Astro Photo Question

Capt_Spaulding

Folks,

After almost two decades living under seriously light polluted skies, the wife and I made a trip out to a pretty remote part of west Texas. I haven't done any "real" astrophotography in a very long time as after a lot of research and questioning, I took a gamble on a Rokinon 20mm f1.8 and bought it for the trip.

We found a particularly dark site and I exposed about 20 frames. Looking at the results, I'm torn. On one hand I like the way the unenlarged images turned out. On the other hand, when I examine them at 1:1 the stars are not the pinpricks I was hoping for. I'm attaching a couple of representative pictures. I someone with more experience than I (that will be a large universe) wouldn't mind taking a look at them and tell me if I should be able to expect more out of the camera (a D750) and how I might go about getting it. Many thanks in advance.

mhedges

I think your shutter speed is too long. @spraynpray told me of the "200 rule" where 200/focal length is max shutter speed for best astro. So that would be 10 seconds in your case.

spraynpray

Hi Capt, mhedges has it right, but it does depend on what you want to do with your pictures. If you want to post smaller files on here and facebook, you can use the 500 rule but if you want to blow them up on your wall, IMHO you have to do the 200 rule with all of its attendant difficulties.

Another option is to take as long an exposure as you want to get the foreground how you like it (perhaps with a little light-painting) then use a tracking head which not only allows you to shoot very long exposures, but multiple very long exposures then do noise averaging. All of the things mentioned above are attainable for the normal photographers like us, you don't have to be a rocket scientist or to have many thousands invested in gear to do it.

Oh, and don't worry about the D750, it is the best camera Nikon have made so far for that type of photography. Changing to another body will not greatly change your results with all else remaining the same.

Another thing to consider is that the D750 is full frame so the edge aberrations are always going to be a problem, and the wider angle lens, the worse they are too.

I don't mean to put you off, I love night photography but it is certainly challenging!

Your images are good!

spraynpray

I've had a good look at your images now and the milky way shot is just a little off critical focus. Focusing is for me, the most difficult part of the process.

ggbutcher

Nice, really. I take it you didn't stack any of them... ?? Would be interested in your exposure/ISO settings.

dpreview has a decent Astrophotography forum, very knowledgeable people who don't mind giving good basic advice.

Capt_Spaulding

Thanks very much all for the comments. I really appreciate them. The focus issue is troubling. Even with the screen brightness at its highest value, live view doesn't provide much on which to focus. Most of the stars are outside the sensor's ability to see. In this case, the brightest thing in the frame is Jupiter. But it's not a point source.

The exposure times on flickr are about 3 seconds longer than the actual exposure times. I tried a few using a three second exposure delay, but just to make sure, I used the "black hat" shutter method and waited several seconds after the shutter opened to move the hat.

I did a bit of this stuff back in the film days with an SRT-101 and Royal Gold 1000. I really enjoyed it. My biggest hurdle now is finding dark skies. These were 600 miles from home.

@ggbutcher I took most of them at ISO 1250 between 15 and 20 seconds. No stacking.

One quick question. Looking at these in 1:1 raw should stars be pinpoints, or is there bleed over to adjacent receptors? I guess what I'm asking is, "how do I know when I get it right?"

Going back out there as soon as possible. Thanks again.

daveznspace

With that combo you should be able to go up to 25 seconds on shutterspeed. However, I always take 2-3 seconds of any formula as I've found the formula's exact numbers tend to produce slight star streaks.

I'd say it's either not quite focused right or tripod movement or mirror slapping or you're using your finger to fire the shutter on the camera and not with a remote or on a timer or even better a remote with mirror up.

Get it focused then remember to turn off AF. Either zoom all the way in on a star in LV and get focus or a very far away light or the moon if it's out or better yet do it during the day and mark the spot on the lens.

Now here's what I'd do in your situation with a D750. I use both a D750 and D700 to shoot astro and I still prefer the d700 for it .

The D750's sensor gives you a lot of leeway with options.
1. The ISO on these shots are too low, you barely have it at 1600 and in areas with little or no light pollution you're going to need more exposure to compensate.
2. You can raise the ISO up to say 6400 and it will look better but since it's a D750 it's almost better to shoot it at ISO 100-400 than in post push your exposure slider up until you like it. The reason and/or benefit to this is the ISO-INVARIANT sensor will let you push exposure in post without adding additional noise but doing this gives you more DR, better colors and contrast.

