Astronomy Test - Stacked Short Subs vs 1 Long Exposure
Vložit
- čas přidán 5. 09. 2024
- I demonstrate that astrophotography thing you've never had time to test yourself!
No sponsors today :) Just doing it for the love of the game.
Support the channel by buying Dylan's Telescopes, Cameras & Equipment using the links below!
EQUIPMENT LIST
----
Celestron 11" Edge HD Telescope
USA : bit.ly/3JB95JE
Celestron RASA F2 11" Telescope
USA : bit.ly/3Jy4Ezl
International : amzn.to/3ABsVBz
QHY268M Mono CMOS Camera
USA : bit.ly/33u4k5w
Skywatcher EQ8Rh Pro Mount
USA : bit.ly/3gTdWJX
AUS : bit.ly/3dxdRv1
NexDome Observatory by Sidereal Trading (Australia)
www.siderealtr...
Celestron RASA 8" F2 Telescope
USA : bit.ly/3GTUwiU
International : amzn.to/3G4nlbP
Celestron NexYZ 3-Axis Smartphone Adapter
USA : bit.ly/351K2ki
International : amzn.to/3g20wLd
Celestron CGX Mount
USA : bit.ly/3BtPVCF
Lunt40mm Solar Telescope
USA : bit.ly/3sP8KMX
Sky-Watcher Star Adventurer 2i Pro Pack
USA : bit.ly/3uUVw3O
International : amzn.to/3ABLAwQ
ZWO ASI 174MM (Planetary / Solar Camera)
USA : bit.ly/3uXpEM8
International : amzn.to/3rXqJAj
ZWO ASI 120MM (Guide Camera)
USA : bit.ly/352iBXg
ZWO ASI 1600MM (Deep Space Camera)
USA : bit.ly/3HVtvg1
International : amzn.to/3rXqJAj
Celestron 0.7x reducer for 11" Edge HD
USA : bit.ly/3gTDJSy
Celestron 0.7x reducer for 9.25” Edge HD
USA : bit.ly/3sNyY2o
Celestron 9.25” Edge HD
USA : bit.ly/3LISupw
Canon 6D mkII DSLR
International : amzn.to/2CbiVER
Tokina 16-28mm Lens
International : amzn.to/3fdFJCo
Canon EF 16-35mm f/2.8L III USM Ultra Wide Angle Zoom Full Frame
International : amzn.to/3G9kjmP
Canon EF 100-400mm f/4.5-5.6L IS II USM Lens
International : amzn.to/3H9yPfA
Support my alcoholism by getting some Star Stuff stuff MERCH:
/ @dylanodonnell
Dylan’s links
----
PATREON : / dylanodonnell
WEB : deography.com
FACEBOOK : / dylanodonnell
TWITTER : / erfmufn
INSTAGRAM : / dylan_odonnell_
TWITCH : / dylan_odonnell_
APPLE MUSIC : / dylan-odonnell
SPOTIFY : open.spotify.c...
Intro Music - "Moving the Ocean (Aerologic Edit)" by Blastculture.
Otis McDonald - Treat Me This Way | Royalty Free Music - No Copyright Music | goo.gl/Fv1DxA | CZcams Music |
Mr Dylan, I love your videos but this appears to be confusing and misleading people.
What was the goal of the video?
I think it was meant to be 'what is the best SNR I can get for 5mins of accumulated exposure time and how does sub length figure in this?'. This is after all what we are trying to do? However, you didnt vary the gain (and therefore optimise the individual subs) so the actual outcome was excellent proof that failing to adequetly swamp the read noise really hurts SNR? It did not show that you can use short subs (with a cmos camera) and get essentially the same result as longer.
You have a heap load of followers who dont understand the difference between cmos and ccd and as a consequence they arent understanding that your video is showing the consequence of under-exposure, not the accumulated exposure time. It looks like you used the same gain for each which means you failed to optimise each sub by swamping the read noise adequetely in the short exposures hence the differences? A 10s exposure does not need to be underexposed if you adjust the gain and adequately swamp the read noise.
It also looks like you used a low gain setting because of the FPN and the data was narrowband? Optimum narrowband SNR with the ASI1600 is achieved with higher gain and shorter subs (it eliminates FPN and minimises glow - the drawbacks of a cmos camera).
If you had done this video with a fixed gain ccd I think the result would be valid but you didnt and people have concluded that they need to shoot > 300s subs if they want better SNR even if they have a cmos camera. This is almost always not the case with a cmos camera.
