Oversampling explained like you’ve never heard before.

oh my god I've just got first light on this new qy2 200m Dyan from the baron Bay observ here how are you it's been a while it's good to see you you're looking well how is your astrophotography going in how is your hot cousin does she ask about me does she ask it's planetary season and I love planetary season but you know what bugs me about it is that I spend my whole year leveling up the deep space stuff I'm taking photos of galaxies and nebula and faint exotic targets and then all of a sudden all my learning all my upgrading all my stuff it just goes out the window because planetary is a whole different ball game taking photos of planets is a completely different set of acquisition processing stacking skills that just really don't apply to deep space astrophotography and although some of the algorithms are similar like with stacking and drizzling really it's a different kind of approach to astrophotography in general people spend their whole lives just focusing on planetary or deep space and ignoring planetary all together together that's no fun I want to do both of them and I know you do too also planetry is like really Hands-On you can't automate most of it you do have to sit there and Fiddle with the focus and Chase the sing and Chase the orbit of the planet and the moons looking for transits and things like that it can get really complicated really quick it's not just like sticking a camera on telescope towards a planet and taking a photo it just doesn't work that way now I've spent years taking photos of the planet so I've been able to share some of my knowledge with you guys and I'll share more of that knowledge today however I do want to talk about one aspect of that which is the other thing that really bugs me about this is that my really expensive cameras my deep spaced cooled High bit depth cameras suddenly aren't good for planetary planetary photography requires a whole different set of camera parameters and what makes a good deep space camera doesn't necessarily translate for planetary work so if you want to take a photo of Jupiter Saturn or Mars I'm going to give you some tips on the technical background for what makes a good planet camera and what's going to be the best camera for you so stick around for that and in amongst the battery died but you know if there's one thing I've leared at being an astronomer being an astrophotographer is patience I have the patience the Zen master you know what I'm talking about because you do it too as I was saying in this video I will get eyes on Jupiter hopefully and I will test out a new planetary camera for myself because I've been on the same camera for years because it just worked but surely there are better cameras around these days so let's do some Hands-On experiments and see if we can figure this out my name is Dylan O'Donnell and you're watching Star [Music] stuff okay we want to take a photo of a planet uh but we don't have a space agency or an Orbiter that orbits around the planet we have a backyard telescope where do we start well Jupiter is at opposition right now which means it's 49 let's say 50 AR seconds across end to end and here's my telescope the Celestron 11in Edge HD uh and I can see that the resolving limit uh whether we calculate doors or Rel is about half an arkc so extrapolating that out to the Jupiter image uh that means it's about 100x 100 right that's the maximum amount of detail I'm going to get out of this telescope no matter the eyepiece no matter the camera as long as it's good seeing I should be able to get good detail on any features up to that half an arcc limit but let's test that theory it's easy enough to just get a you know big version of Jupiter from Hubble or JW us and downscale it to what we just worked out 100x 100 for my particular setup so that should give a good approximation of the maximum possible detail I should be able to get from the telescope and would you look at that it's pretty bang spot on to what I get on a really good night of seeing it's uh one for one apart from the fact that mine looks a little bit softer that's because the one on the left the simulated view is the perfect sort of sampling cuz I've just taken a high resolution image and made it smaller the run on the right is oversampled so I've got extra pixels with no extra detail and no extra information but it gives me a smoother result what's this so I used the planet and work backwards which I think is pretty cool and it illustrates the point I'm trying to make about the maximum level of detail you can get out of your planetary shops and it's good to know that but how do you know that well I've done all the hard work for you if you go to the show sponsor B that's www.intel.com I've developed a calculator where you can actually simulate cameras and your telescope combination it will show you the maximum resolving limit and you can see whether your images are under sampled or oversampled for deep space astrophotography you want to be in the sweet spot but for planetary photography it doesn't matter you can actually be oversampled in fact it's better to be oversampled but not tooo much there is a point where those pixels are not doing anything at all hopefully when this video goes up the new bintel rolled out and you'll be able to see your telescope simulation by simply looking for the telescope on the bintel website and clicking the simulate link this also works for cameras as well don't worry if they don't stock what you're currently using you can actually enter in the details of any particular telescope or any camera