Aviutl Tips, Tricks and Support thread

[MaverickTse]

[QUOTE=Freelife2345;2378623]

For vegas and mkv, you can try mkvtomp4

If you want it side by side, you can temporarily resize the 2 video layers and put them side by side

mkvtomp4 got my file to work on vegas but somehow I am not able to get the two files playing. I read somewhere I have to choose multi camera or something. I can hear the sound but image just remain static.

As for the qaac encoder you recommend using itune, I download it but I still see messages saying failing to export to x264. Is there something still missing or not done properly

You have to CHOOSE the [qaac] encoder in the setting dialog! (on the right side)

[Freelife2345]

I attached a print screen. Took me awhile to find that hidden setting. I think that is what you meant on the top right about choosing the encoder.

[Freelife2345]

I am a newbie on this and I am looking through the instructions. I can't seem to find how to have two videos one on top of another to compare. I see the four layers on the bottom (two for video and two for audio) but in terms of comparison, I only can see one video at a time.

For reference the specs of the video are in post 15 in this thread
https://forum.videohelp.com/threads/370556-Need-help-in-editing-two-videos-of-different...ngth?p=2378161



You actually don't care about the audio for this editing part - you're just using the video 2 as a visual guide to edit video 1. But you might still use the audio to assist your editing. You can disable tracks by clicking the layer number (the track will become "greyed out" )

Import video1 (drag & drop) to the timeline and it will ask if you want to match the specs in the check box, check yes. This means the timeline will be 848x480, 29.97fps just like video1. If you have audio grouped, track 1 will be video 1, track 2 will be video 1's audio

Then drag video2 to the timeline (track 3, and 4 for audio)

The layer order is reversed in aviutl - layers with higher numbers are actually "on top". So Video2 on Track3 will "cover up" Video1 on Track 1 . Resize Video2 to the same dimensions as Video1 (you're only using it as a guide)

The "opacity" parameter is adjusted by"clearness" in aviutl. 100% clearness is 100% transparent. So if you adjust video2 to 50% clear, you should "see" a 50-50 mix of video 1 and 2. You want to make edits to video 1 by cutting out parts that are absent in video 2. You would do that by marking in / out by splitting ("s" key ) the in and out points and hitting delete. Then sliding the remaining video over . Unfortunately aviutl doesn't have "ripple deleteing" like retail NLE's which automatically slides the video over after you delete

When you're done, you can disable the unwanted track 2 and export the video only. Then use a muxer like mkvtoolnix, mp4box etc... to mux the newly edited video with original Japanese audio from video #2.

The editing itself will be easier in a retail NLE - But aviutl can import almost anything because of L-smash. You'll have great difficulty importing a MKV as-is into a retail NLE. (you would need workarounds like muxing into MP4, or re-encoding to a lossless intermediate). You can also export with x264cli directly in aviutl. It's more difficult in retail NLE's to do that, and only x264vfw

I am almost close to the finish product. Now, it is just about exporting the right layers. Laid out both videos on the timeline. File 1's video is on layer 1, audio (Korean) is on layer 2, file 2's video is on layer 3 and audio (Japanese) is on layer 4. I did editing on layer 1 and some editing on the audio on layer 4. These are the two layers I really want to keep when I export. I try exporting the file into mp4 with two layers enabled and I ended up getting a mute file. I suspect the audio in layer 4 needs to be moved to layer 2? I even gave that a try but my export file ended up being in the korean audio that I didn't want.

I hope I can just export one file with both layers I want. If not, I don't mind exporting the video and audio separately and then use mkvtoolnix or muxer to merge them. Please let me know what I can do.

[poisondeathray]

I am almost close to the finish product. Now, it is just about exporting the right layers. Laid out both videos on the timeline. File 1's video is on layer 1, audio (Korean) is on layer 2, file 2's video is on layer 3 and audio (Japanese) is on layer 4. I did editing on layer 1 and some editing on the audio on layer 4. These are the two layers I really want to keep when I export. I try exporting the file into mp4 with two layers enabled and I ended up getting a mute file. I suspect the audio in layer 4 needs to be moved to layer 2? I even gave that a try but my export file ended up being in the korean audio that I didn't want.

I hope I can just export one file with both layers I want. If not, I don't mind exporting the video and audio separately and then use mkvtoolnix or muxer to merge them. Please let me know what I can do.

