@Koz
You raise some important points, but there are ways around it.
- Mixed frame rates:
Many video editor programs handle this just fine.
When you import the first video, it looks at it’s frame rate and that is what the project is set up to.
(You can then change it later).
Once you add a second video and it’s not the same frame rate, it will ask you if you want to convert it.
If yes, it re-render’s it, if not, then you must expect out of synch audio and choppy video as it will play
everything at the “master’s i.e the first imported video” frame rate.
The better video editing pgms, will even check for videos with variable frame rate, which are not good for
editing purposes and recommend it does a temp re-export to an intermediate format with a fixed frame rate.
Other things that can be done to enhance the editing experience and put less load on the CPU are,
Check GOP size, ideally 1 - 12 is good for editing and greater values are mostly for streaming and can be
re-exported to all “I” frames.
This makes scrubbing back and forth much, much smoother as all frames are key frames and the pgm
does not have to “re-build” the frames in RAM.
To give you an example, I can quite easily edit a Pro-Res video on a Celeron if it’s been re-encoded as all “I” frames.
Smooth as butter.
Must also add that audio pgms like Protools, can only handle one video at a time.
After all, they are audio editors not video NLE’s.
Most common usage would be, bring in a video, do whatever needs to be done to it’s audio/s,
re-export, bring it next video…etc
In this case, no need to worry about mixed frame rates or DF, bring in the video and convert to all “I” frames if need be.
Since only the audio is being worked on and not video effects, fades etc, it would be possible to
work on a lower res proxy to lighten the load even more.
The audio would still be in synch and only on final export, is the full res video used.
- Extra resources required.
Of course if you want to play video, you will need extra CPU/GPU power.
(Points made in (1) above, will also help in minimizing the load.
The devs can always do something similar to Reaper, the video is only played
if a video window is enabled, if not, then it just extracts the audio and plays that.
Just like what Audacity does with ffmpeg import.
- Multiple audio tracks.
Many videos, include multi-track audio such as surround 7.1 (8 channels) or broadcast clips
that can have up to 8 mono audios or 4 stereo pairs.
These extra audios could be placed on separate tracks just like what Audacity currently does if you import
a multichannel .mogg file.
I have many .mogg files with up to 24 tracks, Audacity opens them and plays them perfectly.
Below, Audacity playing a 12 channel .mogg audio file:
Plus, audacity can already give a duration and current position count relative to video frame rate, in either PAL or ATSC DF.