3. Stop it down!!!! With a good sensor like the d750 has you really don't need to shoot it wide open. Stop it down to 2.8 will give you sharper stars and you have plenty of leeway with that camera and I wouldn't even hesitate to get to f4. I lost my astro lens in a car wreck and short on short lenses and I have one of the worst nikons lenses ever (the version 1 of the 24-120mm 3.5) and I'll use that all night long and get good results.

I live up in the middle of nowhere wyoming by yellowstone and there's no noise pollution and it was very obvious to me that it's tougher in really really dark spots.


What I've noticed more than anything on not-quite-sharp stars is tripod movement. Like my 3 legged thing POFS tripods are so crappy that I doubt they could hold a paperclip steady, I can go back and look at the pix and you can actually see the shots getting lower and lower which means the head is slipping. Now, If I use my manfroto tripod (with costs less than the crappy 3lt ones), it's rock steady, the difference is easy to see.

gl

mhedges

daveznspace said:

With that combo you should be able to go up to 25 seconds on shutterspeed. However, I always take 2-3 seconds of any formula as I've found the formula's exact numbers tend to produce slight star streaks.

25 seconds with a 20mm lens? Really?

daveznspace

mhedges said:

daveznspace said:

With that combo you should be able to go up to 25 seconds on shutterspeed. However, I always take 2-3 seconds of any formula as I've found the formula's exact numbers tend to produce slight star streaks.

25 seconds with a 20mm lens? Really?

yes.

look at the excellent mw calc

https://lonelyspeck.com/milky-way-exposure-calculator/

mhedges

daveznspace said:

mhedges said:

daveznspace said:

With that combo you should be able to go up to 25 seconds on shutterspeed. However, I always take 2-3 seconds of any formula as I've found the formula's exact numbers tend to produce slight star streaks.

25 seconds with a 20mm lens? Really?

yes.

look at the excellent mw calc

https://lonelyspeck.com/milky-way-exposure-calculator/

Neat link thanks!

I'm going on a beach trip in August. May try to rent a fast wide angle and try some astro.

Mark

tc88

I say forget about the calculators. Shutter speed is easy to decide for self. You will have plenty of time, just expose at different speeds, and zoom in and decide how much star trail is acceptable. Different people make different trade offs between star trail, noise and coma. I typically shoot 8s on 20mm on D850, but I know I'm on one end of the spectrum and have more tolerance of noise than trail.

Having said that, since this is facing southeast, and I don't really see movement of stars at the center, and Capt_Spaulding is using D750, I can't tell whether it's trail or not. Assuming this is not trail which Capt_Spaulding can rule in/out by experimenting with the speeds, then here is my guess.

The top left looks to be coma which wasn't bad for f/1.8. The top right looks a little different than the top left though. So maybe some of the lens imperfections/decentering which f/1.8 will magnify. Honestly, I have seen some Nikon 20mm performing worse at f/1.8. This can well be within the manufacturer tolerance.

spraynpray

daveznspace said:

mhedges said:

daveznspace said:

With that combo you should be able to go up to 25 seconds on shutterspeed. However, I always take 2-3 seconds of any formula as I've found the formula's exact numbers tend to produce slight star streaks.

25 seconds with a 20mm lens? Really?

yes.

look at the excellent mw calc

https://lonelyspeck.com/milky-way-exposure-calculator/

You will get any answer you want on the net if you look long enough. The 200 rule is for critical examination at large print sizes, the 500 rule is good for facebook etc.

Nobody has mentioned the differential speed of stars in the night sky yet. If the image includes the celestial equator (most do due to the location of interesting bodies and the wide angles off the lenses used), then you have to use a shorter exposure to avoid slotting. Just test it yourself. Set your OK button to give you 100% zoom and shoot a 200 rule then a 500 rule shot and compare.

tc88

@spraynpray, that's a good point. Different areas of sky, the stars move at different rate in the sky.

