Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
14th March 2019, 11:31 | #841 | Link |
Noob
Join Date: Mar 2017
Posts: 221
|
can someone help me in removing ghosting/blending in this clip ? i dont understand exactly what type of artifact is this. i tried plain qtgmc + srestore and
QTGMC(Preset = "Slower", InputType=0, NoiseProcess=0) srestore(mode=4, thresh=44) but cant fix it completely or reduce it to minimum. heres the sample source and encoded clip https://www34.zippyshare.com/v/wpy3Lzha/file.html |
14th March 2019, 20:11 | #842 | Link |
Registered User
Join Date: Feb 2017
Posts: 145
|
I didn't look, but I can immediately recommend not using QTGMC as your detection clip for sretore. The best results come from simpler bobs. However, you can still using the frames generated by QTGMC by running both at the same time like so:
FFmpegSource2("....") d=TDeint(mode=1, type=2, full=false, tryWeave=true) QTGMC(Preset = "Slower", InputType=0, NoiseProcess=0) srestore(frate=23.976, speed=-1, thresh=1, dclip=d, mode=4) Yes, these are aggressive srestore parameters, so I'd suggest bumping up the thresh value until the filter stops working well. You might want to try mode=-4 also, since sometimes srestore will work better by analyzing the chroma instead of the luma. It's kind of an art using srestore since no one method seems to work best all the time. |
14th March 2019, 20:26 | #843 | Link |
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,797
|
I've also had better results sometimes by downscaling the analysis clip with ReduceBy2() after bobbing it.
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... |
15th March 2019, 11:27 | #844 | Link |
Pig on the wing
Join Date: Mar 2002
Location: Finland
Posts: 5,797
|
Here's one strange case I've been trying to fix.
https://drive.google.com/open?id=1oI...PD71KtYKHKyZsp (original clip) https://drive.google.com/open?id=1O1...dNAplNFP5oUVnw (bobbed clip) I tried figuring out a pattern with a that bobbed clip and it looks like 4 good frames, 2 blended ones. Then if you look closer, it looks more like 3 good frames and 3 blended ones where the middle one is a real blended mess and the surrounding frames much less so Is it something that cannot be handled by SRestore as the origins seem to be an interlaced clip?
__________________
And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... |
4th August 2019, 21:49 | #845 | Link |
Registered User
Join Date: Feb 2003
Posts: 124
|
I'm using the latest version of the script (2.796) with the latest avisynth+ x64 and all the latest necessary plugins (masktools, etc) and I'm getting an error complaining "I don't know what 'avsplusversionnumber' means". I tried usiing 2.795 but with that one it complains about not knowing what 'dm' means. I looked at the script and both things are specified so it should 'know' what it means. What am i doing wrong? I can't find any mention of this in the previous entries of this thread.
|
4th August 2019, 21:56 | #846 | Link | |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
Quote:
__________________
See My Avisynth Stuff |
|
4th August 2019, 23:03 | #847 | Link | |
Registered User
Join Date: Mar 2011
Posts: 4,877
|
Quote:
Here's what you're missing if you don't want to download the latest SMDegrain, you could add it to the end of the SRestore script if you have it in the auto-loading Avisynth plugins folder, or you could create a separate avsi file etc. Code:
function IsAvsNeo(){ FindStr(VersionString, "AviSynth Neo") != 0 } function IsAvsPlus(){ FindStr(VersionString, "AviSynth+") != 0 || IsAvsNeo } function AvsPlusVersionNumber(){ \ IsAvsNeo ? eval(MidStr(VersionString(),20,4)) : \ IsAvsPlus ? val(MidStr(VersionString(),17,4)) : 0 } Last edited by hello_hello; 7th August 2019 at 17:13. |
|
5th August 2019, 19:29 | #848 | Link |
Registered User
Join Date: Feb 2003
Posts: 124
|
Thanks for letting me know. I thought I had latest one (smdegrain_v3.1.2.100s) but guess not. Downloading SMDegrain v3.1.2.104s solved the 'avsplusversionnumber' error but now it's throwing up "I don't know what 'dm' means" error. Anyone know what that one's about?
|
5th August 2019, 20:03 | #849 | Link | |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
Quote:
__________________
See My Avisynth Stuff |
|
6th August 2019, 01:40 | #851 | Link |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
work fine for me
edit: are you using MT Prefetch() in avs+? if yes then that known thing that mt not work (yet) with real-time functions like srestore
__________________
See My Avisynth Stuff Last edited by real.finder; 6th August 2019 at 02:02. |
6th August 2019, 05:46 | #853 | Link | |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
Quote:
__________________
See My Avisynth Stuff |
|
16th March 2020, 00:08 | #854 | Link |
Registered User
Join Date: Aug 2018
Posts: 25
|
PAL with blends.. hard to deal with..
hi folks I came across two PAL dvds, both with blends.. the first one is progessive but I tried many filters can't fix it, seems to be hard double-telecine? the second is interlaced with more blends, even more messy the first one is TFF, second one BFF here're the samples, thanks a lot http://www.mediafire.com/file/biwgkp...PAL_blends.rar |
24th May 2020, 23:40 | #855 | Link |
Registered User
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,099
|
Follow-up to this old post:
https://forum.doom9.org/showthread.p...34#post1844334 So using an old MaskTools version together with a new SRestore script will not work, but what about the other way around? Using an old SRestore script (v 2.7e) with he latest pinterf MaskTools2 version? Is this safe if I stay away from Hi BitDepth and Hi Color? |
25th May 2020, 04:48 | #856 | Link | |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
Quote:
anyway it should work, new SRestore should also work with old masktools2 in avs 2.5 but new SRestore in avs 2.6 or avs+ will need new masktools2
__________________
See My Avisynth Stuff |
|
25th May 2020, 12:39 | #857 | Link |
Registered User
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,099
|
The reason because I want to keep the old SRestore script is that I use at least three different computers for my video conversions, and one of them is my ancient WinXP machine with with a P3 Coppermine CPU (no SSE2), and I do not want to maintain different plugin folders with different versions of the filters. Least common denominator, and this means using older plugins from the pre "modernization effort" era. And this old SRestore version 2.7e has always worked well for me... |
25th May 2020, 12:53 | #858 | Link | |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
Quote:
anyway you have others like 2.7i or older like 2.7g, should work with older masktools
__________________
See My Avisynth Stuff |
|
31st May 2020, 14:47 | #859 | Link | ||
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,589
|
Quote:
Quote:
it's the last version but build with /arch:SSE
__________________
See My Avisynth Stuff |
||
31st May 2020, 17:47 | #860 | Link |
Registered User
Join Date: Sep 2003
Location: Berlin, Germany
Posts: 3,099
|
Thanks for this build, but it does not work on my ancient WinXP machine.
I removed the old mt-masktools.dll and copied your build of masktools2 into my plugin folder. Then replaced the old SRestore.avsi with your latest version, then I also needed the file Zs_RF_Shared.avsi in the plugins folder. When calling SRestore I got this error: "There is no function MT_Makediff" So it looks like my Non-SSE2 machine does not like this build... |
Tags |
mrestore, srestore |
Thread Tools | Search this Thread |
Display Modes | |
|
|