AnimeIVTC() - All-in-one solution for interlacing and blending - v2.00 [2010-01-07]

[Haunther]

[ Info sul file ]

Nome: Anime.mp4
Data: 16/03/2009 18:38:56
Dimensione: 236,634,347 bytes (225.672 MB)

[ Info generiche ]

Durata: 00:23:05 (1385.108333 s)
Tipo di contenitore: MP4/MOV
Major brand: ISO Base Media version 1
Data creazione: 15/03/2009 01:14:54 UTC
Data modifica: 15/03/2009 01:14:54 UTC
Streams totali: 2
Tipo stream n. 1: video (avc1) {GPAC ISO Video Handler}
Tipo stream n. 2: audio (AAC Low complexity) {default}
Audio streams: 1

[ Dati rilevanti ]

Risoluzione: 1280 x 720
Larghezza: multipla di 32
Altezza: multipla di 16
DRF medio: 20.851056
Deviazione standard: 2.261513
Media pesata dev. std.: 2.195549

[ Traccia video ]

Codec: avc1
Risoluzione: 1280 x 720
Frame aspect ratio: 16:9 = 1.777777
Pixel aspect ratio: 1:1 = 1
Display aspect ratio: 16:9 = 1.777777
Framerate: 28.323090 fps
Frames totali: 39223
Bitrate: 1237.732802 kbps

[ Traccia audio ]

Codec: AAC Low complexity
Bitrate: 125.879004 kbps
Canali: 2
Frequenza: 48000 Hz

[ Info sulla codifica H.264 ]

User data: x264
User data: core 59 r839+10 8134e7a
User data: H.264/MPEG-4 AVC codec
User data: Copyleft 2003-2008
User data: http://www.videolan.org/x264.html
User data: cabac=1
User data: ref=5
User data: deblock=1:-1:-1
User data: analyse=0x3:0x113
User data: me=umh
User data: subme=6
User data: brdo=1
User data: mixed_ref=1
User data: me_range=16
User data: chroma_me=1
User data: trellis=1
User data: 8x8dct=1
User data: cqm=1
User data: deadzone=21,11
User data: chroma_qp_offset=0
User data: threads=6
User data: nr=0
User data: decimate=1
User data: mbaff=0
User data: bframes=8
User data: b_pyramid=0
User data: b_adapt=1
User data: b_bias=0
User data: direct=1
User data: wpredb=1
User data: bime=1
User data: keyint=240
User data: keyint_min=1
User data: scenecut=60(pre)
User data: rc=crf
User data: crf=22.0
User data: rceq='blurCplx^(1-qComp)'
User data: qcomp=0.60
User data: qpmin=8
User data: qpmax=41
User data: qpstep=8
User data: vbv_maxrate=40000
User data: vbv_bufsize=8192
User data: ip_ratio=1.40
User data: pb_ratio=1.30
User data: aq=0
User data: x264
User data: core 59 r839 27c3071
User data: H.264/MPEG-4 AVC codec
User data: Copyleft 2003-2008
User data: http://www.videolan.org/x264.html
User data: cabac=1
User data: ref=5
User data: deblock=1:-1:-1
User data: analyse=0x3:0x113
User data: me=umh
User data: subme=6
User data: brdo=1
User data: mixed_ref=1
User data: me_range=16
User data: chroma_me=1
User data: trellis=1
User data: 8x8dct=1
User data: cqm=1
User data: deadzone=21,11
User data: chroma_qp_offset=0
User data: threads=6
User data: nr=0
User data: decimate=1
User data: mbaff=0
User data: bframes=8
User data: b_pyramid=0
User data: b_adapt=1
User data: b_bias=0
User data: direct=1
User data: wpredb=1
User data: bime=1
User data: keyint=240
User data: keyint_min=1
User data: scenecut=60(pre)
User data: rc=crf
User data: crf=22.0
User data: rceq='blurCplx^(1-qComp)'
User data: qcomp=0.60
User data: qpmin=8
User data: qpmax=41
User data: qpstep=8
User data: vbv_maxrate=40000
User data: vbv_bufsize=8192
User data: ip_ratio=1.40
User data: pb_ratio=1.30
User data: aq=0
SPS id: 0
Profile: High@L4.1
Num ref frames: 5
Chroma format idc: YUV 4:2:0
PPS id: 0 (SPS: 0)
Codifica entropica: CABAC
Weighted prediction: No
Weighted bipred idc: B slices - implicit weighted prediction
8x8dct: Sì
Frames totali: 39223
Drop/delay frames: 0
Frames corrotti: 0

