For the longest time, I’ve been saying (perpetuating the popular adage?) that cropping is bad. I’ve touched on the reason in previous articles – notably these two on compositional building blocks, and proper perspective practice – But I don’t think I’ve really explained why. There are several reasons; I’ll go through these in some detail over the course of this article, and finish with a commentary on what really happens in the commercial industry – and why we photographers frequently want to strangle the creative directors, agency people or layout artists on the other end of the shoot.
But firstly, an example. Look at the image below: compositionally, it works, yes? What focal length was it shot with? The EXIF data says 16.8mm – oops, that’s a small sensor, so it’s probably about 100 or 150mm or something right? After all, the perspective fits.
Fear and exhilaration. RX100
Think again.
And again.
And yet again.
Oh, whoops. It seems that perhaps 16.8mm is actually somewhere closer to 40mm or thereabouts. At this size, with no other visual cues like depth of field to give away the technical properties of the image, there’s really no way to tell from the crop. This matters nothing to the viewer: each of these crops works compositionally, and doesn’t feel as though it was a compromise or chopped down from something larger. (Perhaps this is one of the hallmarks of a strong image period, I haven’t given it that much thought.) So long as them reproduction size doesn’t exceed the amount of resolution you’ve got to work with, there really is no way for the viewer of the image to know that what he or she is seeing wasn’t the original intention of the photographer.
Frequently, in the world of commercial photography, this is precisely what happens. In my earlier days, I spent a lot of time getting my compositions perfect with the intention that the images be used as-is, without cropping; of course I would leave space for copy etc. depending on the requirements of the client or the final aspect ratio. The intention was to preserve as much image quality as possible for whatever enlargements – sometimes billboard size, certainly poster size – would be required. When working with experienced clients and creative teams, this was certainly a viable approach.
Another example. Senso-ji, Tokyo.
The trouble started when working with less experienced clients who either didn’t have a creative/ PR/ advertising team, or whose team was equally inexperienced: I started getting requests to leave a lot more space around my images ‘for cropping’. I was perplexed by this: on one hand, they wanted the same style and quality of images that were in my portfolio – all of which were perfectly framed and had zero loose space whatsoever – yet they still wanted the extra room to crop. Did they not realize that these two things were at diametric odds with each other? It wasn’t the file sizes that were the issue – good-quality D800E files are usually more than enough for almost any use – it was the space around the outside. I tried to argue the point to no avail, and in the end, the client wins – I leave extra space around the border. This has the upshot of making my compositions look a bit ‘small’. But the client is happy, so I suppose that’s one good bright spot.
Senso-ji, original. Even between the crop – probably about 350mm equivalent – and the origina 90mm equivalent, there’s a lot of difference even though both are ostensibly of a telephoto perspective.
Personally, as an artist, this bothers me. It feels as though I’m ceding control of my composition to somebody else who may not have the same eye for it; almost certainly not a photographer. In fact, it’s exactly what is happening. You have no idea how the final image will be used or cropped; you can only hope it’s in a way that maintains the balance of the composition. It’s like a chef cooking the main portion of the dish, but having the waiter plate it and then running the risk of the diner complaining that the proportions of component sub-dishes are wrong, or the food is ugly. It’s now become such an ingrained habit that I’ve got to be careful with my personal work – not to be sloppy with the edges and empty space in a composition.
I realize I still haven’t explained why dead space and cropping are bad. The former is to do with compositional balance. Empty space can be used as a natural frame to isolate your subject; as a sort of visual cue to signal something anticipatory (if placed in front of your subject), or something historical (if placed behind). Or it can simply be used to show remoteness and emptiness in and of itself. The problem comes when the emptiness is not in harmony with the rest of the elements in the frame: this is what is meant by ‘balance’. A very busy center area and empty borders doesn’t work, because it feels as though the subject in the center has been artificially constrained and not allowed to use all of the space in the frame. Same case if all the action happens heavily to one side of the frame or the other. Clearly, this is not conducive to a strong composition – you’re neither using all of the available space in your frame, nor are perspectives properly employed.
A more extreme example, from my experiments with motion. Definitely a telephoto perspective, right?
