Thank you for this analysis. It is refreshing to see, that some people do understand the concept. I have spent counless hours on Clody Nights and in my local club explaining this to people who simply refuse to let go of this daylight photography concept. It is so frustrating, how many astrophotographers completely misunderstand it. TLDR; f-number is a completely useless metric in astrophotography. Aperture dictates light gathering "speed" and focal lenght dictates the FoV. Optimally you need two telescopes, one with a higher FL for small objects like distant galaxies and one with lower FL for those nice, wiedefield nebulea. The bigger the aperture you can aford on both telescopes the better obviously up to a point, where your mount can support it. (Things like under-/ oversampling given your cameras, seeing, light polution, wind, etc. will obviously play a role in determinig the optimal setup.)
Thank you, I agree with everything you said. I spent more than a week making this video, preparing animations etc and while some people got it, I still see some reactions that make me think I wasn’t able to get my point across.
Daylight and astrophotography f-numbers work exactly the same way and based on undeniable physics. What changes is the mindset and mental assumptions one makes when talking about them (what other parameter are fixed or can vary). In daylight photography it's very clear that the focal length is fixed (as well as the sensor and pixel size), then low f-numbers imply bigger apertures and light gathering. In astronomy, people don't always think this way, hence the confusion. To do so you should always compare a rasa f/2 to other telescopes with the same focal length, which will likely be smaller, with less aperture and have higher f/numbers. If you compare a hyperstar to a native SCT at f/10, so same aperture, then yeah you may be able to (partially) compensate surface brightness differences with pixel sizes, drizzling etc and allow these discussions about speed that become confusing to everyone. But the f-number will always strictly dictate the number of photons / second per unit surface area of the focal plane.
@@albertizard "number of photons / second per unit surface area of the focal plane." Sounds fancy and adds to the confusion. I think I know what you mean but what the hell is "unit surface area of the focal plane." (eschew obfuscation) The f number is nothing more, than the ratio between aperture and focal lenght. The number of photons / area is only determined by aperture. This whole confusion in astrophotography stems from people thinking "speed" means they gather more photons with "faster" scopes which is ridiculous. Anyway, all has been said in the video and my original post.
@@christoph4977 It means photons / second that receives each 1x1 micrometer square of the sensor. We can break down the process of taking a photo in two parts. First the telescope projects an image at the focal plane. If you put a paper on the focal plane (it works well with the moon) you can visualize the image. The brightness of that image only depends on the F-ratio (there is a video by deep sky detail from two months ago that explains this well). And second, is what sensor you put on the focal plane to record that image. The amount of light that receives each let's say 4x4 µm pixel depends just on F-ratio. Same with the total number of photons that collects a let's say full frame sensor. That's because at fixed aperture, a slower telescope has a higher F-ratio, longer focal length, the image is larger and the sensor only collects the photons of a smaller area of the sky, with the parts being cropped resulting in lost photons hitting the inside walls of the telescope instead of the sensor. So at the image plane, brightness depends only on F-ratio. However! When we think in terms of the sky "plane" things change: the number of photons / second per arcsec^2 does only depend on aperture. The video does a great job explaining some of these things but I find it disappointing and a disservice to the community to frame it as demystifying speed. F-ratio is an important number
By watching this video with such an impeccable way of presenting information, you have not only gained a follower on UA-cam, but also a fan. I am starting an astronomy channel in Spanish on UA-cam and I would be grateful if I could use the structure of this video as a reference.
Glad it was helpful! I think it takes a minute to wrap one's head around this concept after hearing "faster is better" everywhere, but I'm so happy my video could get this point across!
The larger the aperture, the smaller the details that the scope can resolve. Slower scopes are also usually very much oversampled using modern Astro cameras with 3 micron pixels. Using my C8 I can easily bin X2 without losing any detail, because I'm oversampling with my 183MC and its 2.4 micron pixels. I'm getting the brighter image, but a smaller view of the sky, meaning I need to mosaic my way to larger objects. Converting my C8 to an F2 with a hyperstar allows me to image a larger portion of the sky only slightly undersampled with my 183MC camera, saving a ton of time imaging anything larger. And because I'm now pretty well matched with resolving power and camera sensor imaging smaller objects oversampled, or cropping the image from the hyperstar make no difference in detail. Aperture is king, but F number only matters in terms that you need to match your camera sensor to it. Plus, a well matched setup always "looks" sharper than an oversampled image.
if you're using a reducer so you are at about 1400mm you are actually only a little bit oversampled which should be the perfect recipe for being able to recover contrast later via deconvolution! If the seeing is bad though, binning is a must for sure!
@the_space_koala yes I'm always shooting with the 0.63 reducer. My seeing is usually ok to good which puts me into the slight oversampled region. My SCT is from the 80s though, so my optics are probably a little worse than the calculator assumes. I can tell that images look "soft" despite perfect collimation. A sign of oversampling. Hyperstar at F2 should put me into the green for pixel size. Making everything tack sharp.
I'm a visual observer now...I use only slow telescopes (f/10-f/15) ...because all my eyepieces work beautifully in the slow system, no flock of seagulls at the edge. Nice video
I am sooo glad you mentioned towards the end about T numbers. I come from photography / videography and transmission is way more important sometimes than F number. The super fast telescopes have such big central obstructions you have to wonder.. what's the point. They're often so unbelievably expensive, you're past the point of an optimised setup. I know the Delta Rho for example is a very specialist telescope, but with the popularisation of fast newts we need to have better understanding. Thank you for this video - I see you on Instagram, but I'm glad to have discovered you on UA-cam. You're well informed and we really need more people like you out here 🙌
Thank you for the kind words! The delta Rho is an awesome example! I think it can be a very cool scope if you pair it with a medium format sensor otherwise it’s really a waste to have that huge ventral obstruction 😁
@@the_space_koala Completely. It's way out of the price range for 99% of people, so really it's good for the research side of things with universities more than private individuals. So I'd assume they'd be equipping it with medium format too.
Great explanation. I think the confusion arises from thinking it is the same as terrestrial camera photography and f stop numbers where the focal length and and sensor are fixed, but the aperture can be controlled to vary the exposure. As the stop is closed or opened the field of view remains the same but the light reaching the sensor varies according to the square of the selected f stop number (ie the area of the aperture).
I agree this is definitely where it comes from but even then it’s my “Case #1” - same focal length different aperture. Because the baffles are literally closing the opening down. But obviously nobody is looking at it this way 😁
Really good content here, but I think the binning discussion could be clarified to mention that 2x binning on CCD camera's still incur the Read Noise penalty on the 4 source pixels instead of just the generic term "noise". So in the 2x binning CMOS case the read noise increases by sqrt(4) = 2x for CMOS but stays 1x for CCD. However the absolute noise levels in CCD technology are much higher than CMOS by a factor of 8x and greater.
You’re right. My thinking was that someone will be using either a CMOS or a CCD and then think of using that camera on different scopes. Also, it would be worth discussing that the absolute noise (the variability of the signal) is still there either way. I think it’s a whole other discussion but I agree
Hi! Interesting statement but I would like to consider What You correctly said from a different angle. It is true that, in terms of "absolute" readout noise (nothing more 'cos there are many other type of noise or better to say "errors" related to the CMOS and CCDs) the CMOS sensor is a "winner". The fact to consider other than that is: how big is that unavoidable noise in percentage Vs. the signal acquired on a single shot? The most desirable sensor of the moment is probably the 2600 and all it's incarnations. The pixel size of that sensor is 3.7x3.7 microns of silica. Now look at the physics and what is the "real" maximum charge retainable but that area of silica. In the best scenario it could be no more than 12/16Ke, very distant from what vendors state in they'r specs. Mabye it is possible that the electronics make an inbetween reading of the data in a buffer and retain these numbers for a final sum prior to the readout but this is, in reality, not a "real" full well and the readout noise should be doubled in this case 'cos of the double lecture. Anyway a bigger pixel means a bigger full well just becouse there is more area of silica to retain the charge of the photoelectric effect. If You have a true full well of 50Ke and a readout noise of 5e You hav a 0.02% of uncertanty in Your data (this becouse the readout could be -5e to +5e so a total maximum fluctuation of 10e). If You have a sensor with 3e readout but on a true full well of 10Ke You have a 0.06% of total error that is three times more! Don't read just the "absolute" numbers on graphs and specs in the CMOS world, they could be marketing tricks! If You read the gain at will the 3e readout happens You will see that the full well is "reduced" from the maximum becouse You are going to accept less charge to generate an ADU or, if You will, You are sacrificing precision counting photons on larger and larger blocks for apparent increased sensitivity... Counting with increasing approximation is an error in digital world and it is exactly what we call "noise" since You choose to introduce approximation and so "uncertanty" in what You read and this is exactly noise in digital... Hope is not too technical and gives a different perception on the "lower noise" in CMOS...
