On 08/06/2021 13:32, Andy Burns wrote:
My linux box with DVB-S2 tuner takes about that long to scan all
transponders, given a "seed" transponder and learning the rest from the
SI tables it sees.
Never took that long for me when I had a 'dish'


--
To ban Christmas, simply give turkeys the vote.

The Natural Philosopher wrote:

Andy Burns wrote:

My linux box with DVB-S2 tuner takes about that long to scan all
transponders, given a "seed" transponder and learning the rest from
the SI tables it sees.

Never took that long for me when I had a 'dish'

I don't remember DVBscan on linux taking so long, but TVheadend does.

"Andy Burns" wrote in message
...
The Natural Philosopher wrote:

Andy Burns wrote:

My linux box with DVB-S2 tuner takes about that long to scan all
transponders, given a "seed" transponder and learning the rest from the
SI tables it sees.

Never took that long for me when I had a 'dish'

I don't remember DVBscan on linux taking so long, but TVheadend does.

Yes it does. I'm not sure whether it's because there is a long just-in-case
timeout before the tuner is allowed to scan the next multiplex (ie probably
a lot of idle time) or whether the tuner is busy scanning right up to the
moment that it is released ready for the next mux.

When I first set up TVHeadend (and I've had to do it three times*) I resign
myself to a long wait, with occasional progress checking, when it is
scanning the satellite decoder. In contrast, scanning the six muxes that I
can receive by terrestrial takes a lot less time *even per multiplex*
(without factoring in that there are a hell of a lot more satellite muxes
than terrestrial ones.

Does the scanning process get the list of multiplexes/transponders by
reading the SI table from an initial "seed" mux? I thought (for both
terrestrial and satellite) it used a hard-coded list of transponders, maybe
with minor tweaks overnight once the initial file-based scan had completed.

Certainly the SI list is used, because even through I've removed a couple of
multiplexes (10758V and 11954H) that no longer exist, they reappear at the
next overnight scan - always with zero services. And I can see that the SI
table does still contain reference to them - by examining a mux with
TSReader, a brilliant tool for the inner-nerd within me ;-)


(*) Once when I set up my original Raspberry Pi, then again when I changed
to a different Pi, and a third time on the new Pi after a kernel upgrade
buggered-up the satellite tuner and I had to reinstall Linux from scratch.