It’s the latter which is the cause of the biggest problems in cropping. If you leave say a quarter of the frame width empty on all sides around a center subject, then you’re in fact cropping the frame down to a quarter of the original area, which is doubling the focal length. A 24mm becomes nearly 50mm. 35mm becomes 70mm. Wide becomes telephoto. Take a 35mm shot: the real questions is, at the time of shooting, is the shot composed as a wide, or a tele? Both call for very different arrangements of foreground, midground and background objects to create an effective composition.
And the original frame – a wide 24mm equivalent.
And here we come to the crux of the problem with cropping: it confuses perspectives. Poor use of perspective in composition is one of the major progenitors of weak images. This is generally more of a problem with wide angle than telephoto perspectives, simply because cropping a wide image can completely alter the angle of view of the shot to become a telephoto image, whereas this does not happen on cropping an image that’s already of a telephoto perspective to begin with (there is little difference in perspective between 300mm and 600mm, for instance). Wide-angle images with no clear foreground subjects result in the action being flattened into the background of the image and seeming very far away; there’s no obvious subject to come forward and attract the attention of the viewer.
At this point, if you haven’t read the earlier linked article on proper perspective practice, I highly recommend you do so now.
In order to make a strong image, you have to use both all of the available space in the frame, as well as the perspective of the lens effectively. And to do that, there has to be a degree of previsualization on the part of the photographer before taking the shot; it might be as simple as consciously putting your subject into the foreground with a wide, and looking for layers with a tele, or as precise as knowing exactly what focal length corresponds to what field of view. With practice and a degree of conscious observation, the former turns into the latter. This results in the photographer having the ability assess a scene and compose images in his or her mind’s eye before even looking through the viewfinder; combine that with a degree of anticipation, and you’ve got the skills to see strong images.
However, if your 28mm lens sometimes yields a 75 degree field of view (which is what 28mm should yield in on 35mm full frame format), sometimes it yields 50 degrees, or 42 degrees at others – how are you going to know where to position your subjects in the field of view arc to use up the full 75 degrees? The answer is, you won’t. And this results in dead or empty space in the frame, which in turn leads to compositional imbalance, which then results in a weak image. Familiarity with how to compose for a given field of view (and focal length) is therefore the key to strong images: frequent cropping makes this completely impossible.
In some ways, this is like an amateur who doesn’t understand perspectives shooting with a zoom: the lens does the framing, rather than the photographer. Sometimes you want to your subject to occupy a large proportion of the frame, but still retain context; the answer is to shoot wide and go closer, of course. Zooming in completely destroys context; cropping after the fact is the same. A competent photographer shooting with a zoom will pick their perspective first, then use their feet to frame – perhaps tweaking very slightly with the zoom to finesse things, but not by more than a few millimetres – any more, and the perspective deviates noticeably from what was intended.
The final reason not to crop has to do with reasons of image quality: why pay for 36 million (or however many) pixels, and throw away half of them? You might as well save yourself some money, processing time and storage overhead and just buy a 18 megapixel camera instead (assuming of course shot discipline is identical in both situations). Basically, you’re shortchanging yourself.
At this point, you are probably wondering if there are any situations in which cropping for composition is acceptable – other than the hack-fisted art director. This might surprise you, but I think the answer is a definite yes. Firstly: if you have an imprecise viewfinder, you might make proper use of the intended perspective, but still be left with a bit extra which you couldn’t see; trimming the edges to what you did see is fine. (Though with experience, you’ll soon learn to compensate for this.) Secondly, if you really, really, really have no choice: the rest of the frame is may be unimportant, and you have no way of getting closer, but this once-in-a-lifetime historical event is happening now in front of you. Then maybe. (But even so, I’d probably find a way of composing to use the rest of the frame as context to the main event.)
