This is a good basic discussion, and it works well for digital cameras as well as film cameras. The difference is that you can take 100 digital photos, see the differences, and it didn't cost you anything to develop them :-).
I encourage new people to photography to try to be disciplined about learning how their camera changes as different settings. Get out of the tyranny of "auto" mode and try all the things. You may find it quite fun.
... for practice, learning, or stationary, long-lasting subjects.
Of course you want to get a good feel for the aperture, shutter speed and sensitivity so that when you see a beautiful bird about to take off for flight, or a butterfly touch down ever so momentarily, you aren't fumbling with settings, and you've already got a good sense of what settings will work.
Definitely agree though - many cameras have "priority" modes where you're really only changing the shutter speed or aperture, and it's still doing the rest automatically for you, so you can focus on whatever that one setting does, while also seeing what the smart camera tells you the other settings should be in those situations.
Though for moving objects, the ability to just fire off a rapid sequence on a digital camera can net you a shot which is at least incrementally better--or even the difference between a great shot and a dud. There were motor drives in film days of course but they weren't as fast as modern digital cameras and unless you were a pro, using them much got expensive in a hurry.
Early on, digital did have a problem with shutter lag--Sony even had a series of ads touting the fact that their camera was faster than others--but it's not a problem now especially with better cameras.
I've had trouble getting out of the local optimum with my mirrorless. For shots that rely on the glass, so to speak, it can't be compared with a phone at all: the shots aren't better, they don't exist on the phone to begin with.
But for the kind of picture my phone can take, it frequently does a better job. I can tell going through my collection when I used the mirrorless immediately, because the result is subjectively different, but we're comparing quality equipment with good defaults with very small (nonetheless excellent) equipment with metric boatloads of machine learning applied.
Getting past this involves an immediate and very steep learning curve which I've only intellectually engaged with; the practice of adjusting the basic parameters consistently gives me worse results, and I haven't put in the time to get past that.
Turning auto off and making your own, deliberate, choices is indeed the way to build up some intuition on how to capture the same scene in different ways. Shooting digital is easier. But often reviewing what you shot is hard on tiny screens in full sunlight.
Depth of field is one of the more obvious things you control with the aperture. But another thing it impacts is sharpness. Depending on your lens, it likely is a little fuzzier wide open than it is stopped down a little. Basically, you are covering part of the lens and controlling the way it distorts the light a little. Wide open it diffuses the light a bit. And it will do so differently in different parts of the photo. Knowing how your lens performs under different conditions is something you need to explore through trial and error. Another thing it impacts is color and contract. Just how, depends on the lens. Some lenses have a clear optimum, others are better across a wider range of apertures. People spend a lot of money on glass to get these effects.
In the same way, the shutter speed is not just about controlling the amount of light it is also about motion and motion blur that can both mess up your photo or be something you use creatively. The best tip you can give a beginning photographer to not too slow. Your camera might be able to handle it and you might have steady hands. But stuff still moves in your scene and it will introduce blur.
Some types of photography you want to have fast shutter speeds to capture really crisp moving subjects (e.g. sports photography, moving kids). Even on a bright day you might want to shoot at higher ISOs just so you can shoot a little faster. A zoom lens stopped down focused on a fast moving subject will do a lot better if you can shoot at 1/2000s of a second rather than getting a blurry thing at 1/30th. But sometimes that kind of effect is what you want when shooting e.g. a static subject against a waterfall.
The ISO dial basically allows you to compensate for loss of light by pretty much multiplying the signal. This is something you can easily do in post as well if you are shooting raw. It just gives you more wiggle room with aperture and exposure. Of course the issue is that you end up boosting the noise along with that as well. More sensitive sensors can be pushed a little further for this. But mostly, lower is better in terms of noise. Basically, you pick the aperture and exposure that produce the effect you want and the iso control allows you to get away with it when that is pushing the limits a bit too far. With some cheaper cameras, you end up trading off noise against creative control. Easier to shoot at night with the lens wide open at a low shutter speed that still works hand held. But less chance of blurry photos if you shoot a bit faster at a higher iso.
Then there are different focal lengths that not only "zoom" in the subject but also change the perspective. Using a very wide or zoomed in perspective and compensating with your feet can produce very different photos even if your subject ends up being roughly the same in the picture.
