Aperture is king…but WHY?
If you ask any seasoned astrophotographer what they want the most out of a lens, they’ll usually say fast glass. Aperture is one of the most important things to consider when choosing a lens for astrophotography.
In normal daytime photography, the focal ratio is most often more important for depth of field reasons, but in astrophotography, we care more about the amount of light we can get onto the sensor.
Everything in the sky is far enough away that we don’t have the ability to do anything but focus on it all at the same time, so astrophotographers could care less about depth of field — we just want as much light to enter the camera as we can get, as quickly as possible!
Why do we need fast glass?
Because the nature of astrophotography means taking long-exposure photos of something we can hardly see with our eyes, we want to maximize the amount of light that enters the lens. We want to make sure as much light as possible hits the camera sensor as fast as possible.
The focal ratio is called such because aperture is a ratio of the focal length of the lens to the size of the opening that allows light to enter the camera and illuminate the sensor.
You may think that getting a fast f/2.8 lens is not much of a step up from your f/3.5 kit lens, it’s only 0.7 f-stop units different! However, that is huge in the aperture world.
Take the following example:
A photograph shot at, say, 18mm and f/3.5 has captured only 64% of the light available to the same lens at f/2.8!
So that means the same photo shot at f/2.8 is 156% as illuminated, or requires a much shorter exposure duration to get the same illumination at f/3.5. A shot of around 16 seconds at f/2.8 is just as bright as a shot of 25 seconds at f/3.5!
Differences in aperture settings aren’t what you may think
You may think that an image captured with a lens at f/2 is twice as bright as the same image captured at f/4, but it’s not. The relationship between the amount of light available to the sensor and the focal ratio is not linear! An image captured at f/2 vs. f/4 on the same lens is far “brighter” than double!
As a more graphic representation, each bold f-stop value below lets in approximately twice the amount of light as the next bolded f-stop.
f/1.4 |
~200% of the light at f/2 |
| f/1.8 | Common consumer lens ~123% of the light at f/2 ~378% of the light at f/3.5 |
f/2 |
~200% of the light at f/2.8 |
f/2.8 |
Common fast lens ~156% of the light at f/3.5 ~200% of the light at f/4 |
| f/3.5 | Common kit lens ~131% of the light at f/4 |
f/4 |
Common kit lens ~200% of the light at f/5.6 |
f/5.6 |
~200% of the light at f/8 |
f/8 |
~200% of the light at f/11 |
The math
The focal ratio for a particular lens is just that: a measure of the focal length of the lens divided by the aperture diameter. It doesn’t stop there, though. That particular number (focal length / focal ratio) indicates the physical diameter of the hole in the lens that lets in the light (aperture diameter). Knowing the diameter of the aperture opening allows us to calculate the approximate area of the hole in the lens that allows light in to hit the camera sensor.
Area of aperture opening = π * ((focal length / focal ratio)/2)2 |
We know that the area of a circle is calculated with the equation π*r2, which allows us to compare the sizes of the aperture openings for all focal ratios available for a given lens.
Why does that matter?
Think of yourself in a dark room with the door closed, with daylight on the other side of the door. The wider you open the door, the more light is let into the room, and the brighter it gets.
The door, in this case, represents the aperture of the lens, with a larger or smaller opening to allow more or less light to illuminate the camera sensor. Viola! – turn up the lights, you get a brighter image.
For a longer explanation of aperture regarding general photography, see this article that provided me with research data about the topic: A Tedious Explanation of the f/stop, by Matthew Cole
The photographic proof
Here we compare the same frame on the same night, minutes apart. Both have identical camera settings and lens with the exception of aperture. First we have f/4 and second, the same shot at f/2.8.
One would ask, why don’t we all use F 1.0 apertures? In the Netherlands we have “Cruijffs law”: “Elk voordeel hep se nadeel” (Every advantage has it’s disadvantage). In this case price, quality (chromatic aberration, coma) and critical focus depth.
Awesome stuff, happy to have F2.8.
The question is this though,
if you are shooting static with no tracking, it makes sense to shoot wide open and the longest exposure stime possible.
But with a tacker, its a game changer, but i have noticed that with trackers some shots are stopped down to f3.2 or even F4 and ofcourse exposure time much much longer, i assume to increase sharpness? Since exposure over time isnt an issue with a tracker, it makes sense to me.
BUT
Would it ever be beneficial to do it at F8.0 with a tracked shot? or would that require a huge amount more of exposure over time?
Keen to her your response.
Hi Alex,
The main reason those tracked shots are often stopped down is for a few reasons:
* Sharpness
* Chromatic Aberration
* Coma
* Misc. lens issues when wide open
Even the best of the best lenses create better looking images when they are not wide open. There is a sweet spot when stopped down a step or two for almost every lens, so that is the main reason.
f/8 is a little much, just a stop or two to f/4 or f/3.2-3.5 is fine in most cases, because you still want light-gathering power. Many telescopes are f5-f11 without an adjustable aperture, but they have a much larger objective, so the light gathering area is huge, unlike the small camera lenses. And remember — the step down in light gathering from f/2.8 to f/8 is far more than linear! So yes, it takes a lot more time.
Cheers,
Cory
I got an Opteka fisheye lens (~$150 USD) as a gift and immediately went out that night to shoot “wide open” (f/3.5) since we were vacationing away from typical suburban light pollution. I was very excited to capture so much of the sky. I attributed the fuzziness of stars to the quality of lens and decided to live with it. But only once I got home, I thought to step down the aperture by one click (f/5.6) to reveal beautiful, crisp, clear stars.
Next time I’m away from light pollution, I’ll try some long, tracked (iOptron SkyTracker) exposures to see how much better it can be.
That coma can really ruin a photo!
Hi Alex,
Another drawbacks for long exposure are “not perfect tracker” and “not perfect sensor”.
The first one may smudge the stars due to some instability in mechanical parts or bad pole alignment.
The second one brings too much noise and you have to deal with it using stacking techniques which requires much more long exposed light and dark frames (time grows very significant and you can’t remove the noise completely).
Hi, I`m a newbie on this issues, but I want to improve my skills. I bought a Canon FD f/1.4 lens and the FD-EOS adapter to use on my Canon T1i. This lens 1.4 is better than f/2.8 for astrophotography? How to focus the stars if this lens is 100 % manual? Please, any help/info will help me a lot.
Hello Elizaul,
Yes, to answer your question as far as aperture is concerned, f/1.4 is better than f/2.8. Especially because you can drop it back to f/2 or so which will sharpen things up a lot, and still be some extremely fast glass!
Focusing when the lens is manual is the way to go with stars anyway. So, to start out, set the focus at infinity and take some test exposures. Move the focus ring each direction *slightly* until you find the sweet spot for the focus of the stars, then you’re good to go.
Cheers,
Cory
I was out last night shooting a very faint Milky (at rise plus light pollution). I began shooting at F2.8, 24mm 20 seconds. The stars looked like lima beans.
After several long minutes in the dark adjusting exposure length, ISO and mm at F2.8, but could’n’t dial in a sharp star. Finally finding F4.0, 30MM at 20 secs ISO 6400 and a few with ISO 3400 I got a couple I can feel comfortable playing with in editing. I was thinking my fastest glass at F2.8 would be the way to go, but since F4 gave the better result, I’m thinking my slower lenses should tag along for the evening.
My buddy told me that sometimes it’s just the glass you are using and no fiddling is going to make it better.
Your thoughts?
My gear last night:
Canon 5DM3,
Tamron 24-70 mm F2.8.
All manual settings using your advice in the “How To Milky” video. So glad that you emphasize the “test test test” prior to shooting all night. Learned that last year after many unsuccessful adventures.
Hi Chilly,
The lens quality makes a huge difference, and I’m not sure about your lens specifically.
But, what I CAN tell you is that the reason f/4 on that f/2.8 lens gave a good result is because it wasn’t wide open. Stopping down the aperture a bit on some lenses is often required to make imperfections in the glass go away.
Next time, try this: Stop the lens down to f/3.5, ISO6400 (on the 5DmkIII, depending on temperature), 24mm, and 20s exposure. Depending on your location, the light pollution may require a lower ISO, but if it’s dark enough, you should be able to shoot at ISO6400 with that sensor and clean it up in post.
Also, curved stars could easily be movement in the tripod or camera during the exposure. Make sure everything is EXTREMELY stable, especially if the stars don’t look the same from one exposure to the next.
Best of luck!
Cheers,
Cory
Thanks for the feedback Corey. Really appreciate all you and T share with us. I’m going to rent a lens and see what happens. Thinking the Canon 16-35mm F2.8 might be the one to play with.
Chilly
Hi Chilly,
I have the 16-35 f/2.8 L2 Canon lens, it’s what I shot almost everything in my new timelapse short film, SUTHERLAND, with. For just AP, though, I prefer the 24-70 f/2.8 L2…it’s near perfect!
Cheers,
Cory
Say for example you were to be mainly shooting at f2.8, 800 iso for 30 seconds using a tracker and stacking your images. Would going to f4 (to improve image quality slightly, distortion in corners etc) need double the exposure time to 1 minute? From reading your explanation is that correct?
Thanks
Hi A.D.,
That is correct. Moving to F/4 on a less-than-stellar lens can drastically improve the image quality and vignetting. But, it will double the exposure time if you keep everything else the same.
So, 30s ISO800 f/2.8 roughly equals 60s ISO800 f/4.
Cheers!
Cory
I’ll be traveling to Nepal in October and hiking Annapurna Circuit and/or Everest Base Camp. I’d like to get some astrophotography shots but think the weight of a dedicated lens might not be justifiable. I was thinking I’d shoot at night with my Canon 16-35 f/4 and an ISO a stop higher than if I had the f/2.8. I know greater ISO will be noisier but are there any other drawbacks? I was thinking of purchasing the Rokinon 14mm f/2.8 just an extra expense and extra weight. Thanks for any help!
Hi Adam,
That all depends on what you’re comfortable with, on a noise level. moving from f/4 to f/2.8 is a huge jump, and makes a big difference. The ISO required would be quite high with f/4 (ISO6400 or more, depending). To be honest, if you’re going to someplace amazing like Nepal, I’d either bring a dedicated lens, or, better yet — rent/buy a new lens that will work for both. Some of the aftermarket lenses (like the Tamron 15-30 f/2.8) can be great for AP and replace your f/4 lens altogether.
All that being said, the Rokinon f/2.8 14mm is a great starter lens for Milky Way shooting (I started with one!). I’ve heard it can be hit or miss regarding quality, but in my experience it was a great lens to start with!
All the best, and best of luck on your upcoming trip!
Cheers,
Cory
I found there is no benefits between F4,F8 or F2 for DS astro imaging. You only can short the exposure time to some level. Simply because in Deep sky imaging you can’t espouse your photo forever. Or your stars will blow up (become to bright) and when you stretch the photo to make the nebula more visible for example you will blow up the stars as well. They will not look tiny and pleasant. For example you got 100 shots 2 min. each in F11 and 200 shots 1 min. each in f 8. You will get 200 minets of exposure time for the same object from both lenses and will get almost equal light for the same period of time. Just the difference will be shorter or longer exposure time for the guiding on your mount.
“The focal ratio for a particular lens is just that: a measure of the focal length of the lens divided by the f-stop. ”
Try again.
The focal ratio is the f-stop or the focal length divided by the diameter of the aperture.
I think you meant.
The aperture diameter for a particular lens is a measure of the focal length of the lens divided by the focal ratio or f-stop.
A lens with a 1000mm focal length and focal ratio of 5 would have a aperture diameter of 200mm.
Hi Clinton,
Aha, I see the confusion. I fixed the wording, thanks for the heads up! Crowdsourced editing FTW!
Cheers,
Cory
New to this and want to start on a budget and eco friendly. (Only second handstuff)
I’m going to buy an Canon 400D body for 50,- (With 3 accu’s and 2 memory cards)
I have found a very expensive lens 50mm F1.8 for 85,- (More then the body)
And have found 200mm F3.3 for 5,- (10% of price of the body)
But do i have to devide 200mm/F3.3 because that’s bigger then 50mm/F1.0 ?
And the brand of the cheap one is not a Canon lens but from Magnum. Does that matter a lot ?
Thanx in advance.
you did not show telescopes and eos camera light collection. telescopes have F#’s also. but no aperture range. so you get all the light in the center of the frame.. but intensity drops off as the cos of the angel out on the 35mm cmos.
spot image sizes are limited by the atmospheric arc seconds dispersion…not by the lens EFL. certainly a 6″ f4 telescope has a lot of collection area.. versus an F2.8 eos lens.
joe