Hi, On Wed, Nov 26, 2008 at 10:56:05PM +0100, Udo Richter wrote: > Artem Makhutov wrote: > > Klaus Schmidinger schrieb: > >> The attached patch adds a capability flag that allows an application > >> to determine whether a particular device can handle "second generation > >> modulation" transponders. > > Wouldn't it be better to add something like this: > > > > FE_CAN_8PSK > > FE_CAN_16APSK > > FE_CAN_32APSK > > Without knowing enough about the differences between -S and -S2 tuners > and whatever cards are on the market, I'll try to step back and give > this my 2c point of view: > > What does a DVB app need to know? A DVB app probably just needs to know > "What devices are capable of tuning to channel XYZ?". The API could > answer this the same way as it would tune to channel XYZ, just without > actually doing it. Try-before-you-buy. > > This would also give maximum flexibility to the driver, as a device that > supports some -S2 features could offer these, or a device that has known > bugs on some tuning modes could also deny these. Non-standard modes > could be offered without requiring yet another FE_CAN_XYZ. If I understand you correct you would like to call an function like: bool ItWorks = DoFakeTune(a lot of Paramters); Sure, this would work, but it would make the implementation more difficult for both the driver and the application. So in this case the application would have to call DoFakeTune() for a lot of different modulations, FEC's and so on and interprete the results. The application must then call DoFakeTune(DVB-S,FEC_AUTO); DoFakeTune(DVB-S,FEC_3_4); DoFakeTune(DVB-S,FEC_2_4); DoFakeTune(DVB-S,FEC_4_5); DoFakeTune(DVB-S2,FEC_AUTO); [...] DoFakeTune(DVB-T,FEC_XYZ); DoFakeTune(DVB-T2,FEC_XYZ); This would not make much fun and this could be achieved much easier using some kind of capability flags. Regards, Artem _______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
