Dithering not working in Deep Sky Sequence Planner with REPEAT

[Cobarcore]

I really love SharpCap and I'm using it also as the main acquisition software for deep sky imaging on the field.
All the wonderful features like Polar Align, Plate Solving, Focusing, etc convinced me using even if deep sky acquisition is not at the same level of user experience.

Anyway, using Deep Sky Sequence Planner i set my frames and set DITHERING every 1 frames. Guide camera connected to PHD2 working fine.

Case 1: working perfectly, I get dithering before each frame (except the first)

Code: Select all

Light 5 frames exp 5s
Repeat the above capture steps (DISABLED)

Case 2: not working! I do not get any dithering!!

Code: Select all

Light 1 frames exp 5s
Repeat the above capture steps: 5 (ENABLED)

Case 3: working perfectly, I get dithering before each frame (except the first of each line)

Code: Select all

Light L filter 5 frames exp 5s
Light R filter 5 frames exp 5s
Light G filter 5 frames exp 5s
Light B filter 5 frames exp 5s
Repeat the above capture steps: (DISABLED)

Case 4: not working! I do not get any dithering!!

Code: Select all

Light L filter 1 frames exp 5s
Light R filter 1 frames exp 5s
Light G filter 1 frames exp 5s
Light B filter 1 frames exp 5s
Repeat the above capture steps: 5 (ENABLED)

I attach the screenshot of setting and sequencer progress log.

How to activate the dithering after each repeat cycle?

Case 3 is my actual choice for acquisition, so that if I need to stop imaging (e.g. for clouds) I have all frames LRGB and, e.g., not only LR while GB clouded.

Regards,
Lorenzo

[admin]

Hi,

thanks for reporting - that's an interesting one. If you set your number of frames in each of the captures to more than one then the problem basically goes away, since the cause is capturing one frame via the sequencer.

Right now, the check to work out whether to dither is done before the frame is captured, so when only one frame is being captured, the code checks and decides that it doesn't need to dither because no frames have been captured yet. Then one frame is captured, but there isn't another check. The next time round we are capturing one frame again and the process repeats with no dithering.

I think the right solution is to move the dithering check to happen after the frame is captured, that way the dither would run even for a single frame being captured (if the frequency for dithering is set to 1 frame). I need to convince myself that nothing will break as a result of this change though before I make it

cheers,

Robin

[Cobarcore]

Dear Robin,
thanks as usual for your great answer.

Yes, setting to more than one can solve temporarily.
E.g.

Code: Select all

L 2 frames (dithering on the 2nd frame)
R 1 frame
G 1 frame
B 1 frame
repeat 5 times

So all frames are dithered filter by filter, except 2nd L respect to 3rd L, and so on.

Better is

Code: Select all

L 2 frames (dithering on the 2nd frame)
R 1 frame
G 1 frame
B 2 frame (dithering on the 2nd frame)
repeat 5 times

so that dithering is done both on L 2nd and on B 2nd, so the L 3rd has been dithered by the B.

Unfortunately doing this will prevent to chose arbitrary numbers of frames and will reduce the mixing of frames.

Solution to move the dither at the end of each capture seems good (Maxim DL works in this way), even if it will cause to dither even if not necessary (e.g. between a L and a R frame).

A more complex solution would be better but I don't know if it can be done in SC.
Dithering should occur only when between two frames of the same filter no other dithering has been done.

Ideally it would be fantastic also if ARBITRARY number of frames can be selected, like in MaximDL sequencer: e.g. L 10 frames, R 4 frames, G 4 frames, B 5 frames, WITH cycling between filters. Maxim automatically decide to take 2 L frames, then 1 R, 1 G, 1 B, then 3 L frames, then 1 R, 1 G, 2 B, etc... so that at the end the requested number of frames has been acquired.
Also a command "Resume sequence" is present in Maxim so that if I stop the sequence I can continue with the right filters.

Regards,
Lorenzo

[admin]

Hi,

the sequencing part of SharpCap works by using fairly simple individual steps that do their own job and are not generally influenced by the steps around them (this keeps things simple and robust). That means that the simple approach is best rather than trying to think about whether a filter change is coming up when deciding whether to dither. So, I will make the change of the dither check happening after the frame rather than before and we will see how that works out - I think the only difference will be that a dither will occur after the last frame in the capture if the number of frames being captured is a multiple of the dithering interval.

It's perfectly possible to do more complicated things like different dithering intervals for different filters if you want, but you have to use the 'Advanced Sequence Editor' to do that, rather than the 'Deep Sky Sequence Planner'. There is a delibrate distinction between doing the commonly used things in the Deep Sky Planner and allowing those who want full customization to use the advanced editor. One thing that helps with this is that there is an 'Advanced Edit' button in the Deep Sky Planner - this takes the choices you have made in the planner and produces a sequence in the advanced editor that will carry out those steps - you can then see how this works and customize it further.

cheers,

Robin

[Cobarcore]

Dear Robin,
I understand the logic behind, and I can work on the advanced sequence editor, however when you're at the mountains in the night at freezing temperature and you need to hurry up to start the deepsky sequence so not to loose precious astronomical night hours, the idea of programming code for taking the sequence is out of the options.

I really hope that one day you'll work on an even advanced deep sky editor

Regards,
Lorenzo

[admin]

Hi Lorenzo,

the idea is not really to do programming on the spot, but to have a pre-prepared 'advanced' sequence that you just fill in some details into (what target, how many frames, etc)...

Actually, having written that, I am now wondering if there is any way to have an 'advanced' sequence that is somehow configured to ask you for those sort of details when you start it. Hmm, an interesting idea!

cheers,

Robin