Hello ,ntdev
I’ve made the sound capture driver based on MSVAD sample.
It just creates two virtual sound cards, and transfers sound
from wave out of the first card to the wave in of the second one
and vice versa. It works just great with WaveIn-WaveOut apps.
But with DirectSound apps the sound is horrible and highly depends
on CPU load. I’ve noticed that DirectSound app just doesn’t supply enough
IOCTLs to get all the sound data, so my buffers are overflowed and sound
is lost.
Here how I test it:
I start GraphEdit and add “WDM Streaming capture device” of the FIRST
card, then add “WDM Streaming render device” of the REAL sound card
and then connect these devices.
Then I start any other sound app and tell it to play sound
through the SECOND virtual sound card. In this scheme I get sound transferred
from SECOND to the FIRST and then to the REAL sound card. After clicking
Play button, my speakers start sounding like lownmower when I move window
with mouse or do anything else, and when I don’t touch anything, it sounds
better, but anyway it clicks and glitches.
Can anybody tell me why this happens? Are there any settings in DirectX to
tell it to buffer more sound or something?..
–
Best regards,
Yura mailto:xxxxx@mail.zp.ua