At the moment, when calling FSOUND_Init with XMMS playing, the init function will not finish, until I press stop on xmms. And I cannot start xmms again, until I have closed down FMOD.
My first guess as to why this is happening is because it is using the OSS driver (which I think has problems like that). I can reconfigure XMMS if it isn’t using ALSA, since I have the source (at least I think I can).
Is there any way to get FMOD to use the ALSA driver? Or would that not be a problem? Any ideas would help.
- adruab asked 14 years ago
Yes, I am using FSOUND_Sample_Load. The only tweaky weird thing that I’m doing that I think might affect the outcome is that I’m loading the sound as a stream first. Would that cause the system by accident reuse the cached sample that the stream is using (I assume this is the sort of architecture given you can get the sample out of the stream).
If that is a problem, should I just manually advance the time, and parse the segment I get from the stream each time, until I’m done? That’s seems kind of painful, but then again, this isn’t exactly a normal application of a sound engine :P.
Either way, there is always a chance that the 4.41 seconds thing is a bug in my calculations somewhere (although it’s pretty straight forward, so I kind of doubt it).
Adruab > Adrian Bentley
Ok, so I’ve tested it out, and it will play the entire song (though slower when I tell it to load it in 8 bit) does imply it’s loading the entire thing in. So it’s probably a bug in my code somewhere that I’m getting 4.41 seconds (which is sad because it seems super simple).
One last question and I should be good, in FMOD only the things that contribute to the sample size are size of individual pulse (in bytes) and number of channels right? I would need the size of a sample to accurately parse the data.
Anyways, Thanks again for all the help, and I’m sorry if I was confusing.
Thanks for the pointer. I didn’t notice that :).
However now FMOD emits an error:
ALSA lib pcm_hw.c:1143:(_snd_pcm_hw_open) Invalid value for card
It seems like the same problem, only I have NO idea how to fix this one. Has this ever happened to anyone else? Or perhaps have any pointers?
Yeah, that does work, I just wanted to try to not have it compete. Thanks, btw.
Oh well, on to another question!
I’m trying, on purpose, to decompress an entire mp3 into memory and then parse it for the wave pattern. But when I load it as a sample it appears to only give me 4.41 seconds of data. I just thought I’d check and make sure that FMOD will actually load the whole sound, and is not limited to the mixing buffer size or something. Is there any cases in which it wouldn’t load the whole thing?
Thanks for the help, I really appreciate it.
Please login first to submit.