[MPlayer-dev-eng] [PATCH] Fix AC3 passthrough on Intel Macs

[Sean McGovern]

Except that this decision is made in libmpcodecs/ad_hwac3.c. Decision is putting it mildly, since it only switches from big-endian to native-endian if it's feeding DTS audio.

Apple is the one with the braindead, aberrant interpretation of the data stream -- I'm assuming CoreAudio has to flip the data a 2nd time before it spits it out the optical port since the receiver at the other end will otherwise play silence or (very painful) noise.  The swap does not appear to be required on any other platform.

As for the complaint about top posting -- I'm responding on a BlackBerry. I don't have a choice in where it lets me write a response. I can't even respond inline.

-- Sean
-----Original Message-----
From: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
Sender: mplayer-dev-eng-bounces at mplayerhq.hu
Date: Tue, 17 Aug 2010 08:37:02 
To: <mplayer-dev-eng at mplayerhq.hu>
Reply-To: mplayer-dev-eng at mplayerhq.hu
Subject: Re: [MPlayer-dev-eng] [PATCH] Fix AC3 passthrough on Intel Macs

On Mon, Aug 16, 2010 at 06:44:51PM -0400, Sean McGovern wrote:
> As requested, here is the patch *sans*-compiler warning fixups.


> +    // AC-3 data is *ALWAYS* big-endian, however CoreAudio on an Intel Mac
> +    // still requires the data to be byte-swapped

No it is not, just set the format to the right one (AF_FORMAT_AC3_LE,
AF_FORMAT_AC3_BE, AF_FORMAT_AC3_NE) to request the conversion instead
of duplicating the format conversion code.
_______________________________________________
MPlayer-dev-eng mailing list
MPlayer-dev-eng at mplayerhq.hu
https://lists.mplayerhq.hu/mailman/listinfo/mplayer-dev-eng