Clever, FREE pulldown-removal & transcoding workflow via AviSynth

mojo43 said:
Have you tried a different codec? I am converting to the canopus codec...
Click to expand...

you mean canopus hq ? I did not try it . I intend to try lossless codec . though I am getting realize that this lossless codec may not be a good choice for editing . yes , will try canopus later . But now I wanna know how to avoid this "lossless" loss .
 
Bumping this thread. Why are all the pay programs for pulldown removal stickied and not this free one? I hope the moderators decide to put this up as a sticky so people know there is an alternative.
 
So I tried it and it turned a 2 minute long 220 meg .mts video into a 11 gigabyte mts.

Did I mess something up? Any ideas?

Tim
 
Make sure that you set your compression settings in the script. Without any compression you will have very large file sizes. Or maybe you aren't compressing enough with your codec choice. Let me know if you need any more help...
 
mutabot said:
NOTE: Win 7 users must disable built-in mp4 decoder for this to work.
Click to expand...

Doea anyone have any idea how to do that?
It seems to work OK for me on Win7 except that I'm still seeing some combing so I guess that's the build in decoder mucking things up?
 
thanks

i'm just a little confused about the whole 24p wrapper thing. What is the advantage of having it checked? is it only that you save time transcoding (sp) it from 60i to 24p....or does it change the quality of the colours etc? prevent jittery action?

Also do you do the pulldown before editing your masterpiece or after?

I use vegas video for example...can i do pulldown in that?

appreciate your patience
 
meinyvr said:
thanks

i'm just a little confused about the whole 24p wrapper thing. What is the advantage of having it checked? is it only that you save time transcoding (sp) it from 60i to 24p....or does it change the quality of the colours etc? prevent jittery action?

Also do you do the pulldown before editing your masterpiece or after?

I use vegas video for example...can i do pulldown in that?

appreciate your patience
Click to expand...

There is a difference. Checking the 24p means that you don't have to remove pulldown, but as I understand it is not stable. I remove pulldown and convert to a canopus codec and then edit. I don't believe that you can remove pulldown in vegas, but I could be wrong. I would just go and try it out and see if you can. Hope that helps!
 
mojo43 said:
There is a difference. Checking the 24p means that you don't have to remove pulldown, but as I understand it is not stable. I remove pulldown and convert to a canopus codec and then edit. I don't believe that you can remove pulldown in vegas, but I could be wrong. I would just go and try it out and see if you can. Hope that helps!
Click to expand...

Thanks - yes that clears it up a bit.
 
nathankw said:
Doea anyone have any idea how to do that?
It seems to work OK for me on Win7 except that I'm still seeing some combing so I guess that's the build in decoder mucking things up?
Click to expand...

I have been using this script for a while and got awesome results, but recently i noticed combing and came back here to figure out what might be wrong, and saw that the author mentioned that win7 codecs had to be disabled for it to work. So i googled "disable win7 codecs" and found a program called win7DSfiltertweaker that disables all the builtin windows codecs, and tried the script and the combings gone! Unfortunately theres still a very strange problem with strobing, almost as if this script is throwing away frames.
 
l.breyers said:
I have been using this script for a while and got awesome results, but recently i noticed combing and came back here to figure out what might be wrong, and saw that the author mentioned that win7 codecs had to be disabled for it to work. So i googled "disable win7 codecs" and found a program called win7DSfiltertweaker that disables all the builtin windows codecs, and tried the script and the combings gone! Unfortunately theres still a very strange problem with strobing, almost as if this script is throwing away frames.
Click to expand...

Hmmm... I have never had any problems with it. Are you using the most recent version of the script?
 
mojo43 said:
Hmmm... I have never had any problems with it. Are you using the most recent version of the script?
Click to expand...

I looked at it and found that the reason strobing looked so bad was the window size i was viewing it at. I scaled the footage down to half its size and the strobing disappeared. Considering that its presence is how fast a feature tracks across the screen, bigger screen = bigger strobing.
 
mutabot said:
This is the algorithm:

x0 = selectevery(10, 0,1).Weave()
x1 = selectevery(10, 2,3).Weave()
x2 = selectevery(10, 5,6).Weave()
x3 = selectevery(10, 8,9).Weave()
Click to expand...

I am not a programmer but I understand this algorithm kinda but I have a question. What is keeping it from starting the pulldown pattern on the wrong frame and therefore throwing away the wrong frame? Is there something in the script that I don't recognize that is allowing it to start in the right place?

Also is this algorithm for removing 2-3-3-2 pulldown? I know this is a panasonic forum but for the canon hv30 that has standard 2-3 pulldown wouldn't I just modify the script so it throws away the fifth field in every group of 5?
 
Last edited:
Pixelznbits said:
I am not a programmer but I understand this algorithm kinda but I have a question. What is keeping it from starting the pulldown pattern on the wrong frame and therefore throwing away the wrong frame? Is there something in the script that I don't recognize that is allowing it to start in the right place?

Also is this algorithm for removing 2-3-3-2 pulldown? I know this is a panasonic forum but for the canon hv30 that has standard 2-3 pulldown wouldn't I just modify the script so it throws away the fifth field in every group of 5?
Click to expand...

You could try asking over at the doom9 forums, they understand this stuff.
 
Pixelznbits said:
I am not a programmer but I understand this algorithm kinda but I have a question. What is keeping it from starting the pulldown pattern on the wrong frame and therefore throwing away the wrong frame? Is there something in the script that I don't recognize that is allowing it to start in the right place?

Also is this algorithm for removing 2-3-3-2 pulldown? I know this is a panasonic forum but for the canon hv30 that has standard 2-3 pulldown wouldn't I just modify the script so it throws away the fifth field in every group of 5?
Click to expand...



A better approach is to use an adaptive algorithm like TIVTC. It finds field matches and decimates the duplicates, so it doesn't matter where you start in the cadence cycle

Download TIVTC from here
http://web.missouri.edu/~kes25c/

e.g
DirectShowSource("video.mts")
TFM()
TDecimate()
 
You must log in or register to reply here.
Back
Top