Thanks Toni,
I'm used to compiling from source -- I have written a couple of X2 plugins for TSX. But my normal IDE is Apple's Xcode, and CMake is not a part of that.

On the other hand I'm no expert in C++, in fact I much prefer the old-fashioned C, so that is another hurdle.

So I will have to investigate how to set things up. But it will have to wait until late on Wednesday as I have a high-priority task right now whose deadline is Wednesday.

Cheers,
Richard

Toni,

One thing that occurred to me. I don't remember in the past EKOS building a model by default. It seems to always build one now, at least in the fact that I see a bunch of points being stored. I can't see how two models is good. In the paramount's case, you might have some error in pointing, and more or less will always be there unless you refine your model in TSX. (or remove all sources of the error in balance, flexure etc).

Can we turn the model off in EKOS? Seems like it shouldn't be active at all.

Just a thought.

Bryan

Hi Bryan
You wrote (I) Select a target either on my phone (connected via SkySafari --> indi, or in Kstars OR in TSX, and issue a slew. Then Ill go in EKOS to the alignment tab and and choose Capture and Solve.
There's a possibility that a slew outside of KStars (say with SkySafari or TSX) does not update the target coordinates and thus leaves the destination with it's invalid initialization values after a [Capture & Solve]. The newest binary should catch this.

It seems to always build one (model) now, at least in the fact that I see a bunch of points being stored.
I'm not aware of a change in this area. I only know the [Mount Model] button at the lower edge of the Align tab. Do you see stored points there? Did you try to purge them? Does Paramount have some options in the INDI Control driver?

Im usually using kstars for selecting targets. Sometimes I would just peruse the starchart in another program that has better graphics. But I could never really use it for framing in EKOS because it would never be spot on so it's easier to do it in kstars.

And yes I have cleared out the points in kstars' model, but I don't think it should be creating one in the first place as it will just build up. Normally there would be a setting to turn off a model. So each time the solve n slew function gets called it's just ignoring what it did before. Even with no modeling on a paramount, and for other EQ mounts I have owned this has been good enough that Ive never bothered with modeling for pointing reasons in the past. The main reason in TSX is the tracking correction (IMO) as once a good tpoint model is created it doesn't have to guide near as often.

One other interesting thing to note, is a while back I was thinking, well it's just the paramount "don't sync to me" issue, so I deleted my model and turned off tpoint altogether, then turned off differential slewing, and I didn't notice a difference. Same problem. Ill have to try that again and send the log.

What changed in the latest binary?

Normally there would be a setting to turn off a model
That's why I mentioned INDI Control Panel. For certain mounts you can turn off the model there.

What changed in the latest binary?
The invalid initialization values of skypoint coordinates after a [Capture&Solve] are intercepted and reported.

I'm still too busy to do anything yet, but assuming I manage to compile the sources, where does the binary go?

I have searched on my Mac earlier for indilib and been unable to find it.

Thanks,
Richard

That's why I mentioned INDI Control Panel. For certain mounts you can turn off the model there.
I can turn off the TSX model in TSX, Im talking about EKOS building a model. Because you really shouldn't have both from my understanding. I can of course clear it, but it shouldn't collect points in the first place if the model is elsewhere.

Last night I turned off the model in TSX, and set EKOS to turn off differential slewing. What is weird then is it had the original problem prior to your binary. Completely lost now. Not sure but shouldn't it have behaved like a mount that doesn't have a model?

Last night I turned off the model in TSX, and set EKOS to turn off differential slewing. What is weird then is it had the original problem
This is totally inapprehensible! I have no idea what could cause such a strange behaviour. The questions is: While having turned off "differential slewing", did you see syncs in the log window?

I know right! I forgot to save the log. Does that console log window save somewhere? I can get it for you, or I can try again tonight since the moon is back, of course since it's clear skies now.

Does that console log window save somewhere?
No, i don't think so.

Attached is a screenshot of how it is. Before it was set to regular. Ive tried verbose before also. We'll see how it goes tonight. The current log is very brief. I think my prior attempts to get the log were foiled by the fact that I was looking for the astrometry/alignement log (see my erroneous subject line

Anyway, on verbose, hopefully it logs what you need. And also i chose the indi driver for paramount as well. Maybe it will provide something interesting. Ill try with it set to sync with TSX tpoint off and the other way with differnential and TSX tpoint turned on.

Thanks.

Am I crazy or was there a screenshot with the log settings? Anyway, now you are saying it won't save. Ill capture it manually out of the console window tonight provided it's clear enough.