@alcomposer "But 32 video channels currently can only be edited together, and viewed one at a time, unlike audio."

So when I do a compositing job with over 50 layers, are you suggesting that I really can't see all my layers and I'm only imagining them? Your knowledge of video systems needs a little refresh!

So when I do a compositing job with over 50 layers, are you suggesting that I really can't see all my layers and I'm only imagining them? Your knowledge of video systems needs a little refresh!

Gosh I didn't know you always do compositing on EVERY multicam shoot you do! Unlike audio which is ALWAYS composited together when using multiple microphones.

Normally I only open up Fusion for VFX work. I didn't think that was the discussion.

Also, normally I don't ALWAYS use video in compositing, sometimes mattes, or deep pixel layers. So no, all those channels are not always going to be seperate video media assets, as they are in audio land.

Also currently most compositors are not optimised for GPU acceleration (CUDA can also be slow here as well, simply using GLSL could solve speed issues). Which is a huge problem. Even simple blend operations are still carried out on the CPU, something that a shader could easily achieve, at 32bit precision, in a fraction of the time.

Even decoding of video could be shifted from CPU to GPU, something that needs to happen asap. There are a few accelerated codecs (such as HAP) however they are not lossless enough for production work, only realtime work. (Look into Touch Designer, Vuo for realtime video mixing / effects processing).

Can we please not get personal? I would love to have a great discussion about this without being questioned, I have never questioned your knowledge. I'm sure we are all professionals here, with great bodies of work. It's not helpful having a discussion in which we just throw random: "your knowledge needs a refresh" comments.

This whole discussion has got out of hand. I've greenscreened 360 VR content with up to 100 layers 4k, and was able to playback in realtime... so no need for a refresh.

I don't understand the strawmanning, considering my last comment on this super offtopic debate:

Actually, I have realised though this discussion that there is a huge problem with my logic: We need a way to quantify the quality of both audio and video capture systems. If audio is "much simpler" then we have to quantify at what "quality" is it simpler. Lets say we capture an SD video image of an orchestra, and a super high audio recording: are they both similar? No. Why? Because the audio recording will sound much more real than the video recording. I would suggest that 4k 60p is the equivalent of a good audio recording in making the viewer feel like they were there "in the room". So with that logic, and definition of similar quality then yep, video wins the complexity race.

All because I would like to use video like audio, I didn't think this was a debatable topic. Video is getting easier to work with, try to do 100 4k layers realtime on a 10 year old PC.

As was mentioned before: video is transitioning into a more robust format similar to the evolution of audio did, (16 bit -> 24 bit).

Also remember 32 bit audio is just around the corner, with AD systems that employ the use of multiple AD stages to make distortion and peeking a thing of the past. Maybe audio is the medium that is playing catch up! ;-)

@caveport i think this thread should be brought back to @Vitaliy_kiselevs topic.

Which personally I think is quite interesting. All computer storage systems have bit depth. Even UTF-8. So this discussion is intrinsic of all computer storage systems, which digital video is just one of.