| Re: [AD] Audio code and endian-ness |
[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
- To: Coordination of admins/developers of the game programming library Allegro <alleg-developers@xxxxxxxxxx>
- Subject: Re: [AD] Audio code and endian-ness
- From: Elias Pschernig <elias@xxxxxxxxxx>
- Date: Fri, 09 Jun 2006 19:11:02 +0200
On Fri, 2006-06-09 at 18:00 +0100, Peter Hull wrote:
> I'm just working on Allegro on the Intel Mac. I notice that playing
> samples doesn't work - it sounds all wrong. In the OSX-specific code,
> there is this (cadigi.m:172)
>
> input_format_desc.mFormatFlags = kAudioFormatFlagIsBigEndian |
> kAudioFormatFlagIsSignedInteger | kLinearPCMFormatFlagIsPacked;
>
> I think this is just an endian issue; when I remove the
> kAudioFormatFlagIsBigEndian flag from the list, it seems to work OK.
> This makes sense as the PPC is big endian and the i386 is not.
>
> However this only makes sense if there's some other endian-switching
> code elsewhere, that was active in the PPC port and is not active in
> the i386 port. But, I can't find it.
>
> Can anyone help?
>
Hm, is Allegro's buffer always native endianess maybe? In that case, it
would make sense - for PPC you need the flag, for Intel not.
Oh, and looking at the code - are there two digi drivers for OSX? What
is the difference bitween them?
--
Elias Pschernig