Adding transitions to videos with free software

[Cyber Akuma]

post a video sample

Err, gimme a second to figure out how to cut a clip from a mp4 file, I used to have no problems doing it with AVI before but im still getting used to MP4 based tools.

In the meantime heres two screenshots taken at the exact same moment, the first one playing the MP4 file directly and the second one through avisynth using nothing more than a single line FFVideoSource to load the video (Yes, I know the video is dark already (hence why I want to prevent it from being DARKER) and of low quality, I have mixed videos from night and day shots, the day shots don't seem to become as dark as the night ones when put through AVIsynth):

http://img257.imageshack.us/img257/9507/sany0144mp4snapshot0005.jpg
http://img513.imageshack.us/img513/4236/test5avssnapshot0005201.jpg


Also, I noticed that Media Player Classic claims it is loading different decoders and colorspaces for each as well, I remember you mentioned this before and I assume its the problem, but I have no idea how to fix it, this is what MPC loads for the MP4 directly and through AVIsynth:


---------------------------------

[poisondeathray]

It might be a levels issue, decoding at different ranges , but it's likely the way you took the screenshot (when you play the file directly it's going through a different chain of filters, renders) . vrm9 (renderless) is required to take screenshots through mpc, so it gives you a washout out look (normally looks bad, but might be beneficial in dark videos, artificially brightening it up) . If you change renderer to overlay mixer , then the graphics card overlay settings come into effect . Change the various renderers and see how the video's appearance "changes" . Note the underlying video hasn't changed at all, just the apperance of it. Prove this for yourself.


This is easy to alter in the script , post the sample and I'll suggest a solution (I believe the underlying video is dark, but the renderer is artificially brightening it up. The dvd will be dark unless you change the levels in the script) . Another way of saying this is your playback issues are from how you have it setup in MPC


You can cut a sample with yamb or avidemux , and upload it here directly if <30MB or a free hosting site (e.g. mediafire.com, sendspace.com)


You might consider cleaning up the video before encoding to mpeg2 for dvd as well

[Cyber Akuma]

Ok, you only need the original MP4 right?

By the way, I tried also just writing the entire AVS I would need in advance for when I figure out the dark issue, and am having a little issue with that as well.

The problem is that I want to join 6 clips together, and put a transition between each, and I am not sure how to put a transition between more than two videos at a time, I attempted to define the transition of the first two videos as a variable, then load THAT as a transition with the third video and so on... but it didn't work, the first transition went off fine but the rest just cut to the next video, do you think you can take a look at that too? If you can, I uploaded it as test6.avs.

[poisondeathray]

what kind of transition did you want ?

[Gavino]

The problem is that I want to join 6 clips together, and put a transition between each, and I am not sure how to put a transition between more than two videos at a time, I attempted to define the transition of the first two videos as a variable, then load THAT as a transition with the third video and so on... but it didn't work, the first transition went off fine but the rest just cut to the next video, do you think you can take a look at that too? If you can, I uploaded it as test6.avs.
...

Eff1=TransVenetianBlinds(Mov1, Mov2,100,type="diam")
Eff2=TransVenetianBlinds(Eff1, Mov3,100,type="diam")
Eff3=TransVenetianBlinds(Eff2, Mov4,100,type="diam")
Eff4=TransVenetianBlinds(Eff3, Mov5,100,type="diam")
Eff5=TransVenetianBlinds(Eff4, Mov6,100,type="diam")

FullMov = Eff1+Eff2+Eff3+Eff4+Eff5

Since each of Eff2, Eff3, etc cumulatively includes the previous ones, you just need
FullMov = Eff5

[poisondeathray]

You're going to have to resize to 720x480 to make it dvd legal, also 59.94 fps isn't legal , so you can interlace it to keep the smoothess


For the dark issue you can use a levels filter, it will expose more noise, so you can use a denoising filter and get rid of that chroma noise as well


e.g. for 1 video, adjust the filters as you see fit

FFVideoSource("SANY0144_001.MP4")
Levels(0,1.3,255,15,255,false)
MCTemporalDenoise(settings="high",chroma=true)
Tweak(sat=1.1,coring=false)
LanczosResize(720,480)

#To interlace it add

AssumeTFF()
SeparateFields()
SelectEvery(4,0,3)
Weave()


