Re: [Linphone-developers] LinPhone

[Gautier Pelloux-Prayer]

Hi,

Well, it is difficult for us to accept a patch on OSS because none of us use it 
now, and we cannot ensure that the patch does improve the situation.
However, if you limit your patch to your specific oss driver, we will be able 
to apply it on Linphone. If any other user encounters the problem again with 
another driver, we might then reconsider your patch and apply it globally.

There is a mechanism to drop data from time to time which is the jitter buffer 
(see jitterctl.c in mediastreamer2 sources).

Regards,

Gautier Pelloux-Prayer
Software Engineer @ Belledonne Communications

> On 16 Nov 2014, at 16:50, Hans Petter Selasky <address@hidden> wrote:
> 
> On 11/01/14 15:18, Hans Petter Selasky wrote:
>> On 10/30/14 13:53, Gautier Pelloux-Prayer wrote:
>>> Hi Hans,
>>> 
>>> Thanks for your contribution! However, there are lots of OSS
>>> configurations and since your patch is changing several mechanisms, we
>>> cannot ensure that it will not worsen the situation with some OSS
>>> configurations/drivers.
>>> Current solution has been tested for several years now and even if
>>> this is not perfect, this works enough in all of them.
>>> 
>> 
>> Hi,
>> 
>> What should I do in order for you to accept a change to the OSS driver.
>> Right now I see two problems:
>> 
>> 1) OSS driver induced latency
>> 2) Linphone buffer latency
>> 
>> Does linphone have a mechanism to drop audio data, if the buffers grow
>> too big? I see latency in the order of 1 second somtimes, and that is
>> simply unacceptable!
>> 
> 
> Ping - any more comments on this issue?
> 
> --HPS
> 
> 
> _______________________________________________
> Linphone-developers mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/linphone-developers