[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
On Thursday 18 March 2004 05:53, Chris wrote:
> Ah, ok. 24-bit mixing in 32-bit buffers. Gotcha. That's definately
> doable, just gotta write the mixers for 24-bit voices. The only thing
> though, is that the audio drivers will need to be changed so they send
> the proper bit depth (8, 16, or 24) to _mixer_init and possibly
> _mix_some_samples (I would assume they'd be the same?), instead of the
> current yes/no 16-bit flag. Unless you just want to stick with 8/16-bit
> output but support 24-bit streams internally. Or just do 24-bit mixing
> but leave everything else to use 8/16-bit.
DUMB used to keep everything in this 32/24-bit format and convert to 8-bit or
16-bit only at the final stage (most sound cards won't accept more than
16-bit AFAIK). Now (as of CVS I think), it mixes from 8-bit or 16-bit into
the 32/24-bit format, and converts at the end as before - so it doesn't have
to waste memory on holding the samples.
> Hmm.. they're coming out properly dated here. Your guys' posts are
> coming from the future to me, though. Do I have something set wrong? The
> clock on my kde taskbar is right.
It'll probably be a time zone issue. Is your time zone set correctly?
Ben