count-down issue

[Jasem Mutlaq]

I think INDI behavior is correct. if nothing changed, don't send setXXX for
it. Therefore, we should fix it in Ekos with a timer.

--
Best Regards,
Jasem Mutlaq



On Wed, Feb 16, 2022 at 6:33 PM Wolfgang Reissenberger <
sterne-jaeger at openfuture.de> wrote:

> I think I could explain it. Running with the Telescope Simulator, the old
> and the new value at least on RA is *slightly* different in a range of
> 10^-8 and therefore at least for the simulator the coordinates update
> happens all the time.
>
> The question is: should we rework it in Ekos or in INDI? In INDI it’s a
> matter of one single line, in EKOS it’s quite an effort (and we need a new
> timer).
>
> Cheers
> Wolfgang
> —
> Wolfgang Reissenberger
>
> www.sterne-jaeger.de
> TSA-120 + FSQ-85 + epsilon-160 | Avalon Linear + M-zero | ASI 1600mm pro +
> 6200mm pro
>
> Am 16.02.2022 um 09:56 schrieb Jasem Mutlaq <mutlaqja at ikarustech.com>:
>
> Sorry, I was referring to INDI:
> https://github.com/indilib/indi/blob/f479114a60cddb7c4209e83824702de4efd58245/libs/indibase/inditelescope.cpp#L711
>
> --
> Best Regards,
> Jasem Mutlaq
>
>
>
> On Wed, Feb 16, 2022 at 11:28 AM Wolfgang Reissenberger <
> sterne-jaeger at openfuture.de> wrote:
>
>> I’m not sure whether I get it right. Inditelescope.cpp::248 has this
>> entry:
>>
>>         emit newCoords(currentCoords, pierSide(), hourAngle());
>>
>>
>> This is unconditional, there is no check whether currentCoords equal the
>> previous one.
>>
>> Or are you talking about another place?
>>
>> Am 16.02.2022 um 09:22 schrieb Jasem Mutlaq <mutlaqja at ikarustech.com>:
>>
>> I don't think Ekos should depend on this behavior. INDI::Telescope
>> doesn't send updates when the previous values are identical.
>>
>> --
>> Best Regards,
>> Jasem Mutlaq
>>
>>
>> On Wed, Feb 16, 2022 at 11:09 AM Wolfgang Reissenberger
>> <sterne-jaeger at openfuture.de> wrote:
>>
>>
>> Ekos relies upon the INDI driver that it regularly sends the
>>  EQUATORIAL_EOD_COORD coordinates - no matter whether they change or not.
>> The frequency is defined by the polling period defined in the INDI driver.
>>
>> If for whatever reason the INDI driver does not send them, the MF counter
>> won’t be updated.
>>
>> We could indeed discuss whether it makes sense that Ekos relies upon a
>> regular update even if values do not change. But currently, at least for
>> the scope coordinates, it does.
>>
>> HTH
>> Wolfgang
>>
>> Am 16.02.2022 um 08:49 schrieb Jasem Mutlaq <mutlaqja at ikarustech.com>:
>>
>> Hello Hy,
>>
>> I don't believe this is an INDI driver issue. The driver is behaving
>> as it should. Ekos should do this. Also, I've been using this with
>> EQMod for many years and it works just fine, so what gives?
>>
>> --
>> Best Regards,
>> Jasem Mutlaq
>>
>>
>> On Wed, Feb 16, 2022 at 1:38 AM Hy Murveit <murveit at gmail.com> wrote:
>>
>>
>> I was noticing on the forum a thread you responded to where users were
>> complaining that the meridian-flip count-down wasn't working for their
>> Onstep mount.
>>
>> I had/fixed/hacked the same problem on my A-P driver.
>>
>> Here's what I believe is likely happening to them.
>>
>> The MF count-down is updated in a callback method that's watching the
>> RA/DEC from indi.
>> That is, in mount.cpp, updateTelescopeCoords calls checkMeridianFlip()
>>  which does the updating of the counter.  However, updateTelescopeCoords is
>> only called when the slot newCoords is hit. That signal is likely only sent
>> when the RA/DEC change, but when the scope is tracking, the RA and DEC
>> don't change!
>>
>> I saw that my driver wasn't updating its MF counter either (and
>> presumably, therefore wouldn't flip), but I know my old Eqmod mount did
>> work fine, so I assumed it was an issue in my A-P driver.  FWIW, I added
>> this hack in my driver to get around the check for newness in newRaDec.
>>
>> Hy
>>
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kstars-devel/attachments/20220216/74a9a6fe/attachment-0001.htm>