[mythtv] Re: Re: [mythtv-commits] Re: Ticket #339: SimplifyTuning process

[John Jurrius]

Mac,

The driver itself should be calculating the signal strength percentage from
the AGC Loop Filter values.  There is no need for the register values
themselves be passed up.  Taylor has the routines to calculate this.

If you have a negative value for SNR then there would be insufficient signal
to ever get a lock any way.  As such I would recommend returning SNR values
only after you have lock so that you are guaranteed a positive and
meaningful value.  Taylor also has the routines to calculate this (also with
not floating point of course) from which I have never seen a negative value
returned.

John

-----Original Message-----
From: Mac Michaels [mailto:wmichaels1 at earthlink.net]
Sent: Friday, September 23, 2005 10:56 AM
To: mythtv-dev at mythtv.org
Cc: John Jurrius
Subject: Re: [mythtv] Re: Re: [mythtv-commits] Re: Ticket #339:
SimplifyTuning process

How is  the driver is supposed to return AGC RF/IF Loop
Filter Accumulator Values? The return value holds one
unsigned16 bit value. There are two signed 12 bit values
for the AGC RF/IF Loop Filter Accumulator. Although Taylor
may know how to compute signal strength, I can not pass him
the data from the driver. So, how is signal strength
calculated? I am not aware of the signal strength formula
for the lgdt3302 and lgdt3303 chips. Please advise and I
will update the driver.

With regards to the snr value. I have already written the
code to compute the snr in db (without using floating
point). It is #ifdef'd out in the driver. Folks complained
that it was wrong to return a signed value in an unsigned
variable. For example mythTV displays it as an unsigned
value. Thus a negative number associated with poor signal
conditions showed up as near 100% while a small positive
number that indicates good signal conditions reports near
0%. Instead I return the truncated raw Equalizer Mean
Square Error value.

I can't find a FE_READ_SIGNAL_QUALITY function call in the
DVB API Version 3.

--Mac

On Thursday 22 September 2005 11:44 am, John Jurrius wrote:
> The Signal Strength data for the lgdt330x is intended to
> be calculated from AGC RF/IF Loop Filter Accumulator
> Values.  The SNR is calculated from the Equalizer Mean
> Square Error.  The routines for calculating both have
> already been provided to Taylor.  If you cannot return
> the SNR in dB perhaps you can do what is done in Windows
> driver for the HD5000 which is convert the SNR to a
> percentage of its value over a the usable range of the
> SNR which is then referred to as Signal Quality.
>
> John
>
> -----Original Message-----
> From: mythtv-dev-bounces at mythtv.org
> [mailto:mythtv-dev-bounces at mythtv.org]On Behalf Of Mac
> Michaels
> Sent: Thursday, September 22, 2005 9:08 AM
> To: mythtv-dev at mythtv.org
> Subject: Re: [mythtv] Re: Re: [mythtv-commits] Re: Ticket
> #339: SimplifyTuning process
>
> The lgdt330x chips do not provide signal strength data.
> It might be possible to compute signal strength from the
> automatic gain control (AGC) data, but the spec sheet
> does not provide a formula. As a result the signal
> strength is always set to 0.
>
> Data is available from which the signal to noise ratio
> (snr) can be calculated. The driver currently returns raw
> data since the unsigned 16 bit return value it not
> appropriate for returning the actual snr in db. Snr must
> be signed because it is negative when the signal is less
> than the noise and positive when the signal is greater
> than the noise.
>
> --Mac
>
> On Thursday 22 September 2005 09:53 am, David Engel wrote:
> > On Thu, Sep 22, 2005 at 10:42:16AM -0400, Taylor Jacob
>
> wrote:
> > > Many drivers just report arbitrary vaules that have
> > > no unit because the author did not have the
> > > datasheet, and this will not change.. Not even if all
> > > manufacturers of demodulators decide to support linux
> > > for some reason..
> >
> > FWIW, my Fusion HDTV5 always reports 0 for signal
> > level.
> >
> > David
>
> _______________________________________________
> mythtv-dev mailing list
> mythtv-dev at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
>
> _______________________________________________
> mythtv-dev mailing list
> mythtv-dev at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev