Hi Christoph, > Hmm, is there really any need to change the interface to the sound > subsystem? kDoSound() just sends commands, after all, and it uses a fairly > simple interface to do this. As far as I can tell, the current > interface should work just fine for a message model; where's the problem?
The interface can be made to work certainly, but there could be a timing problem when some of the sound commands return values. The message has to be sent to the sound server which then has to respond with a message ASAP, because the main game thread will be busy-waiting for it. Also, if there are other messages waiting for the main game thread in the message queue, the queue may have to be searched for the message the main thread wants. However I don't think this will be a problem because you can filter messages, effectively doing the same job. Cheers, Alex.
