Erik Andrén a écrit :
> 2009/1/11 Laurent BARRAILLE <[email protected]>:
>   
>> Erik Andrén a écrit :
>>     
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>>
>>>
>>>
>>> Laurent BARRAILLE wrote:
>>>
>>>       
>>>> Hi,
>>>>
>>>> Thanks,
>>>>
>>>> My sensor is detected again.
>>>>
>>>> I've got always the problem of the red and blue blance, the image is
>>>> very red.
>>>>
>>>> When i set manually (with v4lctrl) this controls to default value it's
>>>> ok.
>>>>
>>>> They don't set in the init sequence however it seems to be necessary to
>>>> set them.
>>>> I saw you removed it from init sequence  (
>>>> http://linuxtv.org/hg/~eandren/gspca-m5602/rev/d22ff6898892 )
>>>>
>>>>
>>>>         
>>> They are now set to the current value after the init sequence (see
>>> po1030_init() not the default.
>>> But I did find a bug where I set the red balance twice and not the
>>> blue balance at all. This is now fixed.
>>>
>>>
>>>       
>> My image is less red, but there is a problem yet.
>> It's the red balance only, it doesn't set right in init or other sequence.
>>
>>     
>
> I don't get this. With the latest version there is:
>
>       262     err = po1030_set_red_balance(&sd->gspca_dev,
>
>       263                                   sensor_settings[RED_BALANCE_IDX]);
>
>       264     if (err < 0)
>
>       265             return err;
>
>       266
>
>       267     err = po1030_set_blue_balance(&sd->gspca_dev,
>
>       268                                   
> sensor_settings[BLUE_BALANCE_IDX]);
>
>
> which should work.
>
>   
Yes, i tried to change the value, it works.
The problem is in the value, maybe a problem in decimal conservion ...
I made some test to find where is really the problem.