For your transitions with the cheezy venetian blinds, the basic syntax for a template (change the names to match)

v1=ffvideosource("SANY0144_001.MP4")
a1=ffaudiosource("SANY0144_001.MP4")
vid1=AudioDub(v1,a1).Levels(0,1.3,255,15,255,false ).MCTemporalDenoise(settings="high",chroma=true).T weak(sat=1.1,coring=false).\
LanczosResize(720,480).AssumeTFF().SeparateFields( ).SelectEvery(4,0,3).Weave()


v2=ffvideosource("SANY0144_001.MP4")
a2=ffaudiosource("SANY0144_001.MP4")
vid2=AudioDub(v2,a2).Levels(0,1.3,255,15,255,false ).MCTemporalDenoise(settings="high",chroma=true).T weak(sat=1.1,coring=false).\
LanczosResize(720,480).AssumeTFF().SeparateFields( ).SelectEvery(4,0,3).Weave()

v3=ffvideosource("SANY0144_001.MP4")
a3=ffaudiosource("SANY0144_001.MP4")
vid3=AudioDub(v3,a3).Levels(0,1.3,255,15,255,false ).MCTemporalDenoise(settings="high",chroma=true).T weak(sat=1.1,coring=false).\
LanczosResize(720,480).AssumeTFF().SeparateFields( ).SelectEvery(4,0,3).Weave()

v4=
a4=
vid4=

.
.
.



a=TransVenetianBlinds(vid1, vid2,100,type="diam")
b=TransVenetianBlinds(a, vid3,100,type="diam")
c=TransVenetianBlinds(b, vid4,100,type="diam")

.
.
.


a++b++c....


etc...

If you didn't want the filters or wanted to adjust the filters feel free

Cheers

[Gavino]

a=TransVenetianBlinds(vid1, vid2,100,type="diam")
b=TransVenetianBlinds(a, vid3,100,type="diam")
c=TransVenetianBlinds(b, vid4,100,type="diam")

a++b++c

No, as I pointed out 'c' (or the last in the chain) already includes everything, so no need to add 'a' and 'b' to it.

[poisondeathray]

of course you're correct gavino

one of these days I'm going to catch one of your mistakes

[Cyber Akuma]

Since each of Eff2, Eff3, etc cumulatively includes the previous ones, you just need
FullMov = Eff5

Thanks, that worked for the effects perfectly

You're going to have to resize to 720x480 to make it dvd legal, also 59.94 fps isn't legal , so you can interlace it to keep the smoothess

Actually, I was going to encode the single video into something like huffyuv to then load into my DVD authoring program and let that take care of resizing, interlacing, and all that itself. I just am using avisynth as a way to add transitions to the videos. Most likely will use DVD Flick unless I find something better.

For the dark issue you can use a levels filter, it will expose more noise, so you can use a denoising filter and get rid of that chroma noise as well

Hmmm, is there any way to just make the avs load with the same colorspace that MP4 files are using instead of attempting to re-adjust the levels to match the MP4?

For your transitions with the cheezy venetian blinds, the basic syntax for a template (change the names to match)

Hey now, no need to be mean

[poisondeathray]

Actually, I was going to encode the single video into something like huffyuv to then load into my DVD authoring program and let that take care of resizing, interlacing, and all that itself. I just am using avisynth as a way to add transitions to the videos. Most likely will use DVD Flick unless I find something better.

Sounds good. But I would still really suggest cleaning up the video beforehand, brightening, denoising etc... You can let the other program do the resizing etc. for dvd

avs2dvd uses hcenc which is a better encoder than dvd flick's encoer (which uses ffmpeg)

Hmmm, is there any way to just make the avs load with the same colorspace that MP4 files are using instead of attempting to re-adjust the levels to match the MP4?

It is the same colorspace, libavcodec decodes it at full range. I recommend using the filters to brighten and clean it up. I posted an example of before and after (you can always tweak it to your taste)

Hey now, no need to be mean

Relax, just a joke

[Cyber Akuma]

Allright, ill try messing with the filters and brightness settings, thank you.

So, since I am planning to let me DVD authoring program take care of formatting the video properly that means I shouldn't use the resize or interlace options? Or should I still use them?

Hmm..... come to think of it, does MeGui even support encoding to huffyuv or would I have to use virtualdub? I am not aware of any other encoders.

[manono]

An authoring program ... authors. You're asking about encoding programs, or all-in-one programs that both encode and author. I don't know or care about MeGUI. If you want to make a lossless AVI before doing the MPEG-2 encoding for DVD, use VDub and one of the lossless codecs you've installed.

And I would do all the resizing, filtering, colorspace conversions (if any) and anything else needed to prepare it for encoding for DVD in the AviSynth script.

