QTGMC Deinterlacing Script (v3.384)

[Boulder]

Originally Posted by LigH

In the possibly quite suddenly changing directions and distances between each motion step?

I'd say that is the problem. Or that the actual motion occurs in every other frame or so.

[colours]

Originally Posted by LigH

If the problem is not in the duplicates, where is it then? In the possibly quite suddenly changing directions and distances between each motion step?

Pretty much, yeah.

Here's a quote from Dark_Shikari (emphasis mine):

There are also a number of problems from an encoder perspective in addition to the issues of the transform itself. Motion is rarely smooth in animated content; since animation is usually done at a much lower framerate than the actual video, an object may alternate motion and non-motion, resulting in motion search being unable to use temporal predictors. Furthermore, when that object jumps 20 pixels to the right, it’s inherently hard to find where it went using normal methods.

[Livesms]

Tried to run QTGMC and get such error (att).
OS: Windows 7 x64 SP1
Avisynth 2.6

All plugins and libs put into the right place.

[Livesms]

Solved CACHE_GETCHILD_CACHE_MODE error

Added "SetMTmode(2)" at the very first line of avs Script ...

[LouieChuckyMerry]

I've found that adding "QTGMC(InputType=1)" in this script as so:

LoadPlugin("F:\[0]StandAloneApps\MeGUI-2500(core)2443(data)0.3.5(libs)[Portable]\tools\DGIndexNV\DGDecodeNV.dll")
DGSource("SourcePath")
### Deinterlace-Match Fields-Decimate-Fix Line Doubled Fields ###
LoadPlugin("F:\[0]StandAloneApps\MeGUI-2500(core)2443(data)0.3.5(libs)[Portable]\tools\avisynth_plugin\TIVTC.dll")
Function FieldMatch(Clip C) {
  Global PP = C.DuplicateFrame(0)
  Global CC = C
  Global NN = C.DeleteFrame(0)
  P2 = PP.SeparateFields()
  C2 = CC.SeparateFields()
  N2 = NN.SeparateFields()
  Global PC = Interleave(P2.SelectEven(),C2.SelectOdd()).Weave()
  Global CP = Interleave(C2.SelectEven(),P2.SelectOdd()).Weave()
  Global CN = Interleave(C2.SelectEven(),N2.SelectOdd()).Weave()
  Global NC = Interleave(N2.SelectEven(),C2.SelectOdd()).Weave()
  Global Deint = QTGMC(CC).SelectEven()
  Return ScriptClip(CC, \
    "!CC.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? CC : " + \
    "!NN.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? NN : " + \
    "!CN.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? CN : " + \
    "!NC.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? NC : " + \
    "!PP.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? PP : " + \
    "!CP.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? CP : " + \
    "!PC.IsCombedTIVTC(CThresh=12,Chroma=True,BlockX=16,BlockY=32) ? PC : Deint")
}
TFM(Order=-1,Mode=5,PP=2,Clip2=FieldMatch(),Slow=2,MChroma=False,Ubsco=False,CThresh=12,Chroma=True)
TDecimate(Mode=1)
NNEDI3(Field=-2)
Merge(SelectEven(),SelectOdd())
### Reduce Shimmering ###
QTGMC(InputType=1)
### Stabilize ###
Stab(Mirror=15)
### Crop ###
Crop(8,0,-8,0)
### Gibbs Noise Block ###
Edge=MT_Edge("prewitt",ThY1=20,ThY2=40).RemoveGrain(17)
Mask=MT_Logic(Edge.MT_Expand().MT_Expand().MT_Expand().MT_Expand(),Edge.MT_Inflate().MT_Inpand(),"xor")
MT_Merge(DFTTest(),Mask,Luma=True)
### Overall Temporal Denoise ###
SMDegrain(TR=2,ThSAD=500,ContraSharp=True,RefineMotion=True,Plane=0,Lsb=True,Lsb_Out=True,PreFilter=2,Chroma=False)
### Resize ###
LinearResize(640,480,Lsb_In=True,Lsb_Out=True)
### Darken-Thin Lines ###
F=DitherPost(Mode=-1)
S=F.FastLineDarkenMod(Strength=20,Prot=6).aWarpSharp2(Blur=4,Type=1,Depth=3,Chroma=2)
D=MT_MakeDiff(S,F).Dither_Convert_8_To_16()
Dither_Add16(Last,D,Dif=True,U=2,V=2)
### Deband ###
GradFun3(thR=0.55,SMode=2,Lsb_In=True,Lsb=True,StaticNoise=True)
### Preview Source OR Send 16-bit Output To x264 10-bit ###
# DitherPost()
Dither_Out()

does wonders to minimize the shimmering. Really, the difference is amazing. I was hoping, however, that somebody with an intimate understanding of QTGMC could recommend tweaks to the QTGMC line to increase encoding speed, even if there's a loss of quality. That is, does anyone know any QTGMC parameters I can adjust in conjunction with "InputType=1" to increase the encoding speed with regards to shimmering? I've read the .html file but presently it's above my head. Thanks for any help .

