Steffen Motzer wrote: > This revised patch for dst.c also fixes the tuning problems. It seems > that sometimes the asic acknowledged a command but failed to execute it. > This might have also be the reason why sometimes szap reported that the > tuner supposedly had lost the lock on the signal even though it did not. In order to decrease the delay ASIC_DELAY I did some testing and found out that the absolute minimum with my card is 18ms. With 17ms there are many dst_error_recovery messages in the log but the card does tune after calling dst_error_recovery a couple of times. With any value lower the card almost never tunes. I think 20ms should be a safe value for this card.