For the ASI1600 this link is a brilliant source of info:
www.cloudynights.com/topic/573886-sub-exposure-tables-for-asi-1600-and-maybe-qhy163/
In the interest of science I ran the same test and got very different results. I varied the gain to ensure I did swamp the read noise (so each sub was adequately exposed) but the stacked exposure time was also 5 mins:
10s x 30 @ gain 200, 60s x 5 @ gain 76, 300s x 1 @ gain 0
I actually gathered 2x as much data for each exposure and then picked the best frames for each (as we do when actually imaging).
10s SNR - 1.61
60s SNR - 1.52
300s SNR - 1.47
I was grappling with whispy high clouds so my suspicion is they interfered with the longer subs. The longer subs should have a marginally higher SNR not lower.
It gets really interesting when you measure FWHM:
10s - 2.544
60s - 2.812
300s - 2.882
Resolution is higher in the shorter subs.
Visually they are the same (to my eyes).
I am happy to share the source data with you if you would like to peer-review it.
As a fellow scientist (albeit one with lots of youtube followers!):
Please rerun this test in a more rigarous way.
Please publish/share the data so others can study and learn from it.
Please provide valuable guidance to your followers on how to get the best from their equipment.
I (and I suspect plenty of others) would be happy to participate in such a study so that we can all learn and progress in our quest to get better and possibly find something which isnt meaningless...
First, I like the fact that you pin this particular post Dylan :-). Second, are you a 'fellow' scientist? As for myself, I have a PhD but it's in social sciences which doesn't help me one bit when making CZcams videos about astrophotography :-). Keep uploading and enjoy the Tesla. Cheers
I spent about a year on this topic using a Hyperstar on a C11. And the outcome was? Nothing can beat LOTS of subs at LONG exposure times, it's that simple.
Thanks for doing this! I have been using the (totally unscientific) rule of using the shortest exposure necessary that captures the detail I am looking for. I think I'll try pushing exposures a bit longer.
Your life isn't meaningless to me..... I treasure your uTube presentations.
Excellent Test Dylan! Cleared some of my questions about exposure time!
Thanks Ray! An unproductive evening, but nice to fiddle and test!
I've been following Trevor for some time now and I'm glad I've found this channel because I love astronomy and everything related to it. On the other hand, it's been half an hour and I can't stop laughing at the on screen countdown.... Mate, you are better than half of those "stand up comedians" I've heard about.... Came for the stars, stayed for the laughs.
Haha hey thanks mate :)
“Everything is meaningless, and we’re all going to die” *Brofist* 😂👊
Boom 👊🏼
Translation: "That thing you're stressing about isn't really that important. Chill and enjoy life more because it doesn't last forever"
@@DylanODonnell 👊🏼
@@bitshuffler It's pure existentialism; we are free, in fact, compelled, to create our own meaning and we are all the better off for it. As we are when we are aware of our mortality. Life is made more precious for being finite. Dylan's is actually a very positive statement but it's challenging too. I silently cheer every time I hear it. :)
Nope. Everything is meaningful, and we're immortal. You'll see...
Dylan, great decompilation of speed vs weight. Yes it is hard to be patient and wait for those 300s exposures but it more than makes up for it in post processing headaches. I have done a few 900s exposures if every thing is perfect and sometimes they work and often they do not - in the not to distant past that was the only way to pull out the detail with real film. I don’t think I have ever seen such a simple straight forward comparison of quantity vs quality. Good work. Congrats on soon becoming a bonafide Astronomer! Soon...
Signal to noise ratio is the key. This will vary dependent on the ambient temperature of the light collecting chip. Ergo cooled chips improve the S/N ratio. Each chip setup will have its sweet spot for exposure too.
Definitely.. a good test for everybody to do themselves with their own gear! With my limited time I'm happy to use the compromised middle range but if I ever automated my obs more to leave it going all night - longer is better (that's what she said).
I've been meaning to test this for the last year. Thanks for spending your time so I don't waste mine, and for making it entertaining too.
Was the test a waste of time?
@@carpediem5008 I see what you did there.............
I'm really surprised the noise is less for the 5min exposure compared to 1 minute exposures, that a crazy interesting result, thanks for putting this together, Dylan. Also, congrats oncompleting your degree (I'm going to assume the next four weeks will be fine). I completed my astro degree 15 years back and it's scary how quickly I forgot the math, I couldn't derive the Hubble constant from first principles to save my life now. Which is fine because everything is meaningless :D
You made me buy a rasa. Damn you dylan, damn you.
much love from the usa.
