View Full Version : Decimation Usage
bsmith1593
31st May 2009, 13:48
I'm having a problem with decimating one of my videos. I assumed that I should leave it 29fps because when I used Decimate(), it looks jerky in some parts. After looking through the frames, however, I see some duplicate frames. The problem with them, however, is that there is no set pattern for when they occur. They could go like, after 7 frames, after 8 frames, after 10 frames, after 6 frames, and then so on. Would I have to do a manual decimation to get rid of the duplicate frames? And if I did, what would the framerate turn out to be? Also, I'm pretty sure I wouldn't get the fade in intro and fade-out exit correctly. Can anybody help me?
And if you need to see the video, I'll be glad to upload it...
manono
31st May 2009, 14:14
And if you need to see the video, I'll be glad to upload it...
Yep, 10 seconds or so with movement would be good.
bsmith1593
31st May 2009, 14:17
Yep, 10 seconds or so with movement would be good.
I'll get right on it.
bsmith1593
31st May 2009, 14:42
I did the first 22 seconds of the video: http://eva.adriels.com/miscellaneous/bthe_Doom9.demuxed.m2v
lansing
31st May 2009, 16:48
your source was double hard telecined, so AnimeIVTC(mode=2, aa=0) should do it
bsmith1593
31st May 2009, 17:44
your source was double hard telecined, so AnimeIVTC(mode=2, aa=0) should do it
What filter do I need to have loaded to use AnimeIVTC?
lansing
31st May 2009, 17:58
What filter do I need to have loaded to use AnimeIVTC?
http://forum.doom9.org/showthread.php?t=138305
the plugin package is available for download in the first post, as long as the document
bsmith1593
31st May 2009, 18:08
http://forum.doom9.org/showthread.php?t=138305
the plugin package is available for download in the first post, as long as the document
Thanks
BigDid
31st May 2009, 18:09
What filter do I need to have loaded to use AnimeIVTC?
Hi,
Read the first post of the thread related to animeivtc(), you will find links to guide, set of filters etc...:
http://forum.doom9.org/showthread.php?t=138305
It is a good habit to use the search function of this forum extensively (and from google also) and find your answers directly; and welcome to the forum :)
Did
bsmith1593
31st May 2009, 19:11
How do I load AnimeIVTC into my .avs? Sorry if this is kind of annoying, but there's a lot of plugins and I don't usually load other .avs's into another .avs.
lansing
31st May 2009, 22:10
ok, first download the plugin package, and then put everything that end up with the dll and avsi extension to your Avisynth plugin directory, if you see avs extension, simply change it to avsi and put those in too, this way it will auto load whenever you call it like a regular filter. If you see some files with multiple SSE version, choose only one, SSE2 for Pentium 4 or under, SSE3 for Core Duo.
Remember to download the AnimeIVTC script separately from here (http://forum.doom9.org/showpost.php?p=1193546&postcount=264)to replace the original. And also, put the AvsRecursion.dll file to your C:\windows\system32 folder
manono
31st May 2009, 23:12
It mostly has 2 duplicate frames in every 15, with some variation because of edits breaking the pattern. Unless you're real particular about it, I'd just go with a simple:
TFM()
TDecimate(Mode=0,Cycle=15,CycleR=2)
bsmith1593
3rd June 2009, 02:11
ok, first download the plugin package, and then put everything that end up with the dll and avsi extension to your Avisynth plugin directory, if you see avs extension, simply change it to avsi and put those in too, this way it will auto load whenever you call it like a regular filter. If you see some files with multiple SSE version, choose only one, SSE2 for Pentium 4 or under, SSE3 for Core Duo.
