On Feb 3, 2008 3:17 AM, Rolf Ahrenberg rahrenbe@cc.hut.fi wrote:
The compile time selection between multiproto and normal drivers (as done in Udo's dvb-api-emulate patch) could be used from now on. I don't see any reason to freeze the development at this phase.
The completition of H.264 support would be nice to have in next stable version. The H.264 support isn't related only to DVB-S2 as it's used already in many DVB-C and DVB-T networks nowadays.
Also, I'd vote for making the recording format selectable, PES or TS, as it would make the postprocessing of VDR files with existing Windows/Linux tools a lot easier.
I tend to agree with this. I don't think there should be any rush to release another stable version. I personally would love to see full DVB-S2 and H264 support in the next stable. The switch to multiproto makes sense as well since it should be adopted into the next kernel by the next release or two (leaving time to finish up VDR). Even the kernel maintainer wants multiproto in, and DVB-S2 users certainly do. I don't think theres any harm is having package maintainers wait a little longer. If people are dying that bad to use things like subtitles then there's nothing stopping them from using 1.5.14 for example.
I am also in strong favor of using TS for the recording format as I've heard countless complaints by countless numbers of users who have trouble working with the current PES format. The ability to easily manipulate vdr recordings in other software has been a long-standing desire of many many people.
Overall I'm not in favor of freezing the current development tree. There seems to be good momentum and no solid reasons to interrupt that. I don't think it's a horrible thing that package maintainers would have to wait a while longer, they've certainly waited longer in the past.. Especially if it means the next stable release has all kinds of new goodies wrapped inside!