AnimeIVTC() Help

[MegaTonTerror]

I am going insane trying to use AnimeIVTC(). If it's not one thing it's another. Doom9 is just confusing me. What all avsi and .dll is needed. I am to a point where Resize: Height must be bigger than 4 and I can't solve this error. I see others example scripts using AnimeIVTC and they don't have near as much as what I am asked to have in my script. I've searched for guides and I can't find anything that helps. Can someone guide me in the right direction please.

[poisondeathray]

The 1st post gives you links to a guide , documentation, .dll's , avsi packages, etc..

http://forum.doom9.org/showthread.php?t=138305

If you want more specific help you have to provide more information i.e. your script and exact error message. Although the author is no longer currently active, it would be a good idea to post at Doom9 regarding AnimeIVTC

[MegaTonTerror]

I spent hours at that link and I can't figure it out. sorry =(

After starting with just this
AnimeIVTC(MT=false,mode=2,aa=4,killcomb=2,pass=2)
================================================== ===========================
When I loaded just the AnimeIVTC I got the error " no function for mvanalyse. After I added mvanalyze i was told I had a cdeblend error etc... etc... etc...

mvanalyse(truemotion=true)
Cdeblend(omode=4)
ReduceFlicker(strength=2)
warpsharp()
toon()
limitedsharpenfaster(Smode=3,Smethod=2,Lmode=1,ss_ x=1.0,ss_y=1.0,strength=100,soothe=false)

Here is the link to the error I am getting. I am trying to encode The Transformers from DVD.
http://farm4.static.flickr.com/3419/3373469419_1c0efc6e9a_o.jpg

[manono]

Is there something peculiar to Transformers that the standard methods won't work? In my opinion there's no need for such a complex and very slow script in 99% of the cases. Do you have a small sample you'd like to upload somewhere so we could have a look?

[MegaTonTerror]

It's hybrid footage. I wanted to try out AnimeIVTC thru Megui to see if I can get rid of the jerky video I got using this script via script creator in Megui.
=================================================
# Set DAR in encoder to 144 : 79. The following line is for automatic signalling
global MeGUI_darx = 144
global MeGUI_dary = 79
DGDecode_mpeg2source("C:\Transformers S1 D1\VTS_04_1.d2v", cpu=4, info=3)
ColorMatrix(hints=true, interlaced=true, threads=0)
tfm(order=-1).tdecimate(hybrid=1)
#crop
Lanczos4Resize(720,480) # Lanczos4 (Sharp)
FluxSmoothST(7,7) # Medium Noise
===============================================

I am running encode now with this script. I am a novice so I am sure my script is not in the proper order or I have to much.
DGDecode_mpeg2source("C:\Transformers S1 D1\yo.d2v", cpu=4, info=3)
ColorMatrix(hints=true, interlaced=true, threads=0)
tfm(order=-1).tdecimate(hybrid=1)
AssumeTFF()
crop(6,0,-6,0)
toon()
#tdeint(mode=0,order=-1,full=true,field=-1,mthreshL=-3,mthreshC=-3,type=4,sharp=true,mtnmode=2,cthresh=8,blim=100,m etric=0,slow=2)
hqdn3d()
#unfilter(-5, -5)
Spline16Resize(720,480)
#dfttest()
fastlinedarken()
LSFmod(Smode=3,Smethod=2,Lmode=1,ss_x=1.0,ss_y=1.0 ,strength=100,soothe=false)
__film = last
__t0 = __film.trim(792, 31908)
__t0

[manono]

It's hybrid footage.

How do you know? I'm not saying it's not, but how do you know since you admit to being a novice? I sure wouldn't trust any MeGUI analysis to tell me. Personally, I think you're going a little bit overboard with the filters, another sign of a rookie. Does Transformers really look so bad that it needs all that?

[MegaTonTerror]

I just say I am a novice cause I don't understand a lot of what's going on. I basically read up on what I think needs improved and piece a script together from that. DGIndex switches from interlace/progressive and tells me hybrid. Right now I am trying out RemoveGrain and RemoveDirt without using dfttest. Those 2 (when loaded in Vdubmod) show better results of getting rid of a lot of grain I can see. The only problem is the color loss. here are pics of what I am talking about.
http://farm4.static.flickr.com/3579/3374894832_55e5d80c7e_o.jpg This is from the VOB itself.
http://farm4.static.flickr.com/3634/3374894914_c65c171a90_o.jpg This is from my script with removegrain and remove dirt.

Using Megui's script creator I came up with this. The main thing is to get rid of the grain and not have any jerky and flickering video.

DGDecode_mpeg2source("C:\Transformers S1 D1\VTS_04_1.d2v", cpu=4, info=3)
ColorMatrix(hints=true, interlaced=true, threads=0)
tfm(order=-1).tdecimate(mode=1,hybrid=1)
crop( 6, 0, -4, 0)

Spline36Resize(720,480) # Spline36 (Neutral)
Convolution3D("movielq") # Heavy Noise

__film = last
__t0 = __film.trim(791, 31908)
__t0

[manono]

DGIndex switches from interlace/progressive and tells me hybrid.

And that's what lead you to conclude it's a hybrid? DGIndex told you no such thing. Again, while it still could very well be a hybrid, that's no evidence at all. For all you know from the DGIndex information, it could just need a simple IVTC. For this sort of thing the eyes are your only guide. Make the D2V using Honor Pulldown Flags and have a look without any other filtering. If you only ever see the telltale 3 progressive/2 interlaced pattern of frames, then it only needs an IVTC. Of course, figuring this out is made more difficult when it's anime because of all the duplicate frames.

I do see a lot of grain there, so you're right about that. As for the colors, I don't know if it's from the degraining or other filtering or not, as I don't use those filters much myself. Others would know better about that. One thing for sure, though, is that you don't need the ColorMatrix filter, and if it's actually being used that could account for the dulling of the colors.

But as I said earlier, a sample would be nice. Unless and until one shows up proving it's a hybrid, all anyone can do is guess.

And since you said DGIndex shows at least some of it to be Film/Progressive, your TFM should go:

TFM(D2V="C:\Path\To\Video.d2v")