I'm with you 100%! I don't see any point to spending 10-20+ hours on one target. That's really more a function of automation than photography. I like to capture one to three targets on a clear night capturing about 2-3 hours of data on each one. This is essential for me because I'm 100% mobile. I enjoy the wonder of capturing celestial targets and experience of doing so. For me, focusing on just getting more data would ruin the experience. People who have backyard set ups can more or less "set it and forget it." That's said, I enjoy the work of those who spend 20 to 40 hours on one image, so I don't need to do it. Very happy that people do it better than me. It might be interesting for people just to post the image and not even say how many hours they put into it. Ultimately, in this hobby, to each his own. Astro photography is not my only interest in photography.
That's really good to hear, mate! Finding what actually makes you happy and focusing on that is 100% the right thing to do, it's no good chasing 30 hour integrations if they just don't bring you joy! I'm in the same boat 👍👍 Wishing you all the best for 2024!! :-)
Good video. I would like to see an update factoring in light pollution and integration time of the subs. How deep can you go with a light polluted sky, say Bortle 8 vs number of subs and the diminishing return. When should a user stop.
Sky conditions and bortle scale make such a HUGE impact on this. I’ve had 4 hours of Bortle 2 data look better than 12 hours of Bortle 5 both shot near the new moon. Factoring in the moon is super important as well. I really like your commentary on how some of the tools need more integration time to do an effective job. I think the people who continue to focus on getting quality data, and also incorporate the latest tools will start to separate themselves from people who cut corners and rely heavily on the software
Thank you so much my friend!! I'd love to get out to some darker skies and quite literally broaden my horizons (and experience) by trying out data from better skies captured with my own gear 👍👍 Wishing you all the best for 2024 by the way! Thanks for all your support 🙏
@@lukomatico you were the one who encouraged me to try no filter OSC from the backyard after I saw you have success in Bortle 7 I was blown away by your results. I couldn’t believe how well it worked. Thanks for all you do and I’ve learned a lot for your channel and enjoy your content I can’t wait to see how big your channel gets in 2024!
Excellent explanation and you've reinforced my conviction that I will never, never, never shoot two nights on one target! Especially when I'm lucky to get a single full night in. Thanks for making this.
Myths about Astrophotography Integration Time “I have (fill in the blank) hours on this target! I need more hours to make it better.” How many times have we heard this? Well, it’s both right and wrong, but hardly anyone knows exactly why. What Astrophotographers really want is not more integration time, but a better (higher) signal-to-noise ratio for their final images. Stacking subframes is simply averaging the information for each pixel. One myth is that more subframes get you more signal (with information we want) - this is wrong! It simply creates a signal average with less variability because the information (e.g. a star) is always there at some level in that pixel. The power of stacking is in the fact that, unlike the signal, noise is random. Averaging a random set of numbers tends to create a lower value than that of the signals we want to keep. Let’s break it down using some real mathematics and science: 1. Averaging Signal 1. Statistically speaking the average of measurements in a well-behaved process (also known as “in control”) is well-defined after 30 samples and will not likely change much with subsequent measurements unless the process is changed. 2. At this point the area under the sample distribution within 6 sigma (-3 to +3) represents 99.9999998% of all measurements. In other words, you have a chance of 3.4 in a million (!) of capturing another subframe that falls outside of this distribution and significantly changes the average. 3. Ergo the average star/nebula pixel info (the good stuff) is defined after only 30 subframes. 2. Averaging Noise 1. Because the noise is random, the values can vary wildly from maximum to minimum. 2. By averaging a set of replicate measurements the signal-to-noise ratio (SNR) will be increased, ideally in proportion to the square root of the number of measurements. 3. This means that in order to double the SNR (cut the noise in half), one would need 4x the number of measurements. Note that there is no mention of integration time, just the number of subframes. Of course more subframes equate to more integration time, but the actual amount of time completely depends on the subframe exposure time. 3. Achieving Optimal Subframe Exposures 1. The key to success is figuring out the optimal subframe exposure time in order to take advantage of averaging. Simplistically speaking the optimum exposure is one that minimizes blown-out pixels (pegged at maximum). 2. This is a compromise particularly with very bright stars or light pollution. There are great references on this topic and I would dive deeply into it. Get the maximum subframe exposure without destroying detail. A quick check is to see if there are any visible pixels in your subframe without applying a stretch. 4. Putting it together 1. The darker the sky (e.g. NELM 7.1-7.5), the less subframes you need for a good SNR and you can use a higher the subframe exposure. You could get an excellent image even after 30 subframes of for example 600s. The total integration time is 5hrs. 2. As light pollution worsens (e.g. MELM 5-6)you need many more subframes to average out the increased noise and recover the SNR. Your signal average is still the same as that for the dark sky! 3. To double the SNR (which is noticeable to the eye) in a suburban light polluted area you would need at least 120-150 subframes, to double it again would take 480-600 subframes. You would not be able to use a 600s exposure, but might be around 60-120s. The total integration time is now 4-16hrs to get a dark sky SNR, but your signal average hasn’t changed! 4. This is just an example, you could figure it out more exactly using SQM measurements. 5. Longer exposure times effectively act as noise averaging and will produce higher SNR subframes. 5. Myths 1. Taking 1000 10s subframes is just as good as taking 30 300s shots. Sorry - the average signal from 300s subframes is far superior! The 10s shots will have extremely low SNR. 2. I can get more signal if I capture hundreds of hours. Sorry - no. The signal levels will plateau after 30 subframes; however, your SNR will certainly improve and some will perceive this as more signal. 6. Good Practices 1. Find the darkest sky possible or use narrowband/multiband filters. 2. Find the longest possible subframe exposure for your target given your seeing conditions. 3. Use the stacked image SNR to tell you when you’ve reached the point of diminishing returns with respect to number of subframes. In other words, if you have 60 subframes with an acceptable SNR, keep in mind you’d need 240 total (or 180 more) to cut the noise in half, and you may not even notice it. The signal average will be the same regardless, as long as you took at least 30 subframes.
All true, unless you're not using an averaging algorithm to stack... but you are always fighting a battle with SNR, and that's the key point you're making I think. Everyone should be aware that the 2~3 extra nights they are planning on exposing from their back yard would be better spent spending 2.5hrs driving to a dark sky, and exposing for 2 or 3hrs total from bortle 1, using the longest possible sub exposure lengths their mount can, without saturating the sensor in their camera... Thats how you get the best quality images... 1 -DARK skies (even for narrowband, yes, narrowband makes better images from light pollution, but narrowband from dark skies is FAR better than narrowband from the city) 2 - Longer subs
This video just earned you another subscriber. I really appreciate your philosophy, as I'm only doing this hobby for myself and for the fun. I'm not on social media and other than a couple of my close personal friends, no one else will ever see my photos. I'm 62 and really what I want to do is explore and find as many interesting targets to capture just for the shear fun of doing so. I would have personally been just fine with the time spent on the first photo and then move on. I'm sure I will find targets that I want to spend a lot of time on however... This was a great video in helping me to define why I'm even doing this hobby...Thanks much, I will be going through your other videos now that you have captured my attention with this one. ;-) Cheers from the remote high desert in Utah !
