Re: [Flightgear-devel] Problem with Beaver and sound

2005-01-24 Thread Erik Hofman
Frederic Bouvier wrote:
I just discovered that FG is suggesting me to upgrade my sound driver
after alGenSources failed :-(
This is the first time and all the other aircraft I tried never did the
same. I even remember flying successfully with it not far ago.
This should be fixed in SimGear CVS now.
Erik
___
Flightgear-devel mailing list
Flightgear-devel@flightgear.org
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
2f585eeea02e2c79d7b1d8c4963bae2d


Re: [Flightgear-devel] Problem with Beaver and sound

2005-01-20 Thread Curtis L. Olson
Frederic Bouvier wrote:
I just discovered that FG is suggesting me to upgrade my sound driver
after alGenSources failed :-(
This is the first time and all the other aircraft I tried never did the
same. I even remember flying successfully with it not far ago.
A quick glance with the debugger showed me that alGenSources failed the
33rd time it was called, so I guess my driver/hardware ( yes a realtek
on brand new mainboard with brand new drivers ) is only able to cope
with 32 sources. I also noticed that 20 sources are allocated for the
same sample : Aircraft/dhc2/Sound/click.wav
Question: is it normal that the same sample is being allocated multiple
times ? Couldn't we use a reference counter and share the same
buffer/source ?
It seems that click.wav is bound to panel switches. I guess that a panel
will lots of switches will saturate the sound driver just like too much
texture will flood the graphic card.

I looked at the code a bit and there is some facility to share a sample 
when the name tag is identical, however the dhc2 uses a different name 
for each item that can click, so this means that a new sound sample is 
created for each item that can click.

OpenAL has buffers and sources.  We should be able bind multiple sources 
to a single buffer, but you are saying that it's alGenSources() that is 
failing so that doesn't help us.

The problem is that each alSource specifies it's location and a bunch of 
other parameters.  The way our xml syntax is structured, you can't just 
combine sources because you will overwrite the previous sources location 
and any custom configuration it had.

So it almost seems like we need to solve this on the sound configuration 
side.  Build one sound, with one source, and one location, and then 
trigger it any number of ways ... with a large block of conditions ... 
that is less intuitive from the sound configuration perspective, but it 
might be what we need to do?

Thoughts?
Curt.
--
Curtis Olsonhttp://www.flightgear.org/~curt
HumanFIRST Program  http://www.humanfirst.umn.edu/
FlightGear Project  http://www.flightgear.org
Unique text:2f585eeea02e2c79d7b1d8c4963bae2d
___
Flightgear-devel mailing list
Flightgear-devel@flightgear.org
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
2f585eeea02e2c79d7b1d8c4963bae2d


[Flightgear-devel] Problem with Beaver and sound

2005-01-19 Thread Frederic Bouvier
I just discovered that FG is suggesting me to upgrade my sound driver
after alGenSources failed :-(
This is the first time and all the other aircraft I tried never did the
same. I even remember flying successfully with it not far ago.
A quick glance with the debugger showed me that alGenSources failed the
33rd time it was called, so I guess my driver/hardware ( yes a realtek
on brand new mainboard with brand new drivers ) is only able to cope
with 32 sources. I also noticed that 20 sources are allocated for the
same sample : Aircraft/dhc2/Sound/click.wav
Question: is it normal that the same sample is being allocated multiple
times ? Couldn't we use a reference counter and share the same
buffer/source ?
It seems that click.wav is bound to panel switches. I guess that a panel
will lots of switches will saturate the sound driver just like too much
texture will flood the graphic card.
Regards,
-Fred

___
Flightgear-devel mailing list
Flightgear-devel@flightgear.org
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
2f585eeea02e2c79d7b1d8c4963bae2d