I would have avoided editing the audio if it was possible (from your description , it should have been), because now you have to re-encode both (both will lose quality, and the bitrates / quality looked to be low to begin with)

You don't have to move any layers

Post the error message or screenshot of the console output (the screen while it's encoding, the progress report)

And yes, although it should work, you can still export the audio and video separately then mux them with something else like mkvtoolnix

[MaverickTse]

I am a newbie on this and I am looking through the instructions. I can't seem to find how to have two videos one on top of another to compare. I see the four layers on the bottom (two for video and two for audio) but in terms of comparison, I only can see one video at a time.

For reference the specs of the video are in post 15 in this thread
https://forum.videohelp.com/threads/370556-Need-help-in-editing-two-videos-of-different...ngth?p=2378161



You actually don't care about the audio for this editing part - you're just using the video 2 as a visual guide to edit video 1. But you might still use the audio to assist your editing. You can disable tracks by clicking the layer number (the track will become "greyed out" )

Import video1 (drag & drop) to the timeline and it will ask if you want to match the specs in the check box, check yes. This means the timeline will be 848x480, 29.97fps just like video1. If you have audio grouped, track 1 will be video 1, track 2 will be video 1's audio

Then drag video2 to the timeline (track 3, and 4 for audio)

The layer order is reversed in aviutl - layers with higher numbers are actually "on top". So Video2 on Track3 will "cover up" Video1 on Track 1 . Resize Video2 to the same dimensions as Video1 (you're only using it as a guide)

The "opacity" parameter is adjusted by"clearness" in aviutl. 100% clearness is 100% transparent. So if you adjust video2 to 50% clear, you should "see" a 50-50 mix of video 1 and 2. You want to make edits to video 1 by cutting out parts that are absent in video 2. You would do that by marking in / out by splitting ("s" key ) the in and out points and hitting delete. Then sliding the remaining video over . Unfortunately aviutl doesn't have "ripple deleteing" like retail NLE's which automatically slides the video over after you delete

When you're done, you can disable the unwanted track 2 and export the video only. Then use a muxer like mkvtoolnix, mp4box etc... to mux the newly edited video with original Japanese audio from video #2.

The editing itself will be easier in a retail NLE - But aviutl can import almost anything because of L-smash. You'll have great difficulty importing a MKV as-is into a retail NLE. (you would need workarounds like muxing into MP4, or re-encoding to a lossless intermediate). You can also export with x264cli directly in aviutl. It's more difficult in retail NLE's to do that, and only x264vfw

I am almost close to the finish product. Now, it is just about exporting the right layers. Laid out both videos on the timeline. File 1's video is on layer 1, audio (Korean) is on layer 2, file 2's video is on layer 3 and audio (Japanese) is on layer 4. I did editing on layer 1 and some editing on the audio on layer 4. These are the two layers I really want to keep when I export. I try exporting the file into mp4 with two layers enabled and I ended up getting a mute file. I suspect the audio in layer 4 needs to be moved to layer 2? I even gave that a try but my export file ended up being in the korean audio that I didn't want.

I hope I can just export one file with both layers I want. If not, I don't mind exporting the video and audio separately and then use mkvtoolnix or muxer to merge them. Please let me know what I can do.

For multi-language audio tracks, do not mux them in AviUtl. Everything will get merged into a single track of stereo audio.
If you did not cut the audio tracks, just mux them with the video using MKVToolnix.
If you modfied the audio in AviUtl, DISABLE one of the audio track (by clicking on the layer name), then save it as WAV. Then repeat the process by disabling another audio track.

If your preview has sound but none after encoding, it is likely to be an option at the bottom of the filename selection dialog (No audio) just before encoding

[Freelife2345]

I went back and re did some parts without touching the audio this time. I export the video file successfully even though the audio file failed to export. I was able to still merge the video file created from aviutl and the audio of the original file using mkvtoolnix. I have attached the progress report below.

------------------------------------------------------------------------------------------------------------------------------
[C:\Users\test.mkv]
------------------------------------------------------------------------------------------------------------------------------
auo [info]: x264guiEx 2.24 / Windows 8.1 (x64) / Intel Core i5-4210U @ 1.70GHz [TB: 1.91GHz] (2C/4T)
auo [info]: converting YUY2 -> nv12p, using AVX AVX2
auo [info]: x264 version: 0.142.2491kMod 24e4fed --bit-depth=8 --chroma-format=all
auo [info]: x264 options...
--crf 23 --frames 214896 --input-res 1280x720 --input-csp nv12 --fps 60000/1001 -o "C:\Users\test.mkv" "-"
raw [info]: 1280x720p 0:0 @ 60000/1001 fps (cfr)
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX AVX2 FMA3 LZCNT BMI2
x264 [info]: profile High, level 3.2
x264 [info]: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 fade_compensate=0.00 psy_rd=1.00:0.00
mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2
threads=6 lookahead_threads=1 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0
fgo=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25
scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0000 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4
ip_ratio=1.40 aq=1:1.00
frames fps kb/s elapsed remain size est.size

