Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

So, I do try to physically simulate the stereo signal to some degree with a delay sent to the reverb on my tracks, generally. I didn't end up actually carefully setting up that delay here, however, due to time.

The "closer" something is, the higher the delay I use.

Generally I've been using neoverb, and turning off the DearVR built in reverb, manually setting delay and wetness to what sounds good.

I think the most important things for the illusion are the initial dry signal to each ear being timed right and undergoing the Fourier transform to simulate the shape of the ear reflecting the sound (something UE5 actually has built in).

The reverb is a wetter, more complex signal that comes from too many directions to really easily distinguish between a real space and an artificial one as easily, hence the illusion can still be largely preserved in terms of the physical location of the sound source.

If you were to EQ, you'd want to EQ before DearVR and adjust the EQ after DearVR is in the chain. Mostly saying this as a conscious note to self next time I'm mixing to solidify that point.

What I was curious about was the things you are doing after the DearVR chain in post, considering what those things are, what their benefit is, and if they are necessary or can be put in a different part of the chain as to not screw up the illusion.

(+1)

To your curisoity on my approach: I do quite a bit of context-dependent post-processing on spatialized signals during both the mixing and mastering stages.  Off the top of my head: 

  • Shaping/cleaning transients on percussive elements so they cut through the mix better
  • Bringing air back to vocals (usually a high shelf somewhere around the 10k range) 
  • Just...so much low end processing. I'm always messing with things <150 hz

All of this will in some way ruin the illusion, but it often leads to a mix I enjoy listening to more. I tend to sacrifice realism for a more modern sound, which is probably why both Fox and I find some of the frequency shaping a little annoying.

Your process is probably closer to a more pure, realistic sound. I'd be interested to hear why you have a preference for NeoVerb over, let's say a traditional convolution verb sometime.

If I missed addressing something in this thread,  Hit me up on Discord and I'll be happy to chat about this topic anytime,