combination and it will still give you the results that you need the idea of this calculator is not to simulate every object in the universe it's to give you a good idea of how telescopes and cameras interact and give you a great idea of the sampling so that when you go to buy your next camera you can make sure it's going to work with your stuff I've just just bought a new camera from bintel with my own harde money and I'm going to use that camera now to see if it's a good fit for my particular setup I think it is because I use the [Music] [Music] [Applause] [Music] [Applause] [Music] calculator [Music] I put this LED light in one of mid's dog bags and uh I think it looks good I think it adds like a kind of [Music] ambiance okay let's go back to the theoretical limit of Jupiter in this example let's imagine that there's a monolith a piece of detail which is exactly half an arcc the limit of my particular C camera telescope situation so if the monolith is sitting there at an apparent 0.5 Arc patch of Sky it'll darken one pixel on our perfectly sampled theoretical camera now if we grab a camera with smaller pixels that can actually oversample the image the monolith will now take up four pixels because we've half the size of the pixel this is called critical sampling and this is the actual goal of the 5x Ru what is the 5x rule so what is the ideal sampling or oversampling for your telescope planetary photographers have something called the 5x rule I love rules rules are simple uh I can count to five so that's good right uh the 5x rule is kind of this shorthand rule I'll show you how it works but I think the rule is kind of wrong okay the 5x rule goes something like this you get your pixel size and you times that by five in this this case it's 4 micrometers which equals 20 but 21 it's actually F20 so F20 is the Target that you want for your telescope now I'm using a C11 which is F10 so in order to get my F10 to F20 I want to add a bow maybe a two times bow which would then equal F20 that's the 5x rule how however the 5x rule is wrong astrophotographers swear by the 7x rule so 7 * the 4 micro MIM pixels on the qy 200 in this example would actually equal F28 so if I used a 2.5x bow or Power Mate I would get f25 which is pretty close to that F28 but why what the hell for what does this all LIV mean uh I feel like this is just shorthand for the actual issue at hand which is sampling and because we're using pixels obviously this is all about sampling we want the sampling to be exactly four times the maximum limit of your scope so if our theoretical monolith is the maximum possible detail you can get on your camera we want that monolith to take up a size of four pixels it should actually be called the 4X raw and it shouldn't be using F ratio at all we should just look straight to the pixels and the maximum resolution of your telescope and this is the goal after getting this confusing fact wrong I've added the 5x and the 7x rule to the bintel calculator anyway so you get a green tick to show you which one of these rules you're hitting with your particular sampling but that's not the most important thing the most important thing is the sampling and the oversampling should be four times to reach that that critical sampling so I've added a multiplier in the oversampling results on the calculator so you can see when you hit that 4X Mark that's the right Rule and after talking to some astrophotographers about planetary and getting the best results they all agree that 7x is the one that gets them the best results and this makes sense because the 7x rule is actually the one that gets you to the 4X oversampling so we can keep using the 7x 5x rule but use the 7x version and you can throw all of that away and just use a calculator and make sure you're four times oversampled now I wish I had a money shot to show you at the end of this video but I've been using the qhy 2000m and I did get first light in between the clouds and the rain and man was I pleased I was able to image Saturn and Jupiter in 16bit mode which is actually 12bit but it did it at 100 frames per second and it doesn't drop frames which is something my last planetary camera has been doing and what's more I left this camera accidentally plugged in and running for a week so it was shooting 100 frames a second for a week shooting nothing shooting the wall but it must have taken like 60 million exposures and it was still going and it's still working fine these industrial seos cameras are really a miracle of engineering anyway I got some medium seeing and when I looked at Jupiter I saw that gatam was just gently kissing its limbs it was beautiful and satin was looking good too in fact it a little bit sharper I think because it was higher in the sky at the time so I'm really excited to see what this camera can do with the night a really good scene now bintel has these cameras in stock is this camera for you possibly you should use the camera to work that out there are different cameras pixel sizes that might be better for that critical sampling that you should try and reach with your particular setup if planetary is the goal you want a planetary camera and a barow combination that gets you to that four times oversampling thanks for sitting through all of that while I work that out in my head and hope you enjoyed the calculator upgrades my name is Dylan odonnell you've been watching Star stuff remember everything is meaningless and we're all going to [Music] die [Music]

Channel: Dylan O'Donnell

Views: 27,117

Keywords:

Id: -YSSMxxTz0E

Length: 12min 35sec (755 seconds)

Published: Thu Nov 23 2023