[linux-dvb] Re: SL1935 vs TDA8060TS

[Andrew de Quincey <adq_dvb@lidskialf.net>]

On Sunday 26 Sep 2004 03:24, David Zanetti wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Sun, 26 Sep 2004, Andrew de Quincey wrote:
> > > Eg, 12.456GHz vertical off Optus B1 has a bunch of FTA channels and
> > > there's nothing on horizontal. Tuning vertical gives me:
>
> [tune, some Verror]
>
> > > Tuning horizontal gives me:
>
> [tune, no Verror]
>
> > > And dumping a stream from either tuning gives me channels I expected on
> > > V, with TS continuity errors regardless of which polarisation I choose.
> >
> > Hmmm, hmmm, so its *that* sort of thing - weird effects on the polarity.
> > Well, I've seen THAT before - its always been a site installation problem
> > as opposed to a hardware problem. This is the kind of thing that is very
> > hard to diagnose. Its _probably_ not a hardware problem since tuning
> > works, and you get good reception in windows.
>
> Aye, the "works in win32" for me suggests it's a driver issue, something
> it's not doing which the win32 drivers do.
>
> > OK, first of all, is there a multiswitch of any sort between you and the
> > dish?
>
> It's a shared installation for the building, so I'll have to get back to
> you about the details. However, it does have UHF terrestrial combined into
> the same feed into the appartment, so there will be a device handling
> that, but it's _extremely_ likely to be a single LNB since there's pretty
> much only one useful sat to look at here in NZ: Optus B1, and it's all
> vertical.

Yeah - I see what you mean. http://www.lyngsat.com/optusb1.html. There are a 
couple of H channels mentioned on that page though.

The only time I've seen something like what you're seeing was with a quad LNB 
which had been wired to the multiswitch wrongly - so for example the H/LO on 
the LNB had been connected to the V/HI on the multiswitch and so on. I was 
able to make it work in linux by trying different combinations of the 
polarity switch and hacking about with the SLOF frequency in dvbtune 
(although we then just sorted out the wiring).

The confusing thing was it worked perfectly with commercial satellite decoders 
- I think they must just try all possible cominations of H/V/HI/LO until they 
get something that works. Perhaps the windows software does this as well.

I see from your test you've tried this as well. However, your log for H in 
dvbtunes shows no errors whatsoever - yet you say you definitely get them 
when dumping H. I don't quite understand that...