Advertisement

Handling data flow in highly dynamic scenes

Started by July 12, 2022 03:14 PM
11 comments, last by JoeJ 2 years, 5 months ago

Kleriq said:

So do you assume that the AudioGeometry component just holds a pointer to some sort of mesh resource?

Pretty much. Ideally the mesh should be shared if it is instanced multiple times in the scene (each with separate transform matrix).

@aressera Wow, i had no idea audio is that interesting and advanced! I feel mind blown and ashamed because i have completely ignored this all the years. But i realize this only now. ;D

Thanks for the clear sum up, will read your papers. I always consider my realtime GI system might be eventually useful for audio too, so maybe one day i will come back with related questions…

Regarding the friction example, i did not mean to synthesize sound from materials and physics. I know that's still out of reach.
I meant we have few samples, e.g. wood scraching over metal, and wood bumping on metal. Collisions / contact information from physics would select the ‘proper’ sample.

This topic is closed to new replies.

Advertisement