Haha! I’ll let Celestron know about my commission ;) enjoy!
You get 2 commissions as I bought one also
Larry Marks 😂
Me too! STill in the darn box.
Remember Dylan, everything you say about the Rasa is Meaningful!
Great vid Dylan, Cleared up a few questions I had floating around in my head !!
Good to hear - thanks!
I don't have a mathematical proof of the application of this to the case in hand but, on statistical grounds, to reduce the signal to noise ratio by any factor k, you need the square of k samples (i.e. subs). So, with n subs adding the same light as a single longer exposure, you only get a reduction of the noise (of the shorter subs) by a factor of the square root of n.
So, I bet for longer exposures (not counting the other issues, like having to discard subs, as you explain).
It might no be the whole story (as pointed out by "james 12345"), but Math is on the side of not taking many short subs.
By the way, I have a PhD on Applied Mathematics ... and seven years of doing Astrophotography in the worst imaginable conditions; including not having that much money to spend ;)
Afterthoughts: I'm now thinking that it is more complicated than a simple application of the Central Limit Theorem (the one about the law of squares), because there are several sources of error in a photography ... food for thought.
Congrats for your degree!
Also thanks for all your videos, advice and tips, they are very valuable and an inspiration for many of us.
No worries, thanks Tizoc!
We had a similar debate a month or two ago at my astrophotography club meeting, and basically concluded the same thing w/ a similar graph: 20-30 exposures, anything more and you're wasting time. I tinkered with this last night; revisiting a target I'd shot a year or so ago with 30x3' exposures in Ha and Oiii, versus last night where I did 20x10' each of the same channels. Definitely more detail in the latter. It's a pain in the ass to have to toss a 10m frame, but putting in the time pays off.
Yeh it's great to do these little tests just to figure it out. The SNR return gets really minimal after a certain point so while I can appreciate the effort some people go to to get a gajizzlion hours integration time, I reckon their images would be best served by throwing out the worst 50% of that and stacking the best.
I'd figured it out early on when I went about manually stacking exposures in GIMP and as one adds each sub one has to set the opacity of that sub to 100/n+1 %... so by the time I got past twenty I was realizing that each next sub was only contributing 5% or less to the image - the marginal gains from each next sub starts to drop quickly at 20 and by 25 it is dropping really fast.
Great idea. I wondered the same thing myself. I used to do film astrophotography 20 years ago with hypered film. It wasn't an option to take a bunch of shorter frames and stack them in the darkroom. So most of my shots were 40 minutes or more. Thanks for doing this. My RASA will make life much easier!
Your music is fire bruh
Dont ever stop signing off with EMWAGD!. Still cracks me up after weeks
It makes sense to expose for as long as possible at the lowest ISO for the best signal/noise ration and indeed the integrated/accumulated ratio. I only have a Star Adventurer but have achieved 4 min exposures at ISO400 using a Canon 300 f2.8L series lens with satisfying results. Biggest challenge of all in Scotland is (assuming clear skies lol) is wind. Its the windiest country in Europe and getting flat calm evenings with cloudless skies is rare. But most of the country is Bortle Class 1 or 2 so we're lucky to have not much light pollution. Other folks in hot countries will be fighting noise due to heat. Others live near large light polluted cities so use filters to battle that. I guess you face the challenges based upon where you live and go from there.
I will never buy a Rasa. Until I do.
Dylan. You are one of the great philosophers of our time!
Life is meaningless etc. You are a deep thinker!
Great video about something that I’ve always wondered about. A few times I’ve used shorter/more exposures just for the star background and then layered it in - it’s a bit of a cheat. I, for one, like the “everything is meaningless” sign-off.
Hehe thanks Anthony!
Oh don't worry I managed to recovery the stars from the panel beneath the main bloated one.
Individual exposure time must be long enough to overcome the camera read noise, 10 s exposure has more read noise as 300 s , and the image dim detail will not be distinguishable from the averaged read noise.On the other end if the exposure gets too long you can saturate center of brighter stars and they will have less color. Also the max exposure time is highly sky background level dependent. If you have a lot of sky glow the the dim stuff won’t show even with a very long exposures, and your cameras useable dynamic range will be reduced with no dim detail benefit., the noise will be less overall but you still will not get the dim detail.To estimate optimal exposure length you can measure your sky glow an make your exposure long enough to overwhelm read noise and not saturate or decrease usable dynamic range, at a dark location is where you get the most benefit of long exposures.
