Great video! I was an astro-imager (we called ourselves astrophotographers), in the late 80s thru the mid-90s. I quickly realized that I had to have my own darkroom to process my film and print my images because even the pro-level camera stores had no idea what we were shooting and what we were trying to see in the prints. We typically used exposures of up to 90 minutes for deep-sky objects. Now there's a sub for you! Planetary imaging was a different monster Eventually, stacking negatives became the big thing, pioneered, at the amateur level, by Tony Hallas and a few others. And, as Dr. Glover points out, it was extremely challenging. However, once it was done (and I was only ever to successfully stack 3 negatives), the results were amazing compared to a single negative. Image contrast was increased and film grain was minimized...just what the doctor ordered! That's my two cents! Now, back to the rest of the video!
Very interesting, thanks for posting this. I’ve been learning AP using SharpCap Pro and a cmos camera, I find the documentation for the software excellent and it explains a lot of this in more detail. As a long-time user of scientific multichannel detectors for spectroscopy, I’ve never understood why people use really long exposures when this clearly won’t help decrease the dominant noise source. This confirms that short exposures are sensible in many situations. Thanks again, glad I saw this.
Fantastic video, answered many questions that I had as a beginner/Intermediate Astro Photographer. I started using shorter subs with just above unity gain to get the step effect. The only downside is the number of exposures - they take up a lot of space and PixInsight takes a long time to do all the processing to go from lights,darks,flats,bias all the way up to a stretched image. This hobby is not for the faint of heart, but I love it.
In his history portion of this excellent lecture he left out film hypering .. that made a HUGE difference in the sensitivity of film in the 80/90's and was used by pretty much all serious pretty picture astrophotographers. Also stacking negatives was easy as long as you had a pin registration table and were using film larger than 35mm.
Thanks you for posting this, my wife purchased sharp cap after our free trial with our Altair gpcam 290C ran out. Totally worth it. 👌 I've used it for planetary imaging on Venus, Mars, Jupiter,Saturn & the Moon 🌙 I've recently started remotely operating my setup using Astroberry with Kstars/Ekos through a tablet,,but I allways plug the gpcam into the lap top first & use sharpcap polar alignment,,it's so fast to get " excellent " ,,I just couldn't manage without it .
Hello AstroFarsogaphy, and many many thanks (of course thanks to Dr. Glover too!) for posting this unvaluable scientific distillation of one of the most mythicized theme in astrophotography. This is a must-to-be-seen video for everyone who seriously want to face this journey.
This lecture was excellent! I am waiting on delivery of a ZWO CMOS camera, and had heard discussion about the subexposures being treated differently between CCD and CMOS. I now clearly see the difference. The discussion of ideal subexposure length based on sky conditions was incredibly helpful; Robin presented this beautifully. I love the formulae! Also, until now, I’ve never had a comfortable grasp of shot noise and an understanding that light pollution is, in effect, a form of noise. Thank you for increasing my knowledge. And I've just order SharpCap Pro!
Now I know why the Pixel 4XL that I own does SUCH a great job in low light & I noticed how the time to capture was variable so it's pretty obvious that the phone itself also has the ability to determine how many pics it needs to stack based on an individual time/place where the picture is being taken. Pretty cool to now understand what it has been doing all this time, stacking.
So the theme of this talk was effectively "Contrary to popular opinion, you don't need to be taking really long sub-exposures." Yet the formulas in this talk produce (at least with some of the newer CMOS cameras and the suggested C) what I would call...really long sub-exposure times. Consider the ASI533MC Pro (3.76mm pixel). At a gain of 100, which is just past the cliff in read noise and also happens to be unity gain where most people shoot, the read noise is 1.66e. Assume you have a 72mm f/4.8 refractor and can tolerate 5% noise increase (C = 10). That gives an exposure time of 68.6 seconds with an 80% QE. But now throw a 7nm Ha filter on there. The exposure time jumps to 979.5 seconds! But the QE isn't going to be 80%. It's going to be about 29%, using the color-weighted relative response at that wavelength for this camera. Now you're up to 2700 seconds or over 45 minutes per sub! Take it to a Bortle 2 site and it's 5126 seconds. If that's not a really long exposure, what is? My guess here is that I'm simply missing something or making a bad assumption somewhere or I should be a lot more tolerant with the % of acceptable noise increase until I get to something "reasonable" given the limits of my mount, risk of ruined frames, blown out targets etc. But if I was just going to guess at reasonable, why do the math in the first place instead of just saying "These modern CMOS cameras are getting so close to the ideal camera that they can shoot really long exposures in narrowband that are only really limited by practical concerns"? So am I missing something?
Well, the formula gives you the recomended exposure time, but ofc if the rest of the setup limits you then just go as long as the mount allows you to do safely.
Sharpcap doesn't do sensor analysis at all, at least for me, initially tells me that my light panel isn't bright enough, then can't cope with any alternative brightness, gets stuck in a loop until I give up. So glad I paid for a licence . I use this light panel to take my flat frames, without complaint from NINA.
There actually is a good reason to go beyond recommended exposure time in light polluted skies. I really don't want to stack 8 hours worth of 12 second exposures.
Awesome video. I have maybe a dumb question. Do the final single sub exposure time have any bearing on what is the mag value of the target? For example if I arrive at a number 45s subs, could that number be different for different targets - especially in terms of getting the details. For background noise, I totally understand the math and reasoning. Again - I am asking coz I don’t understand, not questioning the overall premise.
24:16 This is a real eye-opener! So my light pollution, scope and camera combo give 0.92 light pollution pixels, which means there is no point in cooling my ASI2600MC Pro at all!! 47:30 it also justifies my decision to never take longer subs than 35sec (as I don't bother with guiding) 🙂
@@fusion-music no nah not exactly but I did study some videos about how to record audio nicely. Can't say I've exactly studied though. Do plan to study it more
@@AstroFarsography Well, my complements to you. Your limited study has paid off. I would definitely like someone like you to be helping me with sound / video. I've had college/Uni trained guys do work for me and mess it up big time.
@@fusion-music thank you that's incredibly flattering! I'll tell you about it later when I get the chance but thanks for your kind words- it means a lot to me :)
Excellent presentation! Thank you very much. How does the formula change if I use a light pollution filter like an Optolong L-Pro? Do I still have to multiply by 3 or can/should I multiply by a higher number? I really learned a lot today, thanks again.
I so surprised that under light pollution quite short subs already give quite an optimal result. I calculated for Bortle 8, F/6.5, ASI ZWO 224mc with reasonable gain and with 5% extra noise I can work with 2s subs! That means I can use my mount in AZ mode without guiding etc. Even under Bortle 6 it is only about 2.6s... Fascinating!
