On Tue (Tuesday) 08.Jun (June) 2010, 12:14, Bjørn Mork wrote: > Markus Rechberger <mrechberger@xxxxxxxxx> writes: > > > Trident is also still improving the quality of their driver and > > firmware, it very much makes > > sense that they handle their driver especially since those DRX drivers > > are very complex > > (basically too complex for being handled by the community, the drivers > > would just > > end up somewhere unmaintained). > > Ouch. That makes me wonder about the state of the Windows drivers for > those devices... Better stay away from them, I guess. Just to throw this out there, the 'doze support for one such Micronas-based device I have -- the Linux kernel support for which either does not exist or cannot be publicly distributed -- is less than optimal in my experience, which may have nothing to do with reality. While I was able to make a flawless test recording for a few minutes of one medium-bitrate lower-resolution high-definition programme to mislead me into thinking that I'd have success with a full-length programme, for some reason it turned out that my use of the device under 'doze for an extended time on a borrowed 'doze box suffered fairly frequent problems manifested each as a short dropout of the recording. This could also be pilot error, as I remain willfully ignorant of 'doze and its details, but if a machine with CPU horsepower over eight times that (neglecting other acceleration) of my workhorse that routinely makes four simultaneous flawless recordings including some at higher resolution/bitrate, is unable to keep up with the bitstream, then something has got to be seriously wrong, in my opinion. A later recording of a higher bitrate (excellent quality standard- definition video source) stream again exhibited the same problem. Perhaps 'doze can't keep up writing to its own native filesystem as it approaches being full, or if I can't keep my hands away from configuring it to be user-hostile as I prefer. And of course there's the factor of intermediate hardware to be considered -- my device is connected via a USB interface which has caused major filesystem corruption over time with the particular Linux kernel I was using, despite of working flawlessly with a different video card. And 'doze... *shiver* Anyway, I'd be happy to learn that others have had success with the same device, although for me it's no longer a priority to have it working, to say nothing of working perfectly. My testing of the device has been relatively minimal, using it where other tuner cards lack support. barry bouwsma -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