Great information! I’m shooting 20 minute exposures mainly for the challenge. Yes, I do throw away some for tracking errors and satellites, but i enjoy the results on very faint objects.
Very interesting experiment, I never took the time to do it so thanks to have shared your results. I will certainly reconsider my exposure times when I am shooting in a dark sky with my 1600 (from 90s to may be 300s) but I have to check the impact of satellites and other flying objects. Congratulations for your degree Mister the Astronomer! All my best
Hey dylan, what are your thoughts on the sharpcap pro smart histogram feature? it analyses your sky brightness and your various sensor properties and gives you an optimal exposure time depending on your readout noise. Dr Robin Glover has a great presentation on it and it really does debunk the myth that you NEED long exposures.
Yep, this presentation is definitely worth the time ...
czcams.com/video/3RH93UvP358/video.html
Don't forget the missing gain section!
czcams.com/video/ub1HjvlCJ5Y/video.html
Sounds interesting - I'll check it out!
@@DylanODonnell Basically it obviously varies from sensor to sensor. The old school real CCD chips do need longer exposures to swamp the higher read noise they have. However, newer CMOS astro chips, like the 294mc pro, have way lower readout noise so the requisite time to swamp that noise is quite a bit lower.
you are right. long exposure is good in dark sky. staking cant extract extra information.
I’ve wondered the same thing, but I’ve always understood that no amount of short exposure subs will ever bring out detail of faint objects below the detection limit of the exposure setting used, no matter how far one pushes the gain. On the flip side of that, if I expose too long per sub, the bright stars become bloated or, perhaps it’s more a case of the ISO setting I use on my DSLR. Nevertheless, these two scenarios are, of course, way beyond the limits you are proposing. My principal limitation though is the maximum exposure I can achieve without auto-guiding, given that I don’t have a goto mount or high precision stepper motor gears. Still, I love that you put a concept such as this to the test which the rest of us are too lazy to try ourselves, and I also can’t see why any scientist, especially with a penchant for gazing skywards would ever disagree with your catchphrase, one that I’ve ashamedly retold whenever the opportunity arises.
wish you all the best with your studies towards the astronomy degree...i have learned a lot from you in your wonderful and entertaining u tube videos, thanks
Dylan puts the cool factor back into astronomy.
I’d love to see this same test using your good ol’ canon in prime focus.. I’m quite surprised by these results, but having a very high quality Astro cam may be the cinch pin 🤔
Technically the signal that you gather from the object is exactly the same if you do 1x300", 5x60" or 10x30", what changes is the amount of read noise from the camera.
While the SNR from the subject stays exactly the same, with 10 photos instead of 1 you have 3 times the read noise, with 100 photos you have 10 times the read noise and so on.
And on that note, did you change the gain on the camera while doing the shorter subs? Because you should push the gain really high to maximize the SNR with short subs, in that case the difference in overall read noise decreases.
Thank you! I’ve been meaning to do this test for so long!
Hell yes, good luck on those last submissions for your astronomy degree!
There's also an argument depending on dark site vs heavy lp site. It's better to take longer subs in dark site. In heavy lp it isn't as critical to take longer subs. Jon rista explained this on a CN post.
Sir u have where this is posted on CN? Interested to see it
Makes sense!
@@KB2AT I might be remembering incorrectly, but I thought it was Jon. I can't find the exact post but this pretty much explains it: dslr-astrophotography.com/long-exposures-multiple-shorter-exposures/
Punch line is in the Key Takeaways section
Secondary comment. You are not meaningless to me (but we are going to die). You provide good info and entertainment.
You seem like swell mate. Cool that you chill with my fellow Canadian, Trevor from Astrobackyard.
I do like the way you say it though. it seems like a little dig at certain types of people.
Cheers
Didn't realize you were going for an astronomy degree... jealous my good man! Cheers
Awesome idea Dylan. Reminds me of Forest Tanakas Andromeda off a static tripod. Cool video.
Actually when it comes to longe exposures, I can't yet because I don't have the mount, I watched this video to see if I get some sort of results for short exposures, I can only go max 4 second exposures untracked. So thank you for showing that it is possible with brighter targets to get some sort of results.
Dylan, looking at the sub's lack of noise and sky gradients I am assuming these are narrowband images? Because that does change the story dramatically (at least in my case).
The key difference between many short subs and 1 long sub is the added read noise from each sub. There's a key concept called "sky limited exposure" where the read noise becomes negligible (< 5%) compared to the shot noise. PixInsight has a handy script for calculating that for you, and Sharpcap Pro can give measurements to arrive at similar data.