Got a bit confused with exposure time, I'm using the ASI533MC Pro (which has read noise of 1.0e) with L-eXtreme narrowband filter which is 7nm I believe, and WO Z61 (f/5). what should my exposure time be in Bortle 9? I should multiply 0.5 by ..... ?
4 роки тому+1
This is an eye-opener! However, just to understand these exposure settings better, for example, I live in Bortle 9 skies my exposure should be 0.3*3(RGB Camera)=0.9 Seconds? At this exposure, I usually can't see anything. Another question would be, if I have 1 hour of shooting time, should I prioritize shooting 2 minutes * 30 exposures or 1 second * 3600 exposures? That sounds an awful lot of data to process and stack. Takes away the need of auto-guiding at all.
Nice talk. Still a bit confused about it since i find that somewhat long exposures still is king. Sure if you want a low noise picture by all means do so, but i think its more fun to get more information and slapping on more subs doesnt always provide that imo. (I dont know why this is pretty much the oposit to whats explaind here) but i feel that there must be data there in the first place! Orion at my area with 20sec up to 120sec subs provides about what you can see here. While my 300sec subs are AMAZING in comparision! Doesnt matter how many subs id slap on the lower exposure it never reaches the same ammount of details. When my cmos arrives i might rethink things a bit tho.
This was a true eye opener. I never really thought of 5 or 10 min exposures as excessive but they are totally pointless it seems. Sounds like you take 5 hours worth of 10 minute exposures it’ll be no better that if you got 5 hours worth of 1 minute exposures - and you don’t have the risk of having to throw away bad subs that cost you a load of precious time. It’s the total integration time that matters and not so much how you got it. The data that gives you those finer details we see in the 10 min shots is still there once you stack the 1 min shots. The only benefit to the long subs as far as I can tell is that it takes less computer resources to store and stack. Have you come across any other pros/cons in the long vs short exposure argument?
Given the same total exposure time, more exposures = more read noise and quantization error. Ofc in the real world there are many more factors to take into account and that limites maximum exposure time, but in an ideal world longer exposures are better.
I still think you need a go-to mount in many cases because there are many targets you can't even see with the naked eye. I've tried to get these targets manually with a Skyguider Pro single axis non-go-to equatorial mount and have had little to no success with that.
Thank you for sharing dr Glovers presentation. Why does he tell us that longer exposures are required from a dark site than a light polluted site to reduce noise? Seems like exposures could be shorter at dark sites. Warren
Hi! How can I focus my ccd camera in my telescope for deep sky objects? New on this and having a hard time... Planets are some how ok, but can't see deep sky objects
@@AstroFarsography Very much so, it always surpises me firstly how counter intuative this hobby is, secondly how hard it is to find good information backed up up by the secientific theory. Clear Skies.
@@ASBOmarc excellent glad to hear it. I'm in the process of testing this maths and claims as well. Hopefully will have some results soon :D clear skies to you too
@@AstroFarsography So based upon the equations presented by Dr Glover my particular imaging setup and location(f5 trplet APO, ASI 071 cooled OSC, STC Duo narrowband filter, 19.53 SQM) my optimum sub exposure time is 420-425 seconds, with a gain of 110 (unity is 90). For reason best left to myself the number 420 resonates more with me, so I will be be testing 420 second sub exposures as soon as its clear, which will likely be never (thanks UK for the cloud cover, it means a lot...). Clear Skies.
This is great stuff! Thank you for sharing. On a side note... How do you take into account the target being imaged? Meaning, what changes might you make if you are photographing a faint nebula vs a bright galaxy?
I didn't see anything about exposure for a RGB camera with a light pollution filter. Did I miss it? I saw 25x for 12nm and 100x for 3nm filters. Does he mean to multiple 3x for colour sensor and 25x or 100x depending on which light pollution filter you have? Thanks
Would this mean, that with a colour camera in bortle 5 (3x17s=51s) i would get diminishing return after about 100sub a 51s. so after roughly 90min i can go home? versus in bortle 3 data improves in the first 6h?
So what if I’m using a OSC (asi294) and a duo band filter (optolong L-enHance)? After calculating the sky glow electron rate with the sharpcap tool and the recommended sub exposure length for broadband, is there a multiplier for the bandwidth of that filter (roughly 30nm)? Based on his slide that says multiply by 25 for 12nm and 100 for 3nm, I calculate to multiply by 10 for 30nm. But, I don’t know if this applies to OSC or if I still need to multiply by 3 for a OSC sensor in addition to 10 for the filter. Thoughts?
Have you been to the tools.sharpcap website yet? There I would enter all the usual data, but then I would select 'monochrome' and input the total bandpass of the multi-pass filters (for eg 10nm Ha and 20nm OIII&Hb makes 30nm total) to get my sky electron rate. Then from there it's just adding that 'e' value into the formuale. Did that help?
What I can't grasp is dimension and something was done imprecise here: sqrt(R^2+TxP) So P is in electrons per pixel per second, times seconds is ok. R is in electrons per pixel. How can U add them up this way? Some invisible constant should be there to correct for that as now dimensions are awry.
with R (in fact all noise) as e/pixel if you T as secs.e/px then all the equations presented are consistent. At least I convinced myself of this ... the internet is bound to say no of course ...
it would appear so, and elsewhere too, but I think it can be reconciled by expressing noise as e/pixel and the time quantities as seconds.e/pixel. At least I convinced myself this is so ...
One comment on ADU values. This is an averaged value and while it is applicable to most nebulae, galaxies and bright reflection nebula tend to cause that formula to break down. This is because the core will inflate the value leaving fainter parts of the image at or below the read noise. M81 is very bright but if you want to get Holmberg IX you're going to need longer exposures than what that formula suggests. If you want to get the IGF, then you need even longer exposures. It is better to rely on your eyes than theory. Are you clipping the white values or are stars bloating from saturating the cells? If not taking longer exposures are better. The one caveat is with broadband and light pollution because you start to lose SNR with longer exposures. I find 3min is the max with an 11" aperture using LRGB.
Signal always goes up faster than noise, so stacking can still bring the signal out of the noise. The issue is when photons are not adding an ADU due to quantization problems while adding in the read noise. If your taking a sub, and an electron is added, it has to increase the pixel by at least 1 ADU in order for the signal to become apparent after stacking. It may be an absurd number of frames to make it stand out amongst the noise, so it still may not be practical, but at least its possible. This is usually a problem with older cameras or low gain settings. Also, this video is about the background signal of a photograph, not the targets. Longer exposures are usually worse for a multitude of reasons. The primary reason is guiding error. Another reason is environment and seeing.
17:15 I would have thought that "Full Well Depth" (or "filling the bucket") would be a con rather than a pro for long exposures. Am I wrong? Apart from this minor slip, excellent presentation and BRILLIANTLY documented. Many thanks for making it available to the public!!
Good morning, there is something I cannot understand. The Optimum sub exposure length means time. Time is measured in seconds (s). HOw come you can get seconds in a formula where the C parameter is unitless, R (Read noise) is in electrons e- which is the squared (e-^2) and P (e-/px/s). If I make the division I don't get second but something related to e-/px/sec. Am I wrong ?
No one said C was unitless. In fact, since it's defined as inversely proportional to a certain acceptable increase in noise (e-), we can infer its unit is 1/e-.
I agree the units throughout, not just here, appear to be mixed. Given C is related to E, the additional noise we'll tolerate and is stated as a percentage (by construction % is dimensionless) hence so too must be C, which is also implied by the equation for C. There are other equations where the problem manifests itself more dangerously. e.g single frame total noise = sqrt(R^2 + S.P) which appears to add e^2 + e/px. The only way I can reconcile is to have the time variables T, S with units of seconds.e/px and read noise R units as e/px. To me the latter makes sense (R can't be noise of the whole sensor can it surely?) and seconds.e/px is probably fine since we don't care what e/px is, just that all pixels are exposed concurrently (i.e optimum exposure for one px is same as for all px). Doing this, single frame noise has units of sqrt(e^2/px^2 + e/px/secs . secs.e/px) = sqrt(e^2/px^2+e^2/px^2) = e/px. The other noise equations are then consistent too with units e/px. Optimum time has units [C] e^2/px^2 / (e/px/secs) = secs.e/px, the same dimension as T and S above, and C is dimensionless as required. Well, I've convinced myself it's all ok, but I'm sure the internet will have its own thoughts ...
hello.i have a Celestron - NexStar 8SE Telescope - Computerized Telescope .I've up date it with: ZWO ASIAIR PROWiFi controller;ZWO ASI183MC Pro 20.18 MP CMOS Color Astronomy Camera. I've tried to use it and learn how work it for a few week....and I'm not understood it...i can not get a focus.I have seen all the videos on youtube. but I cannot understand anything about the correct use of this camera in my telescope. until today I have not been able to achieve the focus of any planet...what do you think :ill do some thing wrong?or that I need additional in my telescope. I'll appreciate your help
Hey Humberto. I'm not sure about the 8SE but make sure you rack focus in and out fully. Try finding the Moon when it's bright, if you can't reach focus at all at the extent of the focus tube but you can see a blob then you'll require extensions. Or you could try to get a reducer for the scope. For planets you might also want several focus extenders and a good quality Barlow
Interesting lecture indeed, however the formulas presented by Dr Glover fail to take into consideration the amount of subexposure time needed to capture faint nebular details which leads me to believe that this is incomplete, at best
Thank you for clear and informative video. I wonder about exceeding the recommended sub-exposure time (at 51:45 in the video). Although it doesn't reduce stack noise much further, it looks like it doesn't hurt. Can extending sub-exposure time be used to advantage in reducing the number of sub-frames? It seems to me that going to 2x the recommended sub-exposure can cut the number of sub-frames in half to achieve the desired integration time.
You will get better integration with more subframes. Total time is one factor, but total number of subs is also a factor. I've seen another presentation where they show the curve for number of frames, and that doesn't start to tail off until the 100-200 subframe range, and there is still a few % improvement out to about 400-500 frames.
I wish they could have let him have more time. Gain is important... The math doesn't work either for my camera- using his equation my ideal exposure time is 2 seconds. Is there any other video of this full presentation where he isn't rushed?
Love the naysayers hatin on the presentation. We would benefit from you demonstrating in open forum the scientific evidence supporting your position. Even better if you do a presentation to your local astronomy club and share it in youtube. Where's your alternate free software tool we all can benefit from?
Glad you enjoyed it. I personally use Astrophotography Tool to manage my session since it has mount control and platesolving in it and just feels nice to use.
@@Pipe-organ-recordings it's another control software. You can control the camera, set imaging plans (that also utilise different filters and exposures and binning per line on the exposure plan), can warm and cool the camera and can control the mount, playesolve, go-to/go-to++ and more
@@Pipe-organ-recordings unfortunately I can't give a proper comparison since I've never used SGP. I'll look it up and if it isn't too expensive I may buy a copy for review and comparison. I had an issue once with APT and I put it up on the forum and I had a response from the developer within 24 hours so my own short experience with APT support is positive.
Thank you so much Robin. This is a great video! I still have some difficulties to understand all the concepts he is developing. Using an ASI294MC with a William Optics GT81 (F/4,7), under a Bortle 8 sky, the subs exposure should be around 3 seconds (G120). This is very short. To get 4 hours integration time, I would need nearly 5000 subs! Does this make sense?
Yes makes sense. Though if you go to the tools site he mentioned you can get some better numbers. Example my 183m HyperCam has a QE of 84%, pixels of 2.4um, I live in Bortle 6, I use a 7nm narrowband filter and my read noise at the gain I use is 1.8, my sky electron rate from pollution is 0.13e/pixel/s R=1.8^2=3.24 P= 0.13 C=10 3.24/0.13= 24.92 24.92x10= theoretical optimum exposure of 249 seconds I'm not sure what effect light pollution filters have. I keep meaning to ask
@@AstroFarsography Wouldn't light pollution filters have the same effect as narrowband, but not to the same degree of course? There is a calculator out there for optimum subexposure lengths for the ASI183/ASI294, and you can input the bandpass of your filter, so if you just add the bandpasses up for your LP filter you can calculate a good exposure length.
@@creeper360bill that makes sense. I use the calculator a lot to work out exposure lengths but yeah, that's probably exactly that needs to do. I'll have a look about that when I'm home for sure
Thank you for sharing this. There is one thing I don’t get. He said that you don’t need a long exposure to get the best S/N ratio. But I still don’t know if I will get a better signal by going longer. In other words will I get the same signal from 60x1 min exposures as I will from a single 60 min exposure. Is longer still better regarding the final signal?
I believe he mentions that the best you can ever do is 1x60m, but she to risks associated with that we break it down. 60x1m is the best compromise. Going longer than your "ideal sub length" adds diminishing returns. If you want a cleaner signal use a different C factor also in your calculations :)
AstroFarsography thanks. It gives diminishing returns regarding noise reduction but does it give diminishing returns regarding gathering more photons fro dim areas? I also assume the formula gives the exposure time in seconds and not minutes? In other words is there a guide for how long we should go for our total exposure time? The reason I am asking is that using his formula, it seems that I should only be imaging for a few seconds which would be quite a radical change in my practice?
This should deserve more than only 9k views... thank you ! I actually have a defiltered canon 80D and either shoots with my 300mm lens or my 750mm/150Ømm, pixel size is 3.7um, and I've been looking on the internet the values for electron per pixel per second but impossible to find for the 80D. I would love to know what is the best exposure time for my camera, but don't really know how to do it. Do you have a video for this or any tutorial please ? thank you !
Hey there. So without knowing your bortle class I found the 80D seems to 3.33e/sec ( www.dpreview.com/forums/post/57468503 ) Following the formulae displayed on this video, and assuming a C value of 5% noise C=10 Read = 3.33 Light pollution= 3.7e/sec R^2 = 11.0889 11.0889/3.7 = 2.997 10*2.997 = theoretical optimum exposure of 29.97 seconds. But of course using the figures you can get from the website be links to will also give you a much better answer :)
@@AstroFarsography Thank you for that answer, and it was quick ! sorry I didn't precise, I am currently oscillating between bortle 4 and 3, ok for 5% I guess, as you said in the video, it's clearly acceptable. (btw the link on dpreview doesn't work for me, it's a 404 error). Pardon me my maybe stupid question, but if I take a picture with my lens, should I take the F/5.6 of maximum aperture it can get at 300mm ? And I also divide the P by 3 because my canon is a RGB sensor ? I tried to replot your values and found for bortle 3 at F/5.6, P=0.7 approx. so 75sec, and 40sec for bortle 4, 16sec for bortle 5. Am I correct ? =D (I did not divie P by 3 here) thanks !! excel sheet if you want : www.dropbox.com/s/n151wqfzid227xv/Optimum%20sub%20frame%20exp.xlsx?dl=0
@@guillaumefo no problem. Okay so now I know your skies let's try this again For your f/5 newt, using an 80D's quantum efficiency of 54%, read noise of 3.33e/sec (@ISO800) and bortle 4 skies with a colour sensor Your skies, sensor and scope combination using the calculator at www.tools.sharpcap.co.uk comes to a light pollution rate of 0.44e/sec (3.33^2/0.44)=25.2 25.2*10 = 252s (4 mins 12 seconds) Your camera lens is actually 480mm and f/8.9 because of the crop factor (1.6x). That brings your light pollution to 0.14e/sec (3.33^2/0.14)=79.2 79.2*10 = 792s (13 minutes 12 seconds lol) Also I have amended the DP review link now :)
@@AstroFarsography wow ok thank you a lot Robin ! (my 80D is defiltered, does it change the quantum efficiency ?) I don't really understand how you got 0.14 e/sec but as the camera and the lens won't change, I'll just note it somewhere. Ok it will be impossible to make a 13min exposure hahah, my star adventurer astrotracker mini is not precise enough... however I'll try 4min12 exposures with my telescope, thank you again !
@@guillaumefo ohhh I'm not Robin! Haha 😁 It would've got to 0.14e/sec due to the higher f ratio. Let's in less light for the same amount of time :) Hope it was helpful and good luck!
Read noise for example. Read Noise is an error when measuring a value. Sometimes you might read a little more, and other times you read a little less. When you read less, you have a negative noise.
Great lecture but I always skip through the first 5 mins or so. The speakers at these events must think we all read the list of chapters before we read a book lol. Thanks for taking the time to record it and post. :)
Missing Gain Section: ua-cam.com/video/ub1HjvlCJ5Y/v-deo.html
I just learned MORE in 53 minutes than the HOURS/days looking into this topic! Thanks a million!
this geezer is clued up
I am a BIg +1 on that!
Glad this was recorded, thanks and apologies for the shiny reflection of my head in the centre 😄
You and your reflection! 😄 Good to hear you're enjoying the recording!
LOL
Great video! I was an astro-imager (we called ourselves astrophotographers), in the late 80s thru the mid-90s. I quickly realized that I had to have my own darkroom to process my film and print my images because even the pro-level camera stores had no idea what we were shooting and what we were trying to see in the prints. We typically used exposures of up to 90 minutes for deep-sky objects. Now there's a sub for you! Planetary imaging was a different monster Eventually, stacking negatives became the big thing, pioneered, at the amateur level, by Tony Hallas and a few others. And, as Dr. Glover points out, it was extremely challenging. However, once it was done (and I was only ever to successfully stack 3 negatives), the results were amazing compared to a single negative. Image contrast was increased and film grain was minimized...just what the doctor ordered! That's my two cents! Now, back to the rest of the video!
Just saw this. He was able to explain exposures better than anyone else could. Thanks for uploading this.
One of the most useful presentations I've seen for astrophotography so far. Thank you so much.
Really glad you reccorded and shared this! Thank you so much for this.
Best video in this subject I've seen so far
Very informative, thanks for recording this. I have the trial version and will no doubt be purchasing Sharpcap.
Very interesting, thanks for posting this. I’ve been learning AP using SharpCap Pro and a cmos camera, I find the documentation for the software excellent and it explains a lot of this in more detail.
As a long-time user of scientific multichannel detectors for spectroscopy, I’ve never understood why people use really long exposures when this clearly won’t help decrease the dominant noise source. This confirms that short exposures are sensible in many situations.
Thanks again, glad I saw this.
Brilliant presentation. Thank you for recording and most importantly thanks to Robin for the talk.
Fabulous talk. It was a privilege to drop in unannounced to such a well balanced & informative presentation.
Outstanding presentation!!! Thanks a million for all those detailed explanations!
Fantastic video, answered many questions that I had as a beginner/Intermediate Astro Photographer.
I started using shorter subs with just above unity gain to get the step effect. The only downside is the number of exposures - they take up a lot of space and PixInsight takes a long time to do all the processing to go from lights,darks,flats,bias all the way up to a stretched image. This hobby is not for the faint of heart, but I love it.
Thanks for grabbing this info!! one of the best videos on the subject !!!
In his history portion of this excellent lecture he left out film hypering .. that made a HUGE difference in the sensitivity of film in the 80/90's and was used by pretty much all serious pretty picture astrophotographers. Also stacking negatives was easy as long as you had a pin registration table and were using film larger than 35mm.
The missing section on choosing the right gain can be found here - ua-cam.com/video/ub1HjvlCJ5Y/v-deo.html
New to Astrophotography and this was exactly what I've been trying to figure out. Thanks!
Thanks you for posting this, my wife purchased sharp cap after our free trial with our Altair gpcam 290C ran out. Totally worth it. 👌 I've used it for planetary imaging on Venus, Mars, Jupiter,Saturn & the Moon 🌙 I've recently started remotely operating my setup using Astroberry with Kstars/Ekos through a tablet,,but I allways plug the gpcam into the lap top first & use sharpcap polar alignment,,it's so fast to get " excellent " ,,I just couldn't manage without it .
Absolutely brilliant presentation !!! I will recommended this lecture to all my friends. Thanks a lot and all my best from Chile.
Hello AstroFarsogaphy, and many many thanks (of course thanks to Dr. Glover too!) for posting this unvaluable scientific distillation of one of the most mythicized theme in astrophotography. This is a must-to-be-seen video for everyone who seriously want to face this journey.
Robin Glover is awesome, thanks for posting this
Didn't know that these were recorded. I missed out on this talk at the PAS 2019. Thank you so much for capturing the talk.
Yeah. It was only this one and Damian Peach's I got to record. You're welcome though I'm glad it's been a good thing for you that it was recorded :)
Thanks a million, Ruzee! This is great info.
Excellent discussion of the behavior of quantum light effects. This is useful in other optics discussions.
Thank you for recording and sharing this!
Excellent presentation, thanks for sharing.
44:21 Is the most important part about getting the correct exposure.
Thank you so much for this awesome lecture! It's really helpfull
This lecture was excellent! I am waiting on delivery of a ZWO CMOS camera, and had heard discussion about the subexposures being treated differently between CCD and CMOS. I now clearly see the difference. The discussion of ideal subexposure length based on sky conditions was incredibly helpful; Robin presented this beautifully. I love the formulae! Also, until now, I’ve never had a comfortable grasp of shot noise and an understanding that light pollution is, in effect, a form of noise. Thank you for increasing my knowledge. And I've just order SharpCap Pro!
I'm super glad this lecture helped you, Robert! Did you catch the missing gain section as well?
Enjoy your camera and new software as well!
Thanks Ruzeen! Great value.
Thanks, Ruz, great info.
This is brilliant! Thanks so much for sharing.
That was excellent information! Thank you
This was extremely interesting!! Thank you. 🙏
An excellent tutorial which every astro person must know properly
This is really really helpful. Thank you for sharing!
Now I know why the Pixel 4XL that I own does SUCH a great job in low light & I noticed how the time to capture was variable so it's pretty obvious that the phone itself also has the ability to determine how many pics it needs to stack based on an individual time/place where the picture is being taken. Pretty cool to now understand what it has been doing all this time, stacking.
It also dithers! Pretty darn amazing.
Just found this. Thank you for posting. Much appreciated. 👍
Excellent!!
Thank you for the video!
Thanks for the vid. Great info.
Thanks for this.. great talk and well recorded
You're welcome. Thanks for your comment as well :)
So the theme of this talk was effectively "Contrary to popular opinion, you don't need to be taking really long sub-exposures." Yet the formulas in this talk produce (at least with some of the newer CMOS cameras and the suggested C) what I would call...really long sub-exposure times. Consider the ASI533MC Pro (3.76mm pixel). At a gain of 100, which is just past the cliff in read noise and also happens to be unity gain where most people shoot, the read noise is 1.66e. Assume you have a 72mm f/4.8 refractor and can tolerate 5% noise increase (C = 10). That gives an exposure time of 68.6 seconds with an 80% QE. But now throw a 7nm Ha filter on there. The exposure time jumps to 979.5 seconds! But the QE isn't going to be 80%. It's going to be about 29%, using the color-weighted relative response at that wavelength for this camera. Now you're up to 2700 seconds or over 45 minutes per sub! Take it to a Bortle 2 site and it's 5126 seconds. If that's not a really long exposure, what is?
My guess here is that I'm simply missing something or making a bad assumption somewhere or I should be a lot more tolerant with the % of acceptable noise increase until I get to something "reasonable" given the limits of my mount, risk of ruined frames, blown out targets etc. But if I was just going to guess at reasonable, why do the math in the first place instead of just saying "These modern CMOS cameras are getting so close to the ideal camera that they can shoot really long exposures in narrowband that are only really limited by practical concerns"? So am I missing something?
Take a look at ua-cam.com/video/RzvmfNReMOM/v-deo.html about narrowband
Well, the formula gives you the recomended exposure time, but ofc if the rest of the setup limits you then just go as long as the mount allows you to do safely.
Yeah, I live in Bortle 2, and I was looking at over 2 hour subs for narrowband...
Great video! Thanks for sharing!
Thanks for sharing. This video helped me !
Thank you very much, great video, very useful.
Awesome video!! , thanks for sharing.
Awesome share..very valuable..thank you so much. 🙏
This was really interesting. I'd like to see a presentation with updated information with reference to some of the much more modern astrocams.
Amazing cristal clear analysis. Now I really need to do measurements with Sharpcap :)
This was very useful although some of the maths was bit over me it's great to have these references to where to begin with exposure times.👍
This is a fantastic video! Are the charts used available somewhere? I'd love some PDF's of them.
Thanks for sharing!
Very interesting talk. Thanks for sharing. I am going to consider Sharpcap!
Glad you found it useful! It's very good. Did you find the missing gain section okay? It's in the description :)
@@AstroFarsography Yes, thanks for the link!
wonderful presentation
Sharpcap doesn't do sensor analysis at all, at least for me, initially tells me that my light panel isn't bright enough, then can't cope with any alternative brightness, gets stuck in a loop until I give up. So glad I paid for a licence . I use this light panel to take my flat frames, without complaint from NINA.
There actually is a good reason to go beyond recommended exposure time in light polluted skies. I really don't want to stack 8 hours worth of 12 second exposures.
Awesome video. I have maybe a dumb question. Do the final single sub exposure time have any bearing on what is the mag value of the target? For example if I arrive at a number 45s subs, could that number be different for different targets - especially in terms of getting the details. For background noise, I totally understand the math and reasoning. Again - I am asking coz I don’t understand, not questioning the overall premise.
24:16 This is a real eye-opener! So my light pollution, scope and camera combo give 0.92 light pollution pixels, which means there is no point in cooling my ASI2600MC Pro at all!!
47:30 it also justifies my decision to never take longer subs than 35sec (as I don't bother with guiding) 🙂
Perfect recording. Sound is great, unlike many presentations.
Thanks very much! I was determined to get the audio nice and it's good feedback that it's good !
@@AstroFarsography Have you studied sound?
@@fusion-music no nah not exactly but I did study some videos about how to record audio nicely. Can't say I've exactly studied though. Do plan to study it more
@@AstroFarsography Well, my complements to you. Your limited study has paid off. I would definitely like someone like you to be helping me with sound / video. I've had college/Uni trained guys do work for me and mess it up big time.
@@fusion-music thank you that's incredibly flattering! I'll tell you about it later when I get the chance but thanks for your kind words- it means a lot to me :)
Excellent presentation! Thank you very much. How does the formula change if I use a light pollution filter like an Optolong L-Pro? Do I still have to multiply by 3 or can/should I multiply by a higher number? I really learned a lot today, thanks again.
Fantastic!!! By any chance, Did the Dr. Make his slide deck available?
I so surprised that under light pollution quite short subs already give quite an optimal result.
I calculated for Bortle 8, F/6.5, ASI ZWO 224mc with reasonable gain and with 5% extra noise I can work with 2s subs! That means I can use my mount in AZ mode without guiding etc. Even under Bortle 6 it is only about 2.6s...
Fascinating!
Got a bit confused with exposure time, I'm using the ASI533MC Pro (which has read noise of 1.0e) with L-eXtreme narrowband filter which is 7nm I believe, and WO Z61 (f/5). what should my exposure time be in Bortle 9? I should multiply 0.5 by ..... ?
This is an eye-opener! However, just to understand these exposure settings better, for example, I live in Bortle 9 skies my exposure should be 0.3*3(RGB Camera)=0.9 Seconds? At this exposure, I usually can't see anything. Another question would be, if I have 1 hour of shooting time, should I prioritize shooting 2 minutes * 30 exposures or 1 second * 3600 exposures? That sounds an awful lot of data to process and stack. Takes away the need of auto-guiding at all.
Nice talk.
Still a bit confused about it since i find that somewhat long exposures still is king. Sure if you want a low noise picture by all means do so, but i think its more fun to get more information and slapping on more subs doesnt always provide that imo. (I dont know why this is pretty much the oposit to whats explaind here) but i feel that there must be data there in the first place!
Orion at my area with 20sec up to 120sec subs provides about what you can see here. While my 300sec subs are AMAZING in comparision! Doesnt matter how many subs id slap on the lower exposure it never reaches the same ammount of details.
When my cmos arrives i might rethink things a bit tho.
This was a true eye opener. I never really thought of 5 or 10 min exposures as excessive but they are totally pointless it seems. Sounds like you take 5 hours worth of 10 minute exposures it’ll be no better that if you got 5 hours worth of 1 minute exposures - and you don’t have the risk of having to throw away bad subs that cost you a load of precious time. It’s the total integration time that matters and not so much how you got it. The data that gives you those finer details we see in the 10 min shots is still there once you stack the 1 min shots. The only benefit to the long subs as far as I can tell is that it takes less computer resources to store and stack. Have you come across any other pros/cons in the long vs short exposure argument?
Given the same total exposure time, more exposures = more read noise and quantization error.
Ofc in the real world there are many more factors to take into account and that limites maximum exposure time, but in an ideal world longer exposures are better.
nice to him in real time ( video) after talking to him for a few years...
He's a really nice guy. Spoke with him a bit at his stall
I still think you need a go-to mount in many cases because there are many targets you can't even see with the naked eye. I've tried to get these targets manually with a Skyguider Pro single axis non-go-to equatorial mount and have had little to no success with that.
so in Bortle 5 sky f/5 i need 130 subs for L around (x3)=400 for RGB and (x25-50)=3500-6500sek for narrowband!! is my calculating right..
Thank you for sharing dr Glovers presentation. Why does he tell us that longer exposures are required from a dark site than a light polluted site to reduce noise? Seems like exposures could be shorter at dark sites. Warren
Very interesting
Hi! How can I focus my ccd camera in my telescope for deep sky objects? New on this and having a hard time... Planets are some how ok, but can't see deep sky objects
How do i figure out the shot noise of my asi6200MM?
Excellent lecture, thansk for posting this.
Not a problem. Hope you found it useful!
@@AstroFarsography Very much so, it always surpises me firstly how counter intuative this hobby is, secondly how hard it is to find good information backed up up by the secientific theory. Clear Skies.
@@ASBOmarc excellent glad to hear it. I'm in the process of testing this maths and claims as well. Hopefully will have some results soon :D clear skies to you too
@@AstroFarsography So based upon the equations presented by Dr Glover my particular imaging setup and location(f5 trplet APO, ASI 071 cooled OSC, STC Duo narrowband filter, 19.53 SQM) my optimum sub exposure time is 420-425 seconds, with a gain of 110 (unity is 90). For reason best left to myself the number 420 resonates more with me, so I will be be testing 420 second sub exposures as soon as its clear, which will likely be never (thanks UK for the cloud cover, it means a lot...). Clear Skies.
@@ASBOmarc been a couple clear nights. Have you managed to test yet? :) Clear skies to you too
This is great stuff! Thank you for sharing. On a side note... How do you take into account the target being imaged? Meaning, what changes might you make if you are photographing a faint nebula vs a bright galaxy?
Galaxy’s aren’t bright!
I didn't see anything about exposure for a RGB camera with a light pollution filter. Did I miss it?
I saw 25x for 12nm and 100x for 3nm filters. Does he mean to multiple 3x for colour sensor and 25x or 100x depending on which light pollution filter you have? Thanks
Would this mean, that with a colour camera in bortle 5 (3x17s=51s) i would get diminishing return after about 100sub a 51s.
so after roughly 90min i can go home?
versus in bortle 3 data improves in the first 6h?
0:48 - start of Dr. Robin Glovers speech
3 years of personal discovery covered in 50 minutes.
So what if I’m using a OSC (asi294) and a duo band filter (optolong L-enHance)? After calculating the sky glow electron rate with the sharpcap tool and the recommended sub exposure length for broadband, is there a multiplier for the bandwidth of that filter (roughly 30nm)? Based on his slide that says multiply by 25 for 12nm and 100 for 3nm, I calculate to multiply by 10 for 30nm. But, I don’t know if this applies to OSC or if I still need to multiply by 3 for a OSC sensor in addition to 10 for the filter. Thoughts?
Have you been to the tools.sharpcap website yet? There I would enter all the usual data, but then I would select 'monochrome' and input the total bandpass of the multi-pass filters (for eg 10nm Ha and 20nm OIII&Hb makes 30nm total) to get my sky electron rate.
Then from there it's just adding that 'e' value into the formuale.
Did that help?
What I can't grasp is dimension and something was done imprecise here:
sqrt(R^2+TxP)
So P is in electrons per pixel per second, times seconds is ok. R is in electrons per pixel. How can U add them up this way? Some invisible constant should be there to correct for that as now dimensions are awry.
with R (in fact all noise) as e/pixel if you T as secs.e/px then all the equations presented are consistent. At least I convinced myself of this ... the internet is bound to say no of course ...
At t=45min in the video we have this equation:. Sqrt[n*R^2+T*P] . Is this a square root of a mixed unit quantity?
it would appear so, and elsewhere too, but I think it can be reconciled by expressing noise as e/pixel and the time quantities as seconds.e/pixel. At least I convinced myself this is so ...
One comment on ADU values. This is an averaged value and while it is applicable to most nebulae, galaxies and bright reflection nebula tend to cause that formula to break down. This is because the core will inflate the value leaving fainter parts of the image at or below the read noise. M81 is very bright but if you want to get Holmberg IX you're going to need longer exposures than what that formula suggests. If you want to get the IGF, then you need even longer exposures. It is better to rely on your eyes than theory. Are you clipping the white values or are stars bloating from saturating the cells? If not taking longer exposures are better. The one caveat is with broadband and light pollution because you start to lose SNR with longer exposures. I find 3min is the max with an 11" aperture using LRGB.
Signal always goes up faster than noise, so stacking can still bring the signal out of the noise. The issue is when photons are not adding an ADU due to quantization problems while adding in the read noise. If your taking a sub, and an electron is added, it has to increase the pixel by at least 1 ADU in order for the signal to become apparent after stacking. It may be an absurd number of frames to make it stand out amongst the noise, so it still may not be practical, but at least its possible. This is usually a problem with older cameras or low gain settings. Also, this video is about the background signal of a photograph, not the targets.
Longer exposures are usually worse for a multitude of reasons. The primary reason is guiding error. Another reason is environment and seeing.
17:15 I would have thought that "Full Well Depth" (or "filling the bucket") would be a con rather than a pro for long exposures. Am I wrong?
Apart from this minor slip, excellent presentation and BRILLIANTLY documented. Many thanks for making it available to the public!!
Do's this arm make my body look small?
Good morning, there is something I cannot understand. The Optimum sub exposure length means time. Time is measured in seconds (s). HOw come you can get seconds in a formula where the C parameter is unitless, R (Read noise) is in electrons e- which is the squared (e-^2) and P (e-/px/s). If I make the division I don't get second but something related to e-/px/sec. Am I wrong ?
Assuming you're using the right figures, you should be seeing the optimum sub length in seconds depending on your percentage noise tolerance.
No one said C was unitless. In fact, since it's defined as inversely proportional to a certain acceptable increase in noise (e-), we can infer its unit is 1/e-.
I agree the units throughout, not just here, appear to be mixed. Given C is related to E, the additional noise we'll tolerate and is stated as a percentage (by construction % is dimensionless) hence so too must be C, which is also implied by the equation for C. There are other equations where the problem manifests itself more dangerously. e.g single frame total noise = sqrt(R^2 + S.P) which appears to add e^2 + e/px. The only way I can reconcile is to have the time variables T, S with units of seconds.e/px and read noise R units as e/px. To me the latter makes sense (R can't be noise of the whole sensor can it surely?) and seconds.e/px is probably fine since we don't care what e/px is, just that all pixels are exposed concurrently (i.e optimum exposure for one px is same as for all px). Doing this, single frame noise has units of sqrt(e^2/px^2 + e/px/secs . secs.e/px) = sqrt(e^2/px^2+e^2/px^2) = e/px. The other noise equations are then consistent too with units e/px. Optimum time has units [C] e^2/px^2 / (e/px/secs) = secs.e/px, the same dimension as T and S above, and C is dimensionless as required. Well, I've convinced myself it's all ok, but I'm sure the internet will have its own thoughts ...
@@RomanPearah by implication from the equation for C, it must be dimensionless ...
hello.i have a Celestron - NexStar 8SE Telescope - Computerized Telescope .I've up date it with: ZWO ASIAIR PROWiFi controller;ZWO ASI183MC Pro 20.18 MP CMOS Color Astronomy Camera. I've tried to use it and learn how work it for a few week....and I'm not understood it...i can not get a focus.I have seen all the videos on youtube. but I cannot understand anything about the correct use of this camera in my telescope. until today I have not been able to achieve the focus of any planet...what do you think :ill do some thing wrong?or that I need additional in my telescope. I'll appreciate your help
Hey Humberto. I'm not sure about the 8SE but make sure you rack focus in and out fully. Try finding the Moon when it's bright, if you can't reach focus at all at the extent of the focus tube but you can see a blob then you'll require extensions. Or you could try to get a reducer for the scope. For planets you might also want several focus extenders and a good quality Barlow
@@AstroFarsography thank you i'll do.i appreciate your help.
Interesting lecture indeed, however the formulas presented by Dr Glover fail to take into consideration the amount of subexposure time needed to capture faint nebular details which leads me to believe that this is incomplete, at best
Thank you for clear and informative video. I wonder about exceeding the recommended sub-exposure time (at 51:45 in the video). Although it doesn't reduce stack noise much further, it looks like it doesn't hurt. Can extending sub-exposure time be used to advantage in reducing the number of sub-frames? It seems to me that going to 2x the recommended sub-exposure can cut the number of sub-frames in half to achieve the desired integration time.
Same question!
You will get better integration with more subframes. Total time is one factor, but total number of subs is also a factor. I've seen another presentation where they show the curve for number of frames, and that doesn't start to tail off until the 100-200 subframe range, and there is still a few % improvement out to about 400-500 frames.
I wish they could have let him have more time. Gain is important... The math doesn't work either for my camera- using his equation my ideal exposure time is 2 seconds. Is there any other video of this full presentation where he isn't rushed?
What camera do you have? You have to do a lot of math to get the right result, and its easy to do it wrong.
Love the naysayers hatin on the presentation. We would benefit from you demonstrating in open forum the scientific evidence supporting your position. Even better if you do a presentation to your local astronomy club and share it in youtube. Where's your alternate free software tool we all can benefit from?
Great information.... I just wish he would have talked a little faster.
Wow!! I felt that I was at an information auction :)
He explained that he had a limited amount of time available.
UA-cam Playback Speeds of 0.75, 0.5 and 0.25. are available.
UGH?!? he glosses over the ONE thing I wanted most... the gain/offset values??!
He uploaded it separately. Can be found here
ua-cam.com/video/ub1HjvlCJ5Y/v-deo.html
Hey this was excellent. Do you use sequence generator and sharp cap?
Glad you enjoyed it. I personally use Astrophotography Tool to manage my session since it has mount control and platesolving in it and just feels nice to use.
AstroFarsography what is that? I use SPG.
@@Pipe-organ-recordings it's another control software. You can control the camera, set imaging plans (that also utilise different filters and exposures and binning per line on the exposure plan), can warm and cool the camera and can control the mount, playesolve, go-to/go-to++ and more
AstroFarsography how would you compare it to SPG and how is there support? Do they directly answer people who buy their software’s questions?
@@Pipe-organ-recordings unfortunately I can't give a proper comparison since I've never used SGP. I'll look it up and if it isn't too expensive I may buy a copy for review and comparison.
I had an issue once with APT and I put it up on the forum and I had a response from the developer within 24 hours so my own short experience with APT support is positive.
Thank you so much Robin. This is a great video! I still have some difficulties to understand all the concepts he is developing. Using an ASI294MC with a William Optics GT81 (F/4,7), under a Bortle 8 sky, the subs exposure should be around 3 seconds (G120). This is very short. To get 4 hours integration time, I would need nearly 5000 subs! Does this make sense?
Yes makes sense. Though if you go to the tools site he mentioned you can get some better numbers.
Example my 183m HyperCam has a QE of 84%, pixels of 2.4um, I live in Bortle 6, I use a 7nm narrowband filter and my read noise at the gain I use is 1.8, my sky electron rate from pollution is 0.13e/pixel/s
R=1.8^2=3.24
P= 0.13
C=10
3.24/0.13= 24.92
24.92x10= theoretical optimum exposure of 249 seconds
I'm not sure what effect light pollution filters have. I keep meaning to ask
@@AstroFarsography Wouldn't light pollution filters have the same effect as narrowband, but not to the same degree of course? There is a calculator out there for optimum subexposure lengths for the ASI183/ASI294, and you can input the bandpass of your filter, so if you just add the bandpasses up for your LP filter you can calculate a good exposure length.
@@creeper360bill that makes sense. I use the calculator a lot to work out exposure lengths but yeah, that's probably exactly that needs to do. I'll have a look about that when I'm home for sure
@@AstroFarsography how do you get a p of 0.13 ?
Thank you for sharing this. There is one thing I don’t get. He said that you don’t need a long exposure to get the best S/N ratio. But I still don’t know if I will get a better signal by going longer. In other words will I get the same signal from 60x1 min exposures as I will from a single 60 min exposure. Is longer still better regarding the final signal?
I believe he mentions that the best you can ever do is 1x60m, but she to risks associated with that we break it down. 60x1m is the best compromise. Going longer than your "ideal sub length" adds diminishing returns. If you want a cleaner signal use a different C factor also in your calculations :)
AstroFarsography thanks. It gives diminishing returns regarding noise reduction but does it give diminishing returns regarding gathering more photons fro dim areas? I also assume the formula gives the exposure time in seconds and not minutes? In other words is there a guide for how long we should go for our total exposure time? The reason I am asking is that using his formula, it seems that I should only be imaging for a few seconds which would be quite a radical change in my practice?
This should deserve more than only 9k views... thank you !
I actually have a defiltered canon 80D and either shoots with my 300mm lens or my 750mm/150Ømm, pixel size is 3.7um, and I've been looking on the internet the values for electron per pixel per second but impossible to find for the 80D. I would love to know what is the best exposure time for my camera, but don't really know how to do it. Do you have a video for this or any tutorial please ? thank you !
Hey there. So without knowing your bortle class I found the 80D seems to 3.33e/sec (
www.dpreview.com/forums/post/57468503 )
Following the formulae displayed on this video, and assuming a C value of 5% noise
C=10
Read = 3.33
Light pollution= 3.7e/sec
R^2 = 11.0889
11.0889/3.7 = 2.997
10*2.997 = theoretical optimum exposure of 29.97 seconds.
But of course using the figures you can get from the website be links to will also give you a much better answer :)
@@AstroFarsography Thank you for that answer, and it was quick ! sorry I didn't precise, I am currently oscillating between bortle 4 and 3, ok for 5% I guess, as you said in the video, it's clearly acceptable. (btw the link on dpreview doesn't work for me, it's a 404 error). Pardon me my maybe stupid question, but if I take a picture with my lens, should I take the F/5.6 of maximum aperture it can get at 300mm ? And I also divide the P by 3 because my canon is a RGB sensor ? I tried to replot your values and found for bortle 3 at F/5.6, P=0.7 approx. so 75sec, and 40sec for bortle 4, 16sec for bortle 5. Am I correct ? =D (I did not divie P by 3 here)
thanks !!
excel sheet if you want : www.dropbox.com/s/n151wqfzid227xv/Optimum%20sub%20frame%20exp.xlsx?dl=0
@@guillaumefo no problem. Okay so now I know your skies let's try this again
For your f/5 newt, using an 80D's quantum efficiency of 54%, read noise of 3.33e/sec (@ISO800) and bortle 4 skies with a colour sensor
Your skies, sensor and scope combination using the calculator at www.tools.sharpcap.co.uk comes to a light pollution rate of 0.44e/sec
(3.33^2/0.44)=25.2
25.2*10 = 252s (4 mins 12 seconds)
Your camera lens is actually 480mm and f/8.9 because of the crop factor (1.6x). That brings your light pollution to 0.14e/sec
(3.33^2/0.14)=79.2
79.2*10 = 792s (13 minutes 12 seconds lol)
Also I have amended the DP review link now :)
@@AstroFarsography wow ok thank you a lot Robin ! (my 80D is defiltered, does it change the quantum efficiency ?) I don't really understand how you got 0.14 e/sec but as the camera and the lens won't change, I'll just note it somewhere. Ok it will be impossible to make a 13min exposure hahah, my star adventurer astrotracker mini is not precise enough... however I'll try 4min12 exposures with my telescope, thank you again !
@@guillaumefo ohhh I'm not Robin! Haha 😁
It would've got to 0.14e/sec due to the higher f ratio. Let's in less light for the same amount of time :)
Hope it was helpful and good luck!
38 mins in. Negative noise? Anyone know what that is? Know, please, not guess! BobUK.
Read noise for example. Read Noise is an error when measuring a value. Sometimes you might read a little more, and other times you read a little less. When you read less, you have a negative noise.
Great lecture but I always skip through the first 5 mins or so. The speakers at these events must think we all read the list of chapters before we read a book lol. Thanks for taking the time to record it and post. :)
Haha. It was good. I missed on some of it cus I was focusing on the recordings but I'm happy with how it came out. I was so nervous about the sound 😂
its an adult education template., you'll notice this format a lot in adult ed: Tell you what your going to learn. Teach. Tell you what you learned.