Wishing your a happy, healthy, prosperous and as-much-as-possible cloud-free 2024 mate!! As usual, an excellent video tutorial on the Law of Diminishing Returns!! Keep well !!
THESE are the kinda vids that really matter and help - questions we all ask and wonder! Lotsa guys doing GHS and all - but you keep giving us great vids - and THANKYOU once again!!! If you make it to Toronto this summer at all, would love to buy you a steak and some frescas and “talk Astro” for a few hours! (Or Tampa area thru the winter/spring)(i am a Canadian “snowbird”) … thanks Luke!
That's super kind of you my friend, thank you so much!!! Wishing you all the best for 2024, and if ever I find myself in the states I'll bear your kind offer in mind! :-D clear skies and happy new year!!
As you were exclaiming how the difference was so obvious, I was zooming in and struggling to see what you were talking about. I used to give a full night to each target, but with the new RC-Astro tools and Hyperstar find that I can now get a few targets in one night that make me happy once processed. There are so few good dark nights available to a working professional that has to travel to a dark site, that my bar has shifted. The comments about setting your personal goals and doing what makes you happy definitely set us all up for very different capture parameters.
Thanks for the feedback my friend!! - I think much of the observed differences were unfortunately lost to RUclips video compression, my apologies if the comparative segments missed the mark for you! 🙏 I totally agree that with the RC astro tools I can viably get a very nice image in one session with my fast scope, a RASA11, - I do need a bit longer on my RC-TT though for many targets to really get the fainter stuff from the backgrounds! 👍 Clear skies to you!
@dominickzaucha Definitely not ideal, but I would guess fairly common. I intended my comment to be more about balancing life/time priorities than the actual details. The new tools afforded to us these days can shift where we spend time and get value. In my state I couldn't imagine giving up an entire second night to polish a target that already looked fairly good. Perhaps one day I will get to a place where this makes sense... that would be amazing.
@Anatomyofashot oh that went right over my head very sorry! I agree, its quite impressive how much software has improved our hobbies. It really helps narrow the gap down of time/reward. Perhaps one day we can all be blessed with unlimited time, till then this works best.
Where I live, this time of year we only get 2.5-3 hrs of "dark" (although, it's more twilight) and only a few days a month. I aim at 10-12 hrs / target generally.
Great talk my friend!!! As I always do, each night I stack all my “available” data e read the SNR ratio, from that I know where I should stop. Happy new year!!!!!!! 🎉🎉🎉
Excellent video. I think another important corollary is who the intended target is for your photos and how the images are going to be consumed. For example, Instagram? None of those small differences will be discernable. Astrobin? How many of the photos on Astrobin are viewed at 100% by most people? Poster prints? Ect.
Spot on Luke! And very timely for me because tomorrow night looks like it will be the best night for astro in weeks here in the middle of North America. Under my Bortle 6-8 skies, it looks like I will have 3-4 hours of very good skies before the moon rises. And I’m trying to decide if I should add to my two hours of data on the Witch Head Nebula from several weeks ago or go for four hours on the Horsehead Nebula. Thank you again.
Nice comparison Luke. There's no getting around the laws of physics in that integration time is king, but RC's tools do help you bend them a little bit. I think this is more relevant than ever given the weather we've been having (we've just had the 4th big storm of the year up in Scotland and didn't have power for 3 days) as you can create an image that would have been unachievable only a handful of years ago. A case in point is I took 45 minutes of data (using an L-Extreme filter) on the Horse Head nebula a while back as something to do while waiting for my main target to get high enough in the sky. I lost count of how many times I tried to process the data but ended up rage quitting! I have recently tried again using GraXpert, BXT and NXT and now have a pretty good image, it'll never match one taken with more data but I'm more than happy with it given how the raw data looked. The star halos caused by the L-Extreme are still awful but that's a discussion for another day🙂
That's a really interesting way of looking at it mate, and something I've actually almost perfectly mirrored!!! I too took a very short punt at the horsehead and found I could process it actually quite well for such a short integration! As you say quite rightly, it would have been impossible just a few years ago!
As always, great info Luke. The way I look at it, I deliberately bought a colour camera instead of mono because I wanted to maximize my opportunities. Yes, you sacrifice some aspects by not going mono but I can great images from my 2600 MC Pro from my Bortle 2 skies with just one night of imaging (especially in winter when nights are long). RCs tools definitely help a lot too! Integration time is important but I don’t need to squeeze every bit of detail out of an image. For me, just being outside at night under that beautiful sky and knowing I can produce a very cool image of things that are far away is reward enough! Dr B from Manitoba, Canada 🇨🇦
Couldn't agree more mate, beautifully put! That's what it's all about, the whole experience should be rewarding and fulfilling, not just the end result 👍 Clear skies!
I don’t get many astro-nights a year. I’ve come to the conclusion that none of my astro efforts are EVER finished. I just collect photons and keep them. When new software tools come along, I experiment. I keep the data and add to it when I can.
Oh this is well done Mr Luke. I agree with a lot of comments on your bortle class playing a role in diminishing returns. I’m in a 5 and just recently did 24 hours on m33 only to find that my previous 12 hour work looked the same. Thank you for demonstrating this so well. Peace J
Excellent again luke, integration time king ,conditions also come into it ,but most of all....we need some clear skies . I tried the california nebula last month ( last clear night) hoping for 4 hrs got 2 that says it all😂
Well, you've summed that up very well Luke with those four exposure times - always wondered that basic question myself (How much is enough?). Clear Skies in 2024 and looking forward to your videos in the New Year!
Great overview. Have to confess I rarely exceed 6 h on a target unless it’s really faint like the spaghetti nebula in which case it’s 20+ h to do it justice. All the best for the New Year and clear skies.
Great video, Luke! By the way, happy new year, my friend!! The best for you and your significant others for this year that starts. I just shared your video on the Discord server of my astronomy club. We were just discussing about this a few moments ago. Keep this awesome good work, Luke! Cheers!
That's really kind of you Enrique my friend! I hope they enjoyed the video :-D (or at least had a good laugh at me!) Happy new year to you mate, thank you for all your support!! 🙏
I've been working on a long exposure target right now...😂 And I've asked myself the same question recently. As of now I have 14 hours of data on my image... the sky sucks where I live/and have my observatory. However, it's nice to have everything ready to catch a couple of hours here and there.... Great video, and interesting 😊 Happy New Year 🎉
Love the videos Luke. I personally try and spend 10-12 hours on a target. I’d love to get more but clear Skys are like rocking horse poo here in the West Midlands. Plus after the second or third night of shooting I’m to excited to process my image haha. Keep up the good work mate.
You hit it right on the head there mate! Yes eventually you get to that point to where you think most of the noise is gone and nothing else to see then you keeping imaging for 100 hours and end up seeing a huge OIII nebula over Andromeda 😀 I think sometimes it's more about diminishing patience vs returns 🙂 Either way great comparison to drive the point home! Cheers mate!