For my Bortle 7 back yard, on the RASA8 with an ASI183MM the SLE for a broadband image is a few seconds. In Halpha, it is between 60 and 120 seconds. Going longer than the SLE is not going to improve your SNR significantly, more subs will (with diminishing returns yes). In practice I shoot 10-30s broadband subs because the data size becomes unmanageable, and 60-120s subs for narrowband.
"Really faint stuff isn't gonna come out with those shorter exposures" -> the sad truth is, when you are past the point where light pollution noise really dominates the read noise, longer exposures aren't gonna make it come out either. You just need a dark sky then. Since narrowband with a cooled camera has a much lower noise floor than a broadband sub, the story changes dramatically and getting faint glowing gas is possible from the city then with appropriate length subs.
Exactly when I'm fiddling with exposure times for my Tadpoles nebula.
Seems to be 2 minutes are a max for my Bortle 7 skies.
Thanks for the great video again! One reason I started to follow you is because the "Everything in meaningless..." line at the end! And because it's true! :)
It would be awesome if you could make video about GAIN and OFFSET settings in astrophotography cameras (ZWO ASI example), what to use and when etc. I think lots of people would find it useful. This is just a thought. :)
Good suggestion :)
Each camera pixel also has a "Full well capacity" number limit
Thank you carrying out your exposure experiment and share it.
It got me thinking, what about mixing and matching subs with sets of different exposure lenghts. Utilizing many subs for the sharpness and a few longer exposures to bring out the faint details. Would that work or have any benefits at all?
I’m shooting at f/7 so I have to have long exposures on some nebulae. It if can get the dome tracking and the weather is clear, one whole night at 600 seconds is the boss compared to 300 seconds. It makes a big difference.
Best quote on earth “everything is meaningless and were all going to die”
we're.
OK before watching: Some thing in the medium to long wins. 300 1 sec vs 1 300 sec the 300 sec sub crushes it.
My experience started with shooting without a tracker at 300mm on a DSLR where my max sub was 3.2 @ 1600-3200 ISO. I stacked 45 3.2's for about 2 min integration of M31, was happy but it is surpassed now by my single 2 min subs with Star Adventurer tracker and same camera/lens.
Longer is better up to maximizing data vs light pollution. Under ideal dark skies, longer is better up to the limit of tracking errors or risk of mishap (Kicked tripod anyone?!) Of course sensor heat up/ noise will also be a limiting factor.
OK, now I want to see Dylan's test results
Great topic, can't wait to see result.
I experience pretty good results with AZ Astrophotography and 30s exposures. I think the integration time and fast telescope is a very important factor and I have no other option than taking a lot of short exposures in this project but it's working :)
I've had good results on Narrowband targets in Avondale, Az @ 300" on my f/3 Newtonian
I always thought that you will get cleaner images with more subs of total exposure times but more detail with longer subs. Similar amount of photons hit the sensor with say an hours amount of exposure. With longer subs though more photons get to hit and register a signal, therefore you get the finer details as you would if running shorter subs.
No. You will almost certainly get better detail (resolution) from shorter subs. As long as you swamp the read noise adequately the difference between long subs and short is negligible. 1 hour of 10s subs with adequate gain (so you swamp the read noise) will give a near identical SNR to a single 1 hour sub.
You always make me laugh with “Everything’s meaningless and we’re all going to die” fwiw
Hey Dylan, some stacking softwares like Sequator offer the "Accumulation" option instead of the traditional "Averaging". It clearly won't help with noise reduction but it might bring up more detail. You should try that!
.....Man, thank you for making this experiment!! Highly appreciated!
So the rule is, the longest exposures you can do before things start to get overexposed.
If your tracker is up to it
longest before you have a satellite in it, or a plane, or a cloud and - of course - those damn UFOs!
Very helpful video. Thankx
😂..like the circle game symbol put in there lol 👌🏼
Thanks. very useful information with good examples
Very Good for your achievements ,big hug
Likely a very different result for non cooled cameras I suspect Dylan. If you are still bored perhaps repeat the test with a DSLR or leave the cooler off.
