[Dustin Kincaid]

It’s weird, it’s a problem with this workflow.

see

https://forums.creativecow.net/thread/8/1124774#1124774

The solution given there is kind of a no brainer, but not an option when you’re working with terrabytes of data (and it defeats the purpose of having multiple compressor nodes), if I have to batch export on one machine first, I may as well call it a day and be done with it right there – but it would take weeks on one machine.

The workaround (for me) is send in smaller batches and move each job into its own batch. Which as you can imagine is very tedious and almost impossible to manage with the way compressor 3.5 is designed. You can only see so many tabs and it doesn’t let you scroll.

I just want to confirm that it exists and see if anyone else has run into this problem.

Some times its not too big of an issue if my sequences are all the same length… then I keep them all in one job batch and send it like it’s supposed to work. It’s just very difficult if they are all different durations.

[Dustin Kincaid]

Does anyone else have an issue with this process?

This has been my method of exporting for about two years, and there seems to be a bug in compressor.

All my videos will get exported the same length with this process (the length will be determined by the first sequence sent to compressor).

Example:

50 timeline edits – all different durations, right click in FCP and send all of them to compressor – no intermediate encode. (Requires final cut on every cluster machine).

The 50 entries will get added to one compressor batch with 50 jobs – (after you select “export as separate sources” – not merging).

Then, apply the presets and submit to the cluster… (you would think)

BUT…

Whatever the first job entry in compressor seems to determine the length of rest of the videos in the batch. If the first video is longer than the rest, the rest have black footage added until it hits the same timecode, if the rest are shorter, they truncate at the duration of the first video (or a mixture of both scenarios).

I’ve tried reinstalling everything completely from the ground up, doing the same exact thing with a stock Apple template (so I know there’s nothing weird with my custom templates).. yet it still does this.

My workaround is to remove every job entry in the large batch and put each its own individual compressor batch (very time consuming) or send each individual timeline to compressor separately (again, very time consuming).

With my workflow, I need to send 100’s sometimes 1000’s of timeline edits at once to compressor to have it render on about 12 machines. Encoding to a intermediate file is not an option.

Any ideas?