[Cyber Akuma]

Hmm, I can't get the following to work:

Levels(0,1.3,255,15,255,false)
MCTemporalDenoise(settings="high",chroma=true)
Tweak(sat=1.1,coring=false)

First of all, where do I even put these lines since I have all sorts of other code in my avs file? At the end?

When I put them near the beginning it complained that the Levels command was incorrect.
When I put them at the end it complained that there is no MCTemporalDenoise function (A missing plugin?)

I tried googling for MCTemporalDenoise but then it started complaining about LSFMod, I am guessing I need a bunch of plugins to get this to work, is there a plugin pack or anywhere so I can just download everything I need for MCTemporalDenoise?

Sigh, this is why I was trying to just get it to load in the same colorspaces as MP4.

[poisondeathray]

It is the same colorspace, YV12 in and out . Levels are decoded differently because of your media player and how you have your filters setup full range vs. tv range.

You put the lines after the source filter, ffvideosource() or directshowsource() or whatever . You need to load a video before you can use a filter, other wise it doesn't know what it's applying the filter to. In avisynth, order of operations is important

FFVideoSource("SANY0144_001.MP4")
Levels(0,1.3,255,15,255,false)
MCTemporalDenoise(settings="high",chroma=true)
Tweak(sat=1.1,coring=false)



Unfortunately, it's going to take a fair bit of work gathering up plugins. If you use google , it will tell you what you need, required plugins

http://avisynth.org/mediawiki/MCTemporalDenoise

It tells you the required filters in the required filters section. Download mctemporaldenoise.avsi and put that in the plugins directory. The .avsi file is autoloading. You still need the other plugins

Maybe it would be easier for you to use something like convertxtodvd or avs2dvd ?

Good luck

[Cyber Akuma]

It is the same colorspace, YV12 in and out . Levels are decoded differently because of your media player and how you have your filters setup full range vs. tv range.

Shows how much I know about video encoding

You put the lines after the source filter, ffvideosource() or directshowsource() or whatever . You need to load a video before you can use a filter, other wise it doesn't know what it's applying the filter to. In avisynth, order of operations is important

FFVideoSource("SANY0144_001.MP4")
Levels(0,1.3,255,15,255,false)
MCTemporalDenoise(settings="high",chroma=true)
Tweak(sat=1.1,coring=false)

Hmm, it wouldn't load when I tried that. Is it because instead of doing a direct FFVideoSource I do a Vid1=FFVideoSource? And load 12 things at once? (Video and audio for each of the 6 clips)

I tried putting it at the bottom of this, before I used the AudioDub command and it would not work, so I'm confused on where I am supposed to put it.

Unfortunately, it's going to take a fair bit of work gathering up plugins. If you use google , it will tell you what you need, required plugins

http://avisynth.org/mediawiki/MCTemporalDenoise

It tells you the required filters in the required filters section. Download mctemporaldenoise.avsi and put that in the plugins directory. The .avsi file is autoloading. You still need the other plugins

Maybe it would be easier for you to use something like convertxtodvd or avs2dvd ?

Good luck

I see. Yeah, I ran into that site when Googling. Problem is, all it has for download is an .avsi file, the other MCTemporalDenoise download I found had several files in it, I couldn't find anything other thn a single avis file for some of the other plugins like LSFMod either. Is the avis all I need and the pack I found contained extra files or do I need to find the entire thing somehow?

Thanks, and yeah, at this rate I will need luck to finish this.

[poisondeathray]

Shows how much I know about video encoding

don't worry, this is how you learn

You put the lines after the source filter, ffvideosource() or directshowsource() or whatever . You need to load a video before you can use a filter, other wise it doesn't know what it's applying the filter to. In avisynth, order of operations is important

FFVideoSource("SANY0144_001.MP4")
Levels(0,1.3,255,15,255,false)
MCTemporalDenoise(settings="high",chroma=true)
Tweak(sat=1.1,coring=false)

Hmm, it wouldn't load when I tried that. Is it because instead of doing a direct FFVideoSource I do a Vid1=FFVideoSource? And load 12 things at once? (Video and audio for each of the 6 clips)

Go back to the earlier post there is an example there. You can use "." in between commands .
e.g.

v1=ffvideosource("SANY0144_001.MP4")
a1=ffaudiosource("SANY0144_001.MP4")
vid1=AudioDub(v1,a1).Levels(0,1.3,255,15,255,false ).MCTemporalDenoise(settings="high",chroma=true).T weak(sat=1.1,coring=false)

I see. Yeah, I ran into that site when Googling. Problem is, all it has for download is an .avsi file, the other MCTemporalDenoise download I found had several files in it, I couldn't find anything other thn a single avis file for some of the other plugins like LSFMod either. Is the avis all I need and the pack I found contained extra files or do I need to find the entire thing somehow?

