Hybrid DVD

User avatar
x265
Joined: Tue Dec 25, 2012 1:52 pm
Org Profile

Re: Hybrid DVD

Post by x265 » Wed Feb 06, 2013 9:25 am

Code: Select all

MPEG2Source("C:\Users\Abhijith Nair\Desktop\Screenshots\vol1.d2v", cpu=0)
ivtc_txt60mc()<<
When i added this line in avspmod it results in an error saying invalid argument to the function "ivtc_60mc".
“The only true wisdom is knowing that you know nothing.”

Mister Hatt
Joined: Tue Dec 25, 2007 8:26 am
Status: better than you
Contact:
Org Profile

Re: Hybrid DVD

Post by Mister Hatt » Wed Feb 06, 2013 9:37 am

You need the script for it if you're gonna use it, I thought google would have indicated that, or at least a basic understanding of how avisynth works.

User avatar
x265
Joined: Tue Dec 25, 2012 1:52 pm
Org Profile

Re: Hybrid DVD

Post by x265 » Wed Feb 06, 2013 9:44 am

I have the script.
“The only true wisdom is knowing that you know nothing.”

User avatar
mirkosp
The Absolute Mudman
Joined: Mon Apr 24, 2006 6:24 am
Status: (」・ワ・)」(⊃・ワ・)⊃
Location: Gallarate (VA), Italy
Contact:
Org Profile

Re: Hybrid DVD

Post by mirkosp » Wed Feb 06, 2013 10:39 am

You don't use it like that. For starters you use it only where you have the 60i text along with the telecined part, and you MUST specify the 3:2 pattern as a parameter.
For instance, if you trim and the part you apply it on is a cnncc, you'd use ivtc_txt60mc(0) to tell it that pattern (1 for ccnnc, 2 for cccnn, 3 for ncccn, 4 for nnccc ─ going beyond will result in the filter doing a %5, so 5 gets counted as 0, 6 as 1 and so on).
Obviously, if there is a pattern change, you need to make different trims and use it with different values and then put it back together.

But as I said, the easier solution in this specific situation is just to interpolate the text on its own, since you won't need to check the pattern; ivtc the image on its own (even auto-ivtc if you want to keep it simple), and then stack back together. Specifying the wrong pattern in such a case won't harm since ivtc_txt60mc won't actually touch the image on the right just interpolate the text on the left.
Image

User avatar
x265
Joined: Tue Dec 25, 2012 1:52 pm
Org Profile

Re: Hybrid DVD

Post by x265 » Wed Feb 06, 2013 1:56 pm

Please show me a sample code.
“The only true wisdom is knowing that you know nothing.”

User avatar
mirkosp
The Absolute Mudman
Joined: Mon Apr 24, 2006 6:24 am
Status: (」・ワ・)」(⊃・ワ・)⊃
Location: Gallarate (VA), Italy
Contact:
Org Profile

Re: Hybrid DVD

Post by mirkosp » Wed Feb 06, 2013 2:28 pm

There isn't much more to show... I generally do a lossless for the normal ivtc, a separate script for the ivtc_txt60mc, and put them back together in a final script

Code: Select all

DGDecode_Mpeg2Source("source.d2v")
#ivtc here

Code: Select all

DGDecode_Mpeg2Source("source.d2v")
trim(5773,6526).ivtc_txt60mc(0)

Code: Select all

ep    = AVISource("decimated_source.avi")
txt  = AVISource("text_compensated_scene.avi")
ep.trim(0,1127)+txt+ep.trim(5221,0)
Image

User avatar
x265
Joined: Tue Dec 25, 2012 1:52 pm
Org Profile

Re: Hybrid DVD

Post by x265 » Wed Feb 06, 2013 2:43 pm

ivtc_txt60mc(0) << why have you used 0 over here?
“The only true wisdom is knowing that you know nothing.”

User avatar
mirkosp
The Absolute Mudman
Joined: Mon Apr 24, 2006 6:24 am
Status: (」・ワ・)」(⊃・ワ・)⊃
Location: Gallarate (VA), Italy
Contact:
Org Profile

Re: Hybrid DVD

Post by mirkosp » Wed Feb 06, 2013 3:08 pm

mirkosp wrote:For instance, if you trim and the part you apply it on is a cnncc, you'd use ivtc_txt60mc(0) to tell it that pattern (1 for ccnnc, 2 for cccnn, 3 for ncccn, 4 for nnccc ─ going beyond will result in the filter doing a %5, so 5 gets counted as 0, 6 as 1 and so on).
Obviously, if there is a pattern change, you need to make different trims and use it with different values and then put it back together.
Image

User avatar
x265
Joined: Tue Dec 25, 2012 1:52 pm
Org Profile

Re: Hybrid DVD

Post by x265 » Thu Feb 07, 2013 9:41 am

How do i find the pattern?
“The only true wisdom is knowing that you know nothing.”

User avatar
mirkosp
The Absolute Mudman
Joined: Mon Apr 24, 2006 6:24 am
Status: (」・ワ・)」(⊃・ワ・)⊃
Location: Gallarate (VA), Italy
Contact:
Org Profile

Re: Hybrid DVD

Post by mirkosp » Thu Feb 07, 2013 1:24 pm

From GizmoTech's YATTA guide that never got finished:
IVTC Patern Matching:

So you're asking yourself, "Well Bludgers, how do I find a pattern in a scene"? There are really 2 easy ways to quickly find pattern. They involved tracking down an "NcccN" pattern.
First Method: Fast action
Identify a point in the scene (perferably near the start) where a hand/arm/body moves. Identify the first point of movement where it N matches, then set the next 3 frames C and hope that the next frame is a N match. Assuming all goes well, you've just identified a NcccN pattern, which, when moving to the next frame and Activating Auto Patern allows you to just play through the rest of the scene confident in knowing you've achived pattern.
Second Method: Mouth Transitions
The more dificult way, is to identify changes in mouth sizes. Alot of mouth transitions are done in frames of 5, which make identifying mouth transitions usually quite fast. Find a change where a mouth goes from big(er) to small(er) (or vice versa) and make sure it is an N match. Set the next 3 frames to C, and then the 4th frame, in theory, should be a N match on a nother transition. Assuming you were lucky and hit a NcccN pattern, you can now activate auto pattern and finish out the scene automatically.

Now an important point of note. In most cases a scene will NOT change pattern from start to finish, however there are rare occasions where the scene will change pattern mid stream.
Image

Locked

Return to “Video & Audio Help”