View Full Version : I have no idea what I'm doing with stars, with a pic to prove it!
e02937
18th of November 2009 (Wed), 23:34
7d
28 to 135mm at 28mm
ISO 1250
f 5.6
30 Second Exposure
Tripod plus wireless shutter release
Focus set to infinity
IS Off
Mirror Lock up
I just don't get it, how does one focus this kind of shot? I looked through the viewfinder and live view for manual focus. Auto focus not helpful. Fixed white balance in DPP, other than straight out of camera.
http://farm3.static.flickr.com/2690/4116010795_ee8762a46b_o.jpg
trailblazer87
19th of November 2009 (Thu), 00:09
Ways to improve the sharpness of the stars would be:
1. Turn off AF period
2. Turn off IS especially on this lens.
3. Stop down the lens 2 stops from wide open. This lens gives pretty good results stopped down to F/7.1
4. You had the right idea using live view, just use the zoom function and dial it up to 10X. Also stay in live view so the mirror doesn't shake the camera.
5. Try using the 2 second timer for the shot.
6. A sturdy tripod is helpful, if you don't have one try to find a way to brace it so it doesn't vibrate.
Good Luck and have fun. I'm sure the others here can add to my tips.
This is a 30 second shot, so the strrs are coming out a little egg shaped from rotation of the planet, a shorter exposure would help with that.
Adrena1in
19th of November 2009 (Thu), 04:07
Unfortunately, a lens focused right out to infinity doesn't mean it'll be focused on the stars. Lenses focus *past* infinity for some reason. If you can point the camera at the brightest star you can see and frame it in Live View then you're laughing. But at short FLs this can be tricky sometimes. I couldn't get anything to appear in Live View the other day at 18mm, so I went back to my tried and tested routine, thus;
1) On MF, focus as best you can looking through the viewfinder and take a 5-sec shot.
2) Turn the focus a tiny amount by hand and take a second 5-sec shot.
3) Preview the shots. If the second is focused best, turn the focus a bit more in the same direction as in step 2) and take another 5-sec shot. If the focus in the second image is worse, turn the focus back to about where it was, then turn it a tiny bit more in that opposite direction, then take a shot.
4) Repeat until you get a shot where the stars appear to be in focus.
At wide angles this method is good enough for getting sharp stars, and takes less than a minute.
SteveInNZ
19th of November 2009 (Thu), 13:38
With these wider angle shots you can focus on something terrestrial but a mile or two away. It won't move, will most likely be brighter and not a point source so you can see when it's in focus. Focus manually using zoomed live-view as suggested and then point your camera where you want it.
I don't agree with stopping the lens down unless you find that you have a reason to do so. Two stops means 4 times the exposure to get the same number of stars and you're already seeing trailing at 30 seconds.
Steve.
CaptainTonus
19th of November 2009 (Thu), 14:19
There's a lot of really good advice in this thread. The only thing I would add is, I'd recommend using a wider-angle lens than the 28-135. 28mm on a crop body is frankly not very wide at all. For night landscapes like this, especially for avoiding star trails, you're gonna at least want something in the 10-20mm range, and the larger the aperture the better. You'll be amazed at what a difference it makes.-
DanThoman
19th of November 2009 (Thu), 17:52
Don't mean to hijack the thread, but: Trailblazer87 - why step #2? Is something wrong with the IS in the 28-135 lens?
Aweitzel
19th of November 2009 (Thu), 18:11
IS "fights" when its on a tripod. Its best to keep it off when your camera is mounted to a tripod.
It can actually add shake or blur. Im sure someone else can add more and or explain this better then i did. that was how it was explained to me.
trailblazer87
19th of November 2009 (Thu), 19:55
Don't mean to hijack the thread, but: Trailblazer87 - why step #2? Is something wrong with the IS in the 28-135 lens?
This lens is an early IS model and tends to try to compensate for vibration even when there is no vibration. This problem occurs in all of the early IS lenses (100-400L as well), the lens models that were designed later in life (such as the 70-200 2.8L) has this problem resolved but Canon continues to produce this model with the issue, its no problem, just gotta know to turn it off when mounted.
trailblazer87
19th of November 2009 (Thu), 20:00
OP I have been looking at your photo a bit more and it looks like you were focused somewhere between the tree and infinity.
If it helps you can zoom in all of the way to 135mm and try to focus a little more precisely and then zoom out. This lens is close to being a par lens (meaning focus is the same across the zoom range), some minor adjustments may be needed but it will get you close.
I hope this helps.
e02937
19th of November 2009 (Thu), 20:04
Thanks for your help.
Ways to improve the sharpness of the stars would be:
1. Turn off AF period
2. Turn off IS especially on this lens.
3. Stop down the lens 2 stops from wide open. This lens gives pretty good results stopped down to F/7.1
4. You had the right idea using live view, just use the zoom function and dial it up to 10X. Also stay in live view so the mirror doesn't shake the camera.
5. Try using the 2 second timer for the shot.
6. A sturdy tripod is helpful, if you don't have one try to find a way to brace it so it doesn't vibrate.
Good Luck and have fun. I'm sure the others here can add to my tips.
This is a 30 second shot, so the strrs are coming out a little egg shaped from rotation of the planet, a shorter exposure would help with that.