It works, but I've had to discard photos that looked fine in the view finder. It's hard to see all the details. Particularly evaluating focus issues is hard this way.
Certainly. But the rear screen is vastly worse on my camera, so I just wait until I'm at a computer to do any fine sorting. Obviously I can delete grossly out-of-focus shots on the camera.
I don't find the explanation here to be clear. This suffers from the classic issue where they explain the basics and then proceed to give practically no details on how to start applying the knowledge.
For anyone wanting to learn this more in detail I highly recommend the book "Understanding Exposure" by: Bryan Peterson.
My personal approach to taking handheld photos in manual mode using natural light is as follows:
1. Look at the scene and determine what the most important factor is. For example:
* Do I need to freeze motion? Then I need to set a short shutter speed.
* Is the image being taken in low light? Then I probably want to open the aperture and set my shutter speed as long as is suitable for what I can hold.
* Do I need a large depth of field or to maximize sharpness? Then I need to set the aperture to a suitable value to achieve that.
2. After I consider the most important factor and set either shutter speed or aperture to a starting value, set the second most important factor using the same thought process.
3. Adjust the ISO so the image is exposed properly. Knowing your camera, is the ISO going to result in an acceptable image? If not, then go back to the first step and adjust your shutter speed and aperture to bring the ISO down to an acceptable value.
4. Take the photo and review.
Of course, this doesn't apply to every scenario but it generally is a good approach for the types of photos where the exposure settings are critical. If I'm shooting a landscape outdoors, then I'm probably starting off by setting ISO to 100 first and then the aperture to the sharpest value for the particular lens I'm using as I'm sure that the shutter speed will be acceptable regardless of what it ends up having to be or will be what drives me to use a smaller aperture.
Fair enough, I'm definitely coming at it from having a technical background on cameras and optics and to me the article is halfway between one explaining why apertures as so weird (and not even doing that), and how the exposure triangle works.
I've always been interested in photography but really got into the weeds over the pandemic. Learning the exposure triangle (how aperture, shutter speed, and ISO are related/not related) has completely transformed how I think about cameras and taking photos. There is definitely some inherent mathematical/practical beauty in the meaning and relationships between these different camera settings.
I wouldn't want this comment to be interpreted as "shooting in manual is superior", in fact I would suggest against it in almost all cases - the deeper understanding will help you use nearly any camera in existence.
My dad's a hobbyist photographer and through him I've learned to generally shoot in aperture priority mode. For my intents it generally feels 'manual enough'.
Then you leave ISO on auto if it makes sense for the lighting conditions, or manually set it if needed.
> Then you leave ISO on auto if it makes sense for the lighting conditions, or manually set it if needed
Fun thing with modern cameras is that "iso" doesn't actually matter much at all. Typically you can just use one or two fixed iso settings for everything and adjust the exposure in post as needed. Keyword for reading more about this is "iso invariance".
Thanks for the term! That's a rabbit hole we'll venture down next time I see my dad.
I initially had a paragraph half written out in the comment you replied to detailing how my dad's new cameras perform extremely well at absurd (12800???) ISOs, which "iso invariance" would definitely explain.
My camera is an older Canon 5D MK1 which shoots mostly fine at 200 ISO and lower, and gets grainy quite fast at 400 ISO and above. If and when the photography bug bites me harder I'll probably upgrade to a slightly newer body aha.
Bear in mind you need to shoot raw, not jpg, in order to benefit from this.
In a tricky exposure situation, dial exposure compensation way down (-3 to -5 EV) and take the shot. You probably want a low-ish ISO selected too. Then in lightroom/whatever just crank the exposure slider up and fiddle with the levels & curves - easy way to not blow out your highlights.
The problem with this, however, is that most Raw editors only allow you to increase the exposure by +/- 1 or 2 EV. So, say the base ISO for your camera is ISO 100. If you edit it in, say, Lightroom, you can only get it up to ISO 400, whereas in-camera, you could bump it up to 800, 1600 or higher.
Lately I'm close to apperture priority. With Nikon you can over ride the apperture setting in Programm / Automatic mode. Most of the time automatic is just fine. In case I want something "special", longer exposure for stuff like waterfalls or shorter exposure for movement, I just change the apperture until I'm happy with the shutter speed. Preferring low ISO values, that makes you appreciate fast lenses.