I finally settled on my C925 Edge with Hyperstar V4 525mm with 9.25 inch of light gathering. I use an Altair 26C imx571 cmos sensor with IDAS NBZ 2dual narrowband filter which is pre shifted for fast optics.
May I give 2 cents in the collimation of fast telescopes? The difficulty is just if the optics are purely parabolic (newtonian). Hypergraphs (Hyperbolic optics) are Way difficult to collimate than parabolic and spherical are way easier than all above. I usually collimate my C14 in Hyperstar mode (F2) in 2 minutes, while my 6"F6.5 takes almost 15minutes (high precision collimation)... This is due the fact the parabolic and hyperbolic optics are more sensitive to tilt of the mirror while spherical optics can be tilted and the parallel rays will have the same path. Everything above or below sphere (in the colimation point of view), the optical axis of the mirror need to be perfectly aligned.
Yes I agree! My point on the difficulty (or ease) of collimation was mostly for newts. My sharpstar has extremely sensitive optics and not-so-great mechanics to support it...
Thanks for the informative video. However, as someone who images primarily in bortle 9 skies, I find that faster scopes have an advantage of speeding up my total integration time by using short exposures instead of long exposures that achieve the same results.
Focal ration does make a difference with respect to Spatial resolution! (not to be confused with angular resolution) Airy disk (spatial resolution) is directly proportional to focal ratio; this is the resolving limit for an optical system, any smaller detail is lost in the diffraction pattern presented on the sensor; width of disk ~= 2.44 *lambda*f/a; where lambda is frequency of incoming light, f is the focal length, and a is the aperture; f/a is the focal ratio;
I agree, I did briefly mention this at the end quoting faster instruments have a worse performance producing contrast at higher spatial frequencies. However, I believe this is something that could theoretically be recovered using deconvolution algorithms as long as you're at least critically sampled
Great, as always... Smart and gorgeous, an excellent combination! I'd love a video tour of your observing area loaded up with gear sometime, please. Thanks, Michael
@@michaeledmonds3027 thank you Michael! I’m planning on doing that however my place is horribly light polluted it’s a very sad sight with these beautiful telescopes 😁 I go to the mountains to photograph whenever possible
Light pollution seems to be the norm these days. I live in a Bortle 8 suburban area, plus it's heavily wooded. By the way, I appreciate your sense of style in your video staging. I look forward to whatever you may put together, take care. @the_space_koala
Thanks for clarifying this. I am a visual observer and i like my Newt for DSOs as its brighter but i use my Mak for planetary observation as its more high def. My Newt as it has spherical mirror maybe bright but not as hidef at higher magnification as the Mak for planets.
You are to be commended for such a thorough and detailed review. It stands out as one of the best presentations on astronomical equipment. I must admit that at first I wasn't sure such an attractive presenter would get into such detailed scientific explanations. I do apologize for that. Your review ranks up there with the best informational videos available.
You have a 14" EdgeHD? Jealous! Very good video- I get a lot of questions around this in my local astronomy club- I'm going to send them to this video! Clear skies and have a great day.
@@AstroDenny thank you so much! It’s a point I’ve been wanting to make a video on for months but it’s just so hard to get it across! I tried my best but some comments on here and other forums let me know it’s still not clear enough to everybody The C14 EdgeHD is awesome I love it, but it’s so big it’s a 2-person job to mount it 😁
1.25” focuses are small and generally they’ll be good for planetary photography or some deep sky if you’re using a tiny sensor. At least 2” is recommended for medium to larger sensors.
I just ran into your channel - this video showed up in my recommended feed. I think you did a very nice job explaining the pros/cons of "fast" glass. As someone who owns an 8" EdgeHD, an 80mm triplet refactor and a 150mm newt (the same one you've got sitting on the shelf behind you), I can very easily relate to the challenges posed. Collimating my EdgeHD is extremely simple. The f/2.8 newt takes considerably more effort. I actually bought one of those Ocal devices to help me dial it in because the typical Cheshire tubes / lasers weren't good enough. Spacing and tilt are also far more difficult to manage on the f/2.8 scope than on the EdgeHD because the critical focus zone is so much smaller. One point I'd like to bring up regarding your comparisons of drizzled vs non-drizzled data. Because, like everything else in astrophotography, there are _always_ compromises. In the case of drizzle, you're trading resolution for noise. In the 2x drizzle, you're spreading the noise of 1 pixel over 4 pixels. If we assume the same conditions you did in your examples, the 2x drizzled image will have about 4x noise than the non-drizzled one. To get relative parity in noise profiles between the two data sets, you'd need to take 4x as many subs (noise halves for every doubling of exposures) and ensure that data is well dithered. Anyway, dithering, drizzling, sampling are likely a topic for another video (or 3 😀).
Hey congrats on the nice scopes 😁 I 100% agree with your calcs and as far as I can tell this is exactly the information I tried to convey. As it’s extremely difficult to show SNF I showed the drizzled images as being “darker” and said you’ll need to expose 4x as long to get the same amount of light (which is the same as the same level of SNR). Which of the cases was this not clear in? Perhaps I should clarify!
So if time we remove shooting time from the equation because we will always shoot as long as it takes to reach the point of diminishing returns, then what really matters is choosing a focal length that perfectly frames the subject in order to achieve the maximum image quality, no matter what the focal ratio is. Is that correct? So a fast focal ratio only helps when we have a limited shooting time want to maximize the light collection within that time? And even with a limited amount of shooting time, it might still be better to optimize framing the subject over speed and a wide FOV?
Yes and no. You can get to the desired signal to noise ratio with any scope, given enough time. However, you’re still limited by aperture in the resolution that you can get. For example if you shoot a dense section of the Milky Way with a zoom lens vs a large telescope you’ll get very different results. A larger aperture will result in smaller stars and you’ll be able to get a better picture anyway.
Thank you for the very informative and helpful presentation! Is it a correct application that a focal reducer increases FOV but not image brightness? As a side note, your video creation abilities are seriously impressive 🙇♂️
Thank you for saying that! With a reducer you make your overall image brighter but you exchange it for resolution! So you get a smaller, brighter image of the same subject.
benefit of a fast telescope is the option to complete imaging in short time which is extremely good for: - unstable weather area - limited time for people to have a full time job - easier to process frames taken in the same night.
@@the_space_koala watched the whole vid. There are certain benefits as you mentioned for the fast telescope which I totally agree. These benefits can never be replaced by a slow telescope with more zoom power by mosaic or whatsoever, and I just pointed out how important those situations are.
I grew up using film (light sensitive chemical coatings deposited on a thin backing). Focal ratio seems less important to actually capturing an image in 2024, perhaps due to the rise of modern sensors, software and personal computers.
It’s not that much different on film IMO - you still have “pixels” - the little silver grains - even if nobody thinks about them that way. If you have a higher ISO film your pixel size goes up. In my opinion everything in this video is still valid when you put it into this context.
