New versus Old focusing method

Sequence Generator Auto Focus
1

Looking at auto focus during an imaging session I noticed that some of my later subs (6 to 8 hours after the sequence started) were out of focus. Not just one – but a few in a cluster right at the end of the night. When I looked closer at what was happening I found SGP doing the opposite to what I expected……maybe my understanding of the ‘new’ focus regime is incorrect. I’ll provide a link to the full SGP log but a couple of examples of what has me baffled:-
Time 02:01:28 the log shows new method autofocus @ 286, old @273
At 02:02:09 SGP sent the focuser to 273. As I was watching the AF routine & graph low point, I decided to step in and put the focuser manually at 286. The subsequent image was fine.
I changed the Min Star size from 6 to 8.
Again at 03:35:00 the new method shows 362 and the old shows 345. It sent the focuser to 345. Again I stepped in and set it manually to 360. The sub focus was OK (could have been a little better – but was not way off)
So – why is there a new method of focus calc in the system and then SGP uses the old? I have not gone back through the log – these were ones I was watching on the screen and was very aware of where the graph was showing the focus point. I was so surprised to see the focuser being instructed to go to a very different point.
On my next run I will watch this again during the earlier part of the sequence. I did not see any problems in the earlier subs last night – but cannot say one way or the other what instructions were being sent by SGP in relation to the graph low point. You will note in the log probably that focusing (and filter changes) slowed down considerably late in the session – as mentioned in another thread (re Atik EFW2). So I am unsure if comms are getting slow for some reason……but RAM usage was only at about 33% and SGP did not appear to be using significantly more than at other times.
Any ideas – have I misread what was actually happening?

Re slow filter change & focusing: I have no hard evidence - but think this is ongoing for a while now…OK early in the sequence and slower later. I don’t think it is cable. Might it be HUB related? Feels more like a RAM problem - but there is no evidence of that.

Link to the log : Dropbox - File Deleted

2

There are 2 different methods of determining focus. When they don’t agree we generally default to “old” as it is generally more accurate.

New - Linear regression (line intersection) of the V curve.
Old - Weighted lowest local average of 3 best points.

The New method generally works quite well but if a “good” fit can’t be obtained (seeing, etc) we default back to the Old. Also if the New method is out of tolerance with the Old we default to the Old as it’s generally more robust, at least locally.

If you can duplicate these issues it would be helpful to put together some AF Packs and we can take a look at the data and see if we can adjust this some.

Thanks,
Jared

3

Unsure if you mean 3 lowest points - but that was not the case. The line intersection was never far from the low point of the graph. For the times noted in my text, if you had the time , you could probably check back on the numbers…but never mind for now. This weekend I will be taking down my gear before heading off on a short trip. When I come back, I want to do some maintenance on my set up before imaging again. If I run into this again in the future, I will for sure get some AF packs to show what I am experiencing. Until I actually sat watching this in the early hours this morning - I have not noticed this setting the focus point so far away from the bottom of the curve before…it had me very confused re WHY! I think we must be getting some bad electricity here in SE Spain just now :slight_smile: