Quoting Andreas Leitner <[EMAIL PROTECTED]>: > On Tue, 2006-03-21 at 21:18 +0100, Hans Verkuil wrote: >> On Tuesday 21 March 2006 20:37, Andrew Dodd wrote: >> > Quoting Hans Verkuil <[EMAIL PROTECTED]>: >> > > On Tuesday 21 March 2006 19:09, Dave Logan wrote: >> > >> Interesting point. There is a difference between the ivtv tuner module >> and the tuner module from the kernel (at least from 2.6.14, possibly >> earlier too): in ivtv the radio tuner is not put in standby when not in >> use, in the kernel it IS put to sleep when not in use. >> >> Can someone experiment switching between the kernel tuner module and the >> ivtv kernel module? If the problem disappears when the kernel tuner >> module is used, then we've nailed it. Perhaps when the radio tuner is >> in use it interferes with the TV tuner signal. > > I tried it and it had no effect (2.6.15-18-686, Ubuntu Dapper). Note > also that I have a Philips tuner and not a Samsung: > --- > Mar 21 21:46:28 localhost kernel: [4294693.074000] ivtv0: This is the > first unit of a PVR500 > Mar 21 21:46:28 localhost kernel: [4294693.112000] tuner (ivtv): chip > found at addr 0xc0 i2c-bus ivtv i2c driver #0 > Mar 21 21:46:28 localhost kernel: [4294693.113000] TEA5767 detected. > Mar 21 21:46:28 localhost kernel: [4294693.113000] ivtv0: i2c attach to > card #0 ok [client=(tuner unset), addr=60] > Mar 21 21:46:28 localhost kernel: [4294693.114000] tuner: type set to 62 > (Philips TEA5767HN FM Radio) by autodetect > Mar 21 21:46:28 localhost kernel: [4294693.114000] type set to 62 > (Philips TEA5767HN FM Radio) > Mar 21 21:46:28 localhost kernel: [4294693.114000] tuner (ivtv): chip > found at addr 0xc2 i2c-bus ivtv i2c driver #0 > Mar 21 21:46:28 localhost kernel: [4294693.114000] ivtv0: i2c attach to > card #0 ok [client=(tuner unset), addr=61] > ---
Wait a second, are you saying that you are encountering some of the recent issues encountered on Samsung tuner-based boards with a non-Samsung-based board? Your dmesg output that you posted is insufficient, as it only displays info on the FM tuner which is a Philips TEA5767 on all PVR-500 based boards so far, including the ones that have Samsung based TV tuners. i.e. PVR-500 boards should have three actual tuner chips. One TEA5767 FM tuner on the first "sub board" (I really don't know what to call it, as often people refer to the entire tuner + encoder subsystem as a "tuner" but right now we're referring only to the actual tuner ICs/modules as the "tuner".), and a pair of identical TV tuners (Samsung or Philips or possibly another manufacturer), one on each "sub board". One thing Hans didn't say was how we would actually use the in-kernel tuner module with a Samsung-based PVR-500, since 2.6.15 and below didn't support the tuner at all. Of course, the best option at this point looks to be to try 2.6.16 which only needs a bugfix patch for Samsung tuner support with the in-kernel tuner driver. Whether or not I do some board swapping this upcoming weekend, I do intend on upgrading to 2.6.16+0.6.0 in my server box this weekend. One thing that will make things difficult is the fact that it is possible even with the old 2.6.15+0.4.3 drivers for proper operation to occur - My problems with the first tuner disappeared without any explanation 2-3 weeks ago (the current modules/firmware have been loaded continuously for 4-5 weeks at this point.) Perhaps something caused the FM tuner to go into standby mode. Hopefully 2.6.16+0.6.0 will just plain Work Reliably. :) I can't figure out what would explain the issues with slightly degraded reception on us-cable channel 45 on the second tuner other than possibly some sort of board-level EMI problem. I keep forgetting to investigate this on my machine. _______________________________________________ ivtv-users mailing list [email protected] http://ivtvdriver.org/mailman/listinfo/ivtv-users