@@the_space_koala You are a more competent optical scientist than I-which is why I enjoyed viewing your presentation. However, we really did have to off-axis guide, using kludgy cold cameras, attached to precisely collimated scopes, on perfectly balanced and polar-aligned mounts, literally for hours on end, employing dodgy low-res film media of high compromise, pushed to near its limits-and it all could be ruined at any moment by unpredictable weather (no radar or weather satellites for us). Film needed expert chemical pushing during perfect development. Surely for us, if we used a telecompressor to achieve a faster F-ratio we weren’t wrong? I still believe this generation has it easy-which is great! Superb, high-sensitivity electronic digital images produced inexpensively in the backyard by today’s amateur astronomers are just mind-blowing :-)
Tackling this subject without ever talking about image sampling is quite a feat :). Today, CMOS cameras have very small pixels, so the game is not the same as when pixels were 24, 15 or 9 microns wide. If you have a 40cm aperture, at F/8, with 3.76 microns pixels, one pixel is 0.24 arc seconds of the sky. If you shoot in a typical site with 3 arc second seeing, one star image will be 12 pixels large, which is grossly oversampled. Even if you are in paradise and the seeing is 1 arc second, you are still oversampling a bit. But then when the image size starts to be around 1 arc second, you enter in a whole new domain where everything has to be perfect, focus, guiding, optical quality, etc... and there are very few amateur images with sub arc second long exposure images. In practive with today's camera and very long focal length, you don't loose any resolution if you bin 2, or if you have an F/4 telescope. Of course if you shoot with a small focal length (like a 50mm diameter objective), then at F/8 then one pixel is around 2 arc second, and if at F/2 it is 8 arc seconds, and provided the optics are not the limit (i.e. giving 2 arc second images on a wide field) then indeed there is a difference in resolution. Then there is the problem of the optical configuration. Some optical designs are only good on paper, and in practice almost impossible to collimate perfectly on a wide field (i.e. easier to use with a small camera then on a full frame camera). That's why Planewave CdK (with spherical secondary) or Celestron Edge are easy to collimate and all the "corrected" cassegrain systems are a bitch to collimate on a wide field. Then a refractor usually doesn't have these problems, except if they get out of collimation. I have an FSQ85 which was out of collimation and... there is no mechanical way to collimate it, then you are fried (going back to the manufacturer). You can spend quite a lot of nights on a decollimated RC system if you want to cover a full frame image. Even a newtonian, using a field corrector can give problems to use correctly (find the correct backfocus distance, and obtaining sharp images all over the field). Well, image sampling considerations are not the whole story.
I agree with what you’re saying on image sampling, however I did start the whole comparison in the video stating we’re focusing on ideal conditions and perfect seeing. The focus here is on light collecting capabilities only and it still turned out to be 20 minutes long 😁 not to mention you could do the same comparisons with tiny lenses so you’re only ever diffraction limited. There will be a whole different video covering resolution/spatial frequencies which will be a better time to discuss this otherwise important point
In other words: Just get a scope with the FOV you want, get a camera with the resolution that you want. Aperture isn't everything, but it certainly can be important, depending on the rest of your system.
Hi and thank You for this remarkable effort! A well done video indeed. It is a very long time I'm trying to explain that "fast" in the digital era is a very uncorrect term to use... Whan we talk about a telescope otpic that was defined "fast" in the old emulsion era, we are now talking about a wide field telescope. In the past we need to fight the non linearity of the emulsion and the Swartzchild effect (also known as reciprocity failure) but today with linear acquisition devices such as CCDs and CMOS this is no more true. As to help Your nice explanation of the "f-ratyio myth" I would like to say something more. As You stated we are trying to collect as much photons as possible coming from the object of interest. Despite the system of acquisition I would like to point out exactly that: the energy coming from a subject is very dependant from the considered angle of the sky we are trying to measure. Every object as it's own energy emitted on a given angle and no f-ratio could chenge that! So, even considering a "perfect" system, we are totally dependant form the photometry! The photometry is the exact measurement of the flux of photons per second on a considered angle. This means that there is a maximum number of photons You can collect per second on a sigle pixel that define an angle (also called spatial resolution even if it is also correct to call it spatial sampling or angular sampling). For every object at a given wavelenght is already known how much photons You can collect... Really doesn't matter how "fast" you fill Your pixel since the ratio between the unwanted and desired signal is increasing proportionally. You fill Your well (pixel or angle of the sky) faster with signal (but is still a fixed flux that comes in in a poisson way and nothing can change that), AND unwanted noise (which is mostly sky-noise). the ratio between the two does not change at all. The only way to improve the signal considering a fixed angle of the sky is to collect more photons in the same exposure time and, as You explained, the only way is to increase the collecting area (bigger diameter)! Hope this could help to better underestand why f-ratio on linear acquisition devices doesn't matter much if not at all. there are few exceptions to that but they are related to the digital system, not the physics that always rules. Try not to consider the system but what's the reality outside that is, in fact, what You are trying to measure or capture and You will better underestand what You're doing... Thanks a lot and keep on this way! I appreciate it very much!
@@giovannipaglioli2302 thank you so much! Indeed this is a very hard point to get across, even if the video is super long with a million examples, based on many comments here and on other sites I have the feeling the message didn’t go through to many people.
@@the_space_koala keep going! It is something related to traditional photography mabye and, since it is a very diffuse and "known" topic, it is difficult to transfer to people... At the end is only a matter of considering thinks in a different way: we are in the realm of digital and we are no more "taking pictures", we are "making measurements" instead! If You try to look at things in this way mabye it could be easier for anyone to get to the real point. It is sometimes quite counterintuitive but. once acquired, it seems obvious. Thank You!
For me the 'advantage' of a large diameter short focal length telescope is not ending up with a scope twice as long and having to have an extremely expensive mount and enclosure to house it in. An 18" scope with a 5 and a half foot tube is way better than it having an 11 foot tube and needing a motorized lift platform to use it!
That's true! But in the end another fundamental aspect to consider is the resolution per pixel, even if You have a small optics but very high per pixel resolution, You will need an "expensive" mount to track well. For shure You will not need a "big" mount but a very precise one tha could be expensive too... 🥲
@@giovannipaglioli2302 you’re right in principle regarding the precision, however even low-end mounts tend to be quite precise at lower payloads. On the theoretical side, with the same exact image scale, it’ll cost you much less to find a mount that can carry 10kg vs 30kg, to the same precision
Nice video, but I'm not sure I concur that FOV is the driver of interest between fast and slow systems. If you take a fast system, and combine it with small pixels, you'll reach the limit of "seeing limited" resolution (equivalent to what a slow, more magnified system can give), but with many more photons per unit of time. Photons per unit of time roughly equates to SNR. I'd prefer to go fast and deep which will result in nicer images than what most folks are willing to spend with a slower optical setup. I still think faster optics has a significant advantage (even given the drawbacks - which I think you showed well....but which are manageable).
Putting small pixels on a fast scope will make the “photons per pixel” lower. It’s exactly the same as putting larger pixels on a slower scope. As long as the aperture is the same you’re not collecting more light!
@@the_space_koala I don't think we're talking apples to apples here. In your case 2, the apertures are the same diameter, and (while not exactly the case you presented), for an equivalent pixel scale (via different cameras), cropped to the same FOV, the differing f-ratios means the faster system will collect photons in each pixel more quickly as a function of the inversely proportional square f-ratio comparison. In your example 2, using the same camera is where I believe you miss the overall point. When the image scale is the same (via different cameras), the slower f-ratio telescope will take more time to gather the same number of photons in each pixel as the faster setup, so the obvious advantage goes to the faster system by way of less time required on sky for the same resolution (shorter exposures, or ability to gather more SNR over an equivalent time). People pair camera's (image scales) to sampling/seeing, so fast systems will outperform on a time basis. This seems to be an etendue issue more than a FOV issue, which is what is typically advertised in fast astrograph's marketing materials. Do you agree? What am I missing?
First time I have seen such a technical details of Astrophotography. I am a hesitant beginner and want to learn the science of the Astrophotography. Can you or anyone suggest some sources: books etc where I can learn all these concepts.
Hey I know this is not the answer you want to hear but honestly just UA-cam is a very good source. As weird as it is, I don't necessarily recommend books because in astrophotography technology is changing SO quickly, recommendations change every few years as well.
I prefer faster focal ratios merely because I can get more data more quickly, but if the stars are seagulls in the corners and edges faster focal ratios are no benefit.
