tempPreviewVapoursynthFile02_15_28_249.vpy, line 15)
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 2241, in vapoursynth.vpy_evaluateScript
File "/Volumes/temp/Hybrid Temp/tempPreviewVapoursynthFile02_15_28_249.vpy", line 15
clip2clip = core.tdm.TDeintMod(clip=clip2clip, field=1, core.znedi3.nnedi3(clip=clip, field=1))
^
SyntaxError: positional argument follows keyword argumentDeinterlace->VIVTVC (Vapoursynth) -> Misc -> TDeintMod ERROR:
Failed to evaluate the script:
Python exception: name 'clip1core' is not defined
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 2244, in vapoursynth.vpy_evaluateScript
File "src/cython/vapoursynth.pyx", line 2245, in vapoursynth.vpy_evaluateScript
File "/Volumes/temp/Hybrid Temp/tempPreviewVapoursynthFile02_17_43_560.vpy", line 24, in <module>
clipDeint = core.tdm.TDeintMod(clip=clip1core.nnedi3cl.NNEDI3CL(clip=clip, field=1))
NameError: name 'clip1core' is not defined---------------
mClean Error
Python exception: There is no function named Median
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 2244, in vapoursynth.vpy_evaluateScript
File "src/cython/vapoursynth.pyx", line 2245, in vapoursynth.vpy_evaluateScript
File "/private/var/folders/ss/y_mbrf6j0_7cnxvljw_2_wv40000gn/T/tempPreviewVapoursynthFile15_50_34_221.vpy", line 23, in <module>
clip = G41Fun.mClean(clip=clip)
File "/Applications/Hybrid.app/Contents/MacOS/vsscripts/G41Fun.py", line 2517, in mClean
c = core.vcmod.Median(clip, plane=[0, 1, 1]) if chroma else clip
File "src/cython/vapoursynth.pyx", line 1934, in vapoursynth.Plugin.__getattr__
AttributeError: There is no function named Median---------------------
Some strange conflict with Waifu2xNvk/Waifu2x and TimeCube:
Waifu2xNvk outputs RGB24 if i add TimeCube after it:
# adjusting color space from YUV422P10 to RGBS for vsw2xnvk
clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, matrix_in_s="709", range_s="limited")
# resizing using Waifu2xNvk
# resizing to 1440x1152
clip = core.w2xnvk.Waifu2x(clip=clip, scale=2, model=0, tile_size=32)
# adjusting color space from RGBS to RGB24 for vscube
clip = core.resize.Bicubic(clip=clip, format=vs.RGB24, range_s="limited")
# color adjustment using TimeCube
clip = core.timecube.Cube(clip=clip, cube="/Applications/Hybrid.app/Contents/MacOS/TimeCubeFiles/identity.cube")
# adjusting output color from: RGB24 to YUV422P10 for ProResModel (i422)
clip = core.resize.Bicubic(clip=clip, format=vs.YUV422P10, matrix_s="709", range_s="limited")
# Output
clip.set_output()If i turn off TimeCube, script use RGBS output as expected.
# adjusting color space from YUV422P10 to RGBS for vsw2xnvk
clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, matrix_in_s="709", range_s="limited")
# resizing using Waifu2xNvk
# resizing to 1440x1152
clip = core.w2xnvk.Waifu2x(clip=clip, scale=2, model=0, tile_size=32)
# adjusting output color from: RGBS to YUV422P10 for ProResModel (i422)
clip = core.resize.Bicubic(clip=clip, format=vs.YUV422P10, matrix_s="709", range_s="limited")
# Output
clip.set_output()---------------------
LEVELS and Limiter:
Current settings in Levels is total confusion.
Force RGB with "Planes: 0 and 1,2" makes no sense, because it adjusts R and G,B channels.
"do nothing" makes no sense, because all channels YUV is just incorrect mode for this filter and it only produce green/purple artifacts.
It should be just always RGB or selection between two modes:
- RGB input and Planes: all
- YUV input and Planes: luma only
Here Cary Knoop reported that "You mess up your colors by modifying Y'. If you want to do what I suspect you want to do you should use something like L*a*b* but Vapoursynth does not support this color model." https://forum.doom9.org/showthread.php?p=1923885#post1923885
To be honest i can't understand his point. L*a*b* is different color model and Davinci resolve can use it optionally, but Y gamma is not the same as L gamma in L*a*b* in davinci resolve.
There is YRGB color correction concept used in DaVinci Resolve and some other color grading apps:
![[Image: Z6L91hz.jpg]](https://i.imgur.com/Z6L91hz.jpg)
Levels in/out min/max also works correctly only with RGB input and Planes: all.
I tested 10 bit source video (clip=clip, min_in=64, max_in=940, min_out=64, max_out=940, gamma=0.20) and it is equal to Gamma -1.00 in DaVinci Resolve.
