I have a fairly simple flow to transcode an mxf file to mp4 with a watermark and a postroll.
For some reason it fails; 'Validate: Error opening input: Unknown error occurred.'
The flow works when bypassing the 'Insert media' filter.
I've attached the workflow and an error log.
All the help is welcome!
'Insert media' fails
-
- Posts: 13
- Joined: Wed Jun 14, 2017 4:35 pm
'Insert media' fails
- Attachments
-
- workflow.json
- (7.08 KiB) Downloaded 243 times
-
- 20240627-1551-3054-943e-c4814b24e643.json
- (346.91 KiB) Downloaded 220 times
Re: 'Insert media' fails
The logs also say:
Dissolve: The number of audio channels doesn't match
I never used the watermark node unfortunately, guess you found a bug but maybe you try to give the avi same tracks and channels as your source files have if you know it?
Dissolve: The number of audio channels doesn't match
I never used the watermark node unfortunately, guess you found a bug but maybe you try to give the avi same tracks and channels as your source files have if you know it?
emcodem, wrapping since 2009 you got the rhyme?
Re: 'Insert media' fails
Hi Bart,
welcome to the append function in Avisynth ehehehehehehehe
Jokes aside, I took a look at the logs and at the workflow and indeed Avisynth is telling the truth: the number of audio channels don't match.
Keep in mind that what's happening under the hood in the workflow is an append.
Avisynth is a frameserver, which means it serves frames to the final encoder as an uncompressed A/V stream.
In such a stream, the the meaning of "clips" is lost in the sense that clips don't exist, it's just one single uncompressed A/V Stream living in RAM.
When you have a file and use an indexer, your file is decoded in what is a stream and - to perform an append (i.e a "++" in the code) of two streams - those need to be exactly the same (i.e same size, colorspace, sampling, bit depth, framerate, audio channels and audio rate).
Unfortunately, I have no way of knowing how you created "D:\L1 NIEUWS EINDKAART YOUTUBE 2024.avi" but the idea here is that you should add a bunch of logic to the workflow after the A/V decoder so that it makes it match the specs of the .avi clip.
For instance, if we assume that L1 NIEUWS EINDKAART YOUTUBE 2024.avi is 1920x1080 4:2:0 yv12 25p 8bit planar with 2ch stereo 48000Hz 16bit we can change the workflow to work with that.
In the new workflow, we're using the A/V decoder to specify our desired output
followed by a deinterlacing step that deinterlaces the sources (if needed) and a ConvertFPS(25) step that brings everything to 25p before adding the watermark and then appending the media.
You can find the workflow in attachment, please change it according to your needs (i.e according to what the specs of L1 NIEUWS EINDKAART YOUTUBE 2024.avi are).
By the way, for those that are curious, while FFAStrans is processing, you can go to the shared media cache for the job by opening it with the right click in the status_monitor and see the script.![Smile :)](./images/smilies/icon_e_smile.gif)
welcome to the append function in Avisynth ehehehehehehehe
Jokes aside, I took a look at the logs and at the workflow and indeed Avisynth is telling the truth: the number of audio channels don't match.
Keep in mind that what's happening under the hood in the workflow is an append.
Avisynth is a frameserver, which means it serves frames to the final encoder as an uncompressed A/V stream.
In such a stream, the the meaning of "clips" is lost in the sense that clips don't exist, it's just one single uncompressed A/V Stream living in RAM.
When you have a file and use an indexer, your file is decoded in what is a stream and - to perform an append (i.e a "++" in the code) of two streams - those need to be exactly the same (i.e same size, colorspace, sampling, bit depth, framerate, audio channels and audio rate).
Unfortunately, I have no way of knowing how you created "D:\L1 NIEUWS EINDKAART YOUTUBE 2024.avi" but the idea here is that you should add a bunch of logic to the workflow after the A/V decoder so that it makes it match the specs of the .avi clip.
For instance, if we assume that L1 NIEUWS EINDKAART YOUTUBE 2024.avi is 1920x1080 4:2:0 yv12 25p 8bit planar with 2ch stereo 48000Hz 16bit we can change the workflow to work with that.
In the new workflow, we're using the A/V decoder to specify our desired output
followed by a deinterlacing step that deinterlaces the sources (if needed) and a ConvertFPS(25) step that brings everything to 25p before adding the watermark and then appending the media.
You can find the workflow in attachment, please change it according to your needs (i.e according to what the specs of L1 NIEUWS EINDKAART YOUTUBE 2024.avi are).
By the way, for those that are curious, while FFAStrans is processing, you can go to the shared media cache for the job by opening it with the right click in the status_monitor and see the script.
![Smile :)](./images/smilies/icon_e_smile.gif)