AVANTI - FFmpeg/Avisynth GUI (support thread).

[nick_netherlands]

/k ""C:/myVideos/Batch1.bat" "@destination"" from your manual.

google CMD.exe /K:

Quote Characters in a command
If /C or /K is specified, then the remainder of the command line is interpreted as a command and the following logic is used to process quote (") characters:

1. If all of the following conditions are met, then quote characters on the command line are preserved:

- No /S switch (Strip quotes)
- Exactly two quote characters
- No special characters between the two quote characters, where special is one of: & < >( ) @ ^ |
- There are one or more whitespace characters between the the two quote characters
- The string between the two quote characters is the name of an executable file.

2. Otherwise, old behavior is to see if the first character is a quote character and if so, strip the leading character and remove the last quote character on the command line, preserving any text after the last quote character. To negate this behaviour use a double set of quotes "" at the start and end of the command line.

CMD /c ""c:\work\my reports\profit ^& Loss.doc""

____________________________________________

Slash issue fixed???????????? In the next Avanti release???

[Chris K]

/k ""C:/myVideos/Batch1.bat" "@destination"" from your manual.

Aah... now I remember! There are two paths in the example from the manual.

Slash issue fixed???????????? In the next Avanti release???

Yes, in the next release but since you're on XP, I could make a 0.8.4 alpha available if you want to check things out with more complex things. I can't do that for Win7 users because that requires a lot of time to find and fix the issues Avanti has there.

I can send you a PM with a link if you wish.

[nick_netherlands]

Will wait for next version. It is working at this moment.

By the way the AviSynth AssumeFPS is a great tool. Perhaps it is possible with FFmpeq but can it do the magic that easy?

from the Help:
The AssumeFPS filter changes the frame rate without changing the frame count (causing the video to play faster or slower). It only sets the framerate-parameter
If sync_audio is true, it also changes the audio sample rate by the same ratio, the pitch of the resulting audio gets shifted.

Nick

[Chris K]

Will wait for next version. It is working at this moment.

Ok!

By the way the AviSynth AssumeFPS is a great tool. Perhaps it is possible with FFmpeq but can it do the magic that easy?

from the Help:
The AssumeFPS filter changes the frame rate without changing the frame count (causing the video to play faster or slower). It only sets the framerate-parameter
If sync_audio is true, it also changes the audio sample rate by the same ratio, the pitch of the resulting audio gets shifted.

I know! The Avanti frame rate conversion options for video/audio uses it. But there it's only used for speed ups/downs of about 4 percent. I found it a good idea in the script you posted earlier to top-off the high frequencies when the pitch is changed in such a wide range

Chris

[gerrydigits]

Avanti is really impressive and easy to use. There is one thing I cannot find out: How to generate testsrc test patterns in Avanti. I can do it no problem from command line in ffmpeg. Avanti always reports an error No source 1 path found. But of course there is no source path as ffmpeg is generating the source. Example: how would you generate:

ffmpeg -f lavfi -i testsrc -t 30 -pix_fmt yuv420p testsrc.mp4

or
ffmpeg -f lavfi -i color=color=red -t 30 red.mp4

from inside Avanti?

[Chris K]

Avanti always reports an error No source 1 path found. But of course there is no source path as ffmpeg is generating the source. Example: how would you generate:

ffmpeg -f lavfi -i testsrc -t 30 -pix_fmt yuv420p testsrc.mp4

or
ffmpeg -f lavfi -i color=color=red -t 30 red.mp4

from inside Avanti?

Hi gerrydigits,

This is something for the -new option, where you are free to compose your own command line.

Currently you have to load a "dummy" source to let Avanti pass its source properties collection without the path not found error. I think I'll add some kind of nul source option in the next version to overcome this problem.

At the destination field enter your destination path like e.g.; c:\video_test_clips\testsrc.mp4

At the FFmpeg script editor enter at...

[USER_VIDEO_OPTIONS]

-new

-f lavfi -i testsrc -t 30 -pix_fmt yuv420p

-o "@destin1"

Since the -dvs and -das labels are not used, all settings at the main page are ignored.
edit: Only set the container field at the main page to mp4 because this is still functional.

The -o "@destin1" label points to the destination path.