Thanks, and yeah, at this rate I will need luck to finish this.

You have to go through the list of required plugins and do 1 by 1. Maybe someone put together a big bundle, but I don't know of any. Part of the reason is the filters are updated frequently, so each one might be outdated in time etc....

The only files you need are the various .avsi and .dll's for the required plugins. The rest are documentation and instructions.

Maybe just leave out mctemporaldenoise for now, just leave levels and tweak in and see if it's ok

[Cyber Akuma]

Hmmm, I finally got it to work, though some stuff still isn't working that well and I am confused about the different ways to go about it.

By including MCTemporalDenoise, there are some ugly colored lines that appear during the transition effect. Not sure what would be the best way to go about this. Should I encode each video seperately with the levels/denoise thing as huffyuv or something then add them all together into one file with transitions only? Or would that still leave them too dark?

Also, I have more set of videos I need to do this too as well.

Problem is, some of them were day/dusk shots, and they seems to introduce noise when I load them through AviSynth (keep in mind hte only reason I am even using AviSynth is to add transitions, if I didn't need them I would just load the MP4 files into the DVD Authoring program directly).

Another set was also taken at night, but there was a lot of lighting so while still dark, they came out fairly decent unlike the example I posted, when I load this in the script with all these settings, its a little too light (the colors are faded) and a lot of extra noise is introduced. Using mctemporaldenoise blurs/smudges a lot of the video into almost an oil painting sort of image for any details, using just the levels and tweaks introduces a lot of extra noise...

Any suggestions on what I can try?

[poisondeathray]

You have to adjust/tweak the filter settings for each scene. I would fix each of the clips then add the transitions last

Noise is always introduced when you brighten videos (increase levels), and denoising is always damaging. Either live with it dark, or brighten it up and introduce some noise. You're going to have to make some tradeoffs. You should have shot with more light, or used a better camera

The problem is a lot of the data (90%) is Y' <16 (illegal for video, the "superdark" region) . Legal video is 16-235. If you leave it as is, all that stuff in the dark will not be seen on a regular DVD/TV setup. It will look the 1st screenshot in post #40. When you adjust the levels, you will bring out the noise in the dark areas.

This isn't avisynth's or ffmpeg's fault. If you load the clip in premiere or vegas it looks almost completely black (like screenshot #1 in post 40). The problem is most of your data is Y' <16 .

If you loaded the files directly without any filters and made a dvd, you will notice the dvd will be very dark on a regular TV. Your preview in the media player was using VRM9 renderless (which brightens & washes out the video - a dvd player won't be using this renderer - and all that data <16 will not be visible. That's the whole point of bringing the levels up). Test it and see for yourself

You can scale it to legal "TV" levels (16-235) by using levels or smoothlevels. This will be more "automatic" although you might want to make manual adjustments per scene, but the brighter you make it , the more noise will be revealed, and the stronger you will have to denoise (or maybe you can live with the noise)

e.g.
ffvideosource("video.mp4")
smoothlevels(preset="PC2TV")



I appended some screenshots to illustrate what is going on:

1 - this is approximately what it will look like on a dvd. Notice most of the data is Y' <16. The histogram shows 16-235 in black, and the brown bars are "illegal" 0-16 and 235-255. The histogram is rotated 90 degrees to the right




2 - after smoothlevels(preset="pc2tv"). Notice there the data is scaled to "legal" range now, but more noise is revealed in the dark regions




3 - vegas - notice it looks just as dark - most of the data is illegal and will not be seen on a dvd player. (i.e it will look black). The waveform monitor uses 0-100 IRE (below 0 will not be seen, and notice most of the data resides below zero)

[Cyber Akuma]

Quick question:

When I am encoding this n MeGui, it complains that I am not using the command ConvertToYV12, should I use this if I eventually plan to turn it into a DVD playable on a SDTV or no?

UPDATE: So I did a test-encode in huffyuv using VirtualDub and it worked, took about 20 minutes (I saw no difference between using that denoise option or not, so I didn't use it). Encoded the video in about 20 minutes, into a 10 gig file.

Since my DVD authoring program coudl not read this file, I loaded the avs in MeGui to encode it to h264...... and its taking an exstimated 11 hours at .7FPS.