@@the_space_koala so basically you're just spreading the light out,so fewer photons are hitting a pixel so less snr , this really only scales out if you bin, or have larger pixel size in the slower scope vs faster scope
@@BigBadLoneWolf they catch more photons from a larger area of the sky. They catch exactly the same number from a given area. So what changes is only the field of view
I'm about buying a telescope under 10k so I choose 2 telescope edisla astra 100 or fotocart 76AZ who is better and please give me some suggestions....bytw I'm from india
@@RudraEditHub666 I don’t know these models but “az” sounds like it’s on an altazimuth mount. Please know that’s not adequate for astrophotography if that’s your goal you need an equatorial mount
Yes you are right altazimuth mount..I'm a 10th student and I love astrophotography but this is my interest not my journey like youtube so I think AZ mount is enough for my astrophotography
Good video and explanation. But it is "only" first partof the true. then comes the over and under sampling questions, which is also depending the focal lenght and the pixel size. Then comes the atmospheric/seeing conditions. So the astrophotography is not so simple... Anyway, feliratkoztam a csatornádra! :)
Thank you for the explanation! Subscribed and looking forward to the Drizzling video! I seriously wonder though - are you the blonde twin of @AnastasiInTech ?
its all about framing.. using an F4 scope that has your object only covering 20 percent of your frame will not be better than an f7 or higher that frames your subject in the majority of the frame.. its all about framing,, if you get larger diameter you need longer focal lenght unless you want a very large target in frame....the term aperature is king comes at a cost... i would rather do a mosaic and no binning with longer exposures, or exposer time, to catch the entire image if needed.. its all about details and a larger F ratio is king in that department.. aperature is king for visual thats for sure, you can always magnify with the lens,, but not needed for astrophoto.. as most things,, time and patience is the absolute king
I agreed with the first part until I read aperture is not needed for astrophotography. You physically collect more light (plus better theoretical max resolution). Sure time and patience are important but with a larger telescope you still can get the same exact image in a shorter time so surely it's better
Any scope is painful is bortle 7. If you had a faster scope it would also collect more light pollution not just signal. It’s difficult either way (i live in bortle 7 😁)
Telescopes are tools. Unlike other tools like wrenches for example, you can't have the perfect telescope for every thing you watch. You must make compromises. Unless you can buy 50 telescopes, you have to decide what you will do the most with it, and select the best you can afford for that job. Then you have to accommodate this choice in other situations. There is no better or worse. There is what you can afford to buy for the main job and you work with it.🎉
Don’t cite Celestron. Don’t bring example of RASA. Celestron is mass produced astro-equipment. High quality fast telescopes are always better. If long FL is required, increase size of mirror.
the RASA is probably the best known and most accessible example of f/2 optics. In any case whether something is mass produced or not has no impact on its theoretical light gathering capability. The video contains reasoning that is generic (the examples mention no specific model of telescope) and can be applied to the cheapest or the highest-end products as well and provides a framework for people to evaluate their options.
It seems the lady must have missed optics classes. The only information was that the resoluion power is different between the two types of instrument for the same focal length. However, there is also the fact that the spots do not have the same diameter. the idea of coupling photocytes allows us to take into account the size of the Airy spot. Which would also accelerate the speed. The second point is the tolerance for maintaining focus. We talk about scale factor at this level. But certainly at F2 which spots 4 times larger than the diffraction allows, the resolution has become a detail. What matters is the speed of wide field photography. And yet cameras with small pixels now exist. No one seems to know of correctors that reach the diffraction limit. On the Japanese side, there are still companies thinking about it. The third oversight concerns optical aberrations. These kinds of details make amateurs disappointed with their telescope. Certainly having two distinct instruments allows you to combine the strengths of each for beautiful images. And in fact this seems to be the best advice in this video. With the different brands people get lost and traders are unable to understand and offer their customers products that are not compatible with their instruments because in optics you have to know that it is the piece that decides especially if you want to photograph the deep sky. What criterion imposes the choice of a corrector/reducer. Ss reduction of its focal length, or the characteristics of the baffle in a telescope which is sometimes calculated too accurately for a reducer, see which lets in stray light? The camera has its own characteristics given by the manufacturer, which greatly facilitates its best use. And if we're not mistaken about the camera, it's because we know our instrument. It's exactly like photography: The body and its Cmos become secondary to the lens which is placed in front....Thank you for the video
Thank you for your exhaustive comment! There’s definitely an infinite number of factors to consider and it’s simply impossible to discuss every subject in one video - this was already way too long as is. Hence the disclaimer (mentioned twice) in the video - that we’re focusing on equivalent light collecting capabilities only for this comparison. Pixel sizes and camera sensors are also a whole other topic but I got around that by saying we’re using the same exact sensor for the comparisons. The take home message of this video is to make people understand that fast optics trade off image scale for a wider field of view. This is true for any system in general and it takes a while for people to wrap their heads around this - astrophotographers, that is. A visual astronomer would find everything I said extremely obvious - only the diameter will matter 😁
Yes, don’t agree with it. FOV effect compensation with 4 separate pictures is well known no solution. By the way, scientific CMOS are binned different way.
@@anata5127 the whole point of the video is "fast" just means a larger field of view in practice - given the same aperture - and you're saying the same thing. It's mostly aimed at people who put a reducer on a telescope thinking it will collect more light
Incorrect. Fast means more light per pixel. Ok? For example, galaxy emitting the same light, reduction of F distribute light to lesser numbers of pixels. Now, people arguing that resolution is lost. Yes, but I don’t need resolution for vast objects. If I need resolution, then I will rather increase aperture maintaining lower F. What low F scope do you have?
@@anata5127 You'd get exactly the same result with a slower scope using larger pixels or the same small pixels but resampled. The question is how big of a sensor area you're distributing X amount of light over. An f/number tells you nothing without specifying pixel size. And the original amount of light to be distributed is only a function of the diameter of the optics. My fastest telescope is a RASA (f/2) and I also have an f/2.8 reflector.
Wow!!!! As a beginner into deep space photography, this was EXACTLY what I was looking for. Very precise and detailed explanation!!! Thank You!!!
Glad to hear it was helpful!
Thank you! This is by far the best video on this subject! It is really easy to understand what you are explaining. Clear skies!
@@alexandreastronomy8022 thank you so much! Clear skies 🌌
I 100% agree with this comment, thanks for this video :)
I would give this presentation *five* stars!
@@dmpase haha thank you!😊
Thank you for this analysis. It is refreshing to see, that some people do understand the concept. I have spent counless hours on Clody Nights and in my local club explaining this to people who simply refuse to let go of this daylight photography concept. It is so frustrating, how many astrophotographers completely misunderstand it.
TLDR; f-number is a completely useless metric in astrophotography. Aperture dictates light gathering "speed" and focal lenght dictates the FoV.
Optimally you need two telescopes, one with a higher FL for small objects like distant galaxies and one with lower FL for those nice, wiedefield nebulea. The bigger the aperture you can aford on both telescopes the better obviously up to a point, where your mount can support it.
(Things like under-/ oversampling given your cameras, seeing, light polution, wind, etc. will obviously play a role in determinig the optimal setup.)
Thank you, I agree with everything you said. I spent more than a week making this video, preparing animations etc and while some people got it, I still see some reactions that make me think I wasn’t able to get my point across.
Lower focal ratios are literally faster
Daylight and astrophotography f-numbers work exactly the same way and based on undeniable physics. What changes is the mindset and mental assumptions one makes when talking about them (what other parameter are fixed or can vary). In daylight photography it's very clear that the focal length is fixed (as well as the sensor and pixel size), then low f-numbers imply bigger apertures and light gathering. In astronomy, people don't always think this way, hence the confusion. To do so you should always compare a rasa f/2 to other telescopes with the same focal length, which will likely be smaller, with less aperture and have higher f/numbers. If you compare a hyperstar to a native SCT at f/10, so same aperture, then yeah you may be able to (partially) compensate surface brightness differences with pixel sizes, drizzling etc and allow these discussions about speed that become confusing to everyone. But the f-number will always strictly dictate the number of photons / second per unit surface area of the focal plane.
@@albertizard "number of photons / second per unit surface area of the focal plane." Sounds fancy and adds to the confusion. I think I know what you mean but what the hell is "unit surface area of the focal plane." (eschew obfuscation)
The f number is nothing more, than the ratio between aperture and focal lenght.
The number of photons / area is only determined by aperture. This whole confusion in astrophotography stems from people thinking "speed" means they gather more photons with "faster" scopes which is ridiculous.