This generates a nice 30 secs test clip for me

edit2: You could take a look at the tools section here at VideoHelp for a more suitable gui. Thanks for your interest anyway.

Chris

[gerrydigits]

Thanks for that Chris, I tried it and it did work. For me it is a valuable addition to be able to generate correct reference video and audio signals as well as be able to verify them. Hopefully one day everyone will do it G.

[Marchand]

FFMPEG - libx265 - BitDepth 10Bits - ChromaSubsampling (4.2.2. and 4.4.4.) Test in Win8-64bits

TEST: AVANTI - FFmpeg/Avisynth GUI 0.8.5 - yuv422p10le and yuv444p10le

Sample: 00200.v5.mkv

Geral
UniqueID/String : 77891342992512370738587943059064883682 (0x3A9953B578A45CD043EBF6D1B14009E2)
Nome completo : C:\Users\Cracker_Marchand\Videos\00200.v5.mkv
Formato : Matroska
Versão do Formato : Version 4 / Version 2
Tamanho do arquivo : 1.23 MiB
Duração : 7s 550ms
Taxa de Bits Total : 1 362 Kbps
Programa usado : Lavf56.4.101
Biblioteca usada : Lavf56.4.101

Vídeo
ID : 1
Formato : HEVC
Formato/Informações : High Efficiency Video Coding
Perfil do Formato : Unknown@L4.0
ID do Codec : V_MPEGH/ISO/HEVC
Duração : 7s 549ms
Largura : 1 920 pixels
Altura : 1 080 pixels
Proporção da imagem : 16:9
Modo de taxa de quadros : TConstante
Taxa de quadros : 23.976 fps
ColorSpace : YUV
ChromaSubsampling : 4:2:2
BitDepth/String : 10 bits
Biblioteca usada : x265 1.4+1-d2db9c1ab44b95e7:[Windows][GCC 4.8.2][32 bit] 16bpp
Configurações de codificação : wpp / ctu=64 / tu-intra-depth=1 / tu-inter-depth=1 / me=1 / subme=2 / merange=57 / no-rect / no-amp / max-merge=2 / temporal-mvp / no-early-skip / no-fast-cbf / rdpenalty=0 / no-tskip / no-tskip-fast / strong-intra-smoothing / no-lossless / no-cu-lossless / no-constrained-intra / fast-intra / open-gop / interlace=0 / keyint=250 / min-keyint=23 / scenecut=40 / rc-lookahead=15 / bframes=4 / bframe-bias=0 / b-adapt=0 / ref=2 / weightp / no-weightb / aq-mode=2 / aq-strength=1.00 / cbqpoffs=0 / crqpoffs=0 / rd=2 / psy-rd=0.00 / psy-rdoq=0.00 / signhide / lft / sao / no-sao-non-deblock / b-pyramid / cutree / rc=abr / bitrate=1500 / ratetol=1.0 / qcomp=0.60 / qpmin=0 / qpmax=51 / qpstep=4 / ipratio=1.40 / pbratio=1.30
Default : Sim
Forced : Não

Áudio
ID : 2
Formato : AAC
Formato/Informações : Advanced Audio Codec
Perfil do Formato : LC
ID do Codec : A_AAC
Duração : 7s 550ms
Nº de canais : 2 canais
Posições dos canais : Front: L R
Taxa de amostragem : 44.1 KHz
Biblioteca usada : Lavc56.1.100 libfaac
Default : Sim
Forced : Não

Sample: 00200.v6.mkv

Geral
UniqueID/String : 28008950936178548172784644320752761945 (0x151253E63CA3DE9301F0D762F52EFC59)
Nome completo : C:\Users\Cracker_Marchand\Videos\00200.v6.mkv
Formato : Matroska
Versão do Formato : Version 4 / Version 2
Tamanho do arquivo : 1.20 MiB
Duração : 7s 550ms
Taxa de Bits Total : 1 335 Kbps
Programa usado : Lavf56.4.101
Biblioteca usada : Lavf56.4.101