.... ummm, what? I know I set my x264 encoder settings pretty high (one notch below Placebo), but even with high settings, is this even normal? Its a 1.7GHZ dual core laptop with 4 gigs of ram.

[poisondeathray]

you can say yes

it should be already in yv12 space if you are using ffvideosource(), but megui has "bug" that prompts when you use ffvideosource

megui doesn't encode to mpeg2 , so you shouldn't be using it if you are going to regular dvd

[Cyber Akuma]

megui doesn't encode to mpeg2 , so you shouldn't be using it if you are going to regular dvd

The dvd authoring program will be doing the final encode for me, since I had no idea how DVDs are formatted nor how to properly resize all the files to fit on a single DVD, I was just going to encode to a stupidly high bitrate to minimize quality loss as much as possible then load that into the DVD authoring program and let it encode it to however a DVD should be formatted.

P.S. I posted an update about encoding times, I was hoping to update my post before anyone could respond but it seems you were too quick

Can you take a look at the update I posted?

[poisondeathray]

what is your script ?

heavy denoisers will dramatically slow encoding times , and if you use high quality x264 settings it can be very slow as well

you don't need high quality super compressed here, you can compensate with poor analysis and high bitrate, your encode will be 10-20x faster but almost no difference in quality (a larger filesize to compensate)

use a faster preset like --preset very fast , and a lower rate factor like crf ~10

huffyuv should have worked, what program were you using that didn't accept it ?

[budwzr]

Hey, I'm jumping in here late, and can't get up to speed, so I'm jumping back out, isn't that special?

[Cyber Akuma]

Well, that is the first set encoded.

The parts in the middle don't need any transition effects so I can just run them straight into my dvd authoring program without putting them though AviSynth first, ill probably use the textsub feature it to create some titlecards before all the major scenes.... (I'll have to google that)

But I do still have one last (I hope its the last) issue. The last set of videos were also taken in the dark... but there was a lot of light around everyone. I can see a lot more detail when I load these videos with the tweak and levels settings you recommended... but it also washes out some of the colors this way.

FullMov=AudioDub(Vid1,Aud1).Levels(0,1.3,255,15,25 5,false ).Tweak(sat=1.1,coring=false)

I tried to google how the Tweak and Levels commands work, but I am even more confused now. Apparnetly.... the levels command just tells AviSynth what the color level boundaries are and has nothing to do with the brightness or contrast, instead thats the Tweak command? But there are only two arguments you gave me for the tweak command, and one of them is boolean...

I'm confused.

Here is what it looks like normally (played straight instead of through AviSynth):
http://img571.imageshack.us/img571/9374/sany0364mp4snapshot0000.jpg

And here is what it looks like with the levels and tweak command you gave me:
http://img687.imageshack.us/img687/8426/test13avssnapshot000020.jpg

While a lot more detail can be seen, the skintones are a bit washed out as well. Is there any way I can try to adjust brightness, contrast, gamma (whatever that is), etc to see what I can do, or is this a tradeoff I will have to choose between?

huffyuv should have worked, what program were you using that didn't accept it ?

DVD Flick

Before this I was considering DVD Styler but the menu system was far too inflexible, it wouldn't even let you use your own images.

[poisondeathray]

But I do still have one last (I hope its the last) issue. The last set of videos were also taken in the dark... but there was a lot of light around everyone. I can see a lot more detail when I load these videos with the tweak and levels settings you recommended... but it also washes out some of the colors this way.


While a lot more detail can be seen, the skintones are a bit washed out as well. Is there any way I can try to adjust brightness, contrast, gamma (whatever that is), etc to see what I can do, or is this a tradeoff I will have to choose between?

levels affects brightness, gamma and contrast . You can use coloryuv() to change color or luminance, and tweak for saturation , hue , brightness . The boolean false is for coring, which means do not clamp 16-235 (allow overshoots, undershoots)

http://avisynth.org/mediawiki/Levels
http://avisynth.org/mediawiki/ColorYUV
http://avisynth.org/mediawiki/Levels

All this might be difficult to explain unless you are familiar with the concepts or photoshop or similar programs. You have to adjust the parameters and look at a waveform monitor or histogram while you are adjusting it to understand what is going on. Use avsp , use histogram() and hit f5

Or you can do your adjustments in vdub with gradation curves, colormill, levels - this might be easier for you as you can see the preview and what you are doing and the controls are more similar to photoshop and graphic programs

You have to adjust the parameters per scene, or use some "auto" correcting filters, like smoothlevels(preset="pc2tv") , and then some auto brightening filter like hdragc() . You can set parameters like max_gain so it doesn't go overboard. All the filters have documentation, so read the instructions. If you still don't understand after reading the instructions and examples, then ask specific questions

huffyuv should have worked, what program were you using that didn't accept it ?