Hey Luke, thank you for the video, shows very clearly, what intergration time is doing to the data and especially to the amount of residiual noise. Another important aspect is the f ratio i think, many do forget, it plays a huge role in gathering the data, i wish there was a measure for something like overall photons gathered or integration time with f ratio included somehow, to give more of an objective representation of the data gathered. Keep doing, what you are doing Luke, i really enjoy watching your videos and love your perspective on the astrophotography. Cheers and clear skies !
That's a good point mate! 10 hours at f2 is going to look very different than 10 hours at f20 as you quite rightly mention! 👍 Pixel scale is usually a good indicator of overall speed for a given aperture, but it's still just a part of the puzzle hey! Thanks ever so much for watching and giving thoughtful feedback! All the best and happy new year, Luke
Hi, do answer your question. It depends on the aperture 🙂 And then also makes a difference if OSC or mono camera. And many other aspects like eg. Bortle of sky etc. But of caourse you know this 😉 Thank you for a very good comparison. I take for me: BXT and NXT alows us to reduce the exp. time by the half!!! 3h in a cold winter night is enough for me for RGB. And a 2nd night for Luminance (for f/5.5 or faster)
How much time is enough depends a great deal on your sky conditions. For me, in Bortle 5/6 skies with seeing usually between 1.5 and 2 arcseconds, my rule of thumb is to aim for between 15 and 20 hours. More than that, and I don't see any appreciable difference. If I had darker, clearer skies, I'd shoot for longer. People have to experiment to see what works best for them.
Really interesting comparison Luke. For me based on your four comparisons the sweet spot would be the third. I suspect if you're doing 30, 40 50 hours of exposure I doubt each time you're adding you're adding quality to the stack based on different amounts of moonlight, guiding performances, seeing conditions etc. Wishing a very happy, healthy and prosperous new year mate to you and yours 👍
Absolutely true Paul my friend! There's so many variables at play that more isn't automatically better 👍👍 Wishing you all the very best for this new year mate!!!
Very happy new year Luke thanks for the vid I am still very new to this hobby well the photo side of it and it is a question I always ask is how much time to spend on a target and have found this vid very helpful with the weather I have had last year getting an hour on a target has been hard never mind 50 😂 hopefully this year will be clearer thank you
Great video Luke, nicely demonstrated! As you say, not much difference between 66 and 132, but if you’re after that apod, then more is better 😀 Right now I’d be happy with one night of imaging! I Just put out my vid of switching to Nina, my test image was 2.3 hrs! That’s all I could get over two nights!! Frustrating, but optimistic for 2024😀 Happy new year Luke and clear skies! 11:40
Happy new year to you Simon my friend!! Sorry about the late reply, I just want to personally say thank you for all your support through this year mate 🙏 You're appreciated!
Great to see this. I think a very nice add - might be - what SNRs does PI report for each of the four images? Does doubling the time get a 3 dB increase each time (probably not with weighting of images etc)? Is 45 vs 48 dB (or insert a number here) only really getting me that faintest parts to show up? Just a smidge of quantifiable to go with the visual appreciation would get you from an "A" to an "A+", for me. Please consider adding to the comments. Thanks, love your work and Happy New Year!
Thank you so much for the thoughtful comment my friend!! That's very good feedback, - I would offer a consideration that perhaps the visual is the most important feedback, as for example an image binned 4x4 would have a mathematically better snr than an image presented unbinned, but may not actually be aesthetically pleasing to look at! I totally hear what you're saying though, and that's just an extreme 'thought experiment' example 👍👍 Happy new year to you!!
It would be nice if we could get hours & hours on a target lol this season has been shite in the UK lol, still we can hope for the new year. Interesting comparisons though mate so here's wising more clear skies in 2024,
I'm over in Wiltshire and I've literally had just two clear nights since 1st November (with long enough cloudless stretches for 6-8 hours of imaging in one go). The weather has indeed been proper rubbish lately!
Thanks so much for watching mate! It's fun putting comparisons together, I just wish I had some clear skies to gather more data for things like this haha!!
Great topic! Great comparison! Thank you for this and all you’ve given us this past year. One question: I’m wondering if my light polluted skies (Bortle 7-8) would benefit from longer intergration times. Now that Galaxy season is approaching and I switch to LRGB, can light pollution become a deterrent when trying to get integration times of 30 hrs, 40 hrs, or longer? Wishing you and yours a most happy and healthy 2024
Hey Joe!! Happy new year my friend!! :-D I'd always say that if you're not happy with where the image currently is, then double the exposure and see how you feel - there are some sky limitations though for us shooting from very light polluted skies, so managing expectations is also a key skill Clear skies my friend, thanks so much for all you do!
Great video! It would be nice to see the same comparison with exposure duration! Are there differences between images taken with difference exposure times and total exposure time?
I'll do a video on that! :-) I currently have one up on my channel from a while ago about that subject called long vs short exposures or something similar which may be useful in the mean time 👍 Clear skies!
RE: seestar accessories (love yours, btw) - have you thought about making a dew shield or do you think it's useful (more for light block)? I also wonder if they block field of view
Hey thanks mate!! I've given some thought to a dew shield/stray light blocker yeah - I may end up making one available as I have a few prototypes now, watch this space I guess!
great video as always luke, peter zelinka did one on the same subject not so long ago. bortle scale plays a massive part in this, i live in a bortle 7/8 if i did 20 hours on andromeda from home, i would probably only need 3 hours from a bortle 3/4 for the same affect 👍
My problem is at the moment I'm lucky if I get two clear nights within any four week period. Most of the images I have taken this year have not been completed. The ones that have, used a combination of data collected this and the previous year. It hasn't helped that I'm currently in temporary accommodation that only permits imaging of targets that are high in the sky and a couple of hours either side of the meridian.
Thanks for that Luke. You didn't mention the quality of the data also improves the more data you have. It would be an interesting trial to compare the same integration time from bigger data sets. Take 3 hours from 3, 3 hours from 6, 3 hours from 12 etc. Just keep the best data. Another approach would be take a 12 hour or bigger set and use the best 9, 6, 3 hours etc. Assuming not all data is perfect, which of course it isnt. Cheers.
Hey Phil! If I'm not misunderstanding, that's basically what I did with this test - I stacked the best data in all cases, even the largest stack came from an even larger pool of possible data 👍👍 Hope I didn't mistake what you're mentioning my friend!! All the best for the new year!! :-D
@@lukomatico ok, was thinking taking your max integration, stack process, then cut it down, best 90% stack process, repeat till best 10 or 20% then compare the results. Might get more noise but sharper features or will these be ironed out by modern tools? Dylan did this with his Bortle 0 data a few years back and things defo got sharper as the data was pared back to the very sharpest.