Vídeo
ID : 1
Formato : HEVC
Formato/Informações : High Efficiency Video Coding
Perfil do Formato : Unknown@L4.0
ID do Codec : V_MPEGH/ISO/HEVC
Duração : 7s 549ms
Largura : 1 920 pixels
Altura : 1 080 pixels
Proporção da imagem : 16:9
Modo de taxa de quadros : TConstante
Taxa de quadros : 23.976 fps
ColorSpace : YUV
ChromaSubsampling : 4:4:4
BitDepth/String : 10 bits
Biblioteca usada : x265 1.4+1-d2db9c1ab44b95e7:[Windows][GCC 4.8.2][32 bit] 16bpp
Configurações de codificação : wpp / ctu=64 / tu-intra-depth=1 / tu-inter-depth=1 / me=1 / subme=2 / merange=57 / no-rect / no-amp / max-merge=2 / temporal-mvp / no-early-skip / no-fast-cbf / rdpenalty=0 / no-tskip / no-tskip-fast / strong-intra-smoothing / no-lossless / no-cu-lossless / no-constrained-intra / fast-intra / open-gop / interlace=0 / keyint=250 / min-keyint=23 / scenecut=40 / rc-lookahead=15 / bframes=4 / bframe-bias=0 / b-adapt=0 / ref=2 / weightp / no-weightb / aq-mode=2 / aq-strength=1.00 / cbqpoffs=0 / crqpoffs=0 / rd=2 / psy-rd=0.00 / psy-rdoq=0.00 / signhide / lft / sao / no-sao-non-deblock / b-pyramid / cutree / rc=abr / bitrate=1500 / ratetol=1.0 / qcomp=0.60 / qpmin=0 / qpmax=51 / qpstep=4 / ipratio=1.40 / pbratio=1.30
Default : Sim
Forced : Não

Áudio
ID : 2
Formato : AAC
Formato/Informações : Advanced Audio Codec
Perfil do Formato : LC
ID do Codec : A_AAC
Duração : 7s 550ms
Nº de canais : 2 canais
Posições dos canais : Front: L R
Taxa de amostragem : 44.1 KHz
Biblioteca usada : Lavc56.1.100 libfaac
Default : Sim
Forced : Não

[rancur3p1c]

multi ffmpeg instances more efficient than multiple threads on 1 ffmpeg instance; also more effective at maxing out CPU
which is more useful for audio stripping/ripping/encoding/reencoding

plz can haz kthx

[Chris K]

multi ffmpeg instances more efficient than multiple threads on 1 ffmpeg instance; also more effective at maxing out CPU
which is more useful for audio stripping/ripping/encoding/reencoding

plz can haz kthx

Aside from that I don't know how to code that, it also doesn't fit in the current Avanti concept. How would I pipe multiple FFmpeg instances to Avanti and analyse/re-format their data for the log output?

Also, since Avanti doesn't come with a fixed FFmpeg version included, every few months something may become broken because of FFmpeg changes.

At June 2014, I updated templates that failed on new FFmpeg versions. I now found that some already fail again, even with rather cryptic error messages like "[mpeg2video @ 02f7a020] stuffing too large".

There isn't such a thing as a Avanti community that reports these things and possibly comes up with some suggestions to fix it (templates can be easily exchanged).

I guess it's time to hand over the baton to a new generation of gui developers.

Chris

[rancur3p1c]

multi ffmpeg instances more efficient than multiple threads on 1 ffmpeg instance; also more effective at maxing out CPU
which is more useful for audio stripping/ripping/encoding/reencoding

plz can haz kthx

Aside from that I don't know how to code that, it also doesn't fit in the current Avanti concept. How would I pipe multiple FFmpeg instances to Avanti and analyse/re-format their data for the log output?

Also, since Avanti doesn't come with a fixed FFmpeg version included, every few months something may become broken because of FFmpeg changes.

At June 2014, I updated templates that failed on new FFmpeg versions. I now found that some already fail again, even with rather cryptic error messages like "[mpeg2video @ 02f7a020] stuffing too large".

There isn't such a thing as a Avanti community that reports these things and possibly comes up with some suggestions to fix it (templates can be easily exchanged).

I guess it's time to hand over the baton to a new generation of gui developers.

