Let's talk about planning mode

starnet59

I finally had a clear night, the first one in more than a month, so I was able to test the planning mode for the first time. Here are my impressions, the good, the bad and the ugly.

The good, actually the amazing. It works, it's usable and more importantly you can take pictures while sleeping! In my opinion this is the most important new feature released in the last year, and frankly very long overdue. I created a plan with 28 targets, from 10 minutes to 70 minutes, for a total time of 12h40m. The S50 went through it all successfully and powered down on its own at the end. The temperature was -10C at the start of the evening and reached -18C just before dawn. The anti-dew heater was on all the time and I used a 20Ah external USB power module. In the morning the USB module was empty and the S50 battery level was 95%. I also used my own 3D-printed dew and stray light shield, which drops off automatically when the arm closes. My S50 is one year old and I have used it a lot in this configuration during this time. There have been many threads here about powering the S50 with an external power source, battery replacement, tangled wires and magnetically coupled USB cables, operating in sub-zero temperatures, battery lifetime, arm closing with dew shield and so on. All these are non-issues and I hope this test helps address some of those concerns.

The not so good. 7 of the 28 captures have failed. I do not know really why, apart from a very short and cryptic status label. One was of Jupiter and it says "Shoot failed", the other 6 are all "Go To failure". I suspect these 6 are due to the target running into the trees, but there is no way to tell. I was running low on targets while creating the plan, which I did in 20 minutes, just before the start of the night. I can try to reconstruct them in Stellarium and confirm whether that's the case or not. Which brings us to what improvements could be made. Let's see:

10-minute granularity is too large, Dwarf 3 has 1 minute. If you want to catch moons of Jupiter or Saturn even 1 minute can be too long. A user selectable granularity would be best.
This is not a planning issue but planning now makes it easier to compute stacking efficiency - the total time is in the plan and the number of 10-second successfully stacked frames is in the file name. Most of the captures had efficiencies around 65% and none of them went over 75%. The night was perfectly clear, there was no wind and nobody walked around the S50, not at -18C. So when you prepare your plan, you have to take that into account, your mileage will vary.
Preparing a plan in SkyAtlas is not easy. Yes, there are light gray rectangles for all the targets, but they do not have the proper orientation, and more importantly, unless you have a fully unobstructed view in all directions, you cannot be sure whether the target will be visible during the window you select or not. I suspect that's why 6 of my 28 targets failed but I cannot be sure.

The not so pretty. I am still angry it took so long to get this feature, we should have had this 6 months ago, I started asking for some kind of automatic scheduling on this forum in March 2024 I think. This would have saved me so many sleepless nights! It turns out all it took was for Dwarf Lab to announce the Dwarf 3 and ZWO promptly came up with what looks like a carbon copy of the Dwarf 3 planning feature. One year ago ZWO was a trailblazer in the field of consumer grade smart telescopes, now they simply follow.

starnet59

starnet59 Here is one attempt to use the planning mode to create a working mosaic:

The four grey rectangles show the frame rotation if you were to take all pictures right now, they bear no connection to the frame orientation when each actual image will be taken. Since SeeStar knows when each picture will be taken, it can calculate the actual rotation and draw each rectangle accordingly. This is of course more complicated - the orientation of the frame at the start of capture, at the end? For longer captures the frame can rotate a lot, even go back and forth if crossing the meridian.

I think there is a very simple middle ground solution. When the grey rectangles are drawn, in whatever current position, also draw an inscribed circle. Whatever is inside this circle will be captured, no matter what the orientation of the field will be. Here is a mock-up of how this could look like:

This is literally a one-line code change, when you draw the rectangle, also draw a circle. How long can it take to implement something like this?
This would be of great help when trying to create plans to make larger mosaics. Just make sure the circles overlap and you will have full coverage. No need to enter RA/Dec coordinates manually in a cumbersome way, you can make shorter, say 5 or 10 minutes individual captures and spread them over multiple nights, you do not lose hours of data when the clouds roll in or you get a satellite trail. As long as your target is still visible, the same plan can be used on many nights without having to make any changes.
Here is another possible use of the new planning feature. Let's say you want 12 hours of M51. You have to do it over many nights, obviously, but instead of doing 4 or 5 hour captures every night, create a plan of 10 minutes on the same target repeated all night, then even use it multiple nights in a row. Some captures will be better than others, some will have satellite trails, some will be obscured by clouds or the target will run into obstructions, but select the good ones, stack them in say Siril (which is free) and you will be amazed by the results.
What would make this really easy to set up would be a planning feature where you define an all night long single exposure, then have the planner divide it into 15-minute chunks. Then manually delete the chunks you do not want, set up the S50 outside and go to sleep. Seems like another very low hanging fruit to me.

Support@Seestar

I'll send your post to the team, thanks for the advice.
However, due to the holidays, it may not be possible to let the team know in time.