×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Settle Time after Dither: Two different settings, one ignored

  • Posts: 1009
  • Thank you received: 133
Last night I wanted to change the settle time after dither, so I went to Guide->Options->Guide and changed 'Settle'.
But the change was ignored, it continued with the old value :(
Hunting around, I finally found/remembered KStars->Settings->EKOS->Capture, which also has a 'Guiding Settle' setting. Changing that one then indeed did what I wanted.

Why are there two different settings, and what is the actual difference? Somewhat irritating....
4 years 6 months ago #43931

Please Log in or Create an account to join the conversation.

  • Posts: 1957
  • Thank you received: 420
The help of the two options explains it all.

Guide->Options->Guide -> Settle: After dithering is successful, wait this many seconds before proceeding.
KStars->Settings->EKOS->Capture -> Settle: Wait this many seconds after guiding is resumed before starting capture.

In other words, the first option determines how many seconds to wait after dithering before resuming guiding and the second options determines how many seconds to wait after guiding resumes before starting a capture. I must admit that I don't understand the need of the first settle option, but the second one allows for guiding to become stable again after dithering before capturing again.


HTH, Wouter
4 years 6 months ago #43932

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Well, 'explains it all' is maybe a bit optimistic. I didn't notice a difference when changing the first one from 5 to zero....

And the second, 'guiding resumed', I would expect after, e.g., pausing guiding during autofocus, and then re-enabling it. With dither, you don't suspend guiding (else it couldn't settle), so it isn't resumed...

It does make a lot sense to wait (quite) a while when you resume guiding after autofocus, because that can easily offset your pointing when using an OAG. But I don't need that same wait after dither, when guiding is running, trying to correct that 'kick' (i.e., settling). There I *only* might need the first one, depending on how much I trust the settling limits. Using the same value there as for resuming guiding (from unknown offset) dramatically increases the time for dithering, and that does substantially cut the effective exposure time (e.g., I dither every 3 frames, with 45s exposure; If a dither takes 30-40s that is some 20-25% loss...)

I think the current way of applying the settle times isn't intuitive.
4 years 6 months ago #43933

Please Log in or Create an account to join the conversation.

  • Posts: 985
  • Thank you received: 160
The second one is a VERY IMPORTANT setting. Without it, imaging would resume right after dithering. At this point in time though guiding can be WAY off. It takes a couple seconds to get guiding back to normal. I first had this set to 5 seconds but now have it set to 12 seconds since 5 were clearly not enough with my G-11. These two settings should IMO not be found in two totally different places though. I'd prefer to have them both in dithering settings.
Last edit: 4 years 6 months ago by Alfred.
4 years 6 months ago #43951

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Have you really tried that? What are your settings regarding dither settle?
If I dither (via PHD2) the PHD2 display very quickly switches from 'Dither' to 'Guiding' while it catches the star back to nominal position. During that time EKOS waits, until the settle requirement is fulfilled (I think...). At least on my system, when the counter #2 starts after dithering, the guiding is already completely settled.

Are you using the internal guider? Might there be a difference how internal/PHD2 handle things during dither?
4 years 6 months ago #43968

Please Log in or Create an account to join the conversation.

  • Posts: 985
  • Thank you received: 160

What are you suggesting here? Me making up stories?

I report from my own experience, many nights, internal guider. If I had not really tried that I had kept quiet.
4 years 6 months ago #43995

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
So obviously there is a difference between internal and external guider then. With PHD2 enabled it waits until that one is settled, and then it applies the 'resume guiding' wait.
So if I set this to a high value I get unacceptably long dither times, if I set it to a low value dither is fine, but the exposure after an autofocus would be crap.
4 years 6 months ago #43999

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
These options are distinct.

Guide->Options->Guide -> Settle:

In the KStars code, this option is known as ditherSettle.

I can tell you this option is used by both the Internal Guider and by PHD2 Guider. I haven't looked too much into how the internal guider uses it, but PHD2 is sent this parameter BOTH any time it is told to guide AND anytime it is told to dither. The parameter is not optional, so we have to send it even if it is 0. So when using PHD2, KStars doesn't use this for option itself, but tells PHD2 to do it. I know this because I worked on the code that allows Ekos to communicate with PHD2. Here is the documentation which explains how PHD2 uses the parameter. github.com/OpenPHDGuiding/phd2/wiki/EventMonitoring. Now, it is possible that it gets ignored under certain conditions or its also possible that maybe it doesn't always use the full time, or it is possible it was changed in a different version. I don't know, since all I know is what they say in their documentation. PHD2 is a little cryptic about it, but they say "minimum time to be in-range before considering guiding to be stable" and I believe it should be how long they wait until PHD2 reports "SettleDone." We programmed Ekos to wait for the "SettleDone" event before confirming that the dithering is completed or the guiding is started.

KStars->Settings->EKOS->Capture -> Settle:

In the KStars code, this option is known as guideSettle.

I can also tell you about this option. This is an option that is used by the Capture module and is independent of your guiding method. This is implemented entirely in KStars and is not sent to PHD2 to handle. Basically KStars starts a timer after the dither success is reported, and KStars waits this long before it begins taking the next exposure. It is a very simple option.

My thoughts:

It is my belief that what should happen is that the dither command is sent, then the mount moves, then the it waits the ditherSettle time, then the guider starts trying to guide, then it waits the guideSettle time, then the next capture is started. So if both of these options work as advertised, then I think both are necessary.

- The ditherSettle time is needed because you wouldn't really want your guider to try to start guiding until the mount is done its motion.

- The guideSettle time is needed because you wouldn't really want your image to start capturing until the guider has settled down and is guiding well.
The following user(s) said Thank You: Alfred, Wouter van Reeven, Peter Sütterlin
4 years 6 months ago #44002

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Thanks for this in-depth explanation!

But following the link, the time (ditherSettle) sent to PHD2 is the time that the guide star has to be within the settle range. So it is settled when PHD2 reports 'settle done'. The additional wait from EKOS then is a bit like wearing a belt and suspenders?
Nevertheless, I think with this info I can properly tune things now...
4 years 5 months ago #44020

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Yes it could be that both are not really necessary, and maybe one could be removed, but it could also be true that on different systems one or the other setting could be more important. Perhaps it is silly to use both at the same time, or perhaps it is most prudent to divide the time you would have chosen to wait after dithering between the two settings. I don’t really know. I just know that they are very different settings and are used in different ways by the software.

In order to recommend how best to use the settings, I would say someone should extensively test different guiding configurations with the different settings to see what would work best. As of now though I would say both settings should be left alone. Yes they are in two different places and yes that seems odd. Perhaps they should be moved together, but currently the settings are actually organized with the system that uses them.
4 years 5 months ago #44025

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Well I guess the first thing is to assure that the users know that there are two settings, maybe mention each other one in the tool tips as a 'See also'?
4 years 5 months ago #44030

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Actually, I just notice that the tooltip for Guide->Options->Settle says 'After dither is successful, wait this many seconds before proceeding'. Which is more or less the same as in the main EKOS options. But if that is indeed the number sent to PHD2 it rather is 'Error has to be lower than Threshold for this many seconds before dither is assumed as settled'. Might have reduced my confusion.
4 years 5 months ago #44097

Please Log in or Create an account to join the conversation.

Time to create page: 0.763 seconds