Chris

AVANTI has fairly high coupling by design it looks like so yes that makes things difficult, but from the job queue when they're all running on the same template you spin multiple threads that each start an ffmpeg with a new item from the queue. New ff changes shouldn't* affect things like calling ffmpeg multiple times at the same time (once per core) with -threads 1 instead is I believe not just common but recommended.
For my quad core, I think if you were able to do it with 2 instances at once, that would be almost 100% improvement. I guess if I got an FX-8350 I would complain that it needs to do 4 encodes at once.
Could you buffer the output, have more than 1 progress bar, more than one text box in the GUI? Just random ideas I only have a little bit of knowledge to work with here.

*Honestly regular release iterations in FOSS where people change things that breaks random little features that you have to dig through documentation to find "oh yeah, it'll be fixed next release in a couple months probably" is retarded and not the worth the effort to deal with, makes me hate FOSS. I would just recommend builds once every 6 months or year and not worry about it until then. However, how would you personally do that? Without thorough unit tests in place [time]. I have no idea. Let the FF devs figure out what they broke no sense trying to fix your stuff for them.

I'd also like to associate the items in the queue to the output folder; I've had 12 or so folders to go through and I've just been queuing each one sequentially since it seems to associate the entire job to the output folder. There might be way to script it but I'm bad at memorizing CLI stuff

It's by no means a terrible program and it does what I needed and it saved me from having to figure out how to write batch files. So I'm glad I found it

[rancur3p1c]

what language is AVANTI? maybe I can go figure out how to spin multiple threads in it

edit: wait, you said "pipe"...that means it's like something of a script right?

edit2: gui4cli

edit3: http://gui4cli.com/html/manual/Operation.htm at the bottom under "Multiple instances of Gui4Cli" I believe this would be required to run ffmpeg multiple times at once. So, multiple instances, you could split the progress bar, status bar, and log output into a separate window, and each new instance would start its own separate window for the progress/status/log

[rancur3p1c]

oh, and remove requirement to save job before doing job. most jobs are just one time things

again, suggestions, not demands, you could classify them as complaints, but I'm just happy to have this software. Please don't feel bad!

oh, and it looks like with Audio settings codec MP3 where bitrate is set to SOURCE, it puts 'defaultk' into the ffmpeg string because it was thinking there would be a # like "256" to append the 'k' to e.g. 256k.

[rancur3p1c]

also files with unicode (asian characters it seems) in the name cause issues, possibly because 'question mark' can't be part of a filename in Windows

and when that happens the job queue stops processing

[feroblack]

Dear Folks,

I mux two audio streams one video stream but I need to fix audio levels of these two audio file How can I do that? also How can increase audio sound level by db unit?

[Chris K]

@rancur3p1c

That are a lot of posts with subjects to explain and discuss. I'll reply on them later.

[Chris K]

Dear Folks,

I mux two audio streams one video stream but I need to fix audio levels of these two audio file How can I do that? also How can increase audio sound level by db unit?

You need to recode the audio if you want to change the audio levels. Muxing means just copy the stream "as is".

for a single audio stream you would add it to the audio filter chain like;

[USER_AUDIO_OPTIONS]

-afin "volume=6dB"

You can give negative values like -6dB to decrease the volume.

[feroblack]

Dear Sir,

By this command as I understand just increase or decrease volume level. I want to equate both sound db level same

[Chris K]

Dear Sir,

By this command as I understand just increase or decrease volume level. I want to equate both sound db level same

There is nothing available to automate this but FFmpeg has some diagnostic options to get volume info, track_gain and track_peak

http://ffmpeg.mplayerhq.hu/ffmpeg-filters.html#volumedetect

[Parsed_volumedetect_0 @ 02ee25a0] n_samples: 3679488
[Parsed_volumedetect_0 @ 02ee25a0] mean_volume: -20.8 dB
[Parsed_volumedetect_0 @ 02ee25a0] max_volume: -2.2 dB
[Parsed_volumedetect_0 @ 02ee25a0] histogram_2db: 13
[Parsed_volumedetect_0 @ 02ee25a0] histogram_3db: 44
[Parsed_volumedetect_0 @ 02ee25a0] histogram_4db: 206
[Parsed_volumedetect_0 @ 02ee25a0] histogram_5db: 558
[Parsed_volumedetect_0 @ 02ee25a0] histogram_6db: 1602
[Parsed_volumedetect_0 @ 02ee25a0] histogram_7db: 3703

http://ffmpeg.mplayerhq.hu/ffmpeg-filters.html#replaygain

[Parsed_replaygain_0 @ 02dfff60] track_gain = -2.90 dB
[Parsed_replaygain_0 @ 02dfff60] track_peak = 0.941376

You need to get that info for both audio streams and compare them to see what values you have to give for the volume=XdB option.

But I think you should look for a dedicated audio software like perhaps Audacity that is able to normalize audio tracks.

[Marchand]

I guess it's time to hand over the baton to a new generation of gui developers.

Chris

Give up, never .... is an excellent software .... let's testing and adjusting, if necessary ..... light installation program, easy to use, various filtering and with many variations to be used .... That force be with you ... Do not give up.

[Chris K]

I guess it's time to hand over the baton to a new generation of gui developers.

Chris

Give up, never .... is an excellent software .... let's testing and adjusting, if necessary ..... light installation program, easy to use, various filtering and with many variations to be used .... That force be with you ... Do not give up.

Ok, thanks! I will try not to

But there are still the path issues on Windows 7/8 although it looks like it gradually gets more stable.

[Chris K]

AVANTI has fairly high coupling by design it looks like so yes that makes things difficult, but from the job queue when they're all running on the same template you spin multiple threads that each start an ffmpeg with a new item from the queue. New ff changes shouldn't* affect things like calling ffmpeg multiple times at the same time (once per core) with -threads 1 instead is I believe not just common but recommended.

You sure have a point here for audio conversions because I don't know of any audio codec that supports multi-threading. For video it may depend on the codec. As far as I can see x264 uses the four cores on my Phenom effectively. mpeg2video nowdays is also multi-thread and I believe even xvid. I'm not sure because I don't see much activity per core but it suddenly became turbo a few months ago.

http://gui4cli.com/html/manual/Operation.htm at the bottom under "Multiple instances of Gui4Cli" I believe this would be required to run ffmpeg multiple times at once. So, multiple instances, you could split the progress bar, status bar, and log output into a separate window, and each new instance would start its own separate window for the progress/status/log

FFmpeg is not directly launched and controlled by the Gui4Cli scripts. The scripts interface with winapi calls in the avcom32.dll like "CreatePipe" and "CreateProcess". Problem here is that I currently can only run one FFmpeg instance with these functions.

oh, and remove requirement to save job before doing job. most jobs are just one time things

I'm going to take a serious look at this. I need that joblist for processing but I can make it transparent for the user. It's already updated automatically when you press "Process Jobs". I may still keep the option to save a list but if not, it will create and use a default list automatically.

About mixed destination paths. I could change the last column that now contains the destination "name" so that you can set a full destination path at a arbitrary line. Then that path is used for all subsequent jobs until you set another path. Actually the same way as loaded templates work.

I'm also thinking of allowing to use the source path for destinations and add automatic renaming if the file already exists.

also files with unicode (asian characters it seems) in the name cause issues, possibly because 'question mark' can't be part of a filename in Windows and when that happens the job queue stops processing.

I cannot check that out because Windows doesn't allow me to create such a file name for testing. About canceling job processing at fatal errors. I'll see if I can change it so job processing will just mark the job as canceled in the log but continues with the next job.

oh, and it looks like with Audio settings codec MP3 where bitrate is set to SOURCE, it puts 'defaultk' into the ffmpeg string because it was thinking there would be a # like "256" to append the 'k' to e.g. 256k.

I tested it on a folder full of mp3s with both bitrate and sample frequency set to "source" and it did fine. But in contrast with video settings like frame size and frame rate where FFmpeg uses the source settings if you not specify any, for audio it comes up with default settings so I have to extract source audio bitrate and sample frequency myself from the FFmpeg log. Could it be that your source was variable bitrate and that FFmpeg didn't show a bitrate at all?

Chris

[tgbripper]

hi Chris,

thank you for the awesome Avanti tool.

Can I download encrypted HLS streams with it?

