You realize that you learned astrophotography in 7 months when you watch this video and notice that you know all the concepts Cuiv is talking about! Thanks Cuiv, you have been a reliable source of information through my journey and one of the reasons that I started this hobby 7 months ago, as I learned from you that you could do astrophotography from a Bortle 8-9 area like where I live in Orlando FL
Great technical video! This is why I use 64 exposures for each of my calibration frames (Bias, Darks, Flats) because noise is an inverse square function. The Square root of 64 is 8, and 1/8 (the inverse of the square root) is 0.125 or 12.5% which means that 64 calibration frames leaves 12.5% of the noise in the image. To get down to only 10% I'd need 100 calibration frames, and to get down to 1% I'd need 10,000 calibration frames. So 64 for me feels like a good break point in the diminishing returns.
64 frames for flats and biases are okay but 64 for darks are too time consuming if your lights are 2-3 or 5 minutes subs. I hate dark frames the most 😂
@@ssbhide123 This is why having a cooled camera is so helpful, I'm able to build a Dark library at various exposures and at the same temperature that I capture my lights. I haven't shot new darks in months, and when I do I can just do them during the day.
Thanks for the video and especially the link to Vincent's presentation. I believe it is the first time ever I have seen an explanation of the _fundamental_ reason WHY the photons we measure follow a Poisson distribution rather than being given the usual "what we observe fits the criteria for a Poisson distribution, so that's what it is". Pages 42, 43 and 44 are very, very hard to digest fully though. The probabilistic treatment of guiding errors to constrain the optimal exposure parameter is also interesting. I've always felt that Glover's presentation about the "optimal" exposure should be rebadged as "the minimal exposure you can get away with given bad things are likely to happen" - Laurent's optimization strategy is much more comprehensive. Definitely the best out there.
Thanks Cuiv! That is an excellent summary of the challenges involved in reducing noise. Though it was quick, I have enough engineering background and prior knowledge of the subject that I learned something new: How longer exposures overwhelm the effects of read noise - I had not thought of it before. Your quick explanation was fantastic!
Look at the stars... Yeah. It's been a solid cloud cover here since mid December. Forecast for next 10 days, clouds, rain, snow... Daughter gave me a new telescope for Christmas (it was late, didn't get in until last week). I was able to look at the stars for 30 minutes Tuesday. But miracle of all miracles, in that brief 30 minute window, I was able to look at Jupiter, then I had to change to Sirius when clouds covered Jupiter. While gazing at Sirius, a meteor streaked across my view on the telescope! i hadn't seen a meteor with the naked eye in years, and I get one through the telescope! That made up for the short window, needless to say 😍
Excellent work by Devineau! The two main slides are 34 and 38 (from a practical point of view). The other 47 explain why, in a streamlined and logical way.
Great review of important material, Cuiv! Thanks so much for the resource from Laurent as well. A bit complex for me, so I'll have to go through it a couple of times. In case any of your subscribers are interested, I've been reading Charles Bracken's "The Deep-Sky Imaging Primer" (3rd ed.) and the early part of the book describes many of the same concepts but with a bit less mathematics for those of us non-engineers. Thanks again!
Last time I was concerned with Poisson distrubution was when I was estimating the abundance of minke whales in the north-east Atlantic! 😄 Only Japan, Norway and Iceland hunting minke whales. Think I'll pull out some dusty books from my bookshelf! Thanks Cuiv! 😊👍🏻
Thank you for sharing your knowledge Cuiv! I am moving from DSLR to a dedicated astrophotography camera shooting in Bortle 8 with an 8" SCT. I think this will help me figure out how to fight the dreaded light pollution and get better subs. Really appreciate your help navigating all these rabbit holes!
Cuiv, thanks. I love your posted comment at about 15:40 "My overall lesson is: choose the FOV you want for your camera (that determines focal length), then get the biggest aperture you can." Now this is math that I can understand!!! As I tell my friends, if I could have done the math I'd be a physicist not at attorney. Thanks.
I've never taken an astro photography pictures, and I doubt I ever will, but I really enjoy all your videos and discussions. Thank you for all the time you put in to this fascinating subject.
What a superb video Cuiv, I love how you've managed to break down this topic which can be challenging to understand into a bite-sized nugget! 😀 Fabulous work by Laurent too, I've been and checked out the slides and he's done a wonderful job - bravo all around! Clear skies buddy! :-)
Woohoo Luke, thanks for watching and glad it was useful! I love the nerdy stuff in the hobby, and I love to explain, so it's a good combination :p Clear Skies!
