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. |
18th February 2007, 18:31 | #921 | Link |
Registered User
Join Date: Dec 2003
Location: MO, US
Posts: 999
|
@Terranigma
ovr file works fine for me. Could you add display=true to the tdecimate() line, go to the cycle that includes frame 217 of the input (should start at frame 172 of the output), grab a pic of one of the frames, and post it here? It will show '**' next to whichever frame in the cycle has been dropped. |
18th February 2007, 18:44 | #922 | Link | |
*Space Reserved*
Join Date: May 2006
Posts: 953
|
Quote:
I thought the way it works is, it'd drop whatever frame(s) I entered in the ovr file. Thanks for the help! |
|
19th February 2007, 23:44 | #923 | Link | |
Registered User
Join Date: Sep 2005
Location: Vancouver
Posts: 600
|
Quote:
Code:
BlankClip(pixel_type="YV12",length=80304,fps_denominator=1001,fps=30000) TDecimate(cycleR=44,cycle=202,ovr="ovr.txt") |
|
20th February 2007, 01:17 | #924 | Link |
Registered User
Join Date: Dec 2003
Location: MO, US
Posts: 999
|
Thanks, when getting the ovr info it wasn't adjusting the maximum number of frames that could be dropped if the last cycle wasn't full length.
TIVTC v1.0.1, changes: Code:
tfm: + Support d2v file format v16 - fixed a bug in tfmpp destroying hints - fixed generating ovr help info about combed frames in output files when PP=0 tdecimate: - fixed reading in tfm output files created with PP=0 - fixed a bug in destroying hints on frame output - fixed a bug in drop/keep frame overrides in the last cycle of a video if the cycle was not full length |
21st February 2007, 08:44 | #927 | Link |
Registered User
Join Date: Dec 2003
Location: MO, US
Posts: 999
|
Not sure I understand your request, since there shouldn't be any need for a field order specific matching mode. With bff material it only makes sense to match c/n if matching off the top field... just like it only makes sense to match c/n for tff material if matching off the bottom field.
If you have a bff clip and want to match off the top field use: tfm(order=0,field=1,mode=0) in which case tfm will only use c/n matches. |
21st February 2007, 20:11 | #929 | Link |
Registered User
Join Date: Jul 2004
Location: canada B.C.
Posts: 335
|
hi,im trying to clean up this clip...http://rapidshare.com/files/16554591/VTS_01_1.VOB.html.
im a noob to using filters and writing scripts.would TDeint and TIVTC work on this clip and if so how would the script be written.thanks |
23rd February 2007, 16:59 | #930 | Link |
ангел смерти
Join Date: Nov 2004
Location: Lost
Posts: 9,558
|
I posted another compile of MT EEDI2, exact same source, but the latest version of ICL9 works with VC2k5 SP1 now. (That's what the i in imp stood for. ) I patched it so hopefully it works on Athlons, I'd like some feedback, whether it works and if it's any faster than the VC one.
http://foxyshadis.slightlydark.com/random/Eedi2mt.zip |
1st March 2007, 03:56 | #931 | Link |
Registered User
Join Date: Aug 2002
Posts: 467
|
Video speed problem with non-standard frame rate
Using tivtc, I have a video that ends up slightly stretched in time. The video is the correct duration, but the last frame is not at the end of the source. As a result, my audio is out of sync with the video.
Details are in this thread, but the synopsis is: I'm converting an OTA HD TS captured from ABC to HD DivX (our local ABC affiliate converts the 720p to 1080i, hence the need for tivtc). I used DGIndex 1.4.9 beta 14 to extract the AC3 and create the D2V file. I used AVISynth 2.57 and with some help, I was able to determine that the video was telecined at something like 23.2603 fps. For the best results, I'm creating a 23.2603 fps progressive video and remuxing it with the original AC3 using VirtualDubMod and the latest DivX codec. I used the following script: Code:
MPEG2Source("Toy Story (WFAA).d2v") AssumeBFF() Crop(0, 0, 0, -8) tfm() #mode=5 tdecimate(mode=2, rate=23.2603) LanczosResize(1280, 720) This is with your latest release, version 1.0 final. I noticed in playing around, that if I use 23.976, it doesn't seem to end early. FDecimate doesn't seem to do this, but it has serious problems picking the right frames to discard in this noisy (fairly significant pre-processing and MPEG-2 artifacts) source. Xesdeeni |
2nd March 2007, 06:18 | #932 | Link |
Registered User
Join Date: Nov 2005
Posts: 14
|
Hello, I noticed how trimin allows the trimming of the d2v prior tfm operation. Is there a way to add frames to the d2v before tfm instruction to do the exact opposite? Allowing for duplicateframe for instance.
