Re: error -1 returned from alSourcePlay

Thanks so much for the reply.

I have been searching through and revamping my code and I still can't come up with a likely culprit. I moved everything into one thread, so it can't be thread contention (unless its a thread OpenAL creates and uses).

So I just had one more question. Can there be differences between the number of sources you can create and the number of sounds that can play at once?

I create 200 sources at startup and under normal conditions everything is fine. But under stress testing at some point it just seems to not play the sounds, returning this unknown internal error.. They are already being played by other sources, so I know the data is correct. Just wondering if there can be sound playing limits more stringent than the number of sources alone..


Thanks again and Thanks in advance!

steve


On Jan 2, 2013, at 1:28 PM, Eric Wing <[email protected]> wrote:

> On 1/1/13, Steven Fuchs <[email protected]> wrote:
>> Hello All, Thanks in advance for any help that can be given.
>> 
>> Trying to improve a sound engine for an open source game and I run into an
>> odd problem during stress testing. This is that some sources (and associated
>> data) can play jus fine, for a while, but all of a sudden the calls to
>> alSourcePlay start returning -1 (Unknown Internal Error). This doesn't seem
>> to be documented as a possible rerun value, any idea what causes it?
>> 
>> Once this happens it seems to never stop happening. At least for that sound,
>> others do seem to play. I'm under the impression that you can play the same
>> data from two different sources. Is there some sort of timing limit to get
>> this to work? I haven't pinned down the exact situation but so far I've only
>> been able to recreate it in massive battles.
>> 
>> This is on a Macbook Pro running Mountain Lion.
>> 
>> Thanks in Advance.
>> 
>> steve
> 
> Hi Steve,
> I don't think I have encountered or heard of this specific bug, though
> I have had more than my share of bugs with Apple's implementation
> through the years. Many of them seem to be race condition oriented.
> 
> It is possible and allowed to have multiple sources playing the same
> buffer. I do this quite often.
> 
> Assuming this is not a bug in your own code, my recommendation is to
> file a bug with Apple at bugreporter.apple.com. If you can isolate it
> down to a reproducible test case, my experience is that the Core Audio
> team at Apple is very good about fixing these bugs.
> 
> And since you are on Mac, their OpenAL implementation has been open
> sourced (though I don't know if they have posted the latest code
> recently). You might be able to use that to either help track down the
> problem, or fix it directly without waiting for Apple if it is indeed
> an OpenAL implementation bug (though please file the bug for the rest
> of us).
> 
> 
> -Eric
> -- 
> Beginning iPhone Games Development
> http://playcontrol.net/iphonegamebook/

_______________________________________________
Openal mailing list
[email protected]
http://opensource.creative.com/mailman/listinfo/openal