Remember to download the AnimeIVTC script separately from here (http://forum.doom9.org/showpost.php?p=1193546&postcount=264)to replace the original. And also, put the AvsRecursion.dll file to your C:\windows\system32 folder
What do I actually need to put into MY .avs to make it work though? Here's what I have:
LoadPlugin("C:\PROGRA~1\GORDIA~1\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\GORDIA~1\AVISYN~1\Decomb.dll")
***?***
mpeg2source("C:\PROGRA~1\GORDIA~1\AVISYN~1\D2Vs\Behind These Hazel Eyes.d2v")
AssumeTFF()
AnimeIVTC(mode=2,aa=0)
What do I put where it says "***?***" to make AnimeIVTC work?
lansing
3rd June 2009, 03:44
What do I actually need to put into MY .avs to make it work though? Here's what I have:
LoadPlugin("C:\PROGRA~1\GORDIA~1\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\GORDIA~1\AVISYN~1\Decomb.dll")
***?***
mpeg2source("C:\PROGRA~1\GORDIA~1\AVISYN~1\D2Vs\Behind These Hazel Eyes.d2v")
AssumeTFF()
AnimeIVTC(mode=2,aa=0)
What do I put where it says "***?***" to make AnimeIVTC work?
ok, I see your problem. You'd installed an old version of AviSynth from that outdated Gordian Knot package, which is about 5 years old.
That old Avisynth version doesn't have the auto load function, so what you need to do is to uninstall the AviSynth part from the package if you can, or just uninstall the whole package because it's way too old.
Then download and install this (http://sourceforge.net/project/showfiles.php?group_id=57023&package_id=105994&release_id=649730)newer version. Then put all your filter to the directory's plugins folder.
When writing your script, you don't have to load the filter, they will auto load, so the basic script will be like this
mpeg2source("xxx.d2v",info=3)
colormatrix(hints=true) #this will automatically do color correction if needed
AnimeIVTC(mode=2, aa=0) #aa is anti-aliasing, [0-4], 0=off, turn this off if you don't need it
bsmith1593
3rd June 2009, 03:52
ok, I see your problem. You'd installed an old version of AviSynth from that outdated Gordian Knot package, which is about 5 years old.
That old Avisynth version doesn't have the auto load function, so what you need to do is to uninstall the AviSynth part from the package if you can, or just uninstall the whole package because it's way too old.
Then download and install this (http://sourceforge.net/project/showfiles.php?group_id=57023&package_id=105994&release_id=649730)newer version. Then put all your filter to the directory's plugins folder.
When writing your script, you don't have to load the filter, they will auto load, so the basic script will be like this
mpeg2source("xxx.d2v",info=3)
colormatrix(hints=true) #this will automatically do color correction if needed
AnimeIVTC(mode=2, aa=0) #aa is anti-aliasing, [0-4], 0=off, turn this off if you don't need it
Okay, I'll try doing this tomorrow. Yeah, I had installed GordianKnot and its AviSynth. I installed the newest version the other day, but I installed it to the plugin directory in GordianKnot's folder, so that probably messed it up, I'll uninstall both, reinstall GK, delete its AviSynthPlugins folder, install the new AviSynth, and then re-extract all of the files from the requirements package and put them in the right places. I'll tell y'all my results when I eventually get there :). Thanks for being patient.
bsmith1593
4th June 2009, 05:34
I got all the files from the Requirements package moved to my plugins folder. Filters autoload like they should. Now I get this error in VirtualDub:
http://i243.photobucket.com/albums/ff257/Benjamin1593/virtualdub_error.png.
The line it's talking about tries to load "ColorYUY2.dll", which is not in the package :s. Oh, and some files have more than one version, like v.32 and v.54. Should I delete the older ones?
bsmith1593
4th June 2009, 14:25
Okay, I managed to solve the last error by finding ColorYUY2.dll and putting it in the plugins folder, but now I get these:
http://i243.photobucket.com/albums/ff257/Benjamin1593/virtualdub_error2.png
http://i243.photobucket.com/albums/ff257/Benjamin1593/virtualdub_error3.png
Guest
4th June 2009, 14:27
Can't you look at the ColorRGB.auf.avsi script and figure out the problem?
If not, post that script in its entirety.
bsmith1593
4th June 2009, 18:05
Can't you look at the ColorRGB.auf.avsi script and figure out the problem?
If not, post that script in its entirety.