x264 [info]: frame I:1653 Avg QP:19.43 size: 20765
x264 [info]: frame P:82754 Avg QP:22.11 size: 2762
x264 [info]: frame B:130489 Avg QP:19.29 size: 240
x264 [info]: consecutive B-frames: 18.8% 0.0% 2.0% 79.2%
x264 [info]: mb I I16..4: 25.9% 65.9% 8.2%
x264 [info]: mb P I16..4: 1.0% 2.1% 0.1% P16..4: 9.1% 3.4% 1.7% 0.0% 0.0% skip:82.5%
x264 [info]: mb B I16..4: 0.0% 0.0% 0.0% B16..8: 5.7% 0.1% 0.0% direct: 0.0% skip:94.1% L0:69.9% L1:29.0% BI: 1.0%
x264 [info]: 8x8 transform intra:65.7% inter:81.5%
x264 [info]: coded y,uvDC,uvAC intra: 28.3% 36.3% 6.2% inter: 1.3% 1.1% 0.0%
x264 [info]: i16 v,h,dc,p: 50% 28% 6% 17%
x264 [info]: i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 33% 22% 24% 3% 3% 4% 3% 3% 4%
x264 [info]: i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 31% 24% 13% 5% 6% 6% 6% 5% 4%
x264 [info]: i8c dc,h,v,p: 68% 15% 15% 2%
x264 [info]: Weighted P-Frames: Y:0.3% UV:0.2%
x264 [info]: ref P L0: 71.2% 12.3% 10.7% 5.9% 0.0%
x264 [info]: ref B L0: 92.8% 6.1% 1.1%
x264 [info]: ref B L1: 98.8% 1.2%
x264 [info]: kb/s:656.66

encoded 214896 frames, 70.05 fps, 656.67 kb/s, duration 0:51:07.53
auo [info]: CPU Usage: Aviutl: 24.87% / x264: 67.82%
auo [info]: x264 Encode Time: : 0Hr 51Min 14.7Sec
auo [info]: Encoding audio with qaac. ABR (AAC) Bitrate Mode, 0kbps
auo [info]: ".\3rdParty\qaac\qaac.exe" -q 2 --abr 0 --ignorelength -o "C:\Users\test_audio.m4a" "-"
auo [error]: Audio output file is missing. Audio encoding with qaac has failed.
auo [error]: Audio encoding commandline...
auo [error]: ".\3rdParty\qaac\qaac.exe" -q 2 --abr 0 --ignorelength -o "C:\Users\test_audio.m4a" "-"
auo [info]: qaac CPU Usage: 0.00%
auo [info]: Total Encode Time: : 0Hr 56Min 13.8Sec



My video somehow doesn't turn out to be full screen like when I first import the video. I resize everything back to how the print screen looks like when I first import the video but somehow it is still not coming out the same. Is there more settings I need to tweak?

[Freelife2345]

This is the print screen of the settings I'm talking about

[poisondeathray]

That's the way I recommend you do it - so you only have to encode the video, and you don't lose quality on the audio. Aviutl can't passthru audio (stream copy).

The log says you are exporting 1280x720. But Video #1 is 848x480 . So that probably means you have the wrong project size. Recall I told you do drag Video #1 first to the timeline, and match the specs? That would make your timeline 848x480, same as video #1 . Think of the project settings like a canvas for a painting. If you had a 1280x720 project settings, a 848x480 asset would be in a little window and wouldn't fill the frame unless you upscaled it.

[Freelife2345]

That's the way I recommend you do it - so you only have to encode the video, and you don't lose quality on the audio. Aviutl can't passthru audio (stream copy).

The log says you are exporting 1280x720. But Video #1 is 848x480 . So that probably means you have the wrong project size. Recall I told you do drag Video #1 first to the timeline, and match the specs? That would make your timeline 848x480, same as video #1 . Think of the project settings like a canvas for a painting. If you had a 1280x720 project settings, a 848x480 asset would be in a little window and wouldn't fill the frame unless you upscaled it.