![[Image: KXNTMwz.jpg]](https://i.imgur.com/KXNTMwz.jpg)
VapourSynth Levels filter works similar to any Levels color correction tool in any graphic editor. Here is how Levels filter looks in Photoshop:
![[Image: S5HO0EO.jpg]](https://i.imgur.com/S5HO0EO.jpg)
clip = core.resize.Bicubic(clip=clip, format=vs.RGB30, matrix_in_s="470bg", range_s="limited")
# Color Adjustment
clip = core.std.Levels(clip=clip, min_in=64, max_in=940, min_out=64, max_out=940, gamma=1.1)clip = core.resize.Bicubic(clip=clip, format=vs.YUV422P10, matrix_in_s="470bg", range_s="limited")
# Color Adjustment
clip = core.std.Levels(clip=clip, min_in=64, max_in=940, min_out=64, max_out=940, gamma=1.1, planes=[0])---------------------
Limiter filter also produce same green/purple artifacts as Levels in YUV mode. So it should be operated in the same way as Levels:
- RGB input and Planes: all
- YUV input and Planes: luma only
---------------------
Sort TimeCube LUTs list by name
---------------------
Please bring back disableicons in misc.ini
RE: Esxi running macos For selur - Selur - 10.10.2020
Quote:Please bring back disableicons in misc.iniNot happening, because like you yourself made perfectly clear it breaks the general appearance.
(+ I learned from it and will keep silent about other hidden options for the time being.)
Quote:LEVELS and LimiterNo clue what you think would be a better solution.
All those would have been nice before I made a release,..
-> will look at it when I when the time
Cu Selur
RE: Esxi running macos For selur - shijan - 10.10.2020
Quote:No clue what you think would be a better solution.If explain things simple - LEVELS and Limiter need only force RGB mode and always planes: all
Quote:All those would have been nice before I made a release,..
-> will look at it when I when the time
Those problems where found earlier when i still have some hope that disableicons may back in future. As i told before because no disableicons and no UI resize in new versions of Hybrid i will no more participate in bug reports because i don't use newer versions.
RE: Esxi running macos For selur - Selur - 10.10.2020
Quote:Some strange conflict with Waifu2xNvk/Waifu2x and TimeCube:I don't get what's strange.
a. Waifu2xNvk in both cases returns RGBS
b. in the second case Hybrid converts to RGB24 since it assumes TimeCube supports only supports RGB24, RGB27, RGB30, RGB32 and RGB48 don't see any info on the github page (https://github.com/sekrit-twc/timecube).
-> assuming TimeCube supports RGBs I could add it to the list of compatible color spaces for the filter.
Cu Selur
RE: Esxi running macos For selur - Selur - 10.10.2020
Quote:If explain things simple - LEVELS and Limiter need only force RGB mode and always planes: allI thought RGB was only needed when gamma was used?
RE: Esxi running macos For selur - shijan - 10.10.2020
From conversation with TimeCube developer: "The CUBE always operates on full range (actually 0-1 floats) internally." https://github.com/sekrit-twc/timecube/issues/4#issuecomment-685090200
So i guess it can easy support RGBS input.
(10.10.2020, 14:55)Selur Wrote:Quote:If explain things simple - LEVELS and Limiter need only force RGB mode and always planes: allI thought RGB was only needed when gamma was used?
As it appears after testing - it is not only for Gamma. Everything in LEVELS and Limiter require RGB.
RE: Esxi running macos For selur - Selur - 10.10.2020
Quote:Everything in LEVELS and Limiter require RGB.If they require RGB, then the whole planes-parameter doesn't make sense,...
(may be someone over at https://forum.doom9.org/showthread.php?p=1925443#post1925443 can clear this up)
Cu Selur
RE: Esxi running macos For selur - Selur - 10.10.2020
regarding mClean: seems like the temporalmedian library (https://github.com/dubhater/vapoursynth-temporalmedian) is missing.
-> I'll remove it for MacOs until l33tmeatwad adds it to his list.
Cu Selur
RE: Esxi running macos For selur - shijan - 10.10.2020
(10.10.2020, 15:27)Selur Wrote:Quote:Everything in LEVELS and Limiter require RGB.If they require RGB, then the whole planes-parameter doesn't make sense,...
(may be someone over at https://forum.doom9.org/showthread.php?p=1925443#post1925443 can clear this up)
Cu Selur
Juts try to adjust Levels in/out min/max or Limiter in YUV and you will see same Green/Purple color shift artifact.
RE: Esxi running macos For selur - Selur - 10.10.2020
I tried:
clip = core.std.Levels(clip=clip, min_in=0, max_in=255, min_out=40, max_out=220)
clip = core.std.Levels(clip=clip, min_in=40, max_in=220, min_out=40, max_out=220)
both look as expected,...