Do you really need dark frames? I am not so sure

[@AstroOnBudget]

Found this helpful? please subscribe

[@neilluoo1291]

Hi astroonbudget not that this relates to the video but on Canon cameras, is it better to use iso that is higher than 800 because I heard that below 800 you star seeing noise banding?

[@artyombeilis9075]

I'm new and so far I used sv105 and 224mc... With 105 I didn't see much difference but it isn't really camera for deep space. With asi224mc with relatively small sensor 1304x976 using 5s exposures I get a huge amount of hot pixels that show as many-many bright colourful dots (and finally lines due to drift) on the final image. Adding dark eliminates them entirely. So I think lots depends on sensor and camera and setup. Sometimes you don't really need darks and sometimes they are critical.

[@AstroOnBudget]

Agreed! It's really depends on situation. For some cameras, darks are absolutely necessary.

[@ronobjrII]

I could definitely see a slight improvement in gradient with darks. Is it necessary--maybe not...but it did make a slight improvement to my eyes.

[@stevethompson8154]

I used dark frames with my old Nikon 4th generation processor. I now have a Z6ii with 6th generation processor and tried using all calibration frames and more integration time and the results are night and day (pun intended). When the clouds clear I plan on doing more trials. I suspect since noise lives in the darkest areas and in under exposure we all might be better off getting more light frames vs. doing dark frames.

[@AstroOnBudget]

Yeah that makes sense, Steve. I think more lights is better idea than relying overly on darks.

[@deep_space_dave]

I'm glad you made this video as I have been scolded in the past in forums and social media about not needing to use dark frames like it was some kind of sacrilege or something LOL. If you have the following: cooled camera that has NO amp glow, bias frames, and flat frames then you don't need darks. Also to rid yourself of hot pixels, I use Pixinsight's cosmetic correction to remove those. So my frames are very clean without dark frames. I just take bias and flats at the end of each session and additional flats for each filter. If you use a camera that has amp glow then it is necessary but otherwise, with the latest sensors that are cooled it is no longer the case.

[@AstroOnBudget]

Haha. Sorry to hear that you were scolded in the past. Some people can be so passionate that it turns ugly. 100% with what you have said. If Amp glow is an issue, then sadly, there is no way around it. But I am tempted to skip darks if the camera doesn't have amp glow issue.

[@dadwhitsett]

I disagree. How difficult is it with a cooled camera to make a dark frame library? Easy peasy and they last forever. I contend they do make a difference with regards to eliminating hot and cold pixel artefacts. I understand the difficulties with a DSLR but with a cooled camera it is just too easy and I feel the benefit outweighs the negative.

[@crateer]

@@dadwhitsett absolutely, you just have to check whether darks are worth it for your camera or not. I agree with you, that with a cooled camera, they are not an issue as you can create a library once and basically forget about it! With DSLRs, you can easily get around having to take darks for multiple hours each night - simply dither! That alone usually gets rid of all hotpixels. Just check your data and compare with/without darks and decide for yourself. There is no right or wrong, it simply depends on your sensor For me, I tested with multiple Data sets from different nights and adding darks made literally no difference, other than slightly reducing my SNR with darks added. Since I dither anyways, there is no need to take darks for the sensor I use - huge time save and win for me!

[@azrayastrophotography2260]

Great comparison on the darks. I find that when I use my Zwo 294mc-pro camera it has apm glow showing up at the sides. Darks frames get rid of it for my set up and sky conditions anyway. My Zwo 2600 mc camera has no amp glow.

[@AstroOnBudget]

Yes, thats right Az! Its essential if your sensor produces amp glow. To be honest, I was really surprised. I thought darks would make greater difference.

[@dumpydalekobservatory]

Very interesting comparison I generally stich darks in if I have any but there has been times when I've never bothered & the images have turned out ok, I think dithering helps when capturing the subs as that's something I've never done until relatively recently since I started this hobby. I guess every little helps for those who pixel peep but I'm not one those lol.

[@AstroOnBudget]

I think you are absolutely right. Dithering must make up for darks to some extent. Like you, I am not a pixel peeper so that makes the difference very small.

[@mastroitek]

I use a cheap dslr (eos 2000d) and I agree with you. I normally include darks because I have somewhat of a dark library, though in winter I don't see much difference since the outside temp. is around 0°C where I shoot. With that said in summer they help giving more contrast and cleaning a bit the final image, but to be honest now that I use Topaz denoise I'm not sure if darks make a difference at all. They might still be useful on older cameras with weaker sensors but modern gear came a long way and darks don't seem to be essential anymore. Probably the number of lights has quite an impact on the final noise level, I normally prefer to get 1h of short exposure (60-90s) images and then at least 2h of longer exposure (3-4min), for my experience this gives DSS enough data to remove a good amount of noise.