I've used a refractor (4" lens, f7, 714mm focal length) as a telephoto. It's great. I ended up buying a crazy-ass lens and building a solar telescope (look at the sun) which was fun until the focal point melted the enclosure.
Realtalk: focal distance is the only number a beginner needs to learn about. Just set your camera to “Auto” and it will take great photos. This is why we invented computers lol
Letting the camera take control will result in good photos. But great photos are more likely when you take technical control and put thought behind each photo.
Yes, and it’s fine to learn that stuff as an intermediate photographer! I just think framing shots is by far the most important thing for beginners because that provides the motivation for them to keep going out and taking more photos.
Art is often characterized by the specific objectives set by the artist. An automatic computer algorithm to handle the exposure will only typically have one (or a few) preprogrammed set(s) of objectives, thus it adds a constraint to the artist. Great art can be made either way though.
For a beginner I agree learning the intricacies of exposure may not be one of the most pressing concerns, or particularly relevant in many cases, but if someone wants to get more into the hobby for whatever reason (artistic or otherwise) it isn't a bad idea to start experimenting with exposure settings even if the results aren't initially good.
For film cameras, holding exposure time times f-stop constant does not produce the same image density.
Exposure time for film is not straight photon integration, like it is for a CCD array. You can only trigger each film grain once, but new photons can repeatedly add electrons to a single CCD pixel. So there's what's called "reciprocity failure".[1]
What tripped me when shooting and developing film is that it usually works in negative. With digital you are always looking for overexposure, while negative film is OK with overexposure. Light builds silver density, while lack of it leaves a clear base with no information.
Then you get an analogue NOT operation when printing the negative on the enlarger. Two negative processes make a positive.
Film photography is very interesting as it combines art with optics and chemistry. Just pure magic seeing the image fade into the paper when developing.
Anecdotally, I heard that this is the reason why, as recently as 10 years ago, a majority of movies were shot on film: it handles highlights so well. It took a while for digital to catch up.
Oh yes, I dabbled a bit into using motion picture film stock for photography. It is fun to shoot it by the frame and not in feet, if you have a lab close to you to deal with the process.
Vision3 50D was recently(for film) introduced in 2011, and I had amazing results with it. Super smooth with very good latitude.
After one attempt at photography in high school, with a SLR and a darkroom, I decided to wait until I felt that the tech had matured enough so that I never, ever had to deal with physical film media. Reciprocity is a perfect example; when I'm in art mode (or scientific data collection), I never want to think about the nonlinearity of my sensor.
Not quite. Film doesn't get more grain with longer exposures, whereas it does with digital sensors, as you can get [hot pixels](https://photofocus.com/software/what-are-hot-pixels-and-how-...) and other artifacts the longer the sensor is exposing a shot.
Funny, I was just comparing two photos from some holiday snaps I have taken. One on a Google pixel phone and another on an old 100mm lens produced in the late 70s paired with a DSLR. The differences are stark, and remind me that focal length, dynamic range, shutter/aperture settings all have big impacts on how the final image turns out. Particularly important is how dynamic range and tones of the scene are translated to the recorded medium, as decisions always need to be made with a focus on specific tones. HDR goes some way to help with this these days, but artistic decision-making is still a thing.
One little nit is the acceptable shutter speed for a given lens focal length depends _heavily_ on the camera and shutter type being used!
In the past, 1/focal length was perfect advice. You can adjust this down very quite far if you have a lens or body with image stabilization, typically down to as far as 1/30 or even lower.
Also bodies get higher resolution, the amount of blur you can see goes up as well as you start sampling higher frequency movements of the camera body. You probably need 1/2 x focal length above 24MP, unless you have stabilization.
There are also different types of shutters, for example a leaf shutter in the lens vs. a focal plane shutter. Focal plane shutters can inject some shutter shock at lower speeds, where as leaf shutters don't have this effect and can be handheld to much lower speeds, I have gotten very sharp shots as low as 1/8th.
Finally, if your camera has a mirror or not also effects this, as the mirror swing imparts vibration as well.
Knowing exactly how the camera reacts at different speeds will let you tweak this rule a lot!