I ended up using handbrake to reencode it to 848x480. Now it everything looks good. Then I realize, there are some scenes here and there where the audio is off by a sec or two. It wasn't that obvious when I played the file in aviutl. Should I still go back to aviutl to fix it or will you recommend I use audacity or some other software to fix?

[poisondeathray]

Describe what is wrong with it - are some scenes in sync, yet others are not in sync ? When you muxed it at 1280x720 (before re-encoding) was it in sync? Did you use the correct framerate in mkvmerge ? Did you re-encode and it became out of sync in handbrake ? Did you use VFR in handbrake ?

[MaverickTse]

auo [info]: Encoding audio with qaac. ABR (AAC) Bitrate Mode, 0kbps
auo [info]: ".\3rdParty\qaac\qaac.exe" -q 2 --abr 0 --ignorelength -o "C:\Users\test_audio.m4a" "-"

Did you touch something that you shouldn't have?

Anyway, since you seems getting lost with the editing, here is a sample project for your reference.
The first half of the project is a simple overlay, the other one is Side-by-Side.
One of the audio track is muted.
Note that you may have to uncheck "sync with video" option for the audio objects for multi-track audio to work

[MaverickTse]

Describe what is wrong with it - are some scenes in sync, yet others are not in sync ? When you muxed it at 1280x720 (before re-encoding) was it in sync? Did you use the correct framerate in mkvmerge ? Did you re-encode and it became out of sync in handbrake ? Did you use VFR in handbrake ?

I guess it is because one of the source videos is MKV.
MKV container can contains audio-delay info that can screw things up if doing a straight
demux -> mux into another video

Since I don't think excellent audio quality is required in his case, it is fine just to export the audio tracks individually from AviUtl to WAVs.
So from AviUtl, an MP4 + 2x WAV can be obtained.
Then they can be muxed together.

[Freelife2345]

Describe what is wrong with it - are some scenes in sync, yet others are not in sync ? When you muxed it at 1280x720 (before re-encoding) was it in sync? Did you use the correct framerate in mkvmerge ? Did you re-encode and it became out of sync in handbrake ? Did you use VFR in handbrake ?

Looks like it was a bit off already before I reencode using handbrake. Yeah the audio is in sync in majority of the file but there are a few that are slightly off by a sec or two after export it and watch it in media player classic. Somehow when I watch it in aviutl, it looked sync to me or not very obvious. I didn't see or play around with any frame rate option in mkv merge gui. All I did was input the file with the japanese audio and the file from aviutl, and mux.

[poisondeathray]

If it's in sync in aviutl, try what maverick suggested - disable all the other tracks except for the Japanese audio, and export the audio as WAV , then mux with the original export

[gracemichelle]

This software is awesome but the interface looks a bit dated. I just started using this and exploring all the features. I know there are short cuts like pressing the s to split.It was working awhile ago. But, anyone know what can be reasons it will not work? Are there settings that need to be adjusted for short cuts to be enabled? I find myself pressing the s button a few times one night and have no response. I ended up doing a right click to split.

[MaverickTse]

This software is awesome but the interface looks a bit dated. I just started using this and exploring all the features. I know there are short cuts like pressing the s to split.It was working awhile ago. But, anyone know what can be reasons it will not work? Are there settings that need to be adjusted for short cuts to be enabled? I find myself pressing the s button a few times one night and have no response. I ended up doing a right click to split.

It works on my machine.
May be your keyboard starts having problem...

To view and change shortcut keys:
File → SETTINGS → Shortcut Key
(Not All items in this dialog have been translated to English, around 5% remains in Japanese)

This software was first released in Japan around 10 years ago... and yes, it is aging (just like AviSynth and VDub)

[gracemichelle]

No wonder I see some stuff in japanese. When I run the installation from the zip file, it turns out english but when I drag the file onto desktop and then install it comes out in japanese.

by the way, do you know of any free software out there where I can mux the video and audio into a mp4? I know mkvmerge only outputs mkv.

[MaverickTse]

No wonder I see some stuff in japanese. When I run the installation from the zip file, it turns out english but when I drag the file onto desktop and then install it comes out in japanese.

by the way, do you know of any free software out there where I can mux the video and audio into a mp4? I know mkvmerge only outputs mkv.

mp4box, l-smash, ffmpeg...
Note that MKV supports multiple audio tracks, MP4 don't.