I think the only major reason for cropping is to achieve aspect ratios that are non-native to the sensor/ camera combination you’re using; but this of course goes with the intention that you will compose for these aspect ratios at the time of shooting, with the perspective considerations that are implied. Explicitly, this means that the final cropped image must always retain one original dimension from the original source file – height if cropped to a more square aspect ratio, or length if cropped to a more panoramic one. Otherwise, we’re back to throwing away perfectly good pixels again 🙂
My personal shoot philosophy is not to crop with the exception of different aspect ratios. I don’t trim for viewfinder inaccuracy, even when I shoot with rangefinders; I guess I’m trained to ensure that my subject occupies a reasonably central portion of the frame, and stands out strongly enough that the viewer isn’t distracted by any bits that might creep in around the edges. Do commercial requirements bother me? Without a doubt yes, but I do my best to manage the client so they understand why framing and shooting with the final use in mind always delivers the best results. MT
____________
Visit our Teaching Store to up your photographic game – including Photoshop Workflow DVDs and customized Email School of Photography; or go mobile with the Photography Compendium for iPad. You can also get your gear from B&H and Amazon. Prices are the same as normal, however a small portion of your purchase value is referred back to me. Thanks!
Don’t forget to like us on Facebook and join the reader Flickr group!
Images and content copyright Ming Thein | mingthein.com 2012 onwards. All rights reserved
Ming Thein | Photographer 
Cropping is neither “good” nor “bad” in any absolute sense. The fact that Ming is a superb photographers and very knowledgable is no reason to accept his subjective opinions and choices as binding on the rest of us All that matters is the final image. So long as it is fit for purpose it dos not matter how m,any pixels are thrown away.
Sure, but if you can’t visualise that final image, it isn’t going to be fit for purpose. It isn’t about wastage, it’s about seeing. All of photography is subjective, and the internet is on a voluntary participation basis…
Hi Ming – I love going through your archives – you have written an encyclopaedia of photography, over the years!
I am well conscious of the need for a tripod, to minimise the risk of having to trim a shot – or worse, find that while I was busily staring at one corner of the photo, I’d somehow lost an important bit of it. When NOT able to use a tripod, I find it necessary to allow an area that can be trimmed later, to avoid the risk of missing an important element of the subject.
I usually crop only because of the format limitations of the particular camera. The shot may scream for a square format, but most cams are 2×3 or nearby. And if you have a square format, you generally find it necessary to remove one side of the frame.
Your comments on the impact of cropping on perspective are well-aimed. It’s a lot like the problems people create with zoom lenses, when they “make the zoom do the walking”, instead of using the zoom function purely to get the “correct” focal length to take the shot and moving to the “correct” spot to take the shot. Cropping can create the same perspective issues that misuse of zoom lenses leads to.
I am being asked to design something for a deceased photographer, and the family insists that I drastically change one of his photographs by cropping it to remove the watermark, center a tree in the inage, etc. Polite explanations why this is a bad idea have been to no avail.
I feel queasy.
I’m afraid I can’t help you in this case…other than to wish you luck!
I’d like to delineate a typical situation, that I encounter pretty often, when IMO cropping is absolutely necessary. It’s probably not your most typical kind of work, surely not the kind of work that makes you money…
A general subject that I like a lot can be described as “typical human behavior”. I like to register people in spontanuous and typical situations, that can (hopefully strongly) relate to us all, or to a certain group of people. How I proceed is first deciding about a place where chances are big to encounter “a certain kind” of behavior. Then I will look for a good viewpoint, or I will be moving in between people, observing and continuously chosing new viewpoints. Ultimately I’ll decide about a specific subject and shoot. Just like you said that, when shooting cars, the creative process is a matter of feel, I experience the same thing here. Now those shots are not about worldshocking newsfacts, but still, if I wanna shoot scenes like this, it’s impossible to predict them long enough on forehand, to be able to change lenses, position myself in the best viewpoint and carefully frame, because there’s at the most only a few seconds to take the shot. So I need to anticipate a possible “action”, at the same time decide about the viewpoint (the lens is chosen in regard to the environment of course), the subject(s) need(s) to be chosen pretty fast, the framing is done often from a too far distance since there are people all over, and you don’t know what will happen where and when. But in those few seconds I’ll frame and time the shot(s) in the way and at the moment that I know (or hope) I’ll get a strong composition in pp. IMO getting a good picture like this is not at all a matter of luck. You will regularly miss shots in this way, but you will very rarely (if ever) succeed in a good shot by luck.
