I Need Advice with Configuring Ffastrans for Video Processing

[roberrrttt1]

Hello there,

I am working on setting up Ffastrans for video processing, and I have encountered a few issues I hope the community might be able to assist with. I’ve followed the installation instructions and successfully set up the software on my system. Although; I am having trouble with configuring certain workflows for my video files.

Specifically; I am trying to set up a transcoding process that involves converting large .mov files to .mp4 format. While the basic conversion works fine; I have been struggling with ensuring the correct output quality and file size. I am not sure if I’m missing something in the settings or if there is a specific configuration that I need to use to optimize the output without losing too much quality.

Additionally; I have also encountered a problem where Ffastrans intermittently fails to start the queue processing for certain file types. Sometimes; the job will stay in the pending state without progressing. I have checked the logs, but the error messages are not clear enough for me to pinpoint the cause.

Also, I have gone through this post; https://www.ffastrans.com/frm/forum/viewtopic.php?tableau=1276 which definitely helped me out a lot.

Best practices for setting up video transcoding profiles. Troubleshooting tips for jobs that fail to start or get stuck in the queue.

Any settings that could help optimize the output quality for .mp4 conversions.

Thanks in advance for your help and assistance.

[emcodem]

Hi robert, so let me try answer each question separately.

While the basic conversion works fine; I have been struggling with ensuring the correct output quality and file size. I am not sure if I’m missing something in the settings or if there is a specific configuration that I need to use to optimize the output without losing too much quality.

"Quality vs File size".. if that is all you need and you don't have any specific requirement like a constant rate (for streaming) or similar, then you most likely want to use a feature called "crf", to know more about it search for ask chatgpt how "x264 crf" works (constant rate factor). Usual values for crf is 22 or 23, it delivers no really visible artefacts at the smallest size possible. If you go up with the number, the output filesize will become smaller and quality will suffer accordingly.

To use it, in the h264 processor, insert this into the texbox at bottom (x264 opts); it should ignore the "bitrate" setting but instead work with variable bitrate, e.g. the bitrate that is needed to compress the current picture in the desired "crf" factor e.g. quality vs file size.

Code: Select all

crf=22

You can also try this which our pro encoder @FranceBB sent he works for Sky, so he must know what he's doing

Code: Select all

opencl:ref=4:deblock=1,1:crf=22:overscan=show:aud=1

Additionally; I have also encountered a problem where Ffastrans intermittently fails to start the queue processing for certain file types. Sometimes; the job will stay in the pending state without progressing

This is not a "normal issue" that we know or similar so we need to know much more about it from you.
However, from feeling such an issue can't depend on the source file "type" really, what makes you think its the filetype? Can you upload your workflow json and a small source file for us to try reproduce the issue?

Best practices for setting up video transcoding profiles.
Any settings that could help optimize the output quality for .mp4 conversions.

That all depends on the specific case, it is not really possible to give general tipps without knowing whats the goal (e.g. streaming, vod, broadcast, archive, edit etc...)

Troubleshooting tips for jobs that fail to start or get stuck in the queue.

Any complex error can be mitigated by Simplify and Reproduce. Take away complexity, try to reproduce. Find an easy way how to "force" the error.
Once you know how to reproduce the error, tell us