Anyway, all has been said in the video and my original post.
@@christoph4977 It means photons / second that receives each 1x1 micrometer square of the sensor.
We can break down the process of taking a photo in two parts. First the telescope projects an image at the focal plane. If you put a paper on the focal plane (it works well with the moon) you can visualize the image. The brightness of that image only depends on the F-ratio (there is a video by deep sky detail from two months ago that explains this well). And second, is what sensor you put on the focal plane to record that image.
The amount of light that receives each let's say 4x4 µm pixel depends just on F-ratio. Same with the total number of photons that collects a let's say full frame sensor. That's because at fixed aperture, a slower telescope has a higher F-ratio, longer focal length, the image is larger and the sensor only collects the photons of a smaller area of the sky, with the parts being cropped resulting in lost photons hitting the inside walls of the telescope instead of the sensor. So at the image plane, brightness depends only on F-ratio.
However! When we think in terms of the sky "plane" things change: the number of photons / second per arcsec^2 does only depend on aperture.
The video does a great job explaining some of these things but I find it disappointing and a disservice to the community to frame it as demystifying speed. F-ratio is an important number
Superb explanations! Thanks for taking the time.
@@2193191 thanks! I’m glad you think it was worth it
Thank you very much for sharing your knowledge. You are very clear, a true teacher, and you explain such a complex topic clearly.
Thank you for the kind words, I am so glad you found it helpful!
By watching this video with such an impeccable way of presenting information, you have not only gained a follower on UA-cam, but also a fan. I am starting an astronomy channel in Spanish on UA-cam and I would be grateful if I could use the structure of this video as a reference.
@@ElkinGuillermoForeroRosado thank you that is so nice of you to say! Feel free to refer to my video of course
Wow, what a lecture! Thank you so much for the enlightenment regarding this subject. You provide much needed clarification.
Glad it was helpful! I think it takes a minute to wrap one's head around this concept after hearing "faster is better" everywhere, but I'm so happy my video could get this point across!
@@the_space_koala Thanks, I am so glad to find your channel, you are a great teacher. Have a great day.
The larger the aperture, the smaller the details that the scope can resolve.
Slower scopes are also usually very much oversampled using modern Astro cameras with 3 micron pixels.
Using my C8 I can easily bin X2 without losing any detail, because I'm oversampling with my 183MC and its 2.4 micron pixels.
I'm getting the brighter image, but a smaller view of the sky, meaning I need to mosaic my way to larger objects.
Converting my C8 to an F2 with a hyperstar allows me to image a larger portion of the sky only slightly undersampled with my 183MC camera, saving a ton of time imaging anything larger. And because I'm now pretty well matched with resolving power and camera sensor imaging smaller objects oversampled, or cropping the image from the hyperstar make no difference in detail.
Aperture is king, but F number only matters in terms that you need to match your camera sensor to it.
Plus, a well matched setup always "looks" sharper than an oversampled image.
if you're using a reducer so you are at about 1400mm you are actually only a little bit oversampled which should be the perfect recipe for being able to recover contrast later via deconvolution! If the seeing is bad though, binning is a must for sure!
@the_space_koala yes I'm always shooting with the 0.63 reducer. My seeing is usually ok to good which puts me into the slight oversampled region. My SCT is from the 80s though, so my optics are probably a little worse than the calculator assumes. I can tell that images look "soft" despite perfect collimation. A sign of oversampling.
Hyperstar at F2 should put me into the green for pixel size. Making everything tack sharp.
Amazing information ! Thanks !
I'm a visual observer now...I use only slow telescopes (f/10-f/15) ...because all my eyepieces work beautifully in the slow system, no flock of seagulls at the edge.
Nice video
All visual astronomers will agree it makes no sense to push the optics to a “fast” focal ratio!
Outstanding explanations of complicated subjects. Thank you!
@@jeffweiss2131 thank you Jeff glad you find it useful
A banger video! Thanks for the explainer, just what I needed.
Thanks, so glad it’s helpful
Bardzo dobry wideoporadnik - tak trzymaj!
Excellent video. Very clear, thorough and thoughtful, thank you for putting this together
Thank you for saying that I’m glad it’s helpful
I am sooo glad you mentioned towards the end about T numbers. I come from photography / videography and transmission is way more important sometimes than F number. The super fast telescopes have such big central obstructions you have to wonder.. what's the point. They're often so unbelievably expensive, you're past the point of an optimised setup. I know the Delta Rho for example is a very specialist telescope, but with the popularisation of fast newts we need to have better understanding. Thank you for this video - I see you on Instagram, but I'm glad to have discovered you on UA-cam. You're well informed and we really need more people like you out here 🙌
Thank you for the kind words! The delta Rho is an awesome example! I think it can be a very cool scope if you pair it with a medium format sensor otherwise it’s really a waste to have that huge ventral obstruction 😁
@@the_space_koala Completely. It's way out of the price range for 99% of people, so really it's good for the research side of things with universities more than private individuals. So I'd assume they'd be equipping it with medium format too.
Great explanation. I think the confusion arises from thinking it is the same as terrestrial camera photography and f stop numbers where the focal length and and sensor are fixed, but the aperture can be controlled to vary the exposure.
As the stop is closed or opened the field of view remains the same but the light reaching the sensor varies according to the square of the selected f stop number (ie the area of the aperture).
I agree this is definitely where it comes from but even then it’s my “Case #1” - same focal length different aperture. Because the baffles are literally closing the opening down. But obviously nobody is looking at it this way 😁
Really good content here, but I think the binning discussion could be clarified to mention that 2x binning on CCD camera's still incur the Read Noise penalty on the 4 source pixels instead of just the generic term "noise". So in the 2x binning CMOS case the read noise increases by sqrt(4) = 2x for CMOS but stays 1x for CCD. However the absolute noise levels in CCD technology are much higher than CMOS by a factor of 8x and greater.
You’re right. My thinking was that someone will be using either a CMOS or a CCD and then think of using that camera on different scopes. Also, it would be worth discussing that the absolute noise (the variability of the signal) is still there either way. I think it’s a whole other discussion but I agree
Hi! Interesting statement but I would like to consider What You correctly said from a different angle. It is true that, in terms of "absolute" readout noise (nothing more 'cos there are many other type of noise or better to say "errors" related to the CMOS and CCDs) the CMOS sensor is a "winner". The fact to consider other than that is: how big is that unavoidable noise in percentage Vs. the signal acquired on a single shot? The most desirable sensor of the moment is probably the 2600 and all it's incarnations. The pixel size of that sensor is 3.7x3.7 microns of silica. Now look at the physics and what is the "real" maximum charge retainable but that area of silica. In the best scenario it could be no more than 12/16Ke, very distant from what vendors state in they'r specs. Mabye it is possible that the electronics make an inbetween reading of the data in a buffer and retain these numbers for a final sum prior to the readout but this is, in reality, not a "real" full well and the readout noise should be doubled in this case 'cos of the double lecture. Anyway a bigger pixel means a bigger full well just becouse there is more area of silica to retain the charge of the photoelectric effect. If You have a true full well of 50Ke and a readout noise of 5e You hav a 0.02% of uncertanty in Your data (this becouse the readout could be -5e to +5e so a total maximum fluctuation of 10e). If You have a sensor with 3e readout but on a true full well of 10Ke You have a 0.06% of total error that is three times more! Don't read just the "absolute" numbers on graphs and specs in the CMOS world, they could be marketing tricks! If You read the gain at will the 3e readout happens You will see that the full well is "reduced" from the maximum becouse You are going to accept less charge to generate an ADU or, if You will, You are sacrificing precision counting photons on larger and larger blocks for apparent increased sensitivity... Counting with increasing approximation is an error in digital world and it is exactly what we call "noise" since You choose to introduce approximation and so "uncertanty" in what You read and this is exactly noise in digital... Hope is not too technical and gives a different perception on the "lower noise" in CMOS...
Great video, the exemples were clear and easy to understand.
Thank you I’m glad you found it easy to follow!
I finally settled on my C925 Edge with Hyperstar V4 525mm with 9.25 inch of light gathering. I use an Altair 26C imx571 cmos sensor with IDAS NBZ 2dual narrowband filter which is pre shifted for fast optics.
clear skies for the light bucket!
