Build Smarter and More Efficient Sequences in NINA
Vložit
- čas přidán 26. 07. 2024
- NINA's Advanced Sequencer can help you take your Astrophotography to the next level! In this video, I will show you how you can improve upon a typical basic imaging sequence and turn it into an efficient imaging machine that will crank out images all night long as efficiently as possible. We will leverage multiple features including intelligent looping criteria, filter offsets, smarter dithering, and even take our backlash compensation mechanism into account. We are going to start using every second of our limited dark skies to our advantage. Let's go!
Lot's of thanks to Isbeorn and the rest of the NINA users/devs on Discord for providing so many great ideas over the last several months!
Like, Subscribe, and turn on Notifications to be alerted when more videos in the series are available.
Software Used in this Video:
NINA: nighttime-imaging.eu/
For more of my Astrophotography related content, go to:
www.PatriotAstro.com
/ patriot_astro
-------CHAPTERS-------
0:00 - Introduction
0:42 - Typical Basic Sequence
4:34 - Simple Advanced Sequencer
6:26 - Understanding the Costs
8:16 - Adding Loops
9:13 - Smarter Dithering
10:25 - Losing the AF with Offsets
11:22 - Saving Even More Time
12:14 - Smart Backlash Usage
14:36 - Adding more L or HA
15:16 - How Much Time Is Gained
UPDATE: NINA Developers made a change behind the scenes that affects the parallel sequence of Filter Change + Dither simultaneously for people who "Disable Guiding on AF". If you have this set, it now also affects filter changes due to how offsets were impacting the process. Based on the fact that the filter change only provides minimal time savings, you can go ahead and remove that from your sequence such that the dither can take affect as expected.
Two years old and still a relevant video, Chad! Moving from Voyager to NINA (and I loved Voyager for its DragScript) - but I'm discovering (as a result of your videos) that much of what I use in Voyager can be done in the advanced sequencer in NINA - conversion accomplished! I'm using 3.1 Beta.
Mind.. BLOWN. Amazing. Wow, You have convinced me to switch to the nightlys and the advanced sequencer!
You won't be sorry. There is always a learning curve, but this is well worth it. The amount of ways NINA can improve your imaging and save you from mid-session trouble is countless.
underrated video.
Thanks!
One of your most valuable videos and they all very usefull.
Thanks!
Your videos are great. This one is outstanding! I've been using the advanced sequencer since it first came out, and while I was able to streamline my sessions, I never really looked at how to optimize them. I'm updating my templates now with your ideas. Brilliant work!
Thank you so much for the nice words! While I’d love to take credit, I’m more or less just the messenger as so many others have had great ideas within the Nina discord through the entire development cycle. Here I’m just trying to piece them to all together and get it to a wider audience. ;)
Your tutorials are really good, right to the point and definitively not boring compared to others. Thanks man !!
Thanks! I 'hope' to be back at it soon. Too much going on and bad weather to boot...
Exactly what I was looking for when trying to figure out how to automate a new mono setup. Thank you!
Glad it helped!
Again, another great NINA tutorial. I've been looking through the advanced sequencer in NINA but I am not comfortable enough to use it on a dedicated project. This tutorial certainly will give me more confidence. One thing for sure, this ADVANCED SEQUENCER has lots of power in it.
Thanks again! Take a look at my earlier Advanced Sequencer Video. Lots more info there. This video here builds upon that one.
BIG THANKS for everything!
Wow, great focused content. Great job.
Thanks! Glad you enjoyed it. Hope it helps!
@@PatriotAstro I think I will be binge watching a lot of your channel content
LOL. Watch ‘em all! No charge! ;)
Boy this is amazing. I was using advance sequencer but not in this way. I will have to add some flats and dark flats to the sequence but this is amazing. Thanks so much. Sharing now!
Honestly, much of this is just stuff I’ve learned from others. Isbeorn has been a great inspiration and creator that is responsible for most of this :)thanks for watching! I’ll see you over in the Discord I’m sure.
Claudio if you haven't already made a library of flats etc I would highly recommend it as a time saver. I have a complete set of 30, 60, 120, 180, 300 and 600 seconds flat, darks and dark flats all made into single masters but only for gain 139 (unity ASI1600) so far. All done with N.I.N.A flats wizard and daylight. Your 294MC Pro would be able to do the same I think.
@@barrytrudgian4514 Hi Stanley I do have a library of darks but flats and flat darks are done after or beforehand imaging session. I wasn't aware that you can do a library of those subs. Don't you get different flats depending on dusk motes on your imaging train?
@@claudioarchi I guess that is correct. My imaging train is mostly kept the same as my coma corrector filter wheel and camera are always connected so I have been using a master flat and flat dark for about 6 months. May be time for me to sort out fresh masters. I like your NGC6960. Am gathering data on that one tonight.
@@barrytrudgian4514 I do have the whole setup always connected, but I move it from indoors to outdoors, so I do flats and dark flats each imaging session normally afterwards. I use an automated flat panel. Thanks for your comment. That image is one of my serious AP. since I started this hobby. Cheers
Excellent video. The idea of using a parallel instruction to combine filter change with dithering was new to me and is something I'm going to incorporate into my sequences from now on. I'm curious on the time savings of ordering filters based on the focuser backlash compensation though. I imagine this depends on the focuser and how much backlash compensation is needed, but I find doing a filter change between adjacent filters, and one where the filter wheel has to skip over a bunch of filters (like from B all the way back to L) tends to take longer than the focuser adjustment. Another big time saver not mentioned is making sure to use the backlash overshoot in the OUT direction as this speeds up the autofocus time, as the overshoot is only applied twice, at the beginning and end of the autofocus run, whereas using a value in IN will force the compensation on each inward step building the autofocus curve. But yeah I feel this is one of those videos I will be linking to people in the future!
Thanks for the comments! You are right on the filter wheel movement. The savings would be even greater if you reordered your filters in the filter wheel as well. Thanks for watching! Glad you found it useful.
Hi, I tried the filter offset and dither as a last part of one loop method, and I found an interesting thing. I have used off-axis guider and turned on the pause during auto focus option. I though this only has an effect when a full Auto focus is kicked in. But, it is also triggered when focuser moves for the filter offset. So 3 exposures -> stop guiding -> filter switch (focus moves for the offset) -> resuming guiding -> setting guiding happens. I guess, the dithering as a last part of one loop does not save time for off-axis guider unless there is a way that NINA does not stop guiding while offset compensation is in-progress.
YES! Great catch on this BTW. I spoke with the lead developer about this and they did make a change behind the scenes that affects this. If you have Disable Guiding on AF set it now also affects filter changes due to how offsets were impacting the process. Based on the fact that the filter change only provides minimal time savings, you can go ahead and remove that from your sequence such that the dither can take affect as expected.
Thanks for this.
For some targets tough, I like to get lum and blue when the target is high up, and then pick up green and red later on when the target gets closer to the horizon. The way I've done this - while avoiding having no R in the morning - without math is (using the same example as you): setup a timed loop for 2 hours with L, another for 2 hours with B, then G, then R. That way if I have underestimated the overhead, the loop will exit when it's share of the time is up and at least I will end up with a similar number of subs. I don't get the dither time reduction you do (but then again, when doing superlum I think I would want the dithers you eliminate) but the autofocus time is also kept under control.
I can completely understand splitting filters in some cases, yep. On the dithering piece, that is actually the 'magic' of my loop. In a normal use case, you could take 60 L frames and dither every 3 images. This means I dither 20 times overall, and when I look at my L frames the next day, I will see I have dithered every 3 as expected. Same for Red, Blue, Green, etc... In my new style loop, I am still dithering each filter every 3 frames. I am just doing this more efficiently. I take 3L, 3R, 3G, and 3B ... then dither all at once. The outcome is the same, I can look at my L the next day and see I have dithered L every 3 frames just like before, ... same for the other filters. :)
@@PatriotAstro true, but like I said with a superlum where you stack L+R+G+B in one stack (for improved SNR over plain L) you end up with a dither every 12 images, which is a bit much for me.
Ah, superlum, got it. What if you created the super lum a little differently? So in my example you would likely do SupL1=L1+R1+G1+B1, and so on as you cycle through. Maybe change my loop to 1 of each exposure, then dither. This way you are dithering every single “filter frame”, but batching them 4 at a time to still cut back on dithering. Then stagger your SuperLum like SupL1=L1+R2+G3+B4 so that you were combining 4 images that have all been dithered differently because they came out of different cycles of the loop. I’m not sure how you’re creating your superlums today, I’ve seen some creative scripting to help with that, and this would certainly require a change in the script. Anyway, just a thought. :) Thanks for watching!
What a useful video, really great!. Any chance on how to do an all night sequence using weather safe instructions? Park, close the roof on clouds... and such? thank you!
I'll think about how to best do it. I was just having a discussion about this in a discord AP channel actually where we were talking about the Safety Monitor Ascom stuff and NINA instructions like "Wait until Safe". Since I don't have a dome, it is a little tricky, BUT I could try to work it all out and test it with the Ascom Dome Simulator I guess.
Please show us how to setup the weather function.
Very clever setup within the advanced sequencer for time management. How confident are you with the new offset calculator plug in to keep the focus without running the AF after each filter change ?.
It works quite well. I would recommend doing more than 3 cycles through the loop. Change that to four or five, or even more. Your calculations will get better this way. Also, try to do it on a night where the seeing will be consistent through the process so you don’t end up with any outliers. All in all it works extremely well.
Thank you for your very good video. I don't understand where is the first auto-focus, could you help me?
So, before I jump in, I just want to make sure we’re clear. In the advanced sequencer, when I’m showing you all the ways to save time, that particular sequence is NOT a complete sequence. I never cool the camera or unpark the scope or set up meridian flips, etc. I’m really just focusing on the imaging cycle and the pieces that allow me to save you time during your session. OK, with that out of the way, When I talk about the first auto focus that I can’t save you time on, this is because no matter what I do to any sequence, advanced or basic, you always have to have an initial auto focus to make sure your session starts in focus. You see, using offsets is a great timesaver, but they only work if you are currently in focus. So since I cannot save you time by excluding the first auto focus run of the night, I never subtract that time as a savings. Make sense? You will always need to focus at least one time.
Thank you very much for this helpful video! I've got a question concerning your sequence: where do you place the instructions "Slew to Ra/Dec", "Slew and center", and "Start Guiding"? If you place them inside your loop they will be executed every 12 images, won't they?
Correct. Normally what you will do is place a Sequential container inside the Deep Space Object container. The sequential container contains your imaging instructions and loop. Just ahead of that container is where you would do things like AF, Slew/Center, etc. All the 'one-time' stuff. If you haven't yet, track down my Updated Shared Advanced Sequences video for a bit more explanation on how I do it. Also my older Advanced Sequencer tutorial may help as well.
I'm using a Hyperstar on my C-11 and I change filters manually. I have NINA set to use the manual filter wheel, which means it tells me to change the filter when it's time. This will still be useful I think, but can you think of a way to optimize my LRGB runs any further? I think the plan you laid out is great if you have an automatic filter wheel but wouldn't help me (as much as I want it to!)
Not a lot of help with automation there BUT you may want to look at the Ground Station Plugin. You could add a Ground Station instruction to email/text/send a push message/flash a smart bulb in the house, etc when it is time to change the filter. Then drop the filter change behind that. This way you will be notified the change is ready to happen via phone, etc. (better than manually monitoring the situation endlessly)
While this approach will lead to each filter being dithered as expected, I wonder if the fact that the filters aren't dithered differently with respect to each other will lead to some degree of fixed pattern reinforcement when they're eventually combined. In other words, each filter stack might end up with similar residual pattern noise. Thoughts?
I thought I had replied to this but it looks like something happened. Sorry! I haven't seen any fixed pattern as a result. I'm pretty sure you wouldn't, but I can completely understand where your question is coming from, there is a strong degree of logic in it. With this method, each individual stack (for mono) ends up independently just looking like a normal dithered stack and the dither+stack combo does its job to clean up the noise and increase SNR. The question then becomes, is there a residual pattern as a result of combining multiple channels of data that contain matching stacks of dithers. Seems the answer is no based on all of my results, and I think that is because each channel's independent dither+stack has done its job and the each stacked channel doesn't have anything residual to include. With this being said, I could see it happening if you have very few images to combine per channel and have not dithered frequently enough. You can see a similar result when try to drizzle with similar data (checkerboard patterns after the drizzle).
I used one of your filter offsets sequences and it saved me 30 min when doing LRGB! Thank you!
btw, is there a way to take specific amount of subs per filter? example: I wanna take 30 subs of RGB and 90 Lum and then end the sequence or move to the next target, how do I do that?
It is going to be sort of a yes and no, or more like 'meet in the middle'. Traditional imaging, historically, was done by setting up image counts and looping through that until moving on to either the next filter or target. This works, but it can either leave you with extra unused time OR over shooting the available time and missing out on expected images. The Advanced Sequencer was designed to avoid that problem. The idea is that you can now loop based on available time rather than an image count. OK, so,... can you loop for a specific number of images? Yes. Either change your loop condition to 'loop for iterations' or set your exposure counts on something like the Smart Exposure instruction. If you are trying to take advantage of Offset/Dither efficiencies in my sequences, you likely need to use the looping condition. If you do this, just have it do 1 R/G/B per cycle and 3xL. Then after 30 iterations you will have 30xRGB and 90xL as you desired. Make Sense?
@@PatriotAstro thanks for the detailed answer! Yeah I understand, is it better to do for example 3x RGB and 9xL or 1x RGB and 3xL or it doesn't matter?
@@Calzune Depends on the rest of the sequence and how often you want to Dither as an example. In your example, it is either a total of 18 exposures in the loop or just 6 exposures. So depending on the exposure length, and how often you want to dither, i'd adjust for that.
@@PatriotAstro Okey thank you 🙌🙌
Hi Chad, this was really helpful - many thanks. Can you tell me if there's a way in the simple sequence to have NINA take 5 - 20 second images, then pause for 10 minutes, then take another set of 5 images and then pause 10 minutes, then repeat this process for say 20 or so cycles? This type of image/pause/image/pause would be useful in particular for Variable Star imaging and capturing exoplanet transits. So I need to go to the advance sequencer or is there a way with simple one to get this functionality? Cheers! Gary
The Advanced Sequencer is much better suited for this. The only way I can think of accomplishing something similar in the Simple Sequencer would be to set the Sequence Mode = Loop, set the Loop counter to some number of iterations, then set your imaging loop to be 5x 20s images, then add another line for 1x 10m image. You will get all the 20s images the way you want BUT you will need to throw away all the 10m images that act as a wait command or "spacer" at the end of each loop iteration. Make sense?
@@PatriotAstro Ok, guess I get that work-around. Is there a way to recycle those 10 minute wasted photons/electrons… sorry, just a long day.
So, Sharpcap would be easier I guess since the simple Capture routine it has allows for delay between groups of 5. I believe Maximdl does as well. I just really like the NINA GUI so I’m trying to centralize everything there except EAA observing.
Does the advanced sequencer really handle this better. I really don’t know it yet but I didn’t see a simple ‘wait’ command unless the wait was for an event like a 10% change in FWHM or some other measurable attribute. Thank you Chad. G
@@gary5051 understood. The Exoplanet plugin uses the advanced sequencer too (in NINA) but also does a lot of extra work for you like: identifying the Target and its (X,Y) as well as Comparisons and Variables complete with (X,Y) as well, it also calculates suitable exposure times based on target ADU measurements.
@@gary5051 A the bottom of the instruction list are some Utility instruction. There are a couple wait commands including once that does exactly what you want (Wait for Time Span).
@@PatriotAstro Great news Chad - thank you !
Since HFR varies by the wavelength of the light, won't skipping "AF After Filter Change" combined with using "AF After HFR Increase" lead to erroneous detection of or failure to detect HFR increases because the trend would be mixing results from different filters? Or does NINA know to keep per-filter trends?
NINA graphs all HFR on a single graph, BUT if you mouse over the graph, you can tell it to only show data by filter. Also, since NINA is in fact tracking HFR by filter, it also looks at the AF triggers in such a way and will force a filter specific AF to occur when needed.
@@PatriotAstro Also, it seems that when I've tried the dither and switch filter as a parallel instruction set, dithering doesn't work because the filter change stops guiding. In fact, it's not clear why you have to explicitly switch the filter back to the starting filter in the loop when the loop itself should cause that to happen, right?
@@neverfox I actually just pinned a comment on this because I ran into the same issue. I spoke to the lead developer about this. They made a change that affects people that are using offsets AND have Disable Guiding on AF set in the profile. (Most commonly SCT users, but some others as well). To have better performance, they also impact Filter Changes as well now if the Disable parameter is configured. This is due to how Offsets move the filter during the filter switch. I will be removing the switch filter instruction from my sequences because of this change (and it didn't save a lot of time anyway.) Getting back to why I was doing the "Switch Filter + Dither" in parallel, it was just to save time on the filter swap, since the dither was already happening and consuming time... I decided to use that time and save on the eventual filter swap time and any assigned settling time as well while the dither occurred... but not any more. :)
What command should I use to go until dawn but not cycle through the smart exposures. I want the various time exposures to be done sequentially.
I may not exactly understand what you are looking for (I have an earlier advanced sequencer video that may help you.) If you choose to loop until dawn, it will cycle through whatever is being looped. I suppose if you just put too much in the loop such that it could never get through it all until dawn, it would just stop there. If I’m not answering the question right, can you clarify a little?
@@PatriotAstro let's say I use a OSC and want to do a bunch of various exposure length to get star colors and the core of M31. I want to do those short exposures as soon as it is dark to get them out if the way and then do only longer exposures for the finer details for the rest of the night. I don't want a loop command but I want nina to stop imaging by nautical dawn. How do I set that time limit?
@@hael8680 To stop at dawn, it’s going to require a loop. In my mind, what you’re trying to accomplish specifically, will be broken into two parts… Short, and then long exposures. The way I would do this for the short exposures at the beginning of the evening, would be to either put a set fixed quantity of short exposures as a single smart exposure command, or create a loop that just loops that exposure over and over and over until maybe your target gets to a specified altitude. This means your short exposures would either be some finite number before moving on or would just take as many as Nina possibly could until the target reached an altitude where you wanted to start doing your long exposures. Then, after that command or looping sequence for the short exposures was completed, I would run another looping sequence that would just loop long exposures repeatedly… With the loop condition being set to time equals either astronomical Dawn or nautical Dawn. Basically the first part of the sequence would be responsible for short exposures until some condition is met whether it is count of images, time, altitude, etc.… Then the second part of the sequence would just take long exposures as long as it possibly could until it ran out of nighttime.