Technical minutia, but I found that I felt a lot more confident once I understood this: The reason f-number is written like so is because it's not a measure of actual aperture diameter but is instead the ratio of aperture diameter to focal length. Specifically, (f-number) = (focal length)/(aperture diameter), or:
N = f/A
equivalent to,
A = f/N
Hence the format f/N, since you (theoretically) use f-number to calculate actual aperture diameter when given a specific focal length. That aperture is in the denominator also explains why smaller f-numbers are brighter.
Why those weird numbers in particular? As in the article, a 'stop' is a doubling or halving of the brightness (technically, 'illuminance') of the image produced by the lens. Since the amount of light that passes through an aperture is proportional to its area, which is in turn proportional to the square of its diameter, one stop corresponds to a sqrt(2) ~ 1.4 ratio between two aperture diameters. Hence, the standard scale of full f-stops is the powers of two and their geometric means:
The full stop scale in the article has some errors: f/1.8 instead of f/1.4, a typo for f/4/0, and a lack of f/45 (though f/1.8 is the stop below f/2 in the finer 1/3- and 1/4-stop scales):
But why measure image brightness using f-number instead of aperture diameter directly? Because brightness also depends on focal length: Doubling the focal length of a lens doubles the size of the image of an object seen through it. Since the light from the object is spread out over four times the area, the image brightness is quartered. This means lenses with equal aperture diameters and different focal lengths will not have the same exposure. (below edited, I confused myself at first!) But photographers would rather not have to consider focal length when choosing an aperture for proper exposure. Designing lenses such that their aperture controls are labelled with f-number enables this, since actual aperture diameter then scales automatically with focal length to cancel out its effects on exposure.
One point here, when talking about the N = f/D formula to calculate the f-number, D is entrance pupil diameter. That is, it is the apparent diameter of the image of the aperture when viewed from the front of the lens, not the actual aperture diameter.
Something worth adding here for those who want to complicate their lives even more, these equivalences get complicated when you need to compare across different sensor formats (full-frame vs crop vs micro-4/3rds). Something I learned recently is that f/5.6 on a 300mm lens on a crop sensor is more similar to something like an f/8 or f/11 on a 450mm on a full frame. Including ISO into this makes that comparison more complicated as well.
There's an old exposure rule that is probably next to useless now, but worked pretty well for film cameras when you didn't have a light meter: exposure of a subject in sunny bright conditions is an aperture of f/16 and shutter speed set to the film's ISO. So a starting place for a bright, sunny day when using Kodachrome 64 film would be f/16 at 1/60th of a second (or an equivalent aperture/shutter speed combination). From there an experienced photographer could estimate exposure for other lighting conditions. Of course, negative film (either color or black and white) is more forgiving of exposure errors than slide film. These days if your exposure meter isn't working, it means your whole camera isn't working...
On modern cameras, you rarely have to worry about the numbers, unless you a deliberately doing something fancy (like continuous waterfalls).
The biggest mistake most people make is not filling the frame, because they are focused on the center. Checking the borders of the picture is a simple habit that can dramatically improve composition.
51 comments
[ 2.8 ms ] story [ 110 ms ] threadI encourage new people to photography to try to be disciplined about learning how their camera changes as different settings. Get out of the tyranny of "auto" mode and try all the things. You may find it quite fun.
... for practice, learning, or stationary, long-lasting subjects.
Of course you want to get a good feel for the aperture, shutter speed and sensitivity so that when you see a beautiful bird about to take off for flight, or a butterfly touch down ever so momentarily, you aren't fumbling with settings, and you've already got a good sense of what settings will work.
Definitely agree though - many cameras have "priority" modes where you're really only changing the shutter speed or aperture, and it's still doing the rest automatically for you, so you can focus on whatever that one setting does, while also seeing what the smart camera tells you the other settings should be in those situations.
Early on, digital did have a problem with shutter lag--Sony even had a series of ads touting the fact that their camera was faster than others--but it's not a problem now especially with better cameras.
Except the time to sort through 300 photos after just an afternoon with your kids at the beach…
But for the kind of picture my phone can take, it frequently does a better job. I can tell going through my collection when I used the mirrorless immediately, because the result is subjectively different, but we're comparing quality equipment with good defaults with very small (nonetheless excellent) equipment with metric boatloads of machine learning applied.
Getting past this involves an immediate and very steep learning curve which I've only intellectually engaged with; the practice of adjusting the basic parameters consistently gives me worse results, and I haven't put in the time to get past that.
Depth of field is one of the more obvious things you control with the aperture. But another thing it impacts is sharpness. Depending on your lens, it likely is a little fuzzier wide open than it is stopped down a little. Basically, you are covering part of the lens and controlling the way it distorts the light a little. Wide open it diffuses the light a bit. And it will do so differently in different parts of the photo. Knowing how your lens performs under different conditions is something you need to explore through trial and error. Another thing it impacts is color and contract. Just how, depends on the lens. Some lenses have a clear optimum, others are better across a wider range of apertures. People spend a lot of money on glass to get these effects.
In the same way, the shutter speed is not just about controlling the amount of light it is also about motion and motion blur that can both mess up your photo or be something you use creatively. The best tip you can give a beginning photographer to not too slow. Your camera might be able to handle it and you might have steady hands. But stuff still moves in your scene and it will introduce blur.
Some types of photography you want to have fast shutter speeds to capture really crisp moving subjects (e.g. sports photography, moving kids). Even on a bright day you might want to shoot at higher ISOs just so you can shoot a little faster. A zoom lens stopped down focused on a fast moving subject will do a lot better if you can shoot at 1/2000s of a second rather than getting a blurry thing at 1/30th. But sometimes that kind of effect is what you want when shooting e.g. a static subject against a waterfall.
The ISO dial basically allows you to compensate for loss of light by pretty much multiplying the signal. This is something you can easily do in post as well if you are shooting raw. It just gives you more wiggle room with aperture and exposure. Of course the issue is that you end up boosting the noise along with that as well. More sensitive sensors can be pushed a little further for this. But mostly, lower is better in terms of noise. Basically, you pick the aperture and exposure that produce the effect you want and the iso control allows you to get away with it when that is pushing the limits a bit too far. With some cheaper cameras, you end up trading off noise against creative control. Easier to shoot at night with the lens wide open at a low shutter speed that still works hand held. But less chance of blurry photos if you shoot a bit faster at a higher iso.
Then there are different focal lengths that not only "zoom" in the subject but also change the perspective. Using a very wide or zoomed in perspective and compensating with your feet can produce very different photos even if your subject ends up being roughly the same in the picture.
If you have a mirrorless camera, you can use the viewfiender to review.
I tend to leave the rear screen on my camera turned off. It just wastes battery life, it's not usually useful for anything.
For anyone wanting to learn this more in detail I highly recommend the book "Understanding Exposure" by: Bryan Peterson.
My personal approach to taking handheld photos in manual mode using natural light is as follows:
1. Look at the scene and determine what the most important factor is. For example: * Do I need to freeze motion? Then I need to set a short shutter speed. * Is the image being taken in low light? Then I probably want to open the aperture and set my shutter speed as long as is suitable for what I can hold. * Do I need a large depth of field or to maximize sharpness? Then I need to set the aperture to a suitable value to achieve that.
2. After I consider the most important factor and set either shutter speed or aperture to a starting value, set the second most important factor using the same thought process.
3. Adjust the ISO so the image is exposed properly. Knowing your camera, is the ISO going to result in an acceptable image? If not, then go back to the first step and adjust your shutter speed and aperture to bring the ISO down to an acceptable value.
4. Take the photo and review.
Of course, this doesn't apply to every scenario but it generally is a good approach for the types of photos where the exposure settings are critical. If I'm shooting a landscape outdoors, then I'm probably starting off by setting ISO to 100 first and then the aperture to the sharpest value for the particular lens I'm using as I'm sure that the shutter speed will be acceptable regardless of what it ends up having to be or will be what drives me to use a smaller aperture.
I wouldn't want this comment to be interpreted as "shooting in manual is superior", in fact I would suggest against it in almost all cases - the deeper understanding will help you use nearly any camera in existence.
Then you leave ISO on auto if it makes sense for the lighting conditions, or manually set it if needed.
Fun thing with modern cameras is that "iso" doesn't actually matter much at all. Typically you can just use one or two fixed iso settings for everything and adjust the exposure in post as needed. Keyword for reading more about this is "iso invariance".
I initially had a paragraph half written out in the comment you replied to detailing how my dad's new cameras perform extremely well at absurd (12800???) ISOs, which "iso invariance" would definitely explain.