P-slices: 12456 ( 31.757 %) ########
B-slices: 26340 ( 67.154 %) #################
I-slices: 427 ( 1.089 %)
SP-slices: 0 ( 0.000 %)
SI-slices: 0 ( 0.000 %)

[ Analisi DRF ]

DRF medio: 20.851056
Deviazione standard: 2.261513
DRF max: 30

DRF<9: 0 ( 0.000 %)
DRF=9: 1 ( 0.003 %)
DRF=10: 2 ( 0.005 %)
DRF=11: 2 ( 0.005 %)
DRF=12: 8 ( 0.020 %)
DRF=13: 27 ( 0.069 %)
DRF=14: 54 ( 0.138 %)
DRF=15: 179 ( 0.456 %)
DRF=16: 575 ( 1.466 %)
DRF=17: 1212 ( 3.090 %) #
DRF=18: 3527 ( 8.992 %) ##
DRF=19: 4804 ( 12.248 %) ###
DRF=20: 7511 ( 19.149 %) #####
DRF=21: 7574 ( 19.310 %) #####
DRF=22: 4863 ( 12.398 %) ###
DRF=23: 4208 ( 10.728 %) ###
DRF=24: 2266 ( 5.777 %) #
DRF=25: 1319 ( 3.363 %) #
DRF=26: 698 ( 1.780 %)
DRF=27: 268 ( 0.683 %)
DRF=28: 93 ( 0.237 %)
DRF=29: 15 ( 0.038 %)
DRF=30: 17 ( 0.043 %)
DRF>30: 0 ( 0.000 %)

P-slices DRF medio: 20.288455
P-slices deviazione std.: 2.441883
P-slices DRF max: 28

B-slices DRF medio: 21.165907
B-slices deviazione std.: 2.081766
B-slices DRF max: 30

I-slices DRF medio: 17.840749
I-slices deviazione std.: 2.028584
I-slices DRF max: 25

[ Profile compliancy ]

Profilo da testare: MTK PAL 6000
Risoluzione: 1280 x 720 > 720 x 576
Framerate: Ok
Buffer underflow: 00:01:27 (frame 2461)
Buffer underflow: 00:01:27 (frame 2472)
Buffer underflow: 00:01:28 (frame 2490)


Enough?

[Guest]

@Haunther

That's a torrent file. We can't help you with that per forum rule 6. Please do not post about such things in the future.

@all

Do not reply to this issue per forum rule 6.

[Haunther]

Is not a Torrent File -___-

[Leak]

Is not a Torrent File -___-
Whatever it is, you sure as heck downloaded it somewhere, unless Italy magically became a part of Japan in the last few days and thus were able to record it off the air...

np: Jared Emerson-Johnson - Death Of A Lagomorph (Sam & Max Season One OST (Disc 1))

[Haunther]

Mah... However I changed the name of the file! That's good or not you the same answer?

[Guest]

No, deception is not the right way to go.

The right way is to discuss only material that you legitimately own.

You're lucky not to get a double strike here, so take our forum rules to heart. Thank you.

[Haunther]

And now how do I get your help?

[BigDid]

And now how do I get your help?
Hi,

You can't:
@Haunther
That's a torrent file. We can't help you with that per forum rule 6. Please do not post about such things in the future.

@all
Do not reply to this issue per forum rule 6.
This is from one of the moderators; you are requested to not post on the subject, we are requested to not help -PERIOD-

Take time for some reading like:
http://forum.doom9.org/showthread.php?t=52597
http://forum.doom9.org/showthread.php?t=121780
http://forum.doom9.org/showthread.php?t=43066
.. and, I hope things will get clearer.

Did

[n2k3]

Anybody ever did HD content (in my case 1080i) with this filter? It seems with or without MT=true, it gives me more memory errors, then it does its work.
2 line script:
MPEG2Source("C:\Encoding\[m2]ts\raw.d2v", cpu=0)
AnimeIVTC(mode=1)
When stepping frames to search for aa mode it gives me an access violation error at random frames.
This doesn't not happen so often (roughly 1-5%) with SD content.

