[vlc-devel] Volume change lag fix for OS X CoreAudio [auhal]

[Jon Stacey]

And here are the two missing patches.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0052-Audio-have-aout_VolumeHardInit-call-the-setter-to-in.patch
Type: application/octet-stream
Size: 1134 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20120307/d40c39c8/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0053-Audio-Core-Audio-Initialize-starting-volume-for-anal.patch
Type: application/octet-stream
Size: 1464 bytes
Desc: not available
URL: <http://mailman.videolan.org/pipermail/vlc-devel/attachments/20120307/d40c39c8/attachment-0001.obj>
-------------- next part --------------


Jon

On Mar 7, 2012, at 1:42 AM, Jon Stacey wrote:

> Following up on this patch, here are two others for your consideration. The problem was pointed out by kdean on trac (#5605): the volume is not initialized at first start. This is not a bug caused by the last committed patch and has existed in master for quite some time from what I can tell. Since I'm in the area though I'd like to squash it.
> 
> There are two possible routes: deal with it in the audio module, or call the volume setter when the audio module registers its volume callback. Personally, I had expected the volume to be initialized automatically when the audio module registered it's callback, but it doesn't look like that happens. I'm not certain how other audio modules are setting their initial volume levels, or if they're even working as expected or not...
> 
> Patch 0052 attacks the problem in aout_VolumeHardInit() by having it immediately call the registered volume setter to initialize the starting volume. It also stops creating the volume and mute variables since these should presumably already exist and is not consistent with aout_VolumeSoftInit().
> 
> As an alternative, patch 0053 attacks the problem in the audio module, auhal.c, and initializes the start volume there.
> 
> I personally prefer 0052, but I do not know what kind of impact this might have on other audio modules and their behavior, so I leave it in the hands of someone who knows more about these internals.
> 
> 
> Cheers,
> 
> Jon
> 
> On Mar 5, 2012, at 10:10 AM, Felix Paul Kühne wrote:
> 
>> Hi Jon,
>> 
>> On 05.03.2012, at 09:25, Jon Stacey wrote:
>> 
>>> 1. Patch is attached to this email.
>> 
>> Your patch was merged as commit [93a0d5ab5b51d369c20f933f379ad4934daeb7bf] and will be back ported to the stable branch after further testing.
>> 
>> Thanks a lot for your contribution!
>> 
>> 
>> Best regards,
>> 
>> Felix
>> _______________________________________________
>> vlc-devel mailing list
>> To unsubscribe or modify your subscription options:
>> http://mailman.videolan.org/listinfo/vlc-devel
> 
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel