add comment to BufferedAsync explaining its value
This commit is contained in:
parent
2a01e01e78
commit
408a4d6102
|
@ -13,6 +13,21 @@ namespace BizHawk.Emulation.Sound
|
|||
// distributing error to most frames tends to result in persistently distorted audio, especially when
|
||||
// sample playback is involved.
|
||||
|
||||
|
||||
/*
|
||||
* Why use this, when each core has it's own Async sound output?
|
||||
*
|
||||
* It is true that each emulation core provides its own async sound output, either through directly
|
||||
* rendering to arbitrary buffers (pce not turbocd, sms), or using one of the metaspus (nes, turbocd).
|
||||
*
|
||||
* Unfortunately, the vecna metaspu is not perfect, and for maintaining near-realtime playback (the usual
|
||||
* situation which we want to optimize for), it simply sounds better with a BufferedAsync on top of it.
|
||||
*
|
||||
* TODO: BufferedAsync has some hard coded parameters that assume FPS = 60. make that more generalized.
|
||||
* TODO: For systems that _really_ don't need BufferedAsync (pce not turbocd, sms), make a way to signal
|
||||
* that and then bypass the BufferedAsync.
|
||||
*/
|
||||
|
||||
public sealed class BufferedAsync : ISoundProvider
|
||||
{
|
||||
public ISoundProvider BaseSoundProvider;
|
||||
|
|
Loading…
Reference in New Issue