107 replies to this topic

[nemosince1996]

Sorry, littlesat, I am at pc just only for a moment. I could return late night or I will provide data tomorrow.

[littlesat]

The data is important for me as I need it (to see possibilities en test) to solve it... especially as I do bot own that insertable tuner..when I have the files I can simulatie it.

Edited by littlesat, 24 July 2019 - 23:57.

[twol]

Does it mean really that there is no choice how to change it? Only VU team can? But strange is that when I tried OpenATV image it is ok in it and dvb-s2 tuner is correctly recognized and I was able to setup what I needed. I expect that drivers are the same for both images. But I need OpenPLI because you have the best configuration and control of FBC DVB-S2(x) tuners. Thank you for your effort. I can use this box so far the same way as Solo2 which I replaced and to wait for repair.

The code for FBC and DVB-S2x tuners is almost (but not quite) the same in OpenViX as OpenPli(unlike OpenATV), so would be interested to know what happens if you try OpenViX - if it works then OpenPli should be able to work.

[malahini]

I have exactly the same problem as nemosince1996! The problem requires a solution.

[littlesat]

Maybe I can add an extra check as from the root fbc tuner we always need at least 7 further tuners... when it are not eight then it is not an fbc tuner and the fbc indication was False....’but it still stays a trick....

[-M-]

It is a version of FBC chip that have 4 demods. Maybe need to think about if doing a workaround for FBC tuners. However I don't know if any OpenPLi supported STB using it.

[littlesat]

As far I can see it is a dual tuner that fits into a ‘fbc’ slot... I hope tomorrow I can find some time to invent a solution

Edited by littlesat, 25 July 2019 - 18:37.

[nemosince1996]

I have tried latest OpenVIX image and it is the same as OpenPLI. See pictures from OpenVIX. Second picture is very strange. DVB-T2 tuner has letters I-P.

Attached Files

•  IMG_20190725_210420.jpg   118.2KB   2 downloads
•  IMG_20190725_210456.jpg   91.13KB   2 downloads

[littlesat]

Not wierd at all as the latest vu drivers say this is a fbc tuner that should have 8 tuners and not 2.... so e2 gives i-p there (ijklmnop) the drivers have folders in /proc/stb/frontend for tuners that do not exist and the first 16 say they are fbc tuners....
I think I can work a round this but still this is absurt... still I think vu should actually solve it as the procs are giving false (static) information!!!! The work a round I can make is add an ectra check if there are really 7 more tuners after that first root tuner.... if so the fbc indication should be False.. but to add this is somehow redigulas. As far I did soms ‘dry’ checks the work-a-round should always solve/work-a-round it, even when you put two of those tuners in the box...
But when you have a True fbc plus such tuner it must be in the second slot.

Edited by littlesat, 25 July 2019 - 22:09.

[twol]

Not wierd at all as the latest vu drivers say this is a fbc tuner that should have 8 tuners and not 2.... so e2 gives i-p there (ijklmnop) the drivers have folders in /proc/stb/frontend for tuners that do not exist and the first 16 say they are fbc tuners....
I think I can work a round this but still this is absurt... still I think vu should actually solve it as the procs are giving false (static) information!!!! The work a round I can make is add an ectra check if there are really 7 more tuners after that first root tuner.... if so the fbc indication should be False.. but to add this is somehow redigulas. As far I did soms ‘dry’ checks the work-a-round should always solve/work-a-round it, even when you put two of those tuners in the box...
But when you have a True fbc plus such tuner it must be in the second slot.

@nemosince1986 ....... thanks for trying
@littlesat:
Vu+ never fail to amaze me ..... e.g. the audio and video procs are pretty accurate for every manufacturer of Arm receivers (or at least so far) except Vu+ where they contain absolute garbage..... so in ViX we have to use hard coded values for the Vu+ boxes - as you say absurd!

[littlesat]

And we refuse if vu then do something else in our code so we always try to find a fits all solution

Edited by littlesat, 26 July 2019 - 08:16.

[littlesat]

I did made some tries to stop it... But it practically not possible to solve it without creating a log of additional code for this dirty VU stuff as it creates simply a fbc_id proc for the first 16 tuners at all time... So as far I can see now VU+ should solve it. 

[littlesat]

Mmmm I found a reasonable solution.. But each nim has then a special parameter for total number of nims which is a bit obvious but does not harm...

[littlesat]

FYI...

 

https://github.com/O...19d712027896057

 

it will be in develop. Can you copy and paste this newer NimManager.py for testing? When it indeed solves the issue we can consider to chery pick into release.

[nemosince1996]

What should I do specifically to test your changes? From the GitHub page do I need to download NimManager.py via Browse files and copy to / usr / lib / enigma2 / python / Components /? Will enigma2 convert a file to NimManager.pyo automatically? I do not have experience with it. Please advice me.

[littlesat]

Yep put it there and then restart enigma2. Due to the restart a new pyo will be created

Edited by littlesat, 26 July 2019 - 18:30.

[nemosince1996]

Look at picture. You have probably made it. Right configuration is there. Now give me some time to connect cables to tuners, to configure them and test them.

 

 

Attached Files

•  With_changed_NimManager.jpg   120.74KB   4 downloads

[nemosince1996]

Littlesat, great, well done! From short test it looks that everything is OK.

[nemosince1996]

I copied changed NimManager.py to my old Ultimo 4K. I have not observed any problems.

[littlesat]

Sounds we can work-a-round it.... wanwizard can sherry pick.... thanks for the feedback as I need it because I cannot test it because I do not own that tuner to insert in my Ultimo4k. I’m also think we’re the first image where this can work... even vu’s image (when I checked their code) should go incorrectly.
Thanks for your positive approach.

Edited by littlesat, 26 July 2019 - 21:56.