Thanks Cuiv both for your video and making available the slides. Speaking for many astrophotographers lacking the maths skill the maths on the slides are inpenetrable but I follow the gist. I like how for the examples given the exposure time exceeding the sensor read noise such as used by SharpCap Pro can be exceeded by x3 as a rule of thumb is something I have practically experienced when imaging. For an imaager not versed in maths to the extent required by the slides your narrative is perhaps the best we can do when imaging though some estimation of exposure length for a given location would be better. Cheers, Steve
Very informative video Cuiv! Yeah, there are so many differing opinions on what is going on, I don't know what to believe. Thanks for sharing Laurent's slides. Cheers Kurt
What's in this video shouldn't be very controversial, I covered the well established stuff! And my guideline of choosing your desired FOV and then picking the largest aperture you can for it should satisfy both aperture fans and focal ratio fans :D
Its true that focal ratio is irrelevant for the total speed of the system, if the faster and slower systems are binned to equivalent resolutions after capture (never a point in doing so at capture with CMOS) so in effect you could get the same SNR without the hyperstar as with it - with a smaller FOV at f/10 of course. Why its so is very simple and the comparison to a focal reducer works well. Your focal reducer does not affect how much light is passed through the aperture, but just how much is squeezed to each pixel. Binning does exactly the same thing by making the pixel bigger by combining a group of them, and if the subexposure has swamped read noise by x5 in the unbinned sub, it will also have done that at a bin level of lets say x4 (any level). It also means you can reduce the time it takes to get an acceptable result by binning. Binning a mono image (OSC depends on method of debayer how efficient but still close to 2x SNR increase) x2 will double your SNR which means the same as imaging for 4x the total time. Your recent 30h rosette could have been a 7,5h rosette with the same SNR had you binned x2. Binning is your friend for sure, wish this part was given more love on youtube as its the easiest way to get more SNR and many cameras can easily be binned 2 or 3x and still produce a very nice FullHD or larger image. But i dont blame you on skipping that, its a minefield in forums as you found out. I feel like i am laying mines as i write this! You can test binning SNR thing by stacking only a small portion of a nights subs and binning to something silly like x6 and processing as you normally would. You should see a very low resolution version of the complete image if processing is equal. Would link an example i have done but youtube would remove the comment so wont. Where the speed comes into play is how difficult it is to swamp the read noise, but with modern 1e- read noise sensors its really not that difficult to swamp read noise even with a slow scope. Your light pollution would trivialize it even at f/10 (except for narrowband where its still an issue and here i think its fair to say the f/2 scope is faster). Wall of text, gongrats if you made it to the end! (great video by the way)
Yesssss you summarized it excellently, and I have the same understanding. But I don't really want to put it out in a video because, as you say, landmines and flame wars! Another issue for me at F10 is that with my terrible seeing I'd be oversampling...... :P
More like this! Shorter time under 15 minutes definitely helps me process content. You do around 30 minute videos. I loose track because my brains "well depth" overflows, haha. More I learn more questions. Keeping it simple is very important. I find too many advanced amateurs talk over me. Thanks so much.
good video buddy! Sharpcap has the sensor analysis feature with which you can measure the sky glow and based on the smart histogram it recommends the number of exposures and length as well as gain... I am experimenting with this feature for both EAA and astrophotography. need some more clear skies :)
@@CuivTheLazyGeek it's a lot going on for me right now as I'm trying to sort out the C11 that I got on a loan from the university for a NEO research... but i will test the settings from Sharpcap with my own 115/800 frac and will share results... weather permitting ... I feel much more comfortable using my own scope when experimenting for it feels way more familiar and at least I hope I know what to expect ... although this is astrophotography ... one never knows wtf else can go wrong ... such a long stream of variables ....
@@CuivTheLazyGeek very much so. Starting with your “make yourself a smart telescope” vid I’ve got to a stage where I’m about to put my toes in the water. I’ve got an AZ-GTI mount, added a Pi4 with StellarMate, and using my old (had it 17years and was second hand) Skywatcher 127mak cas and 9x50 finder I’ve put a ZWO ASI120mmcs on the finder and an EOS600D DSLR for the scope. With a focal reducer it’s ok for starting deep sky, but with a Barlow should be good for solar system stuff which is my interest. I am disabled and need a light ish set up that I can assemble in chunks and operate remotely. Still waiting for my first clear skies with it.
This is wonderful information I really need. I have a Nikon D3300 Digital SLR with 24.2MP and a Celestron 8" Schmidt-Cassegrain telescope to learn the basics before purchasing better equipment. Also, for now I am using free Windows-10 GIMP software with an astrophotography plugin. My wish list is a larger aperture scope and a better camera. 😎 Thank you.
Laurent´s material is awesome!! Could it be possible to have it in pdf rather than in slides? I would love to print it and read it carefully. I´m afraid I´m old school. And thanks Cuiv for your amazing work!!