If you watch my compilation of 2023 on my channel, you'll probably see that I use far too little integration on my images... The compilation was intentionally made with only data of 2023, and it contained lots of unfinished stacks. I should probably be editing and stacking images instead of watching videos ;-). Having said that, I must say that I usually don't like the effects of too little noise. For some reason a lack of noise makes images look watery (for the lack of a better word). But for dim details, one needs the integration time of multiple nights worth of exposures indeed.
Aye Lad you just answered that nagging question from my deep thoughts .. But one thing I was going to do was set 2 identical setups running on the same target for 1 hr.. one doing 120s exposures, while the other 180s exposures tu see tha nagging thought of equal time vs exposure time.. Ok tell me to shudup now by all means ... but great vid mr Luke 😲
I have wondered if using the PixInsight liveStack noisegraph in parallel to a session, and when the noise curve on the livestack flattens out, then that's enough?
For a lot of folks I think they need to ask themselves how much more 'more' will be gained with integration time. For complex targets more integration yields saturation of the common elements of the target and revelation of the not so common up to the ability of the imaging setup and skill of the individual. If you don't have the skill to bring it out or the time to process days of integration then what value really is it to the amateur? For relatively non-complex targets like the example shown, there isn't much to be discovered 'new' between the 30 and 50 hours that can't be 'found' with reasonable skills in the software.
Very interesting comparison Luke. Certainly a dilemma I face each time I image. I found the comparison between the 66min and the 132min perhaps the most informative. These two examples also interestingly to me seemed to show some quite significant differences in detail which only became obvious after the application of BXT and NXT. Don't get me wrong I am a big fan of Russell Croman and love both plugins, but it seems a little worrying that the deconvolution performed on two integrations of the same target produced with the same OTA and camera can produce different details in the nebulosity. I wonder what you think, but to me in the last example you show, there is more contrast in the dark regions of the 66min stack than the 132min stack. Do you always use the same amount of BXT and NXT (default) no matter what the integration time?
Good point you raise there! I think the extra details are being brought out because the underlying data is more mathematically robust, allowing the deconvolution to have a more accurate result along it's whole application - I haven't seen BXT hallucinate in a long time personally, but it could still be possible in the right conditions I'm sure 👍 Happy new year!
Hi Luke, I’ve become a fan of your videos; this one is very interesting, but I’ve heard many people saying that some objects do not need more time, you will always get the same results once you get to a certain final image, so if that object needs for example 5 hours to get an excellent image, adding 10 more hours will not cause an immense change. What do you think about this? Regards from Chile and happy 2024!
Hey there mate!! Thank you and happy new year to you too!! :-D I guess some targets are so photon rich that extra exposure becomes less meaningful, that could certainly be the case for something super bright in a tight field of view, like a photo of just the core of Orion for example? Good thought!
What did you take them with (i.e., focal ratio)? 8" RASA? The question remains: how much of a difference in the reduction of noise comes not so much from the total integration time but the NUMBER of sub frames (as I understand it, whatever your exposure lengths, until you get to about 150 sub frames the gains are considerable, but really taper off after about 200 sub frames, is that fair to say? So in your case, it was predictable that the gains in noise reduction, given the low number of subframes, would be considerable, obvious and noticeable).
That's a great question mate and something I should probably tackle in another video when I can gather the data for it! This data was taken with a 10" f8 telescope :-) Happy new year!!
So how do we know where to start and stop? Just about to be ready to go with my set-up. Is there a “starting point” for choosing exposure time and total time? I’m using a SW Esprit100ED, AM5 mount, AsiAir, 2600MC camera. Coming from the SeeStar. I only had 10, 20 and 30 second options. I’d usually get 1-2 hours worth.
I understand that visual improvement comes in stops. Like you will see improvement like this 1hr > 2hr > 4hr > 8hr > 16hr > 32hr. Now I work with mono data mostly now and I know it's different from OSC, but do those stops apply for each channel or total combined integration?
It generally depends on filters used, sky conditions at the time and your cameras sensitivity to the wavelengths you're shooting, so in a way - it's both! Total time is important of course, but so is balance to a degree, it'd be no good having 50 hours each of RGB then only taking an hour of L, - an extreme example, but you get the idea 👍 Happy new year!
I think that with the maximum shot does it need as much of a noisexterminator as the rest?, If anything it looks too smooth as it was pretty noise free as it was
That's very fair mate yeah, true!! I would usually use less NR, I just thought to keep the playing field slightly leveled by keeping the settings the same though 👍👍
I'm with you 100%! I don't see any point to spending 10-20+ hours on one target. That's really more a function of automation than photography. I like to capture one to three targets on a clear night capturing about 2-3 hours of data on each one. This is essential for me because I'm 100% mobile. I enjoy the wonder of capturing celestial targets and experience of doing so. For me, focusing on just getting more data would ruin the experience.
People who have backyard set ups can more or less "set it and forget it." That's said, I enjoy the work of those who spend 20 to 40 hours on one image, so I don't need to do it. Very happy that people do it better than me.
It might be interesting for people just to post the image and not even say how many hours they put into it. Ultimately, in this hobby, to each his own.
Astro photography is not my only interest in photography.
That's really good to hear, mate! Finding what actually makes you happy and focusing on that is 100% the right thing to do, it's no good chasing 30 hour integrations if they just don't bring you joy! I'm in the same boat 👍👍
Wishing you all the best for 2024!! :-)
Good video. I would like to see an update factoring in light pollution and integration time of the subs. How deep can you go with a light polluted sky, say Bortle 8 vs number of subs and the diminishing return. When should a user stop.
Great idea mate!! Thank you so much 👍👍
Sky conditions and bortle scale make such a HUGE impact on this. I’ve had 4 hours of Bortle 2 data look better than 12 hours of Bortle 5 both shot near the new moon. Factoring in the moon is super important as well.
I really like your commentary on how some of the tools need more integration time to do an effective job.
I think the people who continue to focus on getting quality data, and also incorporate the latest tools will start to separate themselves from people who cut corners and rely heavily on the software
Thank you so much my friend!! I'd love to get out to some darker skies and quite literally broaden my horizons (and experience) by trying out data from better skies captured with my own gear 👍👍
Wishing you all the best for 2024 by the way! Thanks for all your support 🙏
@@lukomatico you were the one who encouraged me to try no filter OSC from the backyard after I saw you have success in Bortle 7 I was blown away by your results. I couldn’t believe how well it worked. Thanks for all you do and I’ve learned a lot for your channel and enjoy your content I can’t wait to see how big your channel gets in 2024!
@@darkrangersinc Ah mate that's really lovely to hear!! Thank you!! :-D
I can't wait to get back at it and make some fresh captures this year 🙏
Excellent explanation and you've reinforced my conviction that I will never, never, never shoot two nights on one target! Especially when I'm lucky to get a single full night in. Thanks for making this.
My pleasure mate, thanks so much for watching and taking the time to share your thoughts! 👍👍
Clear skies!
Myths about Astrophotography Integration Time
“I have (fill in the blank) hours on this target! I need more hours to make it better.”
How many times have we heard this? Well, it’s both right and wrong, but hardly anyone knows exactly why.
What Astrophotographers really want is not more integration time, but a better (higher) signal-to-noise ratio for their final images.
