[linux-dvb] Some tests on Avermedia A700

Matthias Schwarzott zzam at gentoo.org
Tue Feb 12 22:08:07 CET 2008


On Monday 11 February 2008, you wrote:
> En/na Eduard Huguet ha escrit:
> > En/na Matthias Schwarzott ha escrit:
> >> On Samstag, 9. Februar 2008, Eduard Huguet wrote:
> >>> Hi, Matthias
> >>
> >> Hi Eduard!
> >>
> >>>     I've been performing some tests using your patch for this card.
> >>> Right now neither dvbscan nor kaffeine are able to find any channel on
> >>> Astra (the sat. my dish points to).
> >>>
> >>> However, Kaffeine has been giving me some interesting results: with
> >>> your driver "as is" it's getting me a 13-14% signal level and ~52% SNR
> >>> when scanning. Then, thinking that the problem is related to the low
> >>> signal I have I've changed the gain levels used to program the tuner:
> >>> you were using default values of 0 for all (in
> >>> zl1003x_set_gain_params() function, variables "rfg", "ba" and "bg"),
> >>> and I've changed them top the maximum (according to the documentation:
> >>> rfg=1, ba=bg=3). With that, I'm getting a 18% signal level, which is
> >>> higher but still too low apparently to get a lock.
> >>>
> >>> I've stopped here, because I really don't have the necessary background
> >>> to keep tweaking the driver. I just wanted to share it with you, as
> >>> maybe you have some idea on how to continue or what else could be done.
> >>
> >> So I can do only this guess:
> >> I changed demod driver to invert the Polarization voltage for a700 card.
> >> This is controlled by member-variable voltage_inverted.
> >>
> >> static struct mt312_config avertv_a700_mt312 = {
> >>         .demod_address = 0x0e,
> >>         .voltage_inverted = 1,
> >> };
> >>
> >> Can you try to comment the voltage_inverted line here (saa7134-dvb.c:
> >> line 865).
> >>
> >> BUT: If this helps we need to find out how to detect which card needs
> >> this enabled/disabled.
> >>
> >> Regards
> >> Matthias
> >
> > Hi,
> >   Nothing :(. Removing (or setting it to 0) the voltage_inverted
> > member doesn't seem to make any difference. I'm starting to suspect
> > that there is something wrong with my antennae setup, so I'll test it
> > later using an standalone STB or by plugging the card into a Windows
> > computer and using the supplied drivers.
> >
> > Regards,
> >   Eduard
>
> By the way (sorry if I'm being molest...): I will leave the card in this
> PC for now, as it's easier fo me to test and develop. As I have also
> Windows here ¿is there any way we could do any reverse enginnering from
> Windows driver, etc...?
>

I already asked you to compare eeprom output in dmesg. But did you also 
compare GPIO messages - like init-values read after startup (to detect 
different wiring)?
This is from my dmesg output: "saa7133[0]: board init: gpio is a600"

Or just attach dmesg output after a cold boot and loading saa7134 driver (with 
i2c_scan=1).

Maybe I create a patch where you can select the other existing zl10313 driver.
So we can compare the logs / functionality.

Matthias

-- 
Matthias Schwarzott (zzam)



More information about the linux-dvb mailing list