You can open a new thread if you want more details on this. (since this is a general transcoding issue)

[MaverickTse]

AviUtl Combo Installer Test
I have made an experimental installer for AviUtl:

http://1drv.ms/1xmhIw0

This installer allows you to install either English or Japanese version of AviUtl, with plugins, scripts and utilities.
User can customize the components to install, while the installer should automatically handle the 32bit vs 64bit, Win7 vs XP matter and also skip some components if hardware does not support (e.g. AVX,AVX2, XOP, Pixel Shader 3, DirectX).
Apple Application Support will be installed automatically (not the whole iTune) if user select to install qaac.
Readme/License and Publication Sources are pulled together into a single folder (mostly)
Adaptation for 4k video is provided as an option at the end of installation.
Desktop icon can also be created.
An uninstaller will be generated... but removal is not complete.

If anyone does tried this out, please report any issue you encountered.

[gracemichelle]

I notice after outputting my video from aviutl and play it in media player classic the video will sometimes not run smoothly. it will pause a second once awhile. What kind of codec am I missing or is there something wrong with my settings when I output?

[MaverickTse]

I notice after outputting my video from aviutl and play it in media player classic the video will sometimes not run smoothly. it will pause a second once awhile. What kind of codec am I missing or is there something wrong with my settings when I output?

Describe your workflow in details.
Usually what you get from the preview is more or less the same as the encoded output (AVI, MP4, H264 MKV)

[MaverickTse]

Porting a CustomFlare Script...

[racer-x]

I like the effect Mav, are you planning on releasing it?

[MaverickTse]

I like the effect Mav, are you planning on releasing it?

Yup. I'm also porting other Fx. I'll release them when all done.

[poisondeathray]

@Mav - Does it use the custom object => lens flare , or a modification of it ?

[poisondeathray]

Example of animating lights on a still image using aviutl's tracking line as a mask

gif preview


Short description of steps:

1) Still image of foreground sign is cut out in an image editor like gimp, paint.net etc.... You need to place it over where the sign obscures the light. This is included in the zip as "0 dark alpha.png" . The background is "0 dark.png". Those 2 versions were made "darker" to accentuate the effect. The original image is "las-vegas-at-night-q28s82kd.jpg" downloaded from here:
http://galleryhip.com/welcome-to-las-vegas-sign-at-night.html

2) Background on track #1. Place a right click => new media object => custom object => tracking line on a lower track number like 5 . The "clearness" is adjusted on the background on track #1 temporarily so it's partially transparent, so you can see what is going on as you adjust and animate the tracking line

3) Adjust the tracking line parameters so it covers the path of the lights. It has to be white to show areas affected in the mask. You can play with it to make it animate faster, longer line , etc.. whatever you feel like. Copy it and paste it into scene 1 on track 2, where it will function as the mask, then disable the copy in the main root scene. You can readjust the "clearness" of the background layer to full opacity. Back in scene 1 you need to make a "black" background for track 1. Right click=> new media object => graphic => background from the dropdown, make it black

4) Track 2 in the root scene has copy of the background on track 1, but has the mask effect applied. Match with original size is selected, and select from scene => scene 1 where you placed the tracking line layer

5) Apply the glow effect and adjust to your tastes

6) Place to "0 dark alpha.png" on track #3, so it "covers up" the moving lights

Another way of describing it , is the white tracking line is "revealing" parts of the "glowing" version of the image

[poisondeathray]

BTW, in addition to playing with the glow parameters, you can play with the transfer modes to enhance the "glow" effect on track 2 in the last project. "Add" is usually a good one



More Aviutl tracking line fun

stylistic ECG effect

same thing
right click => new media object => custom object => tracking line , adjust the settings

Basically any "path" like effect can be replicated with the tracking line, it can be used as masks like in the previous project, and for other ideas look at the 1st video maverick posted on the tracking line (unfortunately it was in Japanese, but you can still get some ideas)

[racer-x]

Yes tracking line make a good animated mask. I used it a few time on some projects with great results.

[MaverickTse]

New FX Script Release 1: Trump card simulation

By preparing a few symbol images, the script(Custom object) automatically merge them into cards of different patterns

(The 7z file includes the webp demo video so it seems large)
Please view the webp file in a separate browser tab

[MaverickTse]

FX Script Release 2: CustomFlare (with 3D Cam support)

The script itself is VERY COMPLICATED, but you can get the Flare samples as shown in the earlier post by using the included EXA files.

Also included a sample on how to move the flare object around in 3D space.

Note that this Flare object REQUIRES a CameraControl object in the previous track.

Two anchor points can be set to indicate the light source and ray projection direction.