So indeed, I think of a frame within the frame. In those cases, I will mostly crop between one and two thirds of the image. Sometimes, indeed, I will decide to take a shot of a pretty small “micro frame”, because I realy like what’s happening, although, I admit, it’s still not a worldshocking event. Remarkable: I very often think of a square frame in such cases, althoug my camera is set to 3/2.
Bottom line: when things I wanna shoot happen at an unpredictable place and moment and I have only a few seconds to take the shot, the timing will be the most important factor in composing (not the only one of course – for instance changing from eye level to belly level or moving a few steps can be important as well and is very rapidly executed). But cropping and finalising the creative framing process in pp is essential.
You can imagine that I’m very glad that my camera has a 36MP sensor…
To finish: I particularly like this kind of photography because, IMO, the unique power of photography is exactly to “freeze moments”.
I’m still not sure I agree with you, because other than changing aspect ratio – so one dimension of my original frame is maintained – I don’t crop at all, even for reportage and street work. In fact, I’ll throw it away unless it’s pixel perfect, too.
I know you wield only the very highest standards in whatever you do and I can understand that everything less must feel like “blaspheme” to you. But in your work, I think the importance of “the human factor” is somewhat different as what I described, therefore the need to go for that specific moment is, again in my assumption, somewhat less. I’m only wondering if you wouldn’t partly reconsider, and no longer stoically go that far to say that it’s wrong in all circumstances, when somebody else decides to give a particular moment more importance, hence decides to take the shot, throwing the pixels away in the pp instead of in the publishing fase.
I absolutely understand that you often get (almost) crazy when an art director messes up your pictures by not having a clear idea on forehand of what he wants. But when the whole creative process is in the hands of the same person, IMO it doesn’t matter in what fase the final framing is done.
I can also absolutely get the importance that you give to perspective. But, like other commentators, I believe there is no difference between changing lenses to a narrower angle or deciding to crop to a narrower angle, as long as you know, at the moment of the shot, what kind of frame you want to end up with. Shooting too wide has the advantage that the framing of the shot doesn’t need to be done with the same detailed precision, but then, with the shots that I mentioned, there is no time for very precise framing, for nobody, I’m sure even not for Ming Thein. You can come close, but it won’t be perfect. And if you happen to have the wrong lens on your body (happens a lot in that kind of work), then you can forget it all together, since your timeframe last only seconds. So IM very HO, it again comes down to the moment when to throw away the pixels: in pp or when publishing. BTW, I remember in the early days of publishing with DTP programmature (more than 20 years ago), that we often absolutely needed to throw away the surplus of pixels in pp, otherwise the Linotronic couldn’t produce the film for the press and got stuck halfways. In those days, to get digital images, we needed to scan from sometimes even large format film with drumscanner, or from pretty large format photoprints with flatbet scanner…
I didn’t comment on this article yet, but still, I’m stuck with some questions. To make it not too long, I’ll put only one in this comment. But please, Ming, do tell me if we’re elaborating too much. And before all, I want to emphasize the fact (again) that I have the highest respect for your work, both as a photographer and as the author of this website – you know I do – I repeat this mainly for everybody else who would read this comment. But I know that you’re always up for a meaningfull discussion. I hope this is one.
I wonder if it’s really that bad to, as you put it, “throw away perfectly good pixels”. Personally, in almost all cases, I don’t mind, since those pixels are thrown away anyway. Only if we print really big, ànd need all the details (this means that we also need to look from that close a distance from where we’re no longer able to observe the whole of the picture), we will have to use them. But in all other cases, the final medium certainly doesn’t require 36MP. I think you can agree with me that it really doesn’t matter if we throw away pixels in an early or later stage – as far as the wasting of the pixels is concerned. (I don’t speak of other matters here, like perspective and creative process.) So we are stuck with the relevant question: why spend so much money for a 36MP camera, if we use only part of it? BTW, in commercial work, I agree that it’s absolutely required, but the vast majority of the pictures, of the sold camera’s ànd of your readers are not in that case…
Here’s a thought. You recently published an article about car photography. While reading it, I thought of an older article of yours, a shoot of your own car. This is a very beautiful BMW Z4, if I’m not mistaken. Well, Ming, I don’t know if you have any speed limits in Maleysia, but I guess so, as do we. Ours is 120km/h, about half of what your car can do. So why would anybody buy a Z4, always “throwing away so many good HP”? I guess you can easily do 120km/h with only 20% of the available horspower. Yet, I can think of a few good reasons: you can love to know that you always have the potential at hand, even without using it. And you can love the horsepower for it’s accelleration power. And you can simply love to own such a marvel of modern technology… OK. But the same reasons count for our cameras. It feels good to know that you always have the 36MP at hand. And you can love the MPs for their cropping power. And, let’s be honest, it feels different to own a 36MP camera than a 9MP one.