1) I should have mentioned I did turn AF off, wasn't doing anything anyways
2) I had IS off
3) I will give this a shot!
4) I gotta try this a little bit better. I did use mirror lockup anyways, so I don't think the mirror was an issue
5) I have a wireless remote so I wasn't touching the camera
6) I have a solid tripod manfrotto 055xprob and 488rc2 head.
Thanks for all the help. Good suggestions here, I will give them a shot:
1) Live view
2) 15 or 20 second exposures instead of thirty
3) Stop down the aperture if needed
4) Take some 5 second exposures and reiew for focus accuracy
DanThoman
19th of November 2009 (Thu), 20:32
Thanks for all the IS info. IS is (haha) new to me so I appreciate all the explanations, and sorry to (partially) hijack the thread.
e02937
19th of November 2009 (Thu), 20:36
No problem-o
This video shows you how IS works, and once you see that it makes sense why it can actually induce shake when on a tripod:
http://www.youtube.com/watch?v=C60ehMe3wQ8
Adrena1in
20th of November 2009 (Fri), 06:24
I don't agree with stopping the lens down unless you find that you have a reason to do so.
I do with my nifty fifty, but then it's very fast anyway. Wide open the aberrations are terrible, so I always stop it down 3 or 4 times.
KandJinIN
8th of February 2010 (Mon), 16:23
I figure it would be nice to resurrect an old thread that gives me some good info but leads to more questions.
Why does a wider angle change the star trails? Is it just because you will notice the trail more when you zoom in more on one section of the sky, or what?
How will ISO and aperture effect star trails? I am trying to get clear sharp stars, but I have some camera shake due to forgetting to use my timer on my XTi, so any explanation will be helpful.
Thanks
CaptainTonus
8th of February 2010 (Mon), 17:19
Why does a wider angle change the star trails? Is it just because you will notice the trail more when you zoom in more on one section of the sky, or what?
The more zoom you have, the less exposure time it takes to start noticing trails. There is a certain mathematical equation to determine this, but I don't remember what it is. Being further away from the stars, in general, means you need a longer exposure to get trails. I actually try to avoid star trails when I shoot the night sky, thus I always use the widest-angle lens possible.
How will ISO and aperture effect star trails? I am trying to get clear sharp stars, but I have some camera shake due to forgetting to use my timer on my XTi, so any explanation will be helpful.
Thanks
Using a lower ISO like 100 or 200 will require a longer exposure to get a good shot. Using a more narrow aperture like f/5.6 or f/8 will also require a longer exposure. Anythng longer than a 30-second exposure is likely going to produce some trailing, so the idea is to keep it right around there. If you have an XTi, you probably aren't going to get clean shots at anything more than ISO 400. Shoot the widest angle possible at the largest aperture possible, and try ISO 200. What lens are you using to get these shots? Use a remote shutter release to ensure you don't get any camera shake. If you don't have one, buy one. It will be the best $30 you ever spent. Hope this helps.
mpistone
9th of February 2010 (Tue), 22:48
Why does a wider angle change the star trails? Is it just because you will notice the trail more when you zoom in more on one section of the sky, or what?
Pretty much. The stars move at a fixed speed through the sky, so if you're zoomed way in some given star might cross your entire viewfinder in a few minutes. That same star only crosses a tiny fraction of your wide-angle lens' field of view in the same few minutes. Hope that helps :)
MintMark
10th of February 2010 (Wed), 07:47
I figure it would be nice to resurrect an old thread that gives me some good info but leads to more questions.
Why does a wider angle change the star trails? Is it just because you will notice the trail more when you zoom in more on one section of the sky, or what?
How will ISO and aperture effect star trails? I am trying to get clear sharp stars, but I have some camera shake due to forgetting to use my timer on my XTi, so any explanation will be helpful.
Thanks
CaptainTonus has said it already, but I'll try a different way.
The focal length of the lens is related to its field of view (search for "angle of view" to find the formula). A very wide angle lens could capture more than 90 degrees of sky. Knowing the camera sensor resolution, you can work out how many degrees each pixel covers with that lens. This will be larger for wider angle lenses.
The earth rotates 360 degrees in 24 hours, so that's 15 degrees an hour. Now you could work out how long it would take for a star to move from one pixel to the next pixel. This will be longer for wider angle lenses. So that puts a limit on exposure time to avoid star trails two pixels long. In practice you can go longer than this without noticing them.
In that exposure time a wider aperture will let more light in. It means you can detect fainter objects, but you might also capture more light pollution (sky glow) and with many lenses a wider aperture leads to mis-shapen stars at the edge of the picture. That's why many people advise stopping the lens down if you can.
You can increase the ISO to make the sensor more sensitive. This lets you capture fainter objects in the same exposure time but with most cameras high ISO settings give a noisier image. The noise gets worse as temperature and exposure time go up. This is why specialist astro cameras have cooling systems and very sensitive sensors.
So it's a balancing act. The lens can make a difference too... a prime lens will usually give sharper images and allow wider apertures compared to a zoom lens at the same focal length.
Hope that's helpful,
Mark
vBulletin® v3.6.12, Copyright ©2000-2012, Jelsoft Enterprises Ltd.