[vdr] statistic (was - Multiproto updated Re: [ANNOUNCE] VDR developer version1.5.14)

Manu Abraham abraham.manu at gmail.com
Tue Jan 29 23:22:02 CET 2008


Igor wrote:
> Manu
> 
> will be possibility to receive from driver the raw statistic, too ?
> 

As of now, the STB0899 get's the raw statistics does processing
and the processed details are sent to the application.

It's possible to get the same, what you mentioned, just that the
mentioned ioctl needs to be in there. The relative scale would be
just raw statistics (unprocessed)


Regards,
Manu

> Igor
> 
> 
> -----Original Message-----
> From: Manu Abraham <abraham.manu at gmail.com>
> To: VDR Mailing List <vdr at linuxtv.org>
> Date: Mon, 28 Jan 2008 22:22:58 +0400
> Subject: Re: [vdr] Multiproto updated Re: [ANNOUNCE] VDR developer version1.5.14
> 
>> Morfsta wrote:
>>> Manu,
>>>
>>> Firstly, thanks for all your work - it's appreciated.
>>>
>>> Is the interface working properly for reading the signal strength, BER
>>> and status?
>> Currently, there is a small confusion. Most drivers in kernel, just 
>> report some
>> crap as signal strength, snr etc.
>>
>> For the STB0899, STM helped so much, alongwith i had some help from one
>> of the Newtec guys, we were able to get the statistics in some proper form,
>> currently a dBm/10 scale is used.
>>
>> Since this scale is different from the in kernel existing scale of 
>> randomness
>> to a standardized one, currently this change might look a bit nonsense 
>> to you
>> as it will report different signal statistics as reported by the API.
>>
>> There will be a need to add one more ioctl, where the user can request the
>> driver to provide the statistics in a relative scale or an absolute 
>> scale which
>> can be used for measurements too.
>>
>> The result would be that and end application can display the statistics 
>> like
>> any other commercial STB, in a nice and beautiful way, while being quite
>> accurate (depending on the driver) without any hacks or workarounds to be \
>> done in the application
>>
>> (The statistics that you get is very much device specific and not device 
>> independant
>> and hence is not very easy for a user application to get proper 
>> statistics as of now)
>>
>> With the change, all statistics related operations can be implemented 
>> within the driver
>> such that the user is presented with standard and uniform statistics 
>> across multiple
>> devices without the need to do device specific code as done by MythTV 
>> etc, where it
>> looks for specific device drivers (really ugly)
>>
>>> The rotor plugin doesnt return anything and I can't get anything out
>>> of VDR-Femon.
>>>
>>> Is it working OK with the TT 3200?
>>
>> The statistics is working with the STB0899, maybe i didn't follow what 
>> exactly you are
>> looking at. If you can detail a bit, it would be much more helpful. The 
>> STB0899 supports
>> TT S2 3200, KNC1 DVB-S2+, KNC1 DVB-S2, Pinnacle PCTV 450e and the 
>> VP-1041 as of
>> now and the same signal related stuff is applicable to all.
>>
>>> BTW I have a TT3200 on order so I can compare the difference between
>>> the two cards soon.
>> Currently, the STB0899 is _not_ calibrated for actual statistics. I have 
>> requested
>> Azurewave for some sample devices to be sent to STM for proper 
>> calibration of the
>> LUT's for the statistics. (Most probably, tomorrow or so STM will 
>> receive the samples
>> at Grenoble)
> 
> 
> _______________________________________________
> vdr mailing list
> vdr at linuxtv.org
> http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
> 




More information about the vdr mailing list