Bug 414967

Summary: Focuss module issues at F2
Product: [Applications] kstars Reporter: James <jrlangston11>
Component: generalAssignee: Jasem Mutlaq <mutlaqja>
Status: CLOSED FIXED    
Severity: normal    
Priority: NOR    
Version: git   
Target Milestone: ---   
Platform: Compiled Sources   
OS: Linux   
Latest Commit: https://commits.kde.org/kstars/456c445d89b888c223629090a893051435c955f3 Version Fixed In: 3.3.9
Sentry Crash Report:

Description James 2019-12-08 22:51:12 UTC 
When using focuss module with a F2 RASA the underlying algorithms are not suited to developing a vcurve at F2.  Basically the method used tries to establish a curve and if it's having trouble it sets a very high step of around 600 to even 4000! assuming that the system is slow I guess

But at F2 any steps more than say 150 will just produce a large donought that isn't recognised as a star or not even visible.  This occurs with gradient, iterative, polynomial etc.

It seems to do this as a very shallow curve is produced as focuss changes with only a small amount of steps and larger steps result in no stars being recognised. maybe it expects a steeper curve?

A possible solution is to have an input box that specifies the maximum number of steps the software will make in any single focuss movement.  This way if I know steps above 150 will produce donoughts I can limit maximum step to say 100.

Issue has occurred on all versions incl current git using celestron sct indi driver.
Comment 1 Jasem Mutlaq 2019-12-10 07:29:40 UTC 
Git commit 456c445d89b888c223629090a893051435c955f3 by Jasem Mutlaq.
Committed on 10/12/2019 at 07:29.
Pushed by mutlaqja into branch 'master'.

Add option to limit single step commands.

FIXED-IN:3.3.9

M  +14   -0    kstars/ekos/focus/focus.cpp
M  +67   -44   kstars/ekos/focus/focus.ui
M  +4    -0    kstars/kstars.kcfg

https://commits.kde.org/kstars/456c445d89b888c223629090a893051435c955f3
Comment 2 Jasem Mutlaq 2019-12-10 07:30:39 UTC 
Please test this change and see if it works to resolve your issue.
Comment 3 James 2019-12-19 11:25:27 UTC 
Thanks it works much better now