Excellent, many many thanks
Great video 🌸✨
Great explanation!
@@zara8289 thanks for saying that!
Thanks a lot! Great explanation!
@@Nothnágel_Balázs so glad it’s helpful!
Awesome work! :-D Clear skies to you!
thank you very much! :) clear skies are but a dream! to you too!
May I give 2 cents in the collimation of fast telescopes?
The difficulty is just if the optics are purely parabolic (newtonian). Hypergraphs (Hyperbolic optics) are Way difficult to collimate than parabolic and spherical are way easier than all above. I usually collimate my C14 in Hyperstar mode (F2) in 2 minutes, while my 6"F6.5 takes almost 15minutes (high precision collimation)...
This is due the fact the parabolic and hyperbolic optics are more sensitive to tilt of the mirror while spherical optics can be tilted and the parallel rays will have the same path. Everything above or below sphere (in the colimation point of view), the optical axis of the mirror need to be perfectly aligned.
Yes I agree! My point on the difficulty (or ease) of collimation was mostly for newts. My sharpstar has extremely sensitive optics and not-so-great mechanics to support it...
Thanks for the informative video. However, as someone who images primarily in bortle 9 skies, I find that faster scopes have an advantage of speeding up my total integration time by using short exposures instead of long exposures that achieve the same results.
Focal ration does make a difference with respect to Spatial resolution! (not to be confused with angular resolution) Airy disk (spatial resolution) is directly proportional to focal ratio; this is the resolving limit for an optical system, any smaller detail is lost in the diffraction pattern presented on the sensor; width of disk ~= 2.44 *lambda*f/a; where lambda is frequency of incoming light, f is the focal length, and a is the aperture; f/a is the focal ratio;
I agree, I did briefly mention this at the end quoting faster instruments have a worse performance producing contrast at higher spatial frequencies. However, I believe this is something that could theoretically be recovered using deconvolution algorithms as long as you're at least critically sampled
@@the_space_koala Thank you, I understand.
Outstanding! Thank you!
Thank for that, glad it’s helpful!
Great, as always... Smart and gorgeous, an excellent combination!
I'd love a video tour of your observing area loaded up with gear sometime, please. Thanks, Michael
@@michaeledmonds3027 thank you Michael! I’m planning on doing that however my place is horribly light polluted it’s a very sad sight with these beautiful telescopes 😁 I go to the mountains to photograph whenever possible
Light pollution seems to be the norm these days. I live in a Bortle 8 suburban area, plus it's heavily wooded. By the way, I appreciate your sense of style in your video staging. I look forward to whatever you may put together, take care. @the_space_koala
My 8” Orion has me pulling my hair out trying to collimate, now I know why, thank you 🙏
@@woody5109 is that a newt?
Yes, sorry.
Thanks for clarifying this. I am a visual observer and i like my Newt for DSOs as its brighter but i use my Mak for planetary observation as its more high def. My Newt as it has spherical mirror maybe bright but not as hidef at higher magnification as the Mak for planets.
The key is the image domain. The imaging speed is just a bonus.
Kézcsók! Nagyon jó volt a SpaceJunkie-s műsor ;-) És persze ez a videó is nagyon jó. Tiszta eget!
@@lv4218 köszönöm, neked is jó időt!
Thank you for this video.
Glad it was helpful!
You are to be commended for such a thorough and detailed review. It stands out as one of the best presentations on astronomical equipment. I must admit that at first I wasn't sure such an attractive presenter would get into such detailed scientific explanations. I do apologize for that. Your review ranks up there with the best informational videos available.
thank you for the kind words I'm glad it's helpful
Excelent video, congratulations
Thank you very much!
You have a 14" EdgeHD? Jealous! Very good video- I get a lot of questions around this in my local astronomy club- I'm going to send them to this video! Clear skies and have a great day.
@@AstroDenny thank you so much! It’s a point I’ve been wanting to make a video on for months but it’s just so hard to get it across! I tried my best but some comments on here and other forums let me know it’s still not clear enough to everybody
The C14 EdgeHD is awesome I love it, but it’s so big it’s a 2-person job to mount it 😁
very nice video. how about cameras ? and how about focusor ? which is better ? 1 1/2 '' or 2'' for astrophotgraphy ?
1.25” focuses are small and generally they’ll be good for planetary photography or some deep sky if you’re using a tiny sensor. At least 2” is recommended for medium to larger sensors.
Amazing! :)
Thank you!
So helpful. Following you.
Awesome, thank you!
I just ran into your channel - this video showed up in my recommended feed. I think you did a very nice job explaining the pros/cons of "fast" glass. As someone who owns an 8" EdgeHD, an 80mm triplet refactor and a 150mm newt (the same one you've got sitting on the shelf behind you), I can very easily relate to the challenges posed. Collimating my EdgeHD is extremely simple. The f/2.8 newt takes considerably more effort. I actually bought one of those Ocal devices to help me dial it in because the typical Cheshire tubes / lasers weren't good enough. Spacing and tilt are also far more difficult to manage on the f/2.8 scope than on the EdgeHD because the critical focus zone is so much smaller.
One point I'd like to bring up regarding your comparisons of drizzled vs non-drizzled data. Because, like everything else in astrophotography, there are _always_ compromises. In the case of drizzle, you're trading resolution for noise. In the 2x drizzle, you're spreading the noise of 1 pixel over 4 pixels. If we assume the same conditions you did in your examples, the 2x drizzled image will have about 4x noise than the non-drizzled one. To get relative parity in noise profiles between the two data sets, you'd need to take 4x as many subs (noise halves for every doubling of exposures) and ensure that data is well dithered. Anyway, dithering, drizzling, sampling are likely a topic for another video (or 3 😀).
Hey congrats on the nice scopes 😁 I 100% agree with your calcs and as far as I can tell this is exactly the information I tried to convey. As it’s extremely difficult to show SNF I showed the drizzled images as being “darker” and said you’ll need to expose 4x as long to get the same amount of light (which is the same as the same level of SNR). Which of the cases was this not clear in? Perhaps I should clarify!
So if time we remove shooting time from the equation because we will always shoot as long as it takes to reach the point of diminishing returns, then what really matters is choosing a focal length that perfectly frames the subject in order to achieve the maximum image quality, no matter what the focal ratio is. Is that correct? So a fast focal ratio only helps when we have a limited shooting time want to maximize the light collection within that time? And even with a limited amount of shooting time, it might still be better to optimize framing the subject over speed and a wide FOV?
Yes and no. You can get to the desired signal to noise ratio with any scope, given enough time. However, you’re still limited by aperture in the resolution that you can get. For example if you shoot a dense section of the Milky Way with a zoom lens vs a large telescope you’ll get very different results. A larger aperture will result in smaller stars and you’ll be able to get a better picture anyway.
Thank you for the very informative and helpful presentation! Is it a correct application that a focal reducer increases FOV but not image brightness? As a side note, your video creation abilities are seriously impressive 🙇♂️
Thank you for saying that! With a reducer you make your overall image brighter but you exchange it for resolution! So you get a smaller, brighter image of the same subject.
benefit of a fast telescope is the option to complete imaging in short time which is extremely good for:
- unstable weather area
- limited time for people to have a full time job
- easier to process frames taken in the same night.
I don’t think you watched the video 😁 the point is this benefit holds true only under certain circumstances
@@the_space_koala watched the whole vid. There are certain benefits as you mentioned for the fast telescope which I totally agree. These benefits can never be replaced by a slow telescope with more zoom power by mosaic or whatsoever, and I just pointed out how important those situations are.
I grew up using film (light sensitive chemical coatings deposited on a thin backing). Focal ratio seems less important to actually capturing an image in 2024, perhaps due to the rise of modern sensors, software and personal computers.
It’s not that much different on film IMO - you still have “pixels” - the little silver grains - even if nobody thinks about them that way. If you have a higher ISO film your pixel size goes up. In my opinion everything in this video is still valid when you put it into this context.