With or without the SetMemoryMax(x) same results as above, where x can be 512(default), 1024, 1536, 2048
I have 4GB memory, tested with Memtest86+ about two months ago for 8+ hours with no errors.

[Nightshiver]

I don't know about your problem. But I do know that you must specify aa. If you don't want any aa, then set it to aa=0. As for you memory problem, I've never used this filter for hd content.

[n2k3]

I don't know about your problem. But I do know that you must specify aa. If you don't want any aa, then set it to aa=0. As for you memory problem, I've never used this filter for hd content.

I know I need to specify aa, it was a situation of doing simple stuff the filter is supposed to do good but instead its giving me memory errors :).
Seems I've memory problems with more filters

[Blanchimont]

I've tried mode 5 and the results look fine, but what kind of input does mode 6 need as that always ends like this;
Script Error: Invalid arguments to function "vinverse"
(C:\path\AnimeIVTC - v1.06.1.avsi, line 348)
(C:\path\RS.avs, line 43)

Which functions/parameters affect vinverse, as I've been trying to test it by loading almost everything in the requirements folder, and more, manually (that includes vinverse.dll and vinverse.avs...), but the error remains the same ...

Also, it might look nice, but how to use an overrides file with it? With both Telecide/Decimate and TFM/TDecimate it's easy enough, but with AnimeIVTC?

Edit; Never mind, even though the preview looks good, I get an access violation error when trying to actually encode it in Megui...

Edit2; After juggling with different approaches to this I sidetracked to testing a few functions I found along the way, one of them being TempGaussMC_beta1, which also gave similar headache (ie invalid arguments), but actually got solved...
Using "tempgaussmc_beta1()" didn't work. Trying it in AvsP/Megui gave the familiar "invalid arguments" error, however... writing the defaults (the only thing I changed I think was Sbb to 3) into the function in following manner;
tempgaussmc_beta1(tr0=2,tr1=2,tr2=1,rep0=4,rep1=0,rep2=4,EdiMode="EEDI2",EEDI2maxd=8,
sharpness=0.75,Smode=2,SLMode=2,SLRad=1,Sbb=3,SVthin=1.0,Sovs=0,blocksize=16,overlap=8,
truemotion=false,globalmtn=true,search=2,searchparam=2,pelsearch=2,sharp=2,DCT=0,lambda=0,pnew=0,plevel=0,lsad=1600,
SCth1=180,SCth2=98,thSAD1=640,thSAD2=256,border=true,draft=0,pel2hr=0.0)...actually managed to kickstart it (though at this point no idea if I messed up the default values as it's encoding a sample as we speak. Not that it matters as I was just testing it for fun...)
Note to my self; narrow down the parameters next time by reducing them one-by-one until you know which ones at fault instead of just rushing to encode, now I have to do that tomorrow after work and update this post after the fact >.> (and perhaps then manage to figure out the real culprit)...

I have a very strong hunch both are somehow related to the fact my OS (XP) is configured to the japanese language settings, and that's probably spooking here (special characters, like the one in Didee's name, are enough to cause problems, and probably more, though not sure about the specifics...)

TempGaussMC was at least easy as it was just the parameters of the main function, now to do the same where the affected function is nestled deeper inside the main function...
But I do hope to get to test ANIMEIVTC one of these days, no way I've given up...

Haven't found any mention of anyone else mention this issue though?

Edit3; I thought I'd be able to track down the problem, but... after removing from beginning got a few access violation errors around SVThin & Sovs, so put them back, and started decreasing the inputs from the end.
...when only those two were left, took them again out just to confirm the error, however... No error.
tempgaussmc_beta1() worked as it should, even though it hadn't previously. As it doesn't replicate can't give the exact error code either.
It could be some form of memory problem, but as pc hadn't been rebooted since before the previous experiments, I doubt that's the cause either.
So no clue, however I am going to try this same approach with AnimeIVTC, forcefeeding it the parameters, and seeing what happens, and going from there further if need be...

[Umi]

How do I make a VFR mkv with 23.976fps and 59.94fps?
I used Animeivtc(aa=4, mode=5, pass=1) and then run the queue analysis pass. Then I changed to Animeivtc(aa=4, mode=5, pass=2) and then encoded the video. But according to timecodes.txt, there aren't any 59.94fps

[soulstaker]

I need help with a r2 DVD of mine here...