Thank you in advance. |
5th March 2007, 01:12 | #933 | Link |
Registered User
Join Date: Dec 2003
Location: MO, US
Posts: 999
|
@Eggroll
There isn't, but it could be added as part of the trimIn parameter... could you give the exact details of what you are wanting to do? @Xesdeeni I'll try to figure out what's wrong with mode 2. It was never tested much with non-standard frame rates. @radar It's pure interlaced, so a basic script would be: mpeg2source() tdeint() I would recommend that you search the avisynth usage forum and try out some other deinterlacers to see which is best to you in terms of speed/quality. Some common deinterlacers are: leakkerneldeint, tdeint, tdeint+eedi2, securebob, mvbob, mcbob. You'll also have to decide whether or not you want double framerate (bobbed) output or not. Sports footage with fast action and sharp lines/edges is probably the toughest type of video to get good deinterlacing results on. |
5th March 2007, 22:40 | #934 | Link |
*Space Reserved*
Join Date: May 2006
Posts: 953
|
Guys. what would be the correct way deinterlace ntsc film material? Doing a field blend or interpolation? When I do a interpolation deinterlace, the deinterlaced frame becomes a clone of the next frame causing jerky playback, so field blending seems to be the best way to go. Perhaps there's another way I should go about doing this?
|
6th March 2007, 00:13 | #935 | Link |
Registered User
Join Date: Jul 2004
Location: canada B.C.
Posts: 335
|
hi tritical,thank you for your reply.
i have another question and sorry if its a dumb one. i can see why you put tdeint() in the script but what is mpeg2source() for.again sorry if this is a dumb question. thanks i think i answered my question.im using dvd rb with the filter,and rb puts mpeg2source() in for me.if im wrong please correct me.thanks Last edited by radar; 6th March 2007 at 00:38. |
6th March 2007, 00:37 | #936 | Link | |
Registered User
Join Date: Jan 2007
Posts: 943
|
Quote:
http://www.neuron2.net/dgmpgdec/DGIn...#WhatIsDGIndex http://neuron2.net/dgmpgdec/QuickStart.html http://www.neuron2.net/dgmpgdec/DGDe...ml#MPEG2Source |
|
6th March 2007, 10:04 | #937 | Link | |
ffdshow/AviSynth wrangler
Join Date: Feb 2003
Location: Austria
Posts: 2,441
|
Quote:
TFM(order=0 or 1) followed by TDecimate(mode=1,hybrid=0 or 1) should be a good starting point...
__________________
now playing: [artist] - [track] ([album]) |
|
6th March 2007, 15:35 | #938 | Link |
Registered User
Join Date: Nov 2005
Posts: 14
|
@tritical
I wanted to "edit" a d2v so the frames would match the ones from another d2v (r1 vs r2 in this case) prior tfm operation. Since to match r2 with r1 I need to trim some frames here and duplicates some frames there, it was impossible in this situation. I used another approach though. I removed the d2v parameter and simply trim or duplicate frames between mpeg2source and tfm instruction. Works like a charm and I don't have to use a trimin file anymore. I guess it could be a nice feature to add if you've got some time to waste... If you can trim so why not be allow to add duplicate frames? But since I seem to be the first to ask for such a thing and that I found a workaround, I guess there really is no need for such a feature at this time. Anyway, thank you very much. |
Tags |
tdeint, tivtc |
Thread Tools | Search this Thread |
Display Modes | |
|
|