@Cuiv, The Lazy Geek - Greeting from Denmark. So much to learn ehh but thats great :-) I do really appreciate your videos and good explanations. They are really good and the best I have seen👍 I like 'Astronomy Tips & Reviews with Curtis' too. I find that your two channels compliment each others in a great way. - As a side note, have you seen Huygens Optics channel? 🤯🤓It about optics and lights in general. His new 'Imaging at ASML' video is really interesting and kind mindblowing, but the all are. If you haven't seen it I think it will interest you. Cheers 🍺
Hey Cuiv, glad you are feeling better. Sort of unrelated but are you still using the sharpstar 61 edphII? What's the time investment it usually takes for you to stack and edit your images? I got a very, very slow computer, and tried pixinsight, but if you have a recommendation for a reasonable computer to handling image processing under $500 or $1000, or a guide on using cloud computing to handle images, I'd be very interested to hear your thoughts. Thanks for your content, and take care. Kyle
Hey cuiv, something unrelated to this video. Is there a correlation or even casual between the distance from sensor to filter (duo narrowband for example) and the amount of halo that gets produced? Cs Andreas
Hi Cuiv. As you say, for the same object (a galaxy for example) and the same aperture (200 mm for example), with a longer focal length each pixel gets less photons but the image is spread over more pixels. So the resolution is better but the SNR is lower. While binning this image with the correct ratio, you can obtain the same (lower) resolution and the same (better) SNR you would get with a shorter focal length. So I would say that longer focal length gives you more flexibility for choosing the best compromise between resolution and SNR and lthat lower focal length gives you, of course, more field of view. Do you think this is correct? Cheers. Daniel
Well, it is correct if you're fine with the FOV AND your seeing supports the longer FL! Otherwise you're oversampling (thus having less SNR for no reason). So things can always get more complex :)
What do you think about video by Robert Glover on this subject? What Sharpcap recommend you for sub exposure in Tokyo? 3-5 sec? It works in my hand all the time, every time. Now about SNR. There is software that calculates it for you taking into account everything. Did you use it - SkyTool-4? It works for me as well.
@@CuivTheLazyGeek Great! Superb video. I always calculate exposures using SharpCap. For my backyard, it gives 10 -15 sec. I shoot 60 sec in LRGB to “safe” comp from burning down.
In theory, yes. In practice, if your 1 minute exposures already completely overwhelm the read noise (e.g. at least the optimal subexposure time), there is no practical difference. You'll definitely be better off with long exposures in very dark skies or in narrowband.
Many reasons! I don't shoot videos outside when the weather's not right/possibility of rain Sometimes the sky is just overexposed and appears clear when it isn't In summer it's very often clear during the day and cloudy at night
@@CuivTheLazyGeek well im going up to Fenlonfalls On Canada tomorrow night -14C going to try that C/2022 E3 ZTF ... you should try and do some tips on comet shooting if you get any
Hey Cuiv, thanks as always for your videos but this time I have a little critic. I was used to get a reliable and quality source of infos from you, but recently I've the feeling that your videos "tend" more to the hype than to quality contents. Look at what you say about focal ratio: "there's a lot of mess/debate on the internet so I don't know want to enter in that field" (more or less is your message). That's not your usual quality standard... Wait for a complete understanding and then syntethize as you're wonderfully capable of. Noise subject is as complicated and debated (I think even more) as the f/ratio/otical topic. By the way, understanding why f/ratio is not enough is not difficult (as you know for sure): given a f/ratio, large pixels will deliver higher signal then smaller ones, and so better SNR. So f/ratio is not enough, while pixel scale is what matters. Binning goes basically along the same logic. My intent with this post is not of course to criticize your very valuable work, but to stimulate you to come back to what I think were your standards. This is my modest point of view. Anyway thanks again and always for your very hard work, I appreciate it and love seeing you videos!
Hello Davide! You sure know how to write a comment that lives rent-free in my head when I should be doing other stuff... I understand where you're coming from, and it is your feeling after all - a bit of a discouraging feeling though! That said, I will say that it is precisely because I did a lot of research on focal ratio, etc and because I understand the viewpoints and explanations of the flame wars as well as the limitations of some corner cases that I intentionally didn't go into the details - that would deserve a looong video of its own, which I'm not sure I quite want to do at this stage, although it is on my list :-)
Yeah I know... Didn't notice while filming, and since I don't have a script, reshooting often means different content. So I just let it be. If it's any comfort I had to sit through it many times while editing 😂
Salut Cuiv, un lien qui t'intéressera (le même sujet) et en Français. media.afastronomie.fr/RCE/PresentationsRCE2018/Walliang-RCE2018.pdf bat, et encore merci pour tes vidéos que je déguste avec plaisir. Quelle joie de te revoir sur ta chaine!! Stéphane.
