..:: PCSX2 Forums ::..

Full Version: SPU2-X v2
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
SPU2-X RULEZZZ !!! Current "Async mix" mode works like real console - it controls games framerate and gives a frameskipping functions !=D MetalGear3 works awesome with it ! (and many other games)
So if this mode will be polished and optimized - there no need for timestreching anymore =)
Thanks PCSX2 TEAM ! Keep it up ! Smile
Can i enable this?
[Image: debuggy.png]
No, it's not implemented in SPU2-X and in Peops SPU2 it didn't work for ages.
(10-12-2010, 12:52 AM)rama Wrote: [ -> ]No, it's not implemented in SPU2-X and in Peops SPU2 it didn't work for ages.
Sad

I need that debug to see if the music is sequenced or streamed.
And then maybe browse that assembly chunk in realtime durning gameplay.
Compile debug or developer builds of SPU2-X, then enable some of the logging.
It's not a fancy GUI but it tells you easily what kind of sound generation is used.
Can i disable all streamed music and only allow sequenced music to be played?

Streamed music can sound sequenced but are not sequenced.
Sequenced music can sound streamed but are not streamed.

Pretty confusing if i can't debug the music.

Streamed music is wav/mp3 or ogg
Sequenced music is mid/s3m or xm
Just a quick question/suggestion:

Have you ever thought of adding an ASIO module to the plugin? Its usefullness wouldn't just be restricted to those with dedicated sound cards as there are several solutions such as ASIO4ALL that provides a WDM<->ASIO interface. I am sure it would be desirable from your point of view as you wouldn't have to worry about latency as much and theoretically it should give the best quality, especially when using a sound card with native ASIO support - it should be able to give a bit-perfect output in this case.
Jandazekon:
Streamed music comes into SPU2 memory via autodma on channels 4 or 7.
If you disabled adma for them you block that audio.

YukiS:
Nah, the latency exists mostly in the sound generation buffer.
No output module can help with that buffer.
All the output module needs to do is add as low of it's own latency as possible
while still doing the job.
Currently XA2 does that job better even than WASAPI in exclusive mode (kinda like kernel streaming).
ASIO wouldn't be any better than that Tongue2
Are there any recommended settings for the "tightest" sync, or does it vary from game to game? I've changed a few settings and the one that gives the biggest improvement in-game is Async Mix, but this creates .wav files that are out of sync with the .avis that GSdx makes, whereas TimeStretch provides perfectly synchronised recordings but is noticeably off ingame.
If you want good recordings you'll have to use timestretch or no sync mode at all.
Async mixing "works" by processing SPU2 commands completely unsynchronized to IOP.
(Hence it can break games easily.)