I know you have more reasons for disliking cropping, but those are another matter. Here I only wanted to deal with the “waste of pixels” matter.
What do you think, Ming. Do I have a point? Or will you sell your Z4? lol
Depends very much what your final output is. For me personally, I’ll take all the pixels I can get because I can actually use them in the Ultraprints – you can see quite a visible difference between 12, 16, 24, 36 and 52 at the same output size. My clients – well, if they cover the sides of buildings; and they do ask for more. If having medium format at my disposal helps get a job, why not?
FYI I have taken the Z4 up to its limit on more than one occasion. And it’s more than double 120. 😉
I absolutely agree, it depends. And if clients ask for more, you simply must deliver. But in personal work, I think giving more detail only makes sense when an important part of the message lies in the details. Often this isn’t so, on the contrary, often the message is only communicated when looking at the whole picture. In those cases we don’t need to come very close to the image to completely get the message and then an output with less pixels will work just as well. Personally, I prefer offering detail “as an extra” then, but I stay happy with less as long as the message is clear. So I think it’s absolutely not necessary in those cases.
BTW, concerning the car, I have been your age too, and also love(d) fast cars. But here, driving in an illegal way quickly cost you your licence and a very high fine. Furthermore, the roads are getting more and more filled up, even in Germany, making things quite impossible. So there’s only “renting a day at the circuit” left, which I think is pretty crazy – buying a certain car, paying so much extra money, just for some rounds at the circuit from time to time. Bottom line, all that’s left is enjoying the feeling of owning it… and showing it…
I have to say, with the images you provided, it felt like you were shooting yourself in the foot.
I loved the cropped images as well as the originals. I was impressed with how different aspects came to life, different emotions were evoked. In some cases the same one heightened.
How about, instead of finding fault with cropping, you can create more moments within moments, art within art. Therein lies another talent -to see within a single photograph a continum of life within it: the riders on a rollercoaster, the tracks, the sky behind them, or everything all at once.
if you can capture what you envisioned in the exact way you envisioned it, congratulations. However, sometimes cropping can centralize on something else within the photograph you may not have noticied initially.
I don’t shun either. In the end, show the audience what YOU want them to see.
On the contrary – I think you missed the point. The difference in perspective, overall feel etc. is markedly different between the crops and the original; it shows you that you’re losing something every time you throw away a bit of the edge. That said, I don’t compose rubbish, so the sub-components of the image should be able to stand alone – but they are of course stronger as a whole. I address that in this article.
I’m still digesting this article, but when I look at the roller coaster example, I feel that the cropped image is stronger than the whole image, which is the opposite of your point.
I much prefer the 2nd image over the fourth (uncropped) for the following reasons:
– I can see the kids expressions (especially the one on the right) much better in the crop. I really like their contrasting reactions and feel it should be prominently featured.
– In the crop, the main elements are the roller coaster car (and riders) and the cloud – they give me a sense that the riders are really high (can touch the sky); I don’t get the same feeling in the uncropped shot because the sheer mass of support structures (especially the red stuff) draws my attention away from the apex.
– With more of the roller coaster track visible in the uncropped image, I can see that this isn’t a particularly impressive descent, whereas with the crop I’m free to imagine they’re about to go down a terrifying drop, thus further emphasizing the kids’ excited expressions.
