Questions et réponses de la communauté

Bienvenue sur le forum de questions et réponses d'Audiokinetic, propulsé par la communauté. C'est l'endroit où les utilisateurs de Wwise et Strata s'entraident. Pour obtenir une aide directe de notre équipe, veuillez utiliser la page « Tickets de soutien ». Pour signaler un bug, utilisez l'option Bug Report dans l'Audiokinetic Launcher. (Veuillez noter que les rapports de bug soumis au forum questions-réponses seront rejetés. L'utilisation de notre système de rapport de bug dédié garantit que votre rapport est vu par les bonnes personnes et a les meilleures chances d'être corrigé.)

Pour obtenir rapidement les meilleures réponses, suivez ces conseils lorsque vous posez une question :

  • Soyez précis : qu'essayez-vous de réaliser ou quel est le problème spécifique que vous rencontrez ?
  • Pensez à inclure les détails importants : incluez des détails tels que les versions de Wwise et du moteur de jeu, le système d'exploitation, etc.
  • Expliquez ce que vous avez essayé de faire : indiquez aux autres les mesures que vous avez déjà prises pour essayer de résoudre le problème.
  • Concentrez-vous sur les faits : décrivez les aspects techniques de votre problème. Se concentrer sur le problème aide les autres personnes à trouver rapidement une solution.

+1 vote

My setup:

  • I have a 3D Wwise event (game-defined position) 
  • It's playing from an AkAmbient actor in UE4
  • It's routed in Wwise to an ambisonics 3-3 bus
    • which goes to a binaural bus that uses sce_Audio3d (this bus is also set to Ambi 3-3)
      • and then to the master bus as stereo (Wwise audio output set to Stereo Speakers)

What's wrong:

  • When the listener turns away from the emitter, the sound starts to attenuate once it's in the back 120 degrees (roughly) of the listener.
  • When the sound is directly behind the listener, it is completely silent.
 
Some notes:
  • I've tried this on our HRTF bus and our Non-HRTF bus. The Non-HRTF bus is the same channel configuration as the HRTF bus. The only difference is that it doesn't have the SCE_Audio3d object panner on it. The issue exists on both busses.
  • If I set the sound to bypass the ambisonics bus, going straight to its parent, the issue is gone. Only problem is, we want to use this ambisonics bus as a place to do dynamic compression, so we can't just get rid of it. 
 
Anyone have any ideas?
 
dans General Discussion par Ian S. (2.1k points)
Now that GDC is over... anyone have any insight on this?
An update on this:

I tried changing the ambisonics and HRTF busses to 7.1 to see if I was dropping channels when monitoring in stereo. It monitors fine, so I don't think that's the issue.
if you're on 5.1 or 7.1 headset for this check the setup of headphones if they use some kind of software, that can cause trouble
I just came across this same issue. When a sound was panned to the middle of the rears (i.e. x=0, y=-80) on an Ambisonics buss, the sound attenuates to 0.

I found that by changing my Audio setting in Wwise from Stereo Headphone to 5.1 fixed the issue. Is this a bug? My project is being developed for headphone use.
I also have found this issue.
I'm having the same issue with a more harsh effect. The rotation attenuation is too fast, as soon as listener is turned away almost 30 degrees from the emitter,  the opposite ear (stereo) is completely muted. And then the weird thing is that the sound is back in that between +-120 degrees away from the emitter! So it reaches the full volume on both left and right speakers when the emitter is completely in the back of the listener!

1 Réponse

0 votes
Does the sound object have any cone attenuation on it?
par Monty M. (3.1k points)
Thanks for the response!

Yes, it does. However the cone attenuation is working as expected, and it doesn't change the unexpected behavior when I disable it.
Another thing to maybe investigate is channel configuration throughout the chain.  It's almost like the back channels are being excluded somewhere along the line.  This would be analogous to playing 7.1 content on a 5.1 speaker system without down-mixing, the back channels would simply be missing.  Just stabbing in the dark here.  Let us know when you figure it out.
That's what I was thinking, but I don't know where that would be happening. Thanks for the help.
...