DVD Flick

Before this I was considering DVD Styler but the menu system was far too inflexible, it wouldn't even let you use your own images.

You can also try avs2dvd , it seems popular among the free software. It uses a better mpeg2 encoder, hcenc, than dvd flick (which uses ffmpeg)

[Cyber Akuma]

I put a few files through DVD Flick to see how they would come out before I start encoding the rest to make sure everything is ok, and the source video, which is 640x480, came out at 720x480.

Yes, I know that a SDTV can't do 640x480 and it would have had to encode to some other resolution, but I really don't want the video's aspect ratio to be messed with and the entire video squashed or stretched. Would this appear correctly on a SDTV or do I need to mess around with DVD Flick's settings to try to get it to encode it at the proper aspect ratio, whatever that would be?

[poisondeathray]

It would be 4:3 setting - I don't use DVD Flick, so I don't know if the setting is accessible

(for DVD there are only 2 choices : 4:3 and 16:9) , DVD doesn't use "square pixels"

Your original video was 640x480 with square pixels, or 4:3 AR

[Cyber Akuma]

Sigh, I have a new problem due to some new files they want me to add.

I couldn't really figure out the best way to adjust lighting and everything so I am just using the same setting for all the videos and will make a test DVD to see how they will come out.

However, there are some new videos they want me to throw in..... and they were recorded off a cellphone. So they have different lighting, use mono audio, and everything. However, that is not the problem, the problem is their ridiculous framerate of 11.097701 (At least thats what VLC tells me) so anything I try to play the videos after combining them into one in AviSynth makes it go bonkers and think I have a 5 second movie, there is no "fps=" setting for FFVideoSource like there is for DirectShowSource.

[poisondeathray]

you can use some auto filters, like hdragc() or autolevels(), but you should bring them into legal range first. You can place a limit on hdragc, because it will probably be too strong at default settings (and reveal too much noise). Read the instructions, because there are many settings to tweak it better

e.g.
coloryuv(levels="pc->tv")
hdragc(max_gain=1)

You can change the framerate , ffvideosource has fpsden, fpsnum arguments (for denominator, and numerator) - read the documentation; or use other avisynth functions like convertfps() or changefps() . You probably want to use changefps() because it will insert duplicate frames (if the framerate is really 11, you need to insert more than double) . If the real framerate was something else, then you could use the fpsden, fpsnum arguments , or use assumefps() (assumefps doens't change the frame count, it just speeds or slows down)

[Cyber Akuma]

Ummm, wow.

I tried to use ConvertFPS(60) and it is complaining that the minimum framerate I can use is 333.3333.

Is there something wrong with this script or is it the video's fault?

Vid1=FFVideoSource("VIDEO0036.MP4")
Aud1=FFAudioSource("VIDEO0036.MP4")
Vid2=FFVideoSource("VIDEO0037.MP4")
Aud2=FFAudioSource("VIDEO0037.MP4")
Vid3=FFVideoSource("VIDEO0038.MP4")
Aud3=FFAudioSource("VIDEO0038.MP4")
Vid4=FFVideoSource("VIDEO0039.MP4")
Aud4=FFAudioSource("VIDEO0039.MP4")


Mov1=AudioDub(Vid1,Aud1).ConvertFPS(60).Levels(0,1.3,255,15,255,false ).Tweak(sat=1.1,coring=false)
Mov2=AudioDub(Vid2,Aud2).ConvertFPS(60).Levels(0,1.3,255,15,255,false ).Tweak(sat=1.1,coring=false)
Mov3=AudioDub(Vid3,Aud3).ConvertFPS(60).Levels(0,1.3,255,15,255,false ).Tweak(sat=1.1,coring=false)
Mov4=AudioDub(Vid4,Aud4).ConvertFPS(60).Levels(0,1.3,255,15,255,false ).Tweak(sat=1.1,coring=false)

Eff1=TransVenetianBlinds(Mov1, Mov2,100,type="diam")
Eff2=TransVenetianBlinds(Eff1, Mov3,100,type="diam")
Eff3=TransVenetianBlinds(Eff2, Mov4,100,type="diam")

FullMov = Eff3

FadeIO(FullMov, 50)