The offer to machine up a pier adaptor for the CXX is still open should you decide on fitting a pier into the observatory. The plate I made for myself will accommodate CGX and Skywatcher mounts. TBH I kind of regret getting the CGX rather than bleeding a bit more for the EQ8. I really miss EQMod and the ability to initialise the mount remotely without the need of the hand controller, not to mention the need for a remote desktop connection. Indilib rulze:)
It is quite simple Dylan - the information that you have in 30 sec 1 frame is just the same information you have in 30 sec 100 frames - you just get rid of the noise. If you make 10x300 sec which is the same total exposure time, you will have 10 times more information in a single 300 sec frame so 10 times more information in the final stack, but more noise because you have less frames.
All things being equal, there should be no difference. The expected counts in a given pixel is the same.
That makes lots of sense!! thanks for your content.
Is it obvious to say that you must use a star tracker?
Or I could potentially take 10x30 secs exposure without a startracker? and if it so, what the result will be? blurred or usable?
Great channel. Love the aesthetics of your brand too. Modern day astronomer lol. Liked, subbed & notified
Dylan, I was just thinking. This test is only accurate for Narrowband imaging with RASA. For broadband with RASA even after using HDR (Gain =0), a longer exposure like 300 sec / 5 min will blow up the frame with too much brightness. I do recommend HDR (Gain=0) for narrowband if you are going 300 sec exposures to keep the stars happy. Either way, good video to compare the results. Actually during my test I got better results 5 min/300 sec. My frame in narrowband was too bright for 600 sec/10min. So I end up setting myself at 5 min for now!
Yep good to run this test across all scenarios and targets.. but as a general test to see what the chip is doing it still holds. The non-stacked image in this case is the cleanest, which is a meaningful fact to know. In short - take the longest sub you can if you can.
Thanks for sharing your video. Congratulations on your Astronomy Degree!
Thanks mate !
More like this! 🤗
Dear Dylan, there is an error in your video in 7:20. I think that's 30x10 s not 10 X 30. But is really a good video. Thanks
ive been giving this alot of thought .thanks
The fuzz looks good tbh.
Always love your videos. Just wondering what the effect of dithering has on the signal to noise ratio in the final stack. Might it still be better to have more dithered subs than one longer single frame. Might this be another argument for slightly shorter subs with increased frames to stack. Keep em coming.
Great video Dylan.
I was anticipating mention of Starlink in the "In other news in the astronomy world" segment, but instead we got a look in Dylan's sweet new ride. Coincidence?
I have a QSI cc’d and the asi1600mm cool. On the whole I find 300 sec subs work well for most LRGB targets with the ccd camera on my f7 and f6 setups. I tend to go 180 sec for brighter targets like M31/M42 etc. for narrowband I used to go for 420 sec subs but found little advantage so now these are usually 300 sec too. With the cmos camera I’m still experimenting but rarely go beyond 180 sec on both narrowband and broadband. If I’m using Hyperstar at F2 I tend to stick to 120 sec subs. I reduce this to 60 sec for bright targets. Maybe I need to do more experimenting but hard to give up clear imaging nights unless it’s around full moon, but then longer broadband exposures are going to wash out!
By the way, I’m also a guitarist - so many of us Astro guys seem to be into guitar. Is there a link? Spooky!
Excellent video, Dylan! I knew I wasn't crazy for shooting 512 second subs. . .okay maybe I'd be more justified if I wasn't so heavily mounted but still!
Also you totally got me with that countdown toward the beginning. You clever bastard Lol!
Haha glad I helped justify your mounts existence ;) 👌🏼
@@DylanODonnell Small typo- should read undermounted. But eh I make do. . .sorta
Another issue is, its harder to "track" for longer exposure.
Wind? You have an observatory!
You seem to be under the delusion that the wind is completely blocked by the 'observatory'?
Has it escaped your attention, that there is an opening through which the wind can still have an effect on the telescope?
Lol!!! Some people have the complete inability to detect humour & sarcasm!
What are youi using for grease on the dome? I would try white lithium. You don't need much, it lasts a really long time, won't look as yucky until a ton of dirt accumulates. I use it on my rear motorcycle hub and change it when I get a new tire, 10,000+ KM. Just a thought.
Also, I do movember 24/7/365. Please make at least one video featuring the 'stache before you shave it!
Hey Dylan, any chance of doing a video on the CPWI Celestron/Planewave collaboration...? That software is truly aaaaaaaawesome-sauce. I reckon with a tutorial from you, the entire amateur astronomy community might migrate across to it...! 🤣
Hey Dylan, thanks for doing your informative videos. Perhaps we will see you at an astronomer's event in the USA.
Lucky imaging DSO's is the future.
2500 frames though, would much rather live stack with on chip dithering and main sensor guiding...
Interesting test, how about using just a dslr and prime lens without tracking ?
Depends on what ur aiming for and what telescope ur using.
But short answer very hard, i know ive tried lol
thank you for this. I think I owe you a beer or two at least !!
Nice video! Interested to see how this would turn out with an un-cooled camera as exposure length increases.
Just had my first experiences imaging from a dark(er) site vs my backyard. I was shocked to see I could go 300s vs the 30s from my yard. Uncooled unmodified DSLR, shockingly better results. Granted, it's been pretty cold here, around the freezing point so that undoubtedly helps also.
Another great test idea.
Great tips and info here for newbies like me.
Hi Dylan - pop quiz……in the past year, how many evenings have been considered suitable for astrophotographers? Trying to get a feel for cost vs ability to use.
Dylan, I go to your videos because you do them well, but also because I have the Rasa 8, which I want to learn to use. So, if you are a proponent of longer exposures (what's a "sub"?) you must be guiding, right? But everyone talks about the fast Rasa not requiring guiding.
Do you typically guide all the time? Do you recommend guiding with the Rasa? It's a significant question, as these scopes of course don't come with a guide scope or guide camera .
Yep! I'd say guiding is essential .. not simply for tracking, but for dithering. See my "dither or die" video :)
Dylan O'Donnell That’s a meaningful reply. I will watch that this morning.
How about copying 1 x 300 sec exposure 40 times and stacking ?? Often wondered this but never tried
What difference the dome have compared to just having the telescope outside?
Which Astronomy degree Dylan? Ive been hunting for one without too much calculus! The MSc(Astronomy) from Swinburne looks good. Was yours UNE?
I have just bought my first telescope and I am just wondering, could you stack 10 longer exposures with 20 shorter exposures (sorry if this is a stupid question I haven't stacked anything yet)
Gavin McGrath You can but you’re rejection algorithm might have a hard time determining what are outliers, not to mention having two separate calibration iterations(organization nightmare). Better to just stick to longer exposures and stack as many as you can. The only target I’ve ever needed short and long exposures on was M42 because of how bright the core is. In that case, you will stack the short exposures and the long exposures separately and combine the masters in post processing to recover the clipped core area.
In general yes. Also remember you can stack from different sessions, ISO, time of year, location. Even different gear. A lot of it depends on the software you use. It sometimes works better to stack in groups. Then stack the results of each group.
Start by determining the best time, ISO etc for you gear. For me it is 2 minute at ISO 1600. Longer just adds more light pollution and not enough object data. It also means I am just running the tracker, no guiding.
M31 is a good starter. Easy to find if you are anywhere north, and even just a few minutes of data gets you pretty excited. Even a single 10 sec sub at ISO 12800 looks cool, noisy but cool. I shoot a few high ISO subs to check framing and still smile at the result.
Hoping to kill it with Orion this season. Been dreaming of the horse head since a teenager.
Cheers
But isn't it for very fast focal ratio of f/2 even if the exposure time is 30 seconds, there will be much more signal captured than a slower focal ratio? So even if the RASA subs are 30 seconds, they still should capture very good amount of signal...
Hi Dylan, love your videos...especially your catch phrase!!! Anyone that gets upset over something so trivial can stick it where the sun doesn't shine. Speaking of the sun, I am planning on taking some pictures of the Mercury transit next week and wondered if you had any pointers? I have used my 127 MAK/CAS telescope to do some solar photography in the past, but I now also have an 8" Newtonian on an Atlas EQG that I have never used for that purpose. I am wondering which scope to use as I only have one camera presently. I may just set up both and switch the camera over after an hour or so to compare the images. Any input would be greatly appreciated. Anyway, keep up the good work, and try not to run over any Joey's with your new autonomous, battery powered, driving computer.
So, just to be clear: you are pushing ISO way up for the shorter exposures, right? So, basically, if the 300 second exposure is ISO 400, the 30 second exposure in your comparison is around ISO 3600. And if the 300 second exposure itself is already at 800 ISO, we're talking over ISO 7000 for the 30 second exposure... That's why the shorter the exposure, the more noise your samples have?
If that's the case, I'm not at all surprised. Stacking can compensate only so much for the data lost (and noise introduced) with extremely high ISOs, and I think nobody expects to shoot much above say ISO 3200 and get great detail, even with stacking. But I don't think the right takeaway is that the longer the exposure the better for the quality (especially since at any given ISO, a longer exposure means more noise than a shorter exposure; so a 5 minute exposure at ISO 3200 will have more noise than a 30 second exposure at ISO 3200). The question really would be what the optimal balance is, *while* you keep the ISO below the threshold at which the stacking no longer compensates for the increase in ISO. I own a Fuji X-T3, and I think that its threshold is likely round ISO 2000 already.