Tried modes 1,3 and 5, but nothing seems to work. I tried changing the block, vidthresh and hybriddup (using mode 5) and it produced better results. But i'm still getting combing.

Here is a sample: http://www.sendspace.com/file/1qqc8n

Thanks in advance.

[Blanchimont]

I need help with a r2 DVD of mine here...

Tried modes 1,3 and 5, but nothing seems to work. I tried changing the block, vidthresh and hybriddup (using mode 5) and it produced better results. But i'm still getting combing.

Here is a sample:

Thanks in advance.
[url=http://myfreefilehosting.com/f/17b892f442_5.01MB]Here's (http://www.sendspace.com/file/1qqc8n[/url) a sample where I made a d2v of it and run through Telecide/Decimate instead (low settings but enough for this test: Fast HQ/Megui)...
LoadPlugin("C:\Program Files\AviSynth 2.5\plugins\Decomb521VFR1.3_P4-Athlon.dll")
LoadPlugin("C:\Program Files\AviSynth 2.5\plugins\dgdecode.dll")

DGDecode_Mpeg2Source("C:\PL2\knm sample 1\VTS_01_1.demuxed.d2v")

Telecide(order=1,guide=0,vthresh=40,dthresh=5,blend=true,ovr="frames.ovr")
Decimate(mode=4,threshold=1.0, timecodes="timecodes.txt",vfrstats="vfrstats")
Lacks the combing present in the m2v if compared side by side, but if the m2v was derived from the source vob doesn't say much about what results you got from AnimeIVTC... Can you say the exact locations where it misses combing?
If you're referring to the 'falling', I think that's intentional... If there's a way to use an overrides file with animeivtc(?) might use that if it makes mistakes...
Unfortunately can't produce a sample trough animeivtc atm, per previous post...
Edit; When making a d2v-file, I always check it afterwards, because if it starts at the wrong frame, there will be apparent combing left when processed with Telecide/Decomb, in those cases I just shift the start point few frames back/forward in DGIndex for the d2v. Might perhaps be a similar issue in your case?

[TheRyuu]

I need help with a r2 DVD of mine here...

Tried modes 1,3 and 5, but nothing seems to work. I tried changing the block, vidthresh and hybriddup (using mode 5) and it produced better results. But i'm still getting combing.

Here is a sample: http://www.sendspace.com/file/1qqc8n

Thanks in advance.

There was no real major combing problem in the sample you provided. Perhaps you could provide a sample in which it fails it some more horrible way? :p

The following worked fine for me:
Mpeg2Source("Blah.d2v")
tfm(slow=2).tdecimate(mode=1,cycle=5)
I did notice some slight telecide combing on the crossfade right at the beginning but you can vinverse() that away if you want to (set it post telecide).
i.e.
tfm(slow=2).vinverse().tdecimate(mode=1,cycle=5)

Second.
LoadPlugin("C:\Program Files\AviSynth 2.5\plugins\Decomb521VFR1.3_P4-Athlon.dll")
LoadPlugin("C:\Program Files\AviSynth 2.5\plugins\dgdecode.dll")

DGDecode_Mpeg2Source("C:\PL2\knm sample 1\VTS_01_1.demuxed.d2v")

Telecide(order=1,guide=0,vthresh=40,dthresh=5,blend=true,ovr="frames.ovr")
Decimate(mode=4,threshold=1.0, timecodes="timecodes.txt",vfrstats="vfrstats")
Why are we still using decomb.
By itself it sucks, enjoy your mouth combs I guess. Please use TIVTC. And the sample isn't vfr so dunno why you're using that mode either.

[Nightshiver]

Glad some others are joining in at helping :) There's nothing I need to add.

[soulstaker]

There was no real major combing problem in the sample you provided. Perhaps you could provide a sample in which it fails it some more horrible way? :p

Yeah, I guess this isn't combing... i will upload other sample to demonstrate my problem after trying with the settings you guys recommended.

Blanchimont, i can't download the sample you provided. Can you upload it on Sendspace?

