From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail.kapsi.fi ([217.30.184.167]:52228 "EHLO mail.kapsi.fi" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754047Ab1L0QIg (ORCPT ); Tue, 27 Dec 2011 11:08:36 -0500 Message-ID: <4EF9ED7F.7060808@iki.fi> Date: Tue, 27 Dec 2011 18:08:31 +0200 From: Antti Palosaari MIME-Version: 1.0 To: =?ISO-8859-1?Q?Christian_Pr=E4hauser?= CC: linux-media@vger.kernel.org Subject: Re: DVB-S2 multistream support References: <4EF67721.9050102@unixsol.org> <4EF6DD91.2030800@iki.fi> <4EF6F84C.3000307@redhat.com> <4EF7066C.4070806@redhat.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit Sender: linux-media-owner@vger.kernel.org List-ID: On 12/27/2011 12:12 PM, Christian Prähauser wrote: >> >> Yes, I'm meaning something like what it was described there. I think >> that the code written by Christian were never submitted upstream. >> > > Hello Mauro, > > Konstantin drew my attention to this discussion. Indeed, some time ago I wrote > a base-band demux for LinuxDVB. It was part of a project to integrate support > for second-generation IP/DVB encapsulations (GSE). The BB-demux allows to > register filters for different ISIs and data types (raw, generic stream, > transport stream). > > I realized that the repo hosted at our University is down. If there is interest, > I can update my patches to the latest LinuxDVB version and we can put them on a > public repo e.g. at linuxdvb.org. > > Kind regards, > Christian. I have a question which is a little bit off-topic for that thread but I would like to ask since I think you could have some idea. FEC Code Rate is often given as k/n, for example FEC 1/4. But nowadays there is also seen 0.x like FEC 0.8. I have feeling that this is due to new inner coding used, LDPC instead of traditional convolutional codes. When convolution codes were used it was correct to define 1/2 as basic rate and puncture rest from that. But as now with LDPC we have larger blocks we cannot represent so easily. For example DTMB uses LDPC(7488,6016) = 6016/7488 = ~0.8034 => FEC 0.8 is used. I am adding DTMB support for DVB API and that's why I have to think if I extend old k/n FECs or define new ones as FEC 0.4/0.6/0.8. regards Antti -- http://palosaari.fi/