No, the Gain is set the 0 for all test exposures. The result holds that the best use of my 300 seconds is doing one exposure, and not several smaller ones (+ readout and dither time) for a lesser result. I do make the point of opting for the middle ground where necessary. :)
@@DylanODonnell gain is 0 means you didn't boost exposure in post, but what was the ISO setting you took the various shots at? I assume the ISO wasn't the same for the 30 second shot and the 300 second shot (it certainly doesn't look like those would be shot at the same ISO).
There is no ISO .. it’s a ZWO 1600 MM cmos .. not a DSLR. The images are auto stretch previewed in post for a balanced histogram.
Just one thought/question, wouldn't these results vary wildly based on the camera or sensor that you are using? Not all cameras are created equally. So my thought is, these results apply specifically to the 1600MM and will not or may not be the same say for a similar camera from a different manufacture. Just my .02 cents or .029 Australian rounded to .05 of course if I'm paying real cash.
EDIT: After watching further, you do kind of address the above.
I accept PayPal or BTC so you don’t have to round :) And yep .. agree that everyone needs to try tests like these with their setup .. but as a one-for-one test same chip, same gain, same temp, same time, same target, same site etc .. I think it’s a pretty elucidating example!
in between boring time? Sculptor galaxy is big and bright (for a galaxy) and at zenith at a very reasonable time!
Congratulations on completing your Astronomy degree! That’s a bid deal! How did you find the time with all that you do? I use the same ZWO 1600mm that you do and I’ve been striving to lengthen my exposures as much as possible with upgraded precision worms, hyper tuning & cleaning. I’ve been able to take 600s images with very good star eccentricity. I see where Sara Wager routinely takes 1800s images and records extremely faint details in nebulas. Would you agree that an 1800s exposure is a lot of imaging time to lose to some breeze or a few clouds?
Go and read this thread: www.cloudynights.com/topic/573886-sub-exposure-tables-for-asi-1600-and-maybe-qhy163/ The ASI1600mm is a cmos camera with very low read noise and variable gain. Using a typical astro CCD you need really long narrowband exposures to adequately swamp read noise. In cmos land this is detrimental to image quality and sub optimal. For broadband you will likely be around 30-60s and for narrowband 300s. If you can already get to 600s then you are easily exposed long enough (unless you are imaging at F10 or something crazy).
Awesome tips!
Just some of my observations: Nice kitchen, beautiful, but the sound quality....poor. Congrats on the completion of school, now you get to find a job.
Love the tag line, too bad for those who don't. You're not forcing them to listen to you.
Great channel you got, lots of good stuff, Star Stuff!
Yeh I really need an extension cable for the mic or a lapel. These videos are quick and dirty though 😆
My question is, when the benefit / "pain" ratio starts to be too low. Sure I belive, 5 min exposure time is better then 1min BUT are 2min (or 3min) maybe already 95% in quality of the 5min?? That would be my question. Because it can make a difference how long i need to sit outside in a cold winter night.. etc.
And please compare only the same amount of (stacked) exposures (no matter if 20, 30 or 40). THIS final pictures (2, 3 an 5 min) I would like to see side by side! THANKS!
Do Rasa scopes perform well in dampish conditions as we get in the uk.and the dry atmosphere in Australia must be great to work in.
Aled Jones it’s a drought here right now but I live in the subtropics .. it’s 84% humidity right now and hasn’t properly rained for about 6 months! I also live by the coast so all my gear oxidises in the salty air pretty quickly. I think what you see on this channel is a good indication of rasa performance in damp conditions but I can’t help you with the clouds!
Dylan O'Donnell its is good to view the sky after a lot of rain it washes all the pollution out of the sky but damp.but can I ask you one thing as I am planning on buying a scope over the next few years but I do rather the celestron Hd 14 inch.and love the new kitchen
I’d buy one if I could afford it :)
After reading 'james 12345' from 6 months ago, it made a lot of sense to me regarding gain and the inherent Amp-glow of CMOS cameras. I'm wrestling with this very question of optimum exposure length on a ZWO ASI294MC Pro camera that has significant Amp-glow at 300s. I'm curious what your response is to his comments. Trying to learn. Thank you! Cheers!
The amp glow shows up in your darks and master dark so the amp glow basically disappears when you calibrate your lights :)