Stacking subframes is simply averaging the information for each pixel. One myth is that more subframes get you more signal (with information we want) - this is wrong! It simply creates a signal average with less variability because the information (e.g. a star) is always there at some level in that pixel. The power of stacking is in the fact that, unlike the signal, noise is random. Averaging a random set of numbers tends to create a lower value than that of the signals we want to keep.
Let’s break it down using some real mathematics and science:
1. Averaging Signal
1. Statistically speaking the average of measurements in a well-behaved process (also known as “in control”) is well-defined after 30 samples and will not likely change much with subsequent measurements unless the process is changed.
2. At this point the area under the sample distribution within 6 sigma (-3 to +3) represents 99.9999998% of all measurements. In other words, you have a chance of 3.4 in a million (!) of capturing another subframe that falls outside of this distribution and significantly changes the average.
3. Ergo the average star/nebula pixel info (the good stuff) is defined after only 30 subframes.
2. Averaging Noise
1. Because the noise is random, the values can vary wildly from maximum to minimum.
2. By averaging a set of replicate measurements the signal-to-noise ratio (SNR) will be increased, ideally in proportion to the square root of the number of measurements.
3. This means that in order to double the SNR (cut the noise in half), one would need 4x the number of measurements. Note that there is no mention of integration time, just the number of subframes. Of course more subframes equate to more integration time, but the actual amount of time completely depends on the subframe exposure time.
3. Achieving Optimal Subframe Exposures
1. The key to success is figuring out the optimal subframe exposure time in order to take advantage of averaging. Simplistically speaking the optimum exposure is one that minimizes blown-out pixels (pegged at maximum).
2. This is a compromise particularly with very bright stars or light pollution. There are great references on this topic and I would dive deeply into it. Get the maximum subframe exposure without destroying detail. A quick check is to see if there are any visible pixels in your subframe without applying a stretch.
4. Putting it together
1. The darker the sky (e.g. NELM 7.1-7.5), the less subframes you need for a good SNR and you can use a higher the subframe exposure. You could get an excellent image even after 30 subframes of for example 600s. The total integration time is 5hrs.
2. As light pollution worsens (e.g. MELM 5-6)you need many more subframes to average out the increased noise and recover the SNR. Your signal average is still the same as that for the dark sky!
3. To double the SNR (which is noticeable to the eye) in a suburban light polluted area you would need at least 120-150 subframes, to double it again would take 480-600 subframes. You would not be able to use a 600s exposure, but might be around 60-120s. The total integration time is now 4-16hrs to get a dark sky SNR, but your signal average hasn’t changed!
4. This is just an example, you could figure it out more exactly using SQM measurements.
5. Longer exposure times effectively act as noise averaging and will produce higher SNR subframes.
5. Myths
1. Taking 1000 10s subframes is just as good as taking 30 300s shots. Sorry - the average signal from 300s subframes is far superior! The 10s shots will have extremely low SNR.
2. I can get more signal if I capture hundreds of hours. Sorry - no. The signal levels will plateau after 30 subframes; however, your SNR will certainly improve and some will perceive this as more signal.
6. Good Practices
1. Find the darkest sky possible or use narrowband/multiband filters.
2. Find the longest possible subframe exposure for your target given your seeing conditions.
3. Use the stacked image SNR to tell you when you’ve reached the point of diminishing returns with respect to number of subframes. In other words, if you have 60 subframes with an acceptable SNR, keep in mind you’d need 240 total (or 180 more) to cut the noise in half, and you may not even notice it. The signal average will be the same regardless, as long as you took at least 30 subframes.
All true, unless you're not using an averaging algorithm to stack... but you are always fighting a battle with SNR, and that's the key point you're making I think.
Everyone should be aware that the 2~3 extra nights they are planning on exposing from their back yard would be better spent spending 2.5hrs driving to a dark sky, and exposing for 2 or 3hrs total from bortle 1, using the longest possible sub exposure lengths their mount can, without saturating the sensor in their camera... Thats how you get the best quality images...
1 -DARK skies (even for narrowband, yes, narrowband makes better images from light pollution, but narrowband from dark skies is FAR better than narrowband from the city)
2 - Longer subs
Very helpful, thank you Luke, happy 2024 and may you have clear skies forever and a day
Happy new year my friend, thank you so much!! 🙏
This video just earned you another subscriber. I really appreciate your philosophy, as I'm only doing this hobby for myself and for the fun. I'm not on social media and other than a couple of my close personal friends, no one else will ever see my photos. I'm 62 and really what I want to do is explore and find as many interesting targets to capture just for the shear fun of doing so.
I would have personally been just fine with the time spent on the first photo and then move on. I'm sure I will find targets that I want to spend a lot of time on however... This was a great video in helping me to define why I'm even doing this hobby...Thanks much, I will be going through your other videos now that you have captured my attention with this one. ;-) Cheers from the remote high desert in Utah !
That's absolutely wonderful to hear my friend, thank you ever so much for your kind comment and support, - it means a lot!
Clear skies to you! :-)
Thanks for this great comparison! Makes exposure times and modern processing literally very clear!
Glad you enjoyed it! Thanks!
Wishing your a happy, healthy, prosperous and as-much-as-possible cloud-free 2024 mate!!
As usual, an excellent video tutorial on the Law of Diminishing Returns!!
Keep well !!
Happy new year to you my friend, thank you for your support and kind wishes!! 🙏
THESE are the kinda vids that really matter and help - questions we all ask and wonder! Lotsa guys doing GHS and all - but you keep giving us great vids - and THANKYOU once again!!! If you make it to Toronto this summer at all, would love to buy you a steak and some frescas and “talk Astro” for a few hours! (Or Tampa area thru the winter/spring)(i am a Canadian “snowbird”) … thanks Luke!
That's super kind of you my friend, thank you so much!!!
Wishing you all the best for 2024, and if ever I find myself in the states I'll bear your kind offer in mind! :-D clear skies and happy new year!!
these are the kind of the videos that make sense. Point out how OCD people are in this hobby and inform them there is a point of diminishing returns.
Really enjoyed your interview video, keep them coming Luke.
Glad you enjoyed it!! Clear skies my friend :-)
As you were exclaiming how the difference was so obvious, I was zooming in and struggling to see what you were talking about. I used to give a full night to each target, but with the new RC-Astro tools and Hyperstar find that I can now get a few targets in one night that make me happy once processed. There are so few good dark nights available to a working professional that has to travel to a dark site, that my bar has shifted. The comments about setting your personal goals and doing what makes you happy definitely set us all up for very different capture parameters.
Thanks for the feedback my friend!! - I think much of the observed differences were unfortunately lost to RUclips video compression, my apologies if the comparative segments missed the mark for you! 🙏
I totally agree that with the RC astro tools I can viably get a very nice image in one session with my fast scope, a RASA11, - I do need a bit longer on my RC-TT though for many targets to really get the fainter stuff from the backgrounds! 👍
Clear skies to you!