@@the_space_koala You are a more competent optical scientist than I-which is why I enjoyed viewing your presentation. However, we really did have to off-axis guide, using kludgy cold cameras, attached to precisely collimated scopes, on perfectly balanced and polar-aligned mounts, literally for hours on end, employing dodgy low-res film media of high compromise, pushed to near its limits-and it all could be ruined at any moment by unpredictable weather (no radar or weather satellites for us). Film needed expert chemical pushing during perfect development. Surely for us, if we used a telecompressor to achieve a faster F-ratio we weren’t wrong? I still believe this generation has it easy-which is great! Superb, high-sensitivity electronic digital images produced inexpensively in the backyard by today’s amateur astronomers are just mind-blowing :-)
Tackling this subject without ever talking about image sampling is quite a feat :). Today, CMOS cameras have very small pixels, so the game is not the same as when pixels were 24, 15 or 9 microns wide. If you have a 40cm aperture, at F/8, with 3.76 microns pixels, one pixel is 0.24 arc seconds of the sky. If you shoot in a typical site with 3 arc second seeing, one star image will be 12 pixels large, which is grossly oversampled. Even if you are in paradise and the seeing is 1 arc second, you are still oversampling a bit. But then when the image size starts to be around 1 arc second, you enter in a whole new domain where everything has to be perfect, focus, guiding, optical quality, etc... and there are very few amateur images with sub arc second long exposure images. In practive with today's camera and very long focal length, you don't loose any resolution if you bin 2, or if you have an F/4 telescope. Of course if you shoot with a small focal length (like a 50mm diameter objective), then at F/8 then one pixel is around 2 arc second, and if at F/2 it is 8 arc seconds, and provided the optics are not the limit (i.e. giving 2 arc second images on a wide field) then indeed there is a difference in resolution. Then there is the problem of the optical configuration. Some optical designs are only good on paper, and in practice almost impossible to collimate perfectly on a wide field (i.e. easier to use with a small camera then on a full frame camera). That's why Planewave CdK (with spherical secondary) or Celestron Edge are easy to collimate and all the "corrected" cassegrain systems are a bitch to collimate on a wide field. Then a refractor usually doesn't have these problems, except if they get out of collimation. I have an FSQ85 which was out of collimation and... there is no mechanical way to collimate it, then you are fried (going back to the manufacturer). You can spend quite a lot of nights on a decollimated RC system if you want to cover a full frame image. Even a newtonian, using a field corrector can give problems to use correctly (find the correct backfocus distance, and obtaining sharp images all over the field). Well, image sampling considerations are not the whole story.
I agree with what you’re saying on image sampling, however I did start the whole comparison in the video stating we’re focusing on ideal conditions and perfect seeing. The focus here is on light collecting capabilities only and it still turned out to be 20 minutes long 😁 not to mention you could do the same comparisons with tiny lenses so you’re only ever diffraction limited.
There will be a whole different video covering resolution/spatial frequencies which will be a better time to discuss this otherwise important point
In other words: Just get a scope with the FOV you want, get a camera with the resolution that you want. Aperture isn't everything, but it certainly can be important, depending on the rest of your system.
Faster = better unless you start to run into issues. If you have no issues with filters or edge sharpness, go fast.
faster is better given the same FOV. That just translates to a physically larger aperture
Hi and thank You for this remarkable effort! A well done video indeed. It is a very long time I'm trying to explain that "fast" in the digital era is a very uncorrect term to use... Whan we talk about a telescope otpic that was defined "fast" in the old emulsion era, we are now talking about a wide field telescope. In the past we need to fight the non linearity of the emulsion and the Swartzchild effect (also known as reciprocity failure) but today with linear acquisition devices such as CCDs and CMOS this is no more true. As to help Your nice explanation of the "f-ratyio myth" I would like to say something more. As You stated we are trying to collect as much photons as possible coming from the object of interest. Despite the system of acquisition I would like to point out exactly that: the energy coming from a subject is very dependant from the considered angle of the sky we are trying to measure. Every object as it's own energy emitted on a given angle and no f-ratio could chenge that! So, even considering a "perfect" system, we are totally dependant form the photometry! The photometry is the exact measurement of the flux of photons per second on a considered angle. This means that there is a maximum number of photons You can collect per second on a sigle pixel that define an angle (also called spatial resolution even if it is also correct to call it spatial sampling or angular sampling). For every object at a given wavelenght is already known how much photons You can collect... Really doesn't matter how "fast" you fill Your pixel since the ratio between the unwanted and desired signal is increasing proportionally. You fill Your well (pixel or angle of the sky) faster with signal (but is still a fixed flux that comes in in a poisson way and nothing can change that), AND unwanted noise (which is mostly sky-noise). the ratio between the two does not change at all. The only way to improve the signal considering a fixed angle of the sky is to collect more photons in the same exposure time and, as You explained, the only way is to increase the collecting area (bigger diameter)!
Hope this could help to better underestand why f-ratio on linear acquisition devices doesn't matter much if not at all. there are few exceptions to that but they are related to the digital system, not the physics that always rules. Try not to consider the system but what's the reality outside that is, in fact, what You are trying to measure or capture and You will better underestand what You're doing...
Thanks a lot and keep on this way! I appreciate it very much!
@@giovannipaglioli2302 thank you so much! Indeed this is a very hard point to get across, even if the video is super long with a million examples, based on many comments here and on other sites I have the feeling the message didn’t go through to many people.
@@the_space_koala keep going! It is something related to traditional photography mabye and, since it is a very diffuse and "known" topic, it is difficult to transfer to people... At the end is only a matter of considering thinks in a different way: we are in the realm of digital and we are no more "taking pictures", we are "making measurements" instead! If You try to look at things in this way mabye it could be easier for anyone to get to the real point. It is sometimes quite counterintuitive but. once acquired, it seems obvious.
Thank You!
For me the 'advantage' of a large diameter short focal length telescope is not ending up with a scope twice as long and having to have an extremely expensive mount and enclosure to house it in. An 18" scope with a 5 and a half foot tube is way better than it having an 11 foot tube and needing a motorized lift platform to use it!
You’re right! Though I’m pretty sure it would still have to be an expensive mount if you can put an 18” mirror on it 😁
@@the_space_koala Exactly! The mount would probably cost 10X what the optics would!
That's true! But in the end another fundamental aspect to consider is the resolution per pixel, even if You have a small optics but very high per pixel resolution, You will need an "expensive" mount to track well. For shure You will not need a "big" mount but a very precise one tha could be expensive too... 🥲
@@giovannipaglioli2302 you’re right in principle regarding the precision, however even low-end mounts tend to be quite precise at lower payloads. On the theoretical side, with the same exact image scale, it’ll cost you much less to find a mount that can carry 10kg vs 30kg, to the same precision
Nice video, but I'm not sure I concur that FOV is the driver of interest between fast and slow systems. If you take a fast system, and combine it with small pixels, you'll reach the limit of "seeing limited" resolution (equivalent to what a slow, more magnified system can give), but with many more photons per unit of time. Photons per unit of time roughly equates to SNR. I'd prefer to go fast and deep which will result in nicer images than what most folks are willing to spend with a slower optical setup. I still think faster optics has a significant advantage (even given the drawbacks - which I think you showed well....but which are manageable).
Putting small pixels on a fast scope will make the “photons per pixel” lower. It’s exactly the same as putting larger pixels on a slower scope. As long as the aperture is the same you’re not collecting more light!
@@the_space_koala I don't think we're talking apples to apples here. In your case 2, the apertures are the same diameter, and (while not exactly the case you presented), for an equivalent pixel scale (via different cameras), cropped to the same FOV, the differing f-ratios means the faster system will collect photons in each pixel more quickly as a function of the inversely proportional square f-ratio comparison. In your example 2, using the same camera is where I believe you miss the overall point. When the image scale is the same (via different cameras), the slower f-ratio telescope will take more time to gather the same number of photons in each pixel as the faster setup, so the obvious advantage goes to the faster system by way of less time required on sky for the same resolution (shorter exposures, or ability to gather more SNR over an equivalent time). People pair camera's (image scales) to sampling/seeing, so fast systems will outperform on a time basis. This seems to be an etendue issue more than a FOV issue, which is what is typically advertised in fast astrograph's marketing materials. Do you agree? What am I missing?
@@douglassummers8388 watch the video again
Womder if theres a way to slow down the video, it would help the voice to sound much better.
