[dvbtools-devel] dvbtune-0.5 inverts diseqc pol and band bits

[Zoilo G. <zo...@xs...>]

dvbtune-0.5 source apparently contains a bug where it inverts both the 
polarization and high/low band bits in the diseqc command ... as I 
cannot find any info on this anywhere, I am posting my results here.

I discovered this while using a KNC1 DVB-S card to compare diseqc 
handling with VP1034 (because diseqc is not (yet) working on my VP1034). 
I am using a Spaun 17089-NF 16-input switch (4 sats) + 8 outputs, which 
supports up to Diseqc 2.0.

First I got confused, because the Mantis driver simultaneously uses the 
old 18/13V+22kHz method to control polarization and band, as well as the 
Diseqc method. So for a while I believed that Diseqc was (partially) 
working, until I found out that in fact 18/13V+22kHz seems to be 
responsible for the (partial) job!

As a result of no Diseqc-functionality on VP1034, it seems that with 
VP1034 on my switch, sat-selection is always fixed to sat=1 (makes 
sense, = the first 4 inputs, which happens to be Hotbird 13.0E in my 
setup), however input selection from the 4 feeds V/H and H/L works fine 
and correctly, although based on 18/13V+22kHz.

When I compared this with my KNC-1 DVB-S card (to find out why 
Sat-selection is not working with VP1034), I found that with KNC1, 
Sat-selection does switch to Sat=2 (so it must be using diseqc), but 
polarization and band-selection were inverted ... So when I issued the 
following command:

"dvbtune -D 2 -f 10832000 -p H -s 22000"

I found that it was actually selecting sat=2 pol=V band=H ("e0 10 38 
f5"= switch input 7), instead of sat=2 pol=H band=L (= switch input 6 / 
"e0 10 38 f6").

The Diseqc-specs at 
http://www.eutelsat.com/satellites/pdf/Diseqc/Reference%20docs/bus_spec.pdf 
(table on page 13) confirm that dvbtune is in fact sending the wrong 
command.

With a patched dvbtune (pol-bit and band-bit both inverted) it works OK, 
and is also according to the Eutelsat specs.

Hope someone benefits from this.

Z.