I'm not very experienced in AviSynth except for scripts of others I've used and minor stuff. I know what it says is wrong, but I don't know how to fix it, several files have similar code and that's how they're supposed to work I guess.
http://eva.adriels.com/miscellaneous/ColorRGB.auf.avsi
info = LoadPlugin("ColorYUY2.dll")
filt = AvisynthFilters("RGB color adjust filter", info)
func = filt.AddFunction("ColorYUY2")
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddValue(0.0)
func.AddTrack("R(gain)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("R(offs)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("R(gamm)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("R(cont)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("G(gain)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("G(offs)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("G(gamm)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("G(cont)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("B(gain)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("B(offs)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("B(gamm)", 0, -256, 256, -256.0, 256.0)
func.AddTrack("B(cont)", 0, -256, 256, -256.0, 256.0)
func.AddArray("levels", 0, "", "TV->PC", "PC->TV", "PC->TV.Y", "TV->PC.Y", "709->601", "709TV->601PC", "709PC->601TV")
func.AddArray("opt", 0, "", "coring", "1-254")
func.AddArray("matrix", 0, "", "rec709", "rec601s", "rec709s")
func.AddTrack("debug", 0, 0, 5)
func.AddTrack("hscale", 70, 10, 100)
func.AddTrack("X", 0, -1, 1280)
func.AddTrack("Y", 0, -1, 960)
func.AddArray("interpolation", 0, "", "411->422", "411->422R", "411->444", "411->444R", "420->422", "420->444", "422->444")
func.AddArray("output", 0, "", "YUY2", "YV12", "I420", "RGB32", "RGB24")
return filt
Holy carp... after some googling this seems to be some kinda script for AviUtil, not AviSynth.
That stuff is ANCIENT. I doubt you'll get this to run in any way...
np: Sin Fang Bous - Poi Rot (Clangour)
lansing
4th June 2009, 18:19
I got all the files from the Requirements package moved to my plugins folder. Filters autoload like they should. Now I get this error in VirtualDub:
http://i243.photobucket.com/albums/ff257/Benjamin1593/virtualdub_error.png.
The line it's talking about tries to load "ColorYUY2.dll", which is not in the package :s. Oh, and some files have more than one version, like v.32 and v.54. Should I delete the older ones?
I don't recall the AnimeIVTC requirement package have that file. Post your entire script, meanwhile, for a trial and error process, delete that ColorRGB.auf.avsi file and see what happen.
bsmith1593
4th June 2009, 18:37
I don't recall the AnimeIVTC requirement package have that file. Post your entire script, meanwhile, for a trial and error process, delete that ColorRGB.auf.avsi file and see what happen.
Wow, the same kind of code is in all the .auf.avsi and .vdf.avsi files. I delete all of those and I moved the files out that had stuff like LoadPlugin("c:avisynthfilters\***.dll").
So far VirtualDub has almost opened, but freezes.
lansing
4th June 2009, 19:32
Wow, the same kind of code is in all the .auf.avsi and .vdf.avsi files. I delete all of those and I moved the files out that had stuff like LoadPlugin("c:avisynthfilters\***.dll").
So far VirtualDub has almost opened, but freezes.
:script:
bsmith1593
8th June 2009, 15:48
My script that I load into vdub:
mpeg2source("C:\PROGRA~1\GORDIA~1\D2Vs\Behind These Hazel Eyes.d2v")
AssumeTFF()
AnimeIVTC(mode=2,aa=0)
When I open vdub and open this file, it gets so far and then just freezes:
http://i243.photobucket.com/albums/ff257/Benjamin1593/virtualdub_freeze.png
I think the AnimeIVTC function is what's freezing it, because vdub opens w/o freezing on another one of my scripts that doesn't use it:
mpeg2source("C:\PROGRA~1\GORDIA~1\D2Vs\Since U Been Gone.d2v")
AssumeTFF()
Telecide(guide=1,vthresh=80,ovr="subg.tel")
Decimate()
lansing
8th June 2009, 17:03
well, I think your computer was too slow for the script
thetoof
8th June 2009, 19:22
Lower precision= to make it faster.
bsmith1593
9th June 2009, 01:59
It mostly has 2 duplicate frames in every 15, with some variation because of edits breaking the pattern. Unless you're real particular about it, I'd just go with a simple:
TFM()
TDecimate(Mode=0,Cycle=15,CycleR=2)
Well, since my computer seems to be way to slow for AnimeIVTC, I'll try this method combined with Decomb.dll's decimate override to get the missed duplicate frames.
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.