×

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

Bi-monthly release with minor bug fixes and improvements

New Internal Guider Features

  • Posts: 969
  • Thank you received: 94
PHD2's default is 12 iterations to give a 25 pixel move. The pulse length is determined according to guide speed and camera pixel size.

Would a pulse length calculation remove some of the confusion perhaps?

OTOH, I used EKOS' default of a pulse of 1000ms. It gave 3 pulses on each axis. I've reused this calibration 10 times without problems.

Cheers,
Steve
3 years 7 months ago #57624

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

  • Posts: 1119
  • Thank you received: 182

Steve, do you know whether PHD2 performs a regression analysis on those 12 iterations? Would it matter IYO? PHD2 seems so interminably slow to me, that was one reason why I ditched it as fast as I could. The internal guider always worked well enough for me, and especially now with Hy's PEC it shaved another few decimals off the RMS. PHD2 seems unnecessarily complicated to me.
3 years 7 months ago #57626

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

  • Posts: 969
  • Thank you received: 94
Hi. Sorry, no idea about the regression but I too have now switched to Hy's new GPG pec and SEP multistar. That combination has taken my star fwhm consistently lower than PHD2 equivalents.

Probably best to look at the images themselves though rather than RMS.

Cheers and clear skies,
Steve

So much discussion. It must be full moon!
Last edit: 3 years 7 months ago by alacant.
3 years 7 months ago #57627

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

  • Posts: 62
  • Thank you received: 1
Hi, where do you set the proportional rate individually for RA/Dec?
Cheers,
// Åke
3 years 6 months ago #59368

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

  • Posts: 1957
  • Thank you received: 420
To the lower right of the screenshot you see “Calibration Plot” and “Drift Plot”. There is one more tab to the left of that where you can enter those values.


HTH, Wouter
The following user(s) said Thank You: Åke Liljenberg
3 years 6 months ago #59370

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

  • Posts: 152
  • Thank you received: 20

Replied by Greg on topic New Internal Guider Features

Hy,

I had my first nights using SEP Multistar this week on two different rigs. Fantastic work on this and the guide logs.

One one scope, SEP Multistar seems to "just work". The other not so much out the gate, though I will spend more time with it in the next days. Mainly, the system would not find a guide star. I posted another topic about this here:

www.indilib.org/forum/ekos/7772-multista...tar-to-guide-on.html

If Jasem is correct in the star shape being critical, this will make SEP Mulitstar difficult to use for ONAG implementations that have astigmatism as a feature.

You asked "what's next" and I have a few ideas:

- Auto-tuning of the PI controller
- Represent min/max values in arcsecs of movement based on discovered pulse dimensions that are seen in the options dialog for calibration.
- Auto-scaling of the min/max setting in cases where the calibrated values of ms/arcsec change. I've seen them change enough to think this might be a good idea. Maybe they shouldn't change and I need to chase a problem?
- Have proportional gain be a percentage of the ideal for a given mount's rate. For example, a mount that uses a rate of 0.5 == prop gain 133.3, so setting 100% uses that rate, 50% cuts it in half, 150% does the expected, etc.

Also, one thing that I've wished I've had way too many times is the actual guide log relative to a given exposure embedded in the fits header of subs for easy analysis. Eventually we could have a tool that allows you to open a sub and look at what happened with the guiding while it was exposing.

Thanks again for your hard work on this!

Greg
3 years 6 months ago #60373

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

  • Posts: 1208
  • Thank you received: 559
Thanks Greg.

One thing to note (re your last comment):

> we could have a tool that allows you to open a sub and look at what happened with the guiding while it was exposing.

Right now (if you are using experimental nightly releases or source-code) you can use the new Analyze tab
indilib.org/forum/general/7597-new-kstar....html?start=24#60055
to get an idea of the guiding during individual exposures, for the current or past KStars sessions.

Also, re astigmatism, I agree with Jasem that it's better to wait until the StellarSolver comes out before trying to improve
detections in distorted stars. A related issue is improving detections for defocused and doughnut-shaped stars, which
would also be quite useful.

Hy
3 years 6 months ago #60378

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

  • Posts: 152
  • Thank you received: 20

Replied by Greg on topic New Internal Guider Features

Yep, I'm on source builds so I see the Analyze tab. I hadn't had an opportunity to dig into it until just now, and it looks pretty useful. I like that it can show prior sessions, and open subs for inspection. Good stuff.

I'm standing by on StellarSolver. Can/will test it out when it's ready.
3 years 6 months ago #60395

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

  • Posts: 910
  • Thank you received: 86
SEP Multistar is great! I am consistently getting rms <0.5 with my HEQ5 now. It used to be between 1 and 2 before.
Big thanks to Hy and everyone involved!
-- Max S
ZWO AM5. RST-135. AZ-GTI. HEQ5. iOptron SkyTracker.
TPO RC6. FRA400. Rokinon 135 and other lenses.
ZWO ASI2600MC. D5500 modified with UVIR clip-in filter.
ZWO ASI120MM Mini x 2. ZWO 30F4 guider. Orion 50mm guider.
ZWO EAF x 2.
Last edit: 3 years 6 months ago by maxthebuilder.
3 years 6 months ago #60416

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

  • Posts: 985
  • Thank you received: 161

Replied by Alfred on topic New Internal Guider Features

I can only second that. Last weekend I came to use and test the new features for the first time. Slam dunk! They did work perfectly out of the box. Basically you select SEP Multistar guiding and you're done. Of particular interest to me is GPG RA guiding though. My G-11 suffers from a pronounced periodic error in RA and the new feature did work like a charm! After a few minutes I saw RA RMS results that were totally unheard of.



This is a tremendous improvement that removes the need (or wish) for any dedicated 3rd party guiding software altogether. IMO Ekos has taken several leaps forward in recent months. A wholehearted "Thank you!" to all contributors who made it possible and to Hy in particular!
Last edit: 3 years 6 months ago by Alfred.
3 years 6 months ago #60431
Attachments:

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

  • Posts: 1119
  • Thank you received: 182
I totally second that. The GPG algorithm has made a lot of difference for me as well. And I am not even using an OAG, only a cheap ZWO 120 mm focal length guide scope with a CEM25P mount. I expect that work even better on my Orion Atlas mount using a 280 mm guide scope.

The internal guider for all practical purposes is now definitely on par with PHD2.
3 years 6 months ago #60436

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

  • Posts: 1208
  • Thank you received: 559
@schwim
Greg,

I showed your post to a couple friends who use ONAGs, and their opinion was that your guide image is likely out-of-focus.
The more out-of-focus it is, the more extreme the ellipticity on the stars. Could it be the case that you're guider is out-of-focus?

Apparently the ONAG has a helical focus adjustment on its guiding port, that you adjust after you get your main imager in focus,
with the goal of getting the two cameras parfocal.
They say you're supposed to adjust that to get the best possible star shape in the guide image.
If they are right, perhaps with that adjustment the current SEP star detection would work better.

Hy
3 years 6 months ago #60453

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

Time to create page: 0.483 seconds