[Steelunicorn]

Dear all, hope you can help me with IVTC'ing R2J DVD. (sample (http://slil.ru/27480891) or sample (http://rapidshare.com/files/225086545/VTS_01_1.demuxed.rar.html)). Can't get rid of blended frames. Tried modes 1,2,5 - still same result (5 tried because MeGUI said that it's Hybrid video). AnimeIVTC(mode=2, killcomb=2, decimate=4, cdeblendmthr=0.3, dupthr=9) working slightly better, but the result is still *****.

[stpdrgstr]

thetoof, are you still around? Just to be sure, as I love this script, and would like to know if it's going to be updated. :(

[soulstaker]

The following worked fine for me:
Mpeg2Source("Blah.d2v")
tfm(slow=2).tdecimate(mode=1,cycle=5)
I did notice some slight telecide combing on the crossfade right at the beginning but you can vinverse() that away if you want to (set it post telecide).
i.e.
tfm(slow=2).vinverse().tdecimate(mode=1,cycle=5)

Second.

Deinterlaced well, but was a bit too jerky on some parts of the video (not the one i used on the sample)

Then i played a bit with telecide and solved my problem with this:

AssumeTFF()
Telecide(guide=0,post=4,vthresh=40,dthresh=4,show=false,blend=true)
tdecimate(mode=1,cycle=5)

Thanks for your help.

[TheRyuu]

Deinterlaced well, but was a bit too jerky on some parts of the video (not the one i used on the sample)

Then i played a bit with telecide and solved my problem with this:

AssumeTFF()
Telecide(guide=0,post=4,vthresh=40,dthresh=4,show=false,blend=true)
tdecimate(mode=1,cycle=5)

Thanks for your help.

If you want mouth combs and what have you, that's your prerogative.
I'm just warning you that no one should really be using decomb anymore.

Maybe give a sample where the problem occurs?

[thetoof]

thetoof, are you still around? Just to be sure, as I love this script, and would like to know if it's going to be updated. :(
Hey, I'm still *somewhat* around (checking what's going on the forum every now and then), but as busy as ever. What made me quit last time has been resolved and I'm forcing myself to take a three weeks break at the beginning of June. One of the things I'm planning to do then is updating AnimeIVTC to a decent state and doing the first version of the deinterlacing/deblending/ivtc guide with many examples, images and video samples I spoke about with neuron2 sometime somewhere.

So, keep your suggestions/ideas/comments/bugs/insults/requests somewhere in your head and I'll be back one day or the other.

Cheers.

[Nightshiver]

Wow, that's quite possibly the best news I've heard this year :)

[Codex0nz]

Glad to hear you're still around toof.

I set my vcr up to encode transformers for me, using this script:

DGDecode_mpeg2source("D:\Convert\Transformers Season 1\1\vts_02_1.d2v", info=3)
ColorMatrix(hints=true, interlaced=true, threads=0)

#deinterlace
animeivtc(mode=2, aa=4, iregion="PAL", oregion="NTSC", mt=true)

#resize
crop( 12, 2, -12, -2)

the system uses:
E4700
2gig ram

should it really be going at 0.3fps? it took over a day to do one pass and i think i may have done something wrong...

[Nightshiver]

Speed doesn't indicate something is wrong, you just have a crappy system and animeivtc is absurdly slow. Do a lossless rendering pass first if you want something more acceptable.

[Codex0nz]

Speed doesn't indicate something is wrong, you just have a crappy system and animeivtc is absurdly slow. Do a lossless rendering pass first if you want something more acceptable.

It's just a media center not like it needs an 8600 or anything lol and i dont suppose there is any way to pass off some of the load to the 9600gt in it?

I thought i may have used a wrong filter or something, not sure

[Nightshiver]

You are naming video cards. Video cards have nothing to do with the speed you are getting. Anyway, no. It's just simply that animeivtc is invariably slow, no way around that.

[Codex0nz]

You are naming video cards. Video cards have nothing to do with the speed you are getting. Anyway, no. It's just simply that animeivtc is invariably slow, no way around that.

I meant through the use of CUDA or something perhaps.

Ah well such as life, at least the results are fantastic

[Piper]

I've only just picked up on AnimeIVTC and although all my plugins are in line with the documentation, I'm still getting the following error:

Avisynth open failure:
ChangeFPS: FPS value is out of range.

(Cdeblend_v1.1b.avsi, line 132)
(AnimeIVTC - v1.06.1.avsi, line 157)

I'm using AnimeIVTC(mode=5).

Any ideas??

[R3Z]

I searched the forums and found one or two posts without any solution about madman dvd's, in particular the Transformers G1 R4 DVD's.

Heres is a sample (30mb demuxed);

http://www.users.on.net/~inet_s51/doom9/VTS_02_1.demuxed.m2v

The source has blends and interlaced frames every few frames but there doesnt seem to be any consistency (to me at least). Having never dealt with Anime/Cartoons before all i can say is sometimes it looks like telecined, sometimes every second frame in interlaced.

I tried animeivtc(mode=2, aa=0, iregion="PAL", oregion="NTSC", mt=true) which seems to work, but i am wondering if there is anything easier or quicker.

Any help appreciated !

R3

[Codex0nz]

I searched the forums and found one or two posts without any solution about madman dvd's, in particular the Transformers G1 R4 DVD's.

Heres is a sample (30mb demuxed);

http://www.users.on.net/~inet_s51/doom9/VTS_02_1.demuxed.m2v

The source has blends and interlaced frames every few frames but there doesnt seem to be any consistency (to me at least). Having never dealt with Anime/Cartoons before all i can say is sometimes it looks like telecined, sometimes every second frame in interlaced.

I tried animeivtc(mode=2, aa=0, iregion="PAL", oregion="NTSC", mt=true) which seems to work, but i am wondering if there is anything easier or quicker.

Any help appreciated !

R3

Im working on the exact same thing at the moment, (using that almost exact same line) even after this there appears to be a slight loss of saturation and haloing as well (may be wrong but yeah)

Also trying aa=4, the result is nice but there is some detail loss i think, need to play with it more though.

[thetoof]

detail loss is most likely due to way too aggressive warpsharpening. try lowering thin=

[Gengero]

Hello guys.
Recently I'm tryin to work with an anime DVD and AnimeIVTC but I got some trouble with some frames , here's 2 screenshot :

http://img385.imageshack.us/img385/5956/39975581.jpg
http://img385.imageshack.us/img385/8673/92694718.jpg


I just want to know if this is a blend or an other kind of artifect ?

I've read some posts that tells "that is a blend interlacing and there's no way to deal with it" and others posts says "that's a hybrid material and must be done with a vfr"

any suggetion ?

THX

[Nightshiver]

As always, post a video sample.....

[Gengero]

Here I'm , these's a little simple video .
http://www.megaupload.com/?d=5KEZD3LG

[Guest]

We need an unprocessed source VOB sample, not your encode!

[Gengero]

Okay Sir ;)

Here's a m2v demuxed with DGindex http://www.megaupload.com/?d=NX9E3BVE

THX

[Guest]

A 1MB sample is ridiculous. Next time give at least 50-100MB.

From the few frames you gave it looks like simple hard 3:2 pulldown. Did you try a simple IVTC?

[JohannesL]

I'd like to say that you need an mvtools.dll from the "MT=true - MVTools_V1.9.7.7" folder for SharpAAMCmod to work. With the mvtools.dll found in the "MT=false - mvtools-v1.11.4.3" folder, I get this error: "There is no function named MVAnalyseMulti"
Btw, SharpAAMCmod is awesome. It makes edges ridiculously clean and sharp in 3D video game sources as well. :)

[Nightshiver]

It's in the documents that you need that, if you didn't read......

[Gengero]

A 1MB sample is ridiculous. Next time give at least 50-100MB.

From the few frames you gave it looks like simple hard 3:2 pulldown. Did you try a simple IVTC?

Yes, I've try a simple IVTC with tfm(PP=6).
Tdecimate(Mode=1)

and also IVTC vfr with 2 pass , but still the some thing, when I'll back home I'll give you a more big file.

[ghscreenki]

I'm here again. Before anything else, I'm still learning and understanding the terms you guys are using so please excuse me for sounding clueless.

Did lots of VFR encoding on my anime DVDs and since, uh, sorry to say this, I can't really comprehend animeivtc (reading the Documentation without basic knowledge was hell.), I used tritical's tdecimate instead. Tritical's tdecimate worked great on the anime DVDs I backed up but on this particular source, the plugin is kinda useless.