Capt_Spaulding

For those of us in the northern hemisphere the sky revolves around Polaris. The farther south you look the faster the apparent motion of the stars.

daveznspace

tc88 said:

I say forget about the calculators. Shutter speed is easy to decide for self. You will have plenty of time, just expose at different speeds, and zoom in and decide how much star trail is acceptable. Different people make different trade offs between star trail, noise and coma. I typically shoot 8s on 20mm on D850, but I know I'm on one end of the spectrum and have more tolerance of noise than trail.

Having said that, since this is facing southeast, and I don't really see movement of stars at the center, and Capt_Spaulding is using D750, I can't tell whether it's trail or not. Assuming this is not trail which Capt_Spaulding can rule in/out by experimenting with the speeds, then here is my guess.

The top left looks to be coma which wasn't bad for f/1.8. The top right looks a little different than the top left though. So maybe some of the lens imperfections/decentering which f/1.8 will magnify. Honestly, I have seen some Nikon 20mm performing worse at f/1.8. This can well be within the manufacturer tolerance.

Calculators are a good place to start but like I said I take several seconds off the results because I would get star trails, I'd back it off until there isn't star trails.

This is also a good learning experience because it makes you eventually have to think about things but gives you a good starting point.

I don't care about noise either as I will stack them anyway.

As I mentioned as well that I prefer to stop the lens down to get better results. It used to be that you used them wide open to obviously gather the light but with today's sensors that really isn't an issue.

I've gone from 14mm ultra-wides to 24mm wides but now prefer to shoot a 50mm 1.8 stopped down to at least 2.8 and very short shutter speeds, rip off 1-20 shots then move... aka making a pano, IMO everything about it is better, way more detail, sharper, better colors.

Somethings you need to let ppl figure out to get where they want.

daveznspace

Capt_Spaulding said:

For those of us in the northern hemisphere the sky revolves around Polaris. The farther south you look the faster the apparent motion of the stars.

They still move pretty quick but it's part of the learning process where you are.

daveznspace

spraynpray said:

daveznspace said:

You will get any answer you want on the net if you look long enough. The 200 rule is for critical examination at large print sizes, the 500 rule is good for facebook etc.

Nobody has mentioned the differential speed of stars in the night sky yet. If the image includes the celestial equator (most do due to the location of interesting bodies and the wide angles off the lenses used), then you have to use a shorter exposure to avoid slotting. Just test it yourself. Set your OK button to give you 100% zoom and shoot a 200 rule then a 500 rule shot and compare.

Yes, look long enough... but it's a good starting point. The actual formula gets semi-complex and you need technical info about the camera and sensor that sometimes isn't easy to come up with.

As I said, I take several seconds off the calculators time because I was getting star trails and yes it's easy to determine. But like you said if you look long enough....
The problem is that you can go to any thread (just like this one) and get tons of different answers. So when I started instead of trying to figure out who was right or wrong I just started with that calculator (actually the original one before it) and actually learned all the ins and outs. Like I just always figured I'd just use my 14mm but yet I ended up at 50mm. I've also had the luxury of living in 2 different places where light pollution wasn't an issue so it was pretty easy to determine.

For those who don't want to learn how and why all this works then they can just go buy a star tracker and be done with it.

As far as printing goes between w/e methods, I haven't gone huge as I just do my own with a canon pro-100 but 13x19 mine look like they should.

I do want to get a star tracker though as those images are usually so much better.

The other HUGE HUGE HUGE thing that came up that was messing up my shots was the tripod (a 3lt) which of course as we all know I absolutely hate that no integrity company... But anyway, since they lie about EVERYTHING one of them was the supposed weight they could hold. I think they claimed the one I had would hold 48lbs yet when a D750 with a 50mm 1.8G (super light combo) or a D700/D750 with a rokinon 14mm on it you could see a considerable drop in camera angle per few shots (or the camera would just fly off and break) and that would give it weird trails.

I then moved onto a Manfrotto and that issue was gone.

HipShot

Three hours west of Fort Worth: https://en.m.wikipedia.org/wiki/Copper_Breaks_State_Park

Capt_Spaulding

I've found this site https://darksitefinder.com/maps/world.html#12/30.6869/-104.1022 really helpful. It looks like once you get some distance west of Ft. Worth and north of Abilene you get some pretty dark areas.

My plan is to spend some vacation time next year here:

https://tpwd.texas.gov/state-parks/davis-mountains