[@Tony-Elliott]

Interesting test. What camera are you using?

[@AstroOnBudget]

The camera is ZWO ASI 2600MC

[@darrenreely]

The 2600MC is known for super low noise. I’ve seen a few experienced astro KZbinrs say they’re no longer using darks with these new gen CMOS chips. That could change if you’re doing longer exposures or not cooling the camera well. I wonder what the uncooked and cooled results would be imagining in the 90°f California desert.

[@MakeAMark]

Interesting, but there are a lot of details missing. What camera are you using (cooled vs not, amp glow, temperature), how many darks did you use, how long were the exposures... I do appreciated the efficiency of your videos, but perhaps a quick table at the start of the video could present those details. I think that would be helpful. As an aside, I think some DSLR cameras (like the Sony A7 series) regularly take "darks" every week or so automatically when you turn it on (or off), and automatically apply these to the raw images the collect. You can actually here the shutter open/close sound when it does it.

[@AstroOnBudget]

Agree Richard. I have omitted lots of info because I often keep my videos short. But that comes with the disadvantage. But here's the details: Camera used ZWO ASIO 2600MC (Gain 100) which doesn't have amp glow, the temp was -10, 15 lights of 180 sec exposure; 15 darks, 15 flats, 15 biases. Thanks for the tip. I didn't know Sony A7 had that function. Have you used it? How good does it work?

[@1pacarr]

It would be good to know if your evironment ie bortal status would effect this more. As a relative newbie to DSO not needing to take Darks would mean I have more time on other targets. My last session was 90 second frames so I had to save 45 minutes at the end of my session just for 30 Darks which could be 45 minutes of extra data on my target or on another target. I may try my next few sessions without Darks to see how it goes after viewing this.

[@bakerfx4968]

Are you using a dslr and camera lens? Or astronomy camera and telescope?

[@AstroOnBudget]

That would be so interesting to find out! Give it a go and see what happens. I suspect getting more lights data on the object would be better than shooting darks. But as i said, its very much debatable issue.

[@SpectreLX]

IMHO: I think that the need in dark frames is determined by the characteristics of the camera and the exposure modes... 🤷‍♂ I think that there are bias frames more important...

[@AstroOnBudget]

Agreed, biases are so important in whole scheme of things.

[@captaincook6666]

At 4 min subs I got lots of hot pixels even on cooled camera and with hot pixels removed in DSS- tested with and without and only darks got rid of them.

[@AstroOnBudget]

That makes sense! Darks can work miracles on some cameras.

[@tomreid7]

I couldn't see much difference at all, but what do I know, im a newbie.

[@AstroOnBudget]

You are not a newbie! Humble man

[@mhamen2]

There has been a lot of debate about whether using darks frames is necessary for newer sensors like the IMX533 and 571. I've never seen a definitive conclusion given. Personally, I find that dark frames are so mind numbingly easy to take and use (and reuse for many years) that I see no reason not to do it. People can process however they like and if they're happy with results then kudos to them. It takes a lot of time and effort to gather and process my data and I want every advantage I can to get the best result possible. That said, this is an interesting comparison. Do you have any details on the equipment and number of darks you used and how you processed the separate images? I think that is quite important when making a comparison like this.

[@AstroOnBudget]

Helpful comment, Mike. Thank you. Yes you are absolutely right, with cooled cameras its not really hard to take dark frames. Therefore, its not as inconvenient to include them. Makes perfect sense why you'd want to take every advantage. That said, this was just an experiment to encourage those who find it hard to take dark frames. In the end, to my mind, its better to take more lights and worry less about darks. Btw, equipment used was ZWO ASIO 2600MC (Gain 100) 15 lights of 180 sec exposure; 15 darks, 15 flats, 15 biases. Siril was used to both stack and autostretch.

[@UV-NIR-Thermal]

I thought darks get rid of the hot pixels, I know when I stack sometimes, I get a line of hot pixels of how many shots I took. Using a DSLR.

[@AstroOnBudget]

When you dither and enable hot pixel settings in the stacking software, there is almost no need for darks. But that is debatable. Many people think its essential to take darks.

[@datpudding5338]

Would love to try with some of my photos but I kinda don't get how to even apply them xD

[@AstroOnBudget]

:D If you don't feel like there is need then don't apply it

[@datpudding5338]

@@AstroOnBudget oh but there is definitely a need 😬

[@MrGuilletv]

With a old DSLR you really need darks!!

[@AstroOnBudget]

Yeah fair enough. Thanks for engaging. How do you manage to take darks with the same temperature as lights?

[@silverpassion3588]

You shld state how many % zoom in before it started to notice the dark frames are doing its job. And not saying fr "this angle" I can't see it. Its not "angle" its how %age zoom.