@lukomatico It seems on a 14' laptop screen running at 4k the differences are obvious. Perhaps watching it in a cellphone or tablet is not ideal?
@dominickzaucha Definitely not ideal, but I would guess fairly common. I intended my comment to be more about balancing life/time priorities than the actual details. The new tools afforded to us these days can shift where we spend time and get value. In my state I couldn't imagine giving up an entire second night to polish a target that already looked fairly good. Perhaps one day I will get to a place where this makes sense... that would be amazing.
@Anatomyofashot oh that went right over my head very sorry! I agree, its quite impressive how much software has improved our hobbies. It really helps narrow the gap down of time/reward.
Perhaps one day we can all be blessed with unlimited time, till then this works best.
Where I live, this time of year we only get 2.5-3 hrs of "dark" (although, it's more twilight) and only a few days a month. I aim at 10-12 hrs / target generally.
Great talk my friend!!! As I always do, each night I stack all my “available” data e read the SNR ratio, from that I know where I should stop. Happy new year!!!!!!! 🎉🎉🎉
Good stuff! That's a great way to do it mate 👍👍 clear skies to you!
Excellent video. I think another important corollary is who the intended target is for your photos and how the images are going to be consumed. For example, Instagram? None of those small differences will be discernable. Astrobin? How many of the photos on Astrobin are viewed at 100% by most people? Poster prints? Ect.
True mate!! Thanks for offering your thoughts 👍
Spot on Luke! And very timely for me because tomorrow night looks like it will be the best night for astro in weeks here in the middle of North America. Under my Bortle 6-8 skies, it looks like I will have 3-4 hours of very good skies before the moon rises. And I’m trying to decide if I should add to my two hours of data on the Witch Head Nebula from several weeks ago or go for four hours on the Horsehead Nebula. Thank you again.
Hope you had a fun session my friend!!!
All the best and happy new year :-D
So in the end...... the more the better, am glad about this as it means I can continue improving my shots till I pass my rig onto my grandkids.
Multi-generational astrophotography, now there's an idea!! Heirloom stacks? :-D Clear skies mate!
Nice comparison Luke. There's no getting around the laws of physics in that integration time is king, but RC's tools do help you bend them a little bit. I think this is more relevant than ever given the weather we've been having (we've just had the 4th big storm of the year up in Scotland and didn't have power for 3 days) as you can create an image that would have been unachievable only a handful of years ago. A case in point is I took 45 minutes of data (using an L-Extreme filter) on the Horse Head nebula a while back as something to do while waiting for my main target to get high enough in the sky. I lost count of how many times I tried to process the data but ended up rage quitting! I have recently tried again using GraXpert, BXT and NXT and now have a pretty good image, it'll never match one taken with more data but I'm more than happy with it given how the raw data looked. The star halos caused by the L-Extreme are still awful but that's a discussion for another day🙂
That's a really interesting way of looking at it mate, and something I've actually almost perfectly mirrored!!! I too took a very short punt at the horsehead and found I could process it actually quite well for such a short integration! As you say quite rightly, it would have been impossible just a few years ago!
As always, great info Luke. The way I look at it, I deliberately bought a colour camera instead of mono because I wanted to maximize my opportunities. Yes, you sacrifice some aspects by not going mono but I can great images from my 2600 MC Pro from my Bortle 2 skies with just one night of imaging (especially in winter when nights are long). RCs tools definitely help a lot too! Integration time is important but I don’t need to squeeze every bit of detail out of an image. For me, just being outside at night under that beautiful sky and knowing I can produce a very cool image of things that are far away is reward enough! Dr B from Manitoba, Canada 🇨🇦
Couldn't agree more mate, beautifully put! That's what it's all about, the whole experience should be rewarding and fulfilling, not just the end result 👍
Clear skies!
I don’t get many astro-nights a year. I’ve come to the conclusion that none of my astro efforts are EVER finished. I just collect photons and keep them. When new software tools come along, I experiment. I keep the data and add to it when I can.
That's a great way to do it mate!! Excellent 👍👍
Happy new year to you, Luke and family! 🎉🎉✨✨👍
Happy new year my friend!! :-D
Thanks for taking the time to do this, Luke.
My pleasure! Thank you for watching, my friend :-)
Oh this is well done Mr Luke. I agree with a lot of comments on your bortle class playing a role in diminishing returns. I’m in a 5 and just recently did 24 hours on m33 only to find that my previous 12 hour work looked the same. Thank you for demonstrating this so well.
Peace
J
Thanks so much my friend! :-D Glad you enjoyed, peace!
Excellent again luke, integration time king ,conditions also come into it ,but most of all....we need some clear skies . I tried the california nebula last month ( last clear night) hoping for 4 hrs got 2 that says it all😂
It's been a tough year for astro hey mate!! So much cloud, ugh!
Happy new year by the way!
Well, you've summed that up very well Luke with those four exposure times - always wondered that basic question myself (How much is enough?). Clear Skies in 2024 and looking forward to your videos in the New Year!
Glad it was helpful! Thanks so much for watching and happy new year my friend 👍👍
Great overview. Have to confess I rarely exceed 6 h on a target unless it’s really faint like the spaghetti nebula in which case it’s 20+ h to do it justice. All the best for the New Year and clear skies.
Happy new year to you my friend!!! Thanks for sharing, I'm largely the same haha!
Great video, Luke! By the way, happy new year, my friend!! The best for you and your significant others for this year that starts. I just shared your video on the Discord server of my astronomy club. We were just discussing about this a few moments ago. Keep this awesome good work, Luke! Cheers!
That's really kind of you Enrique my friend! I hope they enjoyed the video :-D (or at least had a good laugh at me!)
Happy new year to you mate, thank you for all your support!! 🙏
I've been working on a long exposure target right now...😂 And I've asked myself the same question recently. As of now I have 14 hours of data on my image... the sky sucks where I live/and have my observatory. However, it's nice to have everything ready to catch a couple of hours here and there.... Great video, and interesting 😊
Happy New Year 🎉
Happy new year my friend!!! Hopefully 2024 brings us clearer skies if nothing else haha, what a pain it's been! 👍
Great video, this is something i agonize over every clear night as well, which isn't often in cloudy Oxfordshire
Glad you enjoyed it!! Thanks for watching, my friend! :-)
Love the videos Luke. I personally try and spend 10-12 hours on a target.
I’d love to get more but clear Skys are like rocking horse poo here in the West Midlands. Plus after the second or third night of shooting I’m to excited to process my image haha.
Keep up the good work mate.
Thanks so much mate!! Clear skies to you 👍👍
You hit it right on the head there mate! Yes eventually you get to that point to where you think most of the noise is gone and nothing else to see then you keeping imaging for 100 hours and end up seeing a huge OIII nebula over Andromeda 😀 I think sometimes it's more about diminishing patience vs returns 🙂 Either way great comparison to drive the point home! Cheers mate!
Thanks so much Dave!! I guess there's always more gold if you dig deep enough in cases like this haha! :-D
Clear skies mate!
Hey Luke, thank you for the video, shows very clearly, what intergration time is doing to the data and especially to the amount of residiual noise. Another important aspect is the f ratio i think, many do forget, it plays a huge role in gathering the data, i wish there was a measure for something like overall photons gathered or integration time with f ratio included somehow, to give more of an objective representation of the data gathered. Keep doing, what you are doing Luke, i really enjoy watching your videos and love your perspective on the astrophotography. Cheers and clear skies !
That's a good point mate! 10 hours at f2 is going to look very different than 10 hours at f20 as you quite rightly mention! 👍
Pixel scale is usually a good indicator of overall speed for a given aperture, but it's still just a part of the puzzle hey!
Thanks ever so much for watching and giving thoughtful feedback!
All the best and happy new year,
Luke
Hi, do answer your question. It depends on the aperture 🙂 And then also makes a difference if OSC or mono camera. And many other aspects like eg. Bortle of sky etc. But of caourse you know this 😉
Thank you for a very good comparison. I take for me: BXT and NXT alows us to reduce the exp. time by the half!!! 3h in a cold winter night is enough for me for RGB. And a 2nd night for Luminance (for f/5.5 or faster)
Thanks for sharing! Happy new year my friend 👍👍
Thank you for another great video! Happy New Year!
Happy new year to you my friend!! :-D
Excellent comparison. Really important question. Thanks for not putting insufferable background spa music too :)
Haha!! :-D Thanks ever so much my friend!! Glad you enjoyed, happy new year!
How much time is enough depends a great deal on your sky conditions. For me, in Bortle 5/6 skies with seeing usually between 1.5 and 2 arcseconds, my rule of thumb is to aim for between 15 and 20 hours. More than that, and I don't see any appreciable difference. If I had darker, clearer skies, I'd shoot for longer. People have to experiment to see what works best for them.
Always best to experiment for yourself I totally agree!! 👍
Hello Luke. Another interesting video, as always. A big hug and a Happy New Year 2024 I wish you from Spain.🎆🎆🎆
Happy new year to you my friend, big hugs right back!! Hoping 2024 brings you and your family every happiness 🙏
Really interesting comparison Luke. For me based on your four comparisons the sweet spot would be the third. I suspect if you're doing 30, 40 50 hours of exposure I doubt each time you're adding you're adding quality to the stack based on different amounts of moonlight, guiding performances, seeing conditions etc. Wishing a very happy, healthy and prosperous new year mate to you and yours 👍
Absolutely true Paul my friend! There's so many variables at play that more isn't automatically better 👍👍
Wishing you all the very best for this new year mate!!!
Very happy new year Luke thanks for the vid I am still very new to this hobby well the photo side of it and it is a question I always ask is how much time to spend on a target and have found this vid very helpful with the weather I have had last year getting an hour on a target has been hard never mind 50 😂 hopefully this year will be clearer thank you
Happy new year to you too my friend, wishing you all the very best for 2024!! :-D
Great video Luke, nicely demonstrated! As you say, not much difference between 66 and 132, but if you’re after that apod, then more is better 😀 Right now I’d be happy with one night of imaging! I Just put out my vid of switching to Nina, my test image was 2.3 hrs! That’s all I could get over two nights!! Frustrating, but optimistic for 2024😀 Happy new year Luke and clear skies! 11:40
Happy new year to you Simon my friend!! Sorry about the late reply, I just want to personally say thank you for all your support through this year mate 🙏 You're appreciated!
Great to see this. I think a very nice add - might be - what SNRs does PI report for each of the four images? Does doubling the time get a 3 dB increase each time (probably not with weighting of images etc)? Is 45 vs 48 dB (or insert a number here) only really getting me that faintest parts to show up? Just a smidge of quantifiable to go with the visual appreciation would get you from an "A" to an "A+", for me. Please consider adding to the comments. Thanks, love your work and Happy New Year!
Thank you so much for the thoughtful comment my friend!! That's very good feedback, - I would offer a consideration that perhaps the visual is the most important feedback, as for example an image binned 4x4 would have a mathematically better snr than an image presented unbinned, but may not actually be aesthetically pleasing to look at! I totally hear what you're saying though, and that's just an extreme 'thought experiment' example 👍👍
Happy new year to you!!
It would be nice if we could get hours & hours on a target lol this season has been shite in the UK lol, still we can hope for the new year. Interesting comparisons though mate so here's wising more clear skies in 2024,
I'm over in Wiltshire and I've literally had just two clear nights since 1st November (with long enough cloudless stretches for 6-8 hours of imaging in one go). The weather has indeed been proper rubbish lately!
Thanks so much for watching mate! It's fun putting comparisons together, I just wish I had some clear skies to gather more data for things like this haha!!
Hi Luke another excellent video mate.
Glad you enjoyed it my friend!!
Great topic! Great comparison! Thank you for this and all you’ve given us this past year. One question: I’m wondering if my light polluted skies (Bortle 7-8) would benefit from longer intergration times. Now that Galaxy season is approaching and I switch to LRGB, can light pollution become a deterrent when trying to get integration times of 30 hrs, 40 hrs, or longer? Wishing you and yours a most happy and healthy 2024
Hey Joe!!
Happy new year my friend!! :-D
I'd always say that if you're not happy with where the image currently is, then double the exposure and see how you feel - there are some sky limitations though for us shooting from very light polluted skies, so managing expectations is also a key skill
Clear skies my friend, thanks so much for all you do!
With the weather we’re having in Sydney atm, I’m lucky to get an hour of integration time per month.
I feel that mate, hope it's better for you soon!
Happy new year!
Great video! It would be nice to see the same comparison with exposure duration! Are there differences between images taken with difference exposure times and total exposure time?
I'll do a video on that! :-) I currently have one up on my channel from a while ago about that subject called long vs short exposures or something similar which may be useful in the mean time 👍
Clear skies!
RE: seestar accessories (love yours, btw) - have you thought about making a dew shield or do you think it's useful (more for light block)? I also wonder if they block field of view
Hey thanks mate!!
I've given some thought to a dew shield/stray light blocker yeah - I may end up making one available as I have a few prototypes now, watch this space I guess!
@@lukomatico thanks!
great video as always luke, peter zelinka did one on the same subject not so long ago. bortle scale plays a massive part in this, i live in a bortle 7/8 if i did 20 hours on andromeda from home, i would probably only need 3 hours from a bortle 3/4 for the same affect 👍
Interesting! I'll have to check it out mate, cheers!!
My problem is at the moment I'm lucky if I get two clear nights within any four week period. Most of the images I have taken this year have not been completed. The ones that have, used a combination of data collected this and the previous year. It hasn't helped that I'm currently in temporary accommodation that only permits imaging of targets that are high in the sky and a couple of hours either side of the meridian.
Thanks for that Luke. You didn't mention the quality of the data also improves the more data you have. It would be an interesting trial to compare the same integration time from bigger data sets. Take 3 hours from 3, 3 hours from 6, 3 hours from 12 etc. Just keep the best data. Another approach would be take a 12 hour or bigger set and use the best 9, 6, 3 hours etc. Assuming not all data is perfect, which of course it isnt. Cheers.