Here are some examples :

http://primary.hls-streaming.production.streaming.siriusxm.com/AAC_Data/thepulse/HLS_t...64k_large.m3u8

https://concdnsecure.ribob02.net/content/29bd68da342c48949a3b3fec5bae1c09/28f2d24d5efc...55c8e5a92e6f3a

In addition, does it work with rtmp too?

rtmp://spjfyhfrbp9td.cloudfront.net/cfx/st/mp3:mp3/52cc192eaf21b208234f60a730826fdca88086bb

All the links requires tokens/signatures which expires, but I believe youve got the idea.

Would be awsome with you have some ideas...

thank you very much!

[Chris K]

thank you for the awesome Avanti tool.

Can I download encrypted HLS streams with it?

Not that I know! Avanti does what FFmpeg can do and I don't think it can decrypt HLS streams (or any other).

I know nothing about this subject and only implement features that FFmpeg offers. Perhaps ask here ...

https://forum.videohelp.com/threads/311893-How-to-download-flash-video-RTMP-RTMPE-strea...-free-software

Chris

[vijm1972]

Dear Chris,

Thank you for wonderful Avanti tool.

Wanted to burn source TC on output file, is it possible using Avanti? My source files are in MXF format and wanted transcode them into MP4 format.

Thanks
Vijay

[Chris K]

Wanted to burn source TC on output file, is it possible using Avanti? My source files are in MXF format and wanted transcode them into MP4 format.

Hi Vijay,

I have no experience with that but it seems to be possible, if I look to the suggestions at this link...

http://miblog.alma.ch/2013/08/burnt-in-timecode-with-ffmpeg.html

I copied the web page to take a closer look but I don't have a MXF file with timecode to check things out.

edit: I assumed MXF has timecode as metadata but I now think the example extracts it from the stream?

About your earlier question concerning a watch folder. I'm working on a option to watch a local folder but perhaps you can post some details on what you expect from such a option and how you plan to use it?

Chris

[vijm1972]

Thank you for your reply Chris.
will check and update you. I will give you all the details about WF once I get free from above task.

Vijay

[vijm1972]

Hi Chris,

I had tried above link but no luck. I don't have much experience in any of the command line media tools. Is it possible for you to try that command at your end with any standard source files you have and provide the script to use in Avanti?

Watch Folder: I have to use Avanti to transcode my MXF source files using Watch Folder. I have multiple source files which I wanted to transcode it to .mov format, I wanted to create Job in Avanti which will automatically process these files. While crating the Job I can configure the Inbox and Outbox location. I only need to drop my source files into Watch Folder Inbox location and I will get the transcoded MOV files into outbox location.

Also wanted to know using Watch Folder can I stitch multiple source files in Avanti?

Regards
Vijay

[Chris K]

I found this very simple filter chain example that actually works in Avanti but it might not be exactly what you are looking for. I don't think it's directly related to the actual frame timecode.

-vfin "drawtext=fontfile=/Windows/Fonts/arial.ttf:fontsize=46:x=30:y=30:
rate=29.970:fontcolor=0x000000ff:timecode='00\:00\:00\:00'"

You can put it somewhere in the FFmpeg script. You need to set rate=xxx to your source frame rate. position and font size can be adapted.

Chris

[Chris K]

Watch Folder: I have to use Avanti to transcode my MXF source files using Watch Folder. I have multiple source files which I wanted to transcode it to .mov format, I wanted to create Job in Avanti which will automatically process these files. While crating the Job I can configure the Inbox and Outbox location. I only need to drop my source files into Watch Folder Inbox location and I will get the transcoded MOV files into outbox location.

Also wanted to know using Watch Folder can I stitch multiple source files in Avanti?

Currently my implementation of the watch folder is as follows...

1. You set up the environment (audio and video settings) to your destination format.
2. You select a watch folder and a destination folder.
3. You enable the watch option.
4. You can optionally minimize Avanti to the taskbar.

From then on every source clip (or number of clips) that are copied to the watch folder are converted to the destination folder.

This only works as long as Avanti is running and the watch option is enabled. If you copy clips to the watch folder while Avanti is not running, it ignores these the next time you enable the watch option. In that case you can first drop them on the job list and batch process them before you enable the watch option again.

edit: this is not yet available in the current Avanti version (0.8.6). It's just what I'm working on for the next.

Chris