The rolling credits is jerky even after using tdecimate. Timecodes:

# timecode format v1
Assume 29.970030
# TDecimate v1.0.3 by tritical
# Mode 5 - Auto-generated mkv timecodes file
0,15,23.976024
21,24,23.976024
30,10261,23.976024
10267,11754,23.976024
11760,18467,23.976024
18473,18588,23.976024
18594,18597,23.976024
18608,22879,23.976024
22890,24109,23.976024
24120,27415,23.976024
27421,28264,23.976024
28270,29573,23.976024
29574,29576,17.982018
29577,29584,23.976024
29585,29587,17.982018
29588,29747,23.976024
# vfr stats: 99.81% film 00.19% video
# vfr stats: 37100 - film 70 - video 37170 - total
# vfr stats: longest vid section - 10 frames
# vfr stats: # of detected vid sections - 11

I asked a fellow encoder for some pointers but since he really doesn't want to show me all that he knows, he gave me something to start with and I have to find out the rest myself. He directed me to animeivtc.

He gave me two scripts:

animeivtc(mode=3, omode=2)
first pass

and

AnimeIVTC(mode=3,aa=0,omode=2,mix=4,estart1=0,eend1=26499,estart2=28258,eend2=29735,istart1=66248,iend1=70642)
second pass

The first pass script is not working so I need to add in the mix parameter and check=0 to get the frames I need for pass two. I modified it like this:

animeivtc(mode=3, omode=2, mix=4, check=0)

Did an analysis pass on vdub in hope for vdub to spit out a text file containing the frames or something I will need for pass two. After 3 hours, it didn't. :(

http://img189.imageshack.us/img189/3265/cb7sc.png

Obviously, I'm doing this wrong or I'm missing something.

Thanks and please excuse me for being clueless.

Will provide a source in a few hours.

[Nightshiver]

Thats because mode=3 won't give you timecodes or any text file. You must use the VFR modes on animeivtc to get that, i.e. mode=5. To actually get the files you want, you must have the line read like this:

animeivtc(mode=5, pass=1)

Load that line into vdub and run an analysis pass. This should give you the txt files you are looking for. To actually encode VFR with animeivtc, you will need to change the previous line I gave you with this:

animeivtc(mode=5, pass=2)

Do a lossless rendering pass with this line, and then you can encode after you have your lossless video file. After you encode the video, take the "Timecode.txt" file that animeivtc produced and add it as the timecode for the video stream in mkvmerge.

[ghscreenki]

Thats because mode=3 won't give you timecodes or any text file. You must use the VFR modes on animeivtc to get that, i.e. mode=5. To actually get the files you want, you must have the line read like this:

animeivtc(mode=5, pass=1)

Load that line into vdub and run an analysis pass. This should give you the txt files you are looking for. To actually encode VFR with animeivtc, you will need to change the previous line I gave you with this:

animeivtc(mode=5, pass=2)

Do a lossless rendering pass with this line, and then you can encode after you have your lossless video file. After you encode the video, take the "Timecode.txt" file that animeivtc produced and add it as the timecode for the video stream in mkvmerge.
In case you forgotten, you suggested me this a long time ago. :) Already did that and the output is still the same: Jerky rolling credits.

Source in a few minutes.

EDIT: Source: http://www.megaupload.com/?d=O5O2SLGO

[lansing]

In case you forgotten, you suggested me this a long time ago. :) Already did that and the output is still the same: Jerky rolling credits.

Source in a few minutes.

EDIT: Source: http://www.megaupload.com/?d=O5O2SLGO

come on man, no one is gotta dl a 600MB video just to take a look at it. Cut out only the problematic scene, 1 minute max.

[manono]

# vfr stats: 99.81% film 00.19% video
Just Force Film it and forget about it.

[ghscreenki]

come on man, no one is gotta dl a 600MB video just to take a look at it. Cut out only the problematic scene, 1 minute max.
Please tell me what application I use to do that. :(

Just Force Film it and forget about it.
I'd love to but, that's not how it works for me. :)

[Guest]

Open the VOB in DGIndex. Use the [ and ] buttons to define a range around a section that shows the problem. Then do Save Project and Demux Video. Upload the resulting M2V.

[Nightshiver]

Ok, then if what I suggested doesn't work, and if tritical's version of VFR doesn't work, maybe you should just forget about doing VFR, hmm? Please explain further as to why force film'ing it is "not how it works for me."