DXV4/5 codec idea... multi-layered video file

"Where is Feature X? I need Feature X! How can you not have Feature X?"
Post Reply
User avatar
UON
Met Resolume in a bar the other day
Posts: 7
Joined: Sun Jun 24, 2018 02:03
Location: Vancouver BC

DXV4/5 codec idea... multi-layered video file

Post by UON » Sun Nov 03, 2019 23:02

So here's a really cool idea for the future of the DXV codec! Imagine having a ray-traced prerendered loop that you could control the lighting in as if if was running in realtime?

I make VJ loops that come with several lighting layers you can overlay over each other and control each layer individually to the BPM or EQ of any music. Right now I sell layer packs where they come with seperate DXV files of all the layers. Some of these packs can be over 10 GB of files for the same loop but different colors!

Wouldn't it be awesome if the next generation of DXV codec can fit all the layers into one file? It would save so much space and processing power, because there is so much redundant data between the seperate files. To encode it in Alley, you could just pick the base layer to encode, then have a menu to add any additional layers. The encoding process could compress the file looking for the redundant data and you end up with a file that offers much more customization and music reactivity than ever before!

If you want to see how you can make these layers react to music, check out my examples:

phpBB [video]


phpBB [video]


phpBB [video]


phpBB [video]


phpBB [video]


phpBB [video]


phpBB [video]


phpBB [video]


phpBB [video]


I understand it would be a huge amount of work to add this functionality, but I think the results would be totally worth it :)

User avatar
drazkers
Wants to marry Resolume, and Resolume said "yes!"
Posts: 955
Joined: Wed May 18, 2011 10:54
Location: Brady V up in Canada

Re: DXV4/5 codec idea... multi-layered video file

Post by drazkers » Mon Nov 04, 2019 09:39

If your lighting covers every pixel how would it save space?

At it heart its still multiple videos in a single container. Now instead of picking the files we need for the comp, won't this large single video files put alot more processing and size into one file instead of several? Because when dragging the video in, its going to read every layer inside of it.

I guess the Resolume workflow changes from a layer based system of mixing to a clip based system of mixing with sub mixes.

There are people doing this right now with FFGL effects and UVW remapping and reflection mixing. And its pretty cool.

This all is kinda stemming from your frustration of having to use column triggers to get all your layers to line up right? Thats really whats making the lighting mixing difficult in its current form for you.

Post Reply