[luigizaninoni]

QTGMC(InputType=1,Preset="Medium")
you could even try preset "fast" or "faster", but quality will suffer a bit

If you do not indicate a preset, qtgmc by default uses "slow"

[Taurus]

Why not try a faster preset on QTGMC?
Maybe "Preset=Medium"?
The default preset is IMHO "Preset=Slower".
Changing to "medium", "fast" or "faster" should give you a nice speedboost.

[LouieChuckyMerry]

Originally Posted by luigizaninoni

QTGMC(InputType=1,Preset="Medium")
you could even try preset "fast" or "faster", but quality will suffer a bit

If you do not indicate a preset, qtgmc by default uses "slow"

Originally Posted by Taurus

Why not try a faster preset on QTGMC?
Maybe "Preset=Medium"?
The default preset is IMHO "Preset=Slower".
Changing to "medium", "fast" or "faster" should give you a nice speedboost.

Thank you both for your answers . So obvious it didn't even cross my mind...

[TheSkiller]

real.finder,

using your mod I noticed YUY2 works only as long as I don't use SourceMatch.

If I try to use SourceMatch with YUY2 I get this error message:

"mt_lutxy : unsupported colorspace. masktools only support planar colorspaces (YV12, YV16, YV24)"

Lines 928 and 585.

[real.finder]

Originally Posted by TheSkiller

real.finder,

using your mod I noticed YUY2 works only as long as I don't use SourceMatch.

If I try to use SourceMatch with YUY2 I get this error message:

"mt_lutxy : unsupported colorspace. masktools only support planar colorspaces (YV12, YV16, YV24)"

Lines 928 and 585.

2015 9 10 http://pastebin.com/7zZV7Q3h and in Attachments

2016-01-19 http://pastebin.com/MgZ2pMJm

2016-04-11 http://pastebin.com/TV5YRwyu and in Attachments

2016-12-06 http://pastebin.com/RqB1sf9X and in Attachments

2017-01-01 http://pastebin.com/HAD74H7Z and in Attachments (QTGMC 3.352s)

2017-02-06 http://pastebin.com/2mnirSiN and in Attachments (QTGMC 3.353s)

new updates will be here https://forum.doom9.org/showpost.php...postcount=2277 In future

[TheSkiller]

Thanks for the quick fix, much appreciated.

[Music Fan]

Virtual Dub can't open scripts with v3.33 on my pc, the last version working is 3.32, what could be the reason ?

[LigH]

And what could be the verbose error message?

[Music Fan]

There is no detail, just "unable to open file".

[LigH]

Very strange. That would mean for me that VirtualDub does not understand the whole script anymore just because QTGMC 3.33 is used?! If it could just not load a plugin, I would expect a different error ... More checks required.

Does AvsMeter (32 bit) report a more meaningful error? Did you save the script in Unicode format or "ANSI" (Windows CP1252)? Will AvsPmod or VirtualDubMod open the script and display a preview?

[StainlessS]

Also try MPC-HC or some other media player that normally opens avs files.

And try temporary empty of Plugins32 dir in Vdub, (I've had problems before with avs ffms2 and VDub FFInputDriver.vdplugin dll's
interfering with each other [some kind of change to the dll directories search order]).

[Music Fan]

I tried 3.33s with MPC-HC and AvsMeter and I get this message ;

MDegrain1 does not have a named argument "lsb" (QTGMC.avsi, line 571)

I wonder why VDub does not show this message.
edit : If I empty Vdub's Plugins32 dir (which contains FFInputDriver.vdplugin and its dll's), I get the same message than with MPC-HC and AvsMeter !

By the way, is it possible to export error messages in text ?

[LigH]

I guess AvsMeter would report an error as text in the console.

It seems that you put several different plugins and import scripts into your autoload directory, and now accidently a wrong one is used...

MDegrain1() is a function in mvtools2.dll; having this DLL in the correct (minimum) version is very important for QTGMC.

# --- REQUIREMENTS ---
#
# Input colorspaces: YV12, YUY2
#
# Core plugins:
#	MVTools2 (2.5.11.2 or above)
#	MaskTools v2 (recommend 2.0a45 or above. Must use the 2.5 version with YUY2)
#	NNEDI3 (recommend 0.9.4 or above for speed)
#	RemoveGrain + Repair (several versions of this plugin, use the SSE2 dlls from the file called "RemoveGrain-1.0.rar". Don't use the SSE3 versions )
#   SSE2Tools for YUY2 support (from the earlier 0.9 version of RemoveGrain, use only SSE2Tools.dll from this version. Don't use the SSE3 version)

If you are using AviSynth MT, you will probably need one of the last fixes of ViT's mod or b1 mod by TP7 for MaskTools2 to avoid cache mode exceptions (MVTools2 is a different plugin).

[Music Fan]

I don't use MT version and I have mvtools 2.5.11.3, thus I guess it should work.

I guess AvsMeter would report an error as text in the console.

It can't be copied, and no log is created in this case, but I can propose to AvsMeter's developer to add it.

[LigH]

Ah, your QTGMC version is not 3.33, but probably 3.33s; an "lsb" parameter was introduced rather late. But mvtools-v2.5.11.3.zip offered by Fizick does not document it. There must be another one. Fizick's mvtools2.dll is from 2011-09-19, but I have one from 2012-04-02 too. Both don't seem to support a parameter "lsb", so there seems to be a more recent source.