Hey Phil! If I'm not misunderstanding, that's basically what I did with this test - I stacked the best data in all cases, even the largest stack came from an even larger pool of possible data 👍👍
Hope I didn't mistake what you're mentioning my friend!!
All the best for the new year!! :-D
@@lukomatico ok, was thinking taking your max integration, stack process, then cut it down, best 90% stack process, repeat till best 10 or 20% then compare the results. Might get more noise but sharper features or will these be ironed out by modern tools? Dylan did this with his Bortle 0 data a few years back and things defo got sharper as the data was pared back to the very sharpest.
Luke you must be reading my mind, I did an almost identical comparison yesterday lol.
Thanks so much for watching mate :-D
If you watch my compilation of 2023 on my channel, you'll probably see that I use far too little integration on my images... The compilation was intentionally made with only data of 2023, and it contained lots of unfinished stacks. I should probably be editing and stacking images instead of watching videos ;-). Having said that, I must say that I usually don't like the effects of too little noise. For some reason a lack of noise makes images look watery (for the lack of a better word). But for dim details, one needs the integration time of multiple nights worth of exposures indeed.
I hear you Martin! Too little noise can look a bit 'painted' or something!
Happy new year to you my friend!! 👍👍
I'm new to this hobby and based in the UK and one thing I can't understand is how you actually do it when there are so few clear nights
I ask myself the same question bud! 😂 This last year of weather has really seemed exceptional in terms of cloudy nights.
Cheers!
The answer to 'how much integration time' is surprisingly simple: twice as much as you have now.
Love it, haha! :-D what if I currently have zero though due to all these cloudy skies?!
Aye Lad you just answered that nagging question from my deep thoughts .. But one thing I was going to do was set 2 identical setups running on the same target for 1 hr.. one doing 120s exposures, while the other 180s exposures tu see tha nagging thought of equal time vs exposure time..
Ok tell me to shudup now by all means ... but great vid mr Luke
😲
Happy new year to you my friend, you've given so much support and help this year Chris mate, I appreciate your friendship!!
All the very best :-D
I have wondered if using the PixInsight liveStack noisegraph in parallel to a session, and when the noise curve on the livestack flattens out, then that's enough?
That's a really good idea Dave!! Should be a good way to do it 👍👍
What is the "anti-halo pro"? Very informative video!
It's a 3.5nm Dual-Narrowband Ha/OIII filter by Player One Astronomy :-) Cheers!
For a lot of folks I think they need to ask themselves how much more 'more' will be gained with integration time. For complex targets more integration yields saturation of the common elements of the target and revelation of the not so common up to the ability of the imaging setup and skill of the individual. If you don't have the skill to bring it out or the time to process days of integration then what value really is it to the amateur?
For relatively non-complex targets like the example shown, there isn't much to be discovered 'new' between the 30 and 50 hours that can't be 'found' with reasonable skills in the software.
True mate! Very fair points 👍👍
Thanks so much for watching!
Very interesting comparison Luke. Certainly a dilemma I face each time I image. I found the comparison between the 66min and the 132min perhaps the most informative.
These two examples also interestingly to me seemed to show some quite significant differences in detail which only became obvious after the application of BXT and NXT. Don't get me wrong I am a big fan of Russell Croman and love both plugins, but it seems a little worrying that the deconvolution performed on two integrations of the same target produced with the same OTA and camera can produce different details in the nebulosity. I wonder what you think, but to me in the last example you show, there is more contrast in the dark regions of the 66min stack than the 132min stack. Do you always use the same amount of BXT and NXT (default) no matter what the integration time?
Good point you raise there! I think the extra details are being brought out because the underlying data is more mathematically robust, allowing the deconvolution to have a more accurate result along it's whole application - I haven't seen BXT hallucinate in a long time personally, but it could still be possible in the right conditions I'm sure 👍
Happy new year!
Hi Luke, I’ve become a fan of your videos; this one is very interesting, but I’ve heard many people saying that some objects do not need more time, you will always get the same results once you get to a certain final image, so if that object needs for example 5 hours to get an excellent image, adding 10 more hours will not cause an immense change. What do you think about this? Regards from Chile and happy 2024!
Hey there mate!! Thank you and happy new year to you too!! :-D
I guess some targets are so photon rich that extra exposure becomes less meaningful, that could certainly be the case for something super bright in a tight field of view, like a photo of just the core of Orion for example?
Good thought!
What did you take them with (i.e., focal ratio)? 8" RASA? The question remains: how much of a difference in the reduction of noise comes not so much from the total integration time but the NUMBER of sub frames (as I understand it, whatever your exposure lengths, until you get to about 150 sub frames the gains are considerable, but really taper off after about 200 sub frames, is that fair to say? So in your case, it was predictable that the gains in noise reduction, given the low number of subframes, would be considerable, obvious and noticeable).
That's a great question mate and something I should probably tackle in another video when I can gather the data for it!
This data was taken with a 10" f8 telescope :-)
Happy new year!!
So how do we know where to start and stop? Just about to be ready to go with my set-up. Is there a “starting point” for choosing exposure time and total time? I’m using a SW Esprit100ED, AM5 mount, AsiAir, 2600MC camera. Coming from the SeeStar. I only had 10, 20 and 30 second options. I’d usually get 1-2 hours worth.
I'd usually start with around 90s exposures for broadband, and around 180s for narrowband as a very rough starting point 👍
Thanks!
That's really kind of you, thank you ever so much my friend!! 🙏 All the best and happy new year to you!! :-)
I understand that visual improvement comes in stops. Like you will see improvement like this 1hr > 2hr > 4hr > 8hr > 16hr > 32hr. Now I work with mono data mostly now and I know it's different from OSC, but do those stops apply for each channel or total combined integration?
It generally depends on filters used, sky conditions at the time and your cameras sensitivity to the wavelengths you're shooting, so in a way - it's both!
Total time is important of course, but so is balance to a degree, it'd be no good having 50 hours each of RGB then only taking an hour of L, - an extreme example, but you get the idea 👍
Happy new year!
Integration time is really matter for faint and LRGB targets. Then, these targets could be captured from dark locations.
I sue Sharp Cap and get bored easily, so I do not spend much time per target.
Absolutely perfect approach mate! I love sharpcap imaging too :-) Clear skies!!
good video
Glad you enjoyed mate!
I'd love to get 10-20 hours on a target... but by the time I get a clear night my previous target has always moved out of view!
Oh mate I feel that haha, been there!!! - Hope it's a better 2024, Happy new year!
Quoting the Monty Python supercomputer, the answer is 42 :).
Perfect, haha!
I think that with the maximum shot does it need as much of a noisexterminator as the rest?, If anything it looks too smooth as it was pretty noise free as it was
That's very fair mate yeah, true!! I would usually use less NR, I just thought to keep the playing field slightly leveled by keeping the settings the same though 👍👍
@@lukomatico all good makes sense :)