Regards

Laurent

> Regards,
> Erik
>
>
>   
>> Thanks
>>
>> Regards
>>
>> Laurent
>>     
>>> Thanks for reporting,
>>> Erik
>>>
>>>
>>>       
>>>> Erik Andrén a écrit :
>>>>
>>>>
>>>> Laurent BARRAILLE wrote:
>>>>
>>>>         
>>>>>>> Ok, the sensor dump  seems to work again but the sensor detection
>>>>>>> doesn't work yet.
>>>>>>>
>>>>>>> I attach result of dump_sensor
>>>>>>>
>>>>>>>
>>>>>>>               
>>>> Thanks,
>>>> When correlating your sensor dump with the datasheet it seems like
>>>> all reads are offsets of one. When sensor register address 0x00 is
>>>> read we get the contents of address 0x01 and so on. I have some
>>>> saves register dumps from earlier and it seems like we've always had
>>>> this problem.
>>>>
>>>> The po1030 window driver snoops I have never reads the sensor only
>>>> writes to it so I'll have to think about how to solve this.
>>>>
>>>>
>>>>         
>>>>> Good, i think this will help you to improve the driver.
>>>>>
>>>>>           
>>>>         
>>>>> Regards.
>>>>>      Laurent
>>>>>
>>>>>           
>>>> I've pushed the previous patch plus a temporary solution to the
>>>> probe routine so that you don't need to force the sensor.
>>>>
>>>> Thanks,
>>>> Erik
>>>>
>>>>
>>>>
>>>>         
>>>>>>> Regards
>>>>>>>
>>>>>>> Laurent
>>>>>>>
>>>>>>> Erik Andrén a écrit :
>>>>>>>
>>>>>>>
>>>>>>> Laurent BARRAILLE wrote:
>>>>>>>
>>>>>>>               
>>>>>>>>>> Hi,
>>>>>>>>>>
>>>>>>>>>> This revision is the offender of the bad sensor dump :
>>>>>>>>>>
>>>>>>>>>> http://linuxtv.org/hg/~eandren/gspca-m5602/rev/14a865ca7f2a
>>>>>>>>>>
>>>>>>>>>> Regards
>>>>>>>>>>
>>>>>>>>>> Laurent
>>>>>>>>>>
>>>>>>>>>>                     
>>>>>>> Thanks for your quick work.
>>>>>>> Does the attached patch fix the issue?
>>>>>>>
>>>>>>> Regards,
>>>>>>> Erik
>>>>>>>
>>>>>>>
>>>>>>>               
>>>>>>>>>> Erik Andrén a écrit :
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Laurent BARRAILLE wrote:
>>>>>>>>>>
>>>>>>>>>>                     
>>>>>>>>>>>>> Hi,
>>>>>>>>>>>>>
>>>>>>>>>>>>> The new sensor dump.
>>>>>>>>>>>>>
>>>>>>>>>>>>> Regards
>>>>>>>>>>>>>
>>>>>>>>>>>>> Laurent
>>>>>>>>>>>>>
>>>>>>>>>>>>>                           
>>>>>>>>>> Thanks,
>>>>>>>>>> Obviously I broke the routine reading the po1030 registers as all
>>>>>>>>>> registers now respond with 0xff. Is it possible for you to either
>>>>>>>>>> do
>>>>>>>>>> a bisect of the last commits or back them out one by one doing a
>>>>>>>>>> dump_sensor to see what commit that is the offender?
>>>>>>>>>>
>>>>>>>>>> See the hg manual for more instructions on how to perform these
>>>>>>>>>> routines.
>>>>>>>>>>
>>>>>>>>>> Regards,
>>>>>>>>>> Erik
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>                     
>>>>>>>>>>>>> Erik Andrén a écrit :
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> Laurent BARRAILLE wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>>                           
>>>>>>>>>>>>>>>> Erik Andrén a écrit :
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> Laurent BARRAILLE wrote:
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>>>> Hi,
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> My sensor is no more automaticly recognized.
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>                                       
>>>>>>>>>>>>>>>> Hm, this is bad. If you insert the gspca_m5602 with
>>>>>>>>>>>>>>>> dump_sensor=1,
>>>>>>>>>>>>>>>> could you please send me the produced kernel log.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>> i attached the result of this command :
>>>>>>>>>>>>>>>>>      sudo modprobe gspca_m5602 dump_sensor=1 force_sensor=5
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>                                   
>>>>>>>>>>>>>>>>>>> It works only when i force the sensor in loading module.
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>                                       
>>>>>>>>>>>>> Ok, the attached dump showed that all register values were
>>>>>>>>>>>>> bogus.
>>>>>>>>>>>>> I've now moved the dump_sensor function to a more "safe"
>>>>>>>>>>>>> position.
>>>>>>>>>>>>> Could you please update to the latest driver and attach a new
>>>>>>>>>>>>> dump
>>>>>>>>>>>>> result?
>>>>>>>>>>>>>
>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>> Erik
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>                           
>>>>>>>>>>>>>>>>>>> The auto balance white control has almost no effect on my
>>>>>>>>>>>>>>>>>>> image
>>>>>>>>>>>>>>>>>>> maybe
>>>>>>>>>>>>>>>>>>> it's normal.
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>                                       
>>>>>>>>>>>>>>>> Did you activate it? It's disabled by default.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>> Yes, with vlc and v4lctrl.
>>>>>>>>>>>>>>>>> I've tried many times, there is a very very small change
>>>>>>>>>>>>>>>>> when i
>>>>>>>>>>>>>>>>> activate
>>>>>>>>>>>>>>>>> it but useless.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>                                   
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>> Best Regards
>>>>>>>>>>>>>>>>>      Laurent
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>                                   
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>>>> Unfortunately, i have noticed no really improvement in
>>>>>>>>>>>>>>>>>>> the use
>>>>>>>>>>>>>>>>>>> of my
>>>>>>>>>>>>>>>>>>> webcam.
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> With the old and the new version of the driver, the image
>>>>>>>>>>>>>>>>>>> is
>>>>>>>>>>>>>>>>>>> too red,
>>>>>>>>>>>>>>>>>>> too pink.
>>>>>>>>>>>>>>>>>>> When i set the blue and red balance to 54 ( 36 in hex)
>>>>>>>>>>>>>>>>>>> instead of
>>>>>>>>>>>>>>>>>>> 64 (40
>>>>>>>>>>>>>>>>>>> in hex) the colors seems better.
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>                                       
>>>>>>>>>>>>>>>> Ok, I'll set these as new default values.
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> Thanks for testing!
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>>>>> Erik
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>>>> Best Regards
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> Laurent
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> Erik Andrén a écrit :
>>>>>>>>>>>>>>>>>>> Hi,
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> I've recently been able to aquire the datasheet for the
>>>>>>>>>>>>>>>>>>> PO1030
>>>>>>>>>>>>>>>>>>> from
>>>>>>>>>>>>>>>>>>> pixelplus, unfortunately I'm not allowed to distribute or
>>>>>>>>>>>>>>>>>>> share it
>>>>>>>>>>>>>>>>>>> with anyone. I've pushed some changes to the linuxtv tree
>>>>>>>>>>>>>>>>>>> for the
>>>>>>>>>>>>>>>>>>> po1030.
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> Couly you please test the latest version and see if it
>>>>>>>>>>>>>>>>>>> still
>>>>>>>>>>>>>>>>>>> works
>>>>>>>>>>>>>>>>>>> for you?
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> Thanks,
>>>>>>>>>>>>>>>>>>> Erik
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>                                       
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>                                 
>>>>>>>>>>>>>>>>>>> ------------------------------------------------------------------------------
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> Check out the new SourceForge.net Marketplace.
>>>>>>>>>>>>>>>>>>> It is the best place to buy or sell services for
>>>>>>>>>>>>>>>>>>> just about anything Open Source.
>>>>>>>>>>>>>>>>>>> http://p.sf.net/sfu/Xq1LFB
>>>>>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>>>>>> M560x-driver-devel mailing list
>>>>>>>>>>>>>>>>>>> [email protected]
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/m560x-driver-devel
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>                                       
>>> -----BEGIN PGP SIGNATURE-----
>>> Version: GnuPG v1.4.9 (GNU/Linux)
>>>
>>> iEYEARECAAYFAklqOY4ACgkQN7qBt+4UG0GqnQCaArnwPEALunYZFhGUznjdbXs5
>>> 6AgAoJlcyJ2Wg6utawUmBLpipVzs7Eph
>>> =+rlJ
>>> -----END PGP SIGNATURE-----
>>>
>>>       
>>     


------------------------------------------------------------------------------
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
_______________________________________________
M560x-driver-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/m560x-driver-devel

Reply via email to