First time I have seen such a technical details of Astrophotography. I am a hesitant beginner and want to learn the science of the Astrophotography. Can you or anyone suggest some sources: books etc where I can learn all these concepts.
Hey I know this is not the answer you want to hear but honestly just UA-cam is a very good source. As weird as it is, I don't necessarily recommend books because in astrophotography technology is changing SO quickly, recommendations change every few years as well.
Thanks!!! Appreciate your response!!
I prefer faster focal ratios merely because I can get more data more quickly, but if the stars are seagulls in the corners and edges faster focal ratios are no benefit.
@@taras3702 the whole point of the video is that you don’t necessarily get data quicker
@@the_space_koala so basically you're just spreading the light out,so fewer photons are hitting a pixel so less snr , this really only scales out if you bin, or have larger pixel size in the slower scope vs faster scope
brava, tutto corretto, a volte queste scelte sono anche influenzate dal budget a disposizione e dal tempo che si può dedicare all'hobby.
@@fritzarken74 grazie! Ma il punto era esattamente che il tempo di esposizione non cambia necessariamente
The object of the exercise is to catch photons, fast optics catch more in the same time, given the same aperture
@@BigBadLoneWolf they catch more photons from a larger area of the sky. They catch exactly the same number from a given area. So what changes is only the field of view
I'm about buying a telescope under 10k so I choose 2 telescope edisla astra 100 or fotocart 76AZ who is better and please give me some suggestions....bytw I'm from india
@@RudraEditHub666 I don’t know these models but “az” sounds like it’s on an altazimuth mount. Please know that’s not adequate for astrophotography if that’s your goal you need an equatorial mount
Yes you are right altazimuth mount..I'm a 10th student and I love astrophotography but this is my interest not my journey like youtube so I think AZ mount is enough for my astrophotography
@@RudraEditHub666 good luck! 😊 clear skies
Good video and explanation. But it is "only" first partof the true. then comes the over and under sampling questions, which is also depending the focal lenght and the pixel size. Then comes the atmospheric/seeing conditions. So the astrophotography is not so simple... Anyway, feliratkoztam a csatornádra! :)
Yes but it’s a whole different discussion, nothing to do with how “fast” you can take a picture. Köszi 😁
Thank you for the explanation! Subscribed and looking forward to the Drizzling video! I seriously wonder though - are you the blonde twin of @AnastasiInTech ?
Thank you very much! I didn’t know her channel - thank you it’s interesting ☺️
@@the_space_koala if you do a sideye in you next video, we can gain certainty 😉
Nice.
Thank you! Cheers!
So many fast scopes sacrifice resolution. I’ll stick to my toa-130 at 7.7 with only a flattener.
The best telescope is the one you have and love! ❤️
I love my 250mm Newt at f/3.5 and 150mm Newt at f/3
Faster than my Lambo 🔥
I’m fast AF boi
Great video, but I confess I'll need to watch it again to understand it better.
I really appreciate you saying that. It is a complex concept to get one's mind around and I love if people invest time in understanding
its all about framing.. using an F4 scope that has your object only covering 20 percent of your frame will not be better than an f7 or higher that frames your subject in the majority of the frame.. its all about framing,, if you get larger diameter you need longer focal lenght unless you want a very large target in frame....the term aperature is king comes at a cost... i would rather do a mosaic and no binning with longer exposures, or exposer time, to catch the entire image if needed.. its all about details and a larger F ratio is king in that department.. aperature is king for visual thats for sure, you can always magnify with the lens,, but not needed for astrophoto.. as most things,, time and patience is the absolute king
I agreed with the first part until I read aperture is not needed for astrophotography. You physically collect more light (plus better theoretical max resolution). Sure time and patience are important but with a larger telescope you still can get the same exact image in a shorter time so surely it's better
I like it Fast - save time but can loose quality good to still use other ratio`s.. emm who you waving at? ;)
Ehehe it’s a habit - can’t talk with a firm hand 😁
f10 in bortle 7 is painful
Any scope is painful is bortle 7. If you had a faster scope it would also collect more light pollution not just signal. It’s difficult either way (i live in bortle 7 😁)
Telescopes are tools.
Unlike other tools like wrenches for example, you can't have the perfect telescope for every thing you watch. You must make compromises. Unless you can buy 50 telescopes, you have to decide what you will do the most with it, and select the best you can afford for that job.
Then you have to accommodate this choice in other situations. There is no better or worse. There is what you can afford to buy for the main job and you work with it.🎉
Don’t cite Celestron. Don’t bring example of RASA. Celestron is mass produced astro-equipment.
High quality fast telescopes are always better. If long FL is required, increase size of mirror.
the RASA is probably the best known and most accessible example of f/2 optics. In any case whether something is mass produced or not has no impact on its theoretical light gathering capability. The video contains reasoning that is generic (the examples mention no specific model of telescope) and can be applied to the cheapest or the highest-end products as well and provides a framework for people to evaluate their options.
It seems the lady must have missed optics classes. The only information was that the resoluion power is different between the two types of instrument for the same focal length. However, there is also the fact that the spots do not have the same diameter. the idea of coupling photocytes allows us to take into account the size of the Airy spot. Which would also accelerate the speed. The second point is the tolerance for maintaining focus. We talk about scale factor at this level. But certainly at F2 which spots 4 times larger than the diffraction allows, the resolution has become a detail. What matters is the speed of wide field photography. And yet cameras with small pixels now exist. No one seems to know of correctors that reach the diffraction limit. On the Japanese side, there are still companies thinking about it. The third oversight concerns optical aberrations. These kinds of details make amateurs disappointed with their telescope. Certainly having two distinct instruments allows you to combine the strengths of each for beautiful images. And in fact this seems to be the best advice in this video. With the different brands people get lost and traders are unable to understand and offer their customers products that are not compatible with their instruments because in optics you have to know that it is the piece that decides especially if you want to photograph the deep sky. What criterion imposes the choice of a corrector/reducer. Ss reduction of its focal length, or the characteristics of the baffle in a telescope which is sometimes calculated too accurately for a reducer, see which lets in stray light? The camera has its own characteristics given by the manufacturer, which greatly facilitates its best use. And if we're not mistaken about the camera, it's because we know our instrument. It's exactly like photography: The body and its Cmos become secondary to the lens which is placed in front....Thank you for the video
Thank you for your exhaustive comment! There’s definitely an infinite number of factors to consider and it’s simply impossible to discuss every subject in one video - this was already way too long as is. Hence the disclaimer (mentioned twice) in the video - that we’re focusing on equivalent light collecting capabilities only for this comparison. Pixel sizes and camera sensors are also a whole other topic but I got around that by saying we’re using the same exact sensor for the comparisons. The take home message of this video is to make people understand that fast optics trade off image scale for a wider field of view. This is true for any system in general and it takes a while for people to wrap their heads around this - astrophotographers, that is. A visual astronomer would find everything I said extremely obvious - only the diameter will matter 😁
Trades off everywhere😅
no free lunches :(
Yes, they are much better, if they are high quality astrographs. They cost a lot!😮
did you watch the video?
Yes, don’t agree with it. FOV effect compensation with 4 separate pictures is well known no solution. By the way, scientific CMOS are binned different way.
@@anata5127 the whole point of the video is "fast" just means a larger field of view in practice - given the same aperture - and you're saying the same thing. It's mostly aimed at people who put a reducer on a telescope thinking it will collect more light
Incorrect. Fast means more light per pixel. Ok? For example, galaxy emitting the same light, reduction of F distribute light to lesser numbers of pixels.
Now, people arguing that resolution is lost. Yes, but I don’t need resolution for vast objects. If I need resolution, then I will rather increase aperture maintaining lower F.
What low F scope do you have?
@@anata5127 You'd get exactly the same result with a slower scope using larger pixels or the same small pixels but resampled. The question is how big of a sensor area you're distributing X amount of light over. An f/number tells you nothing without specifying pixel size. And the original amount of light to be distributed is only a function of the diameter of the optics.
My fastest telescope is a RASA (f/2) and I also have an f/2.8 reflector.
You're a myth, five stars!
how nice, thank you for saying that
Very clear video. Very well explained with the animations.
thank you very much, glad you appreciate it