What features next for SharpCap 4.2?

[Howardk]

Flip mirror works wonders for finding a pesky planet

[takuya_t]

Hi Robin,

The irony is that the more beginner-oriented the equipment is, the harder it tends to be to work with.
It's a common phenomenon, but it's especially noticeable in astrophotography.
This feature is really needed for beginner-level equipment, yet that's exactly where implementing it is the most difficult. It’s quite a paradox...

"Reverse" function would undoubtedly be effective if it worked reliably. The problem is that it's far from clear how many mounts can actually return to the exact same position after performing a reverse move. That uncertainty is simply too great.
One practical approach would be…
・Stop the search as soon as we recognize that the target has been overshot… At least we know the mount has moved closer to the target.
・Start a new spiral search centered on the current position.
Ideally, the next search would run at a slower speed to avoid overshooting again.
However, that directly conflicts with the backlash problem, making it a difficult tradeoff.

In any case, I think Spiral Search function would be viable as long as it can satisfy the following two minimum requirements:
1) The spiral pattern must have no gaps.
2) The spiral must expand outward from the current position in all directions. (even if the actual travel distance varies depending on the direction)

Requirement 1) should be achievable by designing the search pattern with sufficient margin based on focal length, pixel size, and slew speed. Differences in mount acceleration and response characteristics may result in more overlap than necessary between successive spiral passes, but at least it should be possible to eliminate gaps.
Requirement 2) is the more difficult challenge, especially at low slew speeds, because of the uncertainty introduced by backlash.
It's just a thought, PHD2 already includes a mechanism for measuring backlash and compensating for it during guiding. I wonder whether a similar approach could be adapted for this purpose.

Even if it takes longer, a search method that users can trust will eventually get them to the target would be a game changer for people who would otherwise be completely stuck and frustrated.

Sorry for the long message.

Takuya

[admin]

Hi,

thanks for the thoughts on the spiral search - always useful to have another view on the issues and possible solutions to them. Another feature that I should consider for the spiral search if/when it happens is an 'auto stop' function - monitor the image and if something large and bright appears, stop straight away. That should be able to respond significantly faster than a human trying to either click a button or release a mouse button once they see the target.

I had also been thinking about some sort of calibration of the actual movement of the mount co-ordinates in response to different speeds/durations of 'MoveAxis' - this doesn't help with true mechanical backlash, but would help calibrate out the effects of speed spin up/slow down and mount reaction time to commands.

cheers,

Robin

[admin]

Thank you everyone for the feedback and suggestions - with the poll completed, the autofocus profiles idea comes out on top and the remote notifications and spiral search are tied for second place.

I already have some partly complete code for the autofocus profiles, so I will try to bring that to completion over the next week or two.

We have already had some posts above in favour of the enhanced spiral search functionality, but I don't think anyone posted a reply in support of the remote notifications... It would be interesting to hear from one or two of the people who did vote for that to hear what they would like such a feature to do.

cheers,

Robin

[Suby81]

Hi Robin,

some after survey comments.

Autofocus profiles are welcome.

Spiral search improvements is a very nice feature to have. I use small sensors and long focal, resulting in difficulties in pointing the planets and the sun. What I usually do, and I'm telling you thinking that it might be automatized, is to strongly increase the gain and reasonably the exposure and let me guide by the histogram. If I get closer to the sun (works also with the planet but much less) the histogram increases even if I don't get the planet in the screen, because I get his glow. When the histogram decreases, I know I'm getting away from it.

Remote notification are also useful. When I do lunar/solar mosaic or a timelapse, or I do several video of a plant to then de-rotate, I don't stay in from of the screen. Receiving a notification on the phone that alerts me about cloud/planet is out of screen/camera connection dropped/frame rate is too low I can try to fix the error and not lose the complete sequence capture.

Best Regards,
Suby

[admin]

Hi,

thanks for the input - I also use the 'move towards the brightness' approach, particularly for finding the sun. I have wondered about automating it, but sometimes you end up at a 'local' maximum brightness which isn't really the target - probably a reflection within the OTA leading to a bright spot some way off of the target. It happens often enough to me that I suspect an automated tool would also frequently get trapped by such a ghost image. That's basically put me off trying to automate the whole thing as I think it would probably not be robust.

It's useful to hear the sort of cases that remote notifications might be useful for - any more detail from anyone on that particular topic is welcome.

thanks,

Robin