My camera is an older Canon 5D MK1 which shoots mostly fine at 200 ISO and lower, and gets grainy quite fast at 400 ISO and above. If and when the photography bug bites me harder I'll probably upgrade to a slightly newer body aha.
In a tricky exposure situation, dial exposure compensation way down (-3 to -5 EV) and take the shot. You probably want a low-ish ISO selected too. Then in lightroom/whatever just crank the exposure slider up and fiddle with the levels & curves - easy way to not blow out your highlights.
https://en.wikipedia.org/wiki/%C6%91/8_and_be_there
https://www.dpreview.com/reviews/nikon-coolpix-p1000
For silly apertures, you want the Pinhole Pro Max.
https://www.pcmag.com/news/hands-on-thingyfys-pinhole-pro-ma...
I see what you did here.
[1] https://www.youtube.com/watch?v=9cT0jXI7l4E
https://www.telescope.com/Astrophotography/Astrophotography-...
Telescopes do tend to prefer specifying girth rather than length, though.
For a beginner I agree learning the intricacies of exposure may not be one of the most pressing concerns, or particularly relevant in many cases, but if someone wants to get more into the hobby for whatever reason (artistic or otherwise) it isn't a bad idea to start experimenting with exposure settings even if the results aren't initially good.
I'd love to have a search engine that just searches writing like this.
For film cameras, holding exposure time times f-stop constant does not produce the same image density. Exposure time for film is not straight photon integration, like it is for a CCD array. You can only trigger each film grain once, but new photons can repeatedly add electrons to a single CCD pixel. So there's what's called "reciprocity failure".[1]
[1] https://en.wikipedia.org/wiki/Reciprocity_(photography)
Then you get an analogue NOT operation when printing the negative on the enlarger. Two negative processes make a positive.
Film photography is very interesting as it combines art with optics and chemistry. Just pure magic seeing the image fade into the paper when developing.
Vision3 50D was recently(for film) introduced in 2011, and I had amazing results with it. Super smooth with very good latitude.
[1] https://photos.app.goo.gl/aeCJPnp5BSZ9aBxT6
In the past, 1/focal length was perfect advice. You can adjust this down very quite far if you have a lens or body with image stabilization, typically down to as far as 1/30 or even lower.
Also bodies get higher resolution, the amount of blur you can see goes up as well as you start sampling higher frequency movements of the camera body. You probably need 1/2 x focal length above 24MP, unless you have stabilization.
There are also different types of shutters, for example a leaf shutter in the lens vs. a focal plane shutter. Focal plane shutters can inject some shutter shock at lower speeds, where as leaf shutters don't have this effect and can be handheld to much lower speeds, I have gotten very sharp shots as low as 1/8th.
Finally, if your camera has a mirror or not also effects this, as the mirror swing imparts vibration as well.
Knowing exactly how the camera reacts at different speeds will let you tweak this rule a lot!
Why those weird numbers in particular? As in the article, a 'stop' is a doubling or halving of the brightness (technically, 'illuminance') of the image produced by the lens. Since the amount of light that passes through an aperture is proportional to its area, which is in turn proportional to the square of its diameter, one stop corresponds to a sqrt(2) ~ 1.4 ratio between two aperture diameters. Hence, the standard scale of full f-stops is the powers of two and their geometric means:
The full stop scale in the article has some errors: f/1.8 instead of f/1.4, a typo for f/4/0, and a lack of f/45 (though f/1.8 is the stop below f/2 in the finer 1/3- and 1/4-stop scales): But why measure image brightness using f-number instead of aperture diameter directly? Because brightness also depends on focal length: Doubling the focal length of a lens doubles the size of the image of an object seen through it. Since the light from the object is spread out over four times the area, the image brightness is quartered. This means lenses with equal aperture diameters and different focal lengths will not have the same exposure. (below edited, I confused myself at first!) But photographers would rather not have to consider focal length when choosing an aperture for proper exposure. Designing lenses such that their aperture controls are labelled with f-number enables this, since actual aperture diameter then scales automatically with focal length to cancel out its effects on exposure.More on that subject https://photographylife.com/equivalence-also-includes-apertu...
The biggest mistake most people make is not filling the frame, because they are focused on the center. Checking the borders of the picture is a simple habit that can dramatically improve composition.