Anyway, I admit that I do a fair share of cropping after the fact, mainly to remove distracting elements I shouldn’t have included in the first place, so I think I understand the point of this article. I should strive to have an optimal composition without cropping, and if I find a better one from cropping, then the original composition wasn’t the best to begin with. I still have a lot of learning to do when it comes to understanding what makes a great image so I really appreciate articles like this.
There’s one thing that changes everything: viewing size. An 800px web jpeg isn’t the intended output size, so yes, details and subtlety would be lost. It’s very different as at least a full size image on screen, better yet as a print.
Ming…I hope you’ll find the time, energy, and interest to continue producing essays of this sort. I find them stimulating and helpful. Plus — it’s always gratifying to follow the threads of conversation they engender and see a level of quality of thought and civility that is rare on the internet.
Thanks Michael. More to come, of course.
I came to your blog because of the Lumix LX7 review you have written. It was interesting. I hope to get the camera soon and get back to photography after a while. I have been browsing a bit on your site.
In this article, you seem to have it backwards. You seem to choose your composition based on your focal length. While it is one way of stating it, one should start with the image one wants to create in ones mind and then use your photographic tools. For instance, isolation of the main subject to enhance it, could be done using a shallow depth of field ( you might use a fast long focal length ), or by surrounding it with a negative space, or creating a contrast around it by placing a light background behind a dark subject. I think while you are being instructive it is a very cookbook kind of approach that will tie a beginner down with formulaic dictats like, if you are using a 24mm you should do so and so. There are ways to achieve compositions based on the lens at hand. For instance if a 50 mm 1.8 is all you own ( used to be the starting kit) , you might have to do things differently. But its not to say that the lens is driving the composition, it’s just that the compositions using that lens have to be done differently. Your tools will limit what you can do but always think of what you want to express and then see how best to create an image that does so using your tools. Dont try to find an image that fits what you want to express. You should “make” an image not “find” one. There are images that are ‘found’ by the great photographers and this is a slippery slope of much debate but for beginners a deliberate approach is best.
Also what you have illustrated in the first set of pictures is an illustration of something altogether different. It goes to show that if you had a 180 degree wide angle of infinite pixel resolution. You could conceivably create all the possible images lenses of all focal lengths would make. Like the small portion you showed in the first one is telephoto like and as you include more and more of the image we tend towards the wideangle lenses. This is a a simplistic way of describing it but it does do it comprehensively.
The cropping part historically was introduced by Henri Cartieer Bressons purist attitude towards visualizing the final image “in camera” ( In his famous book “The Decisive moment”). Since then it has been the source of disagreement between the creators vision and the editors need to use the image to a particular effect.
Best wishes.
No, I choose my perspective THEN my focal length. I think your view is very confused indeed…but if you think I’m wrong, and can consistently deliver images to prove it, then by all means start your own blog…
This is just playing with semantics, if you’ll forgive me for saying so. As we all know, physical distance, framing, and focal length are all mutually dependent. You can start with any one of them when conceptualising the shot, although physical distance seems to me the most important, as that determines the relative sizes of the various elements in the shot, and hence the perspective. Unless you change your position, using a wide-angle lens and then cropping later will provide the same composition (albeit with reduced resolution) as using a longer lens (or zooming) and not cropping.
I don’t think there’s anything wrong with cropping, either by a photographer or by an art director, provided it’s done intelligently. The notion of ditching a client because he/she has the temerity to crop a shot seems somewhat ‘precious’ to me. Photographers might their energies more productively chasing the client so as to get paid on time.
Technically, you’re right. But practically, you’re wrong:
a) Constant cropping kills the ability to previsualize; especially for less proficient photographers. This results in unquestionably weaker compositions because one isn’t making full use of the perspective.
b) Intelligent cropping by art directors/ clients is fine; it’s unintelligent cropping that bothers me, then complaining ‘your images don’t look like your portfolio’ – after they’ve hacked away at your composition!
Just re-read the article. Lots of good information about a wide range of topics. Many Thanks!
Thanks Eric!
I see a lot of frustration and, frankly, ignorance in this article and few valid points. I understand your frustration from clients who don’t know what they want exactly and then request a piece of the compositional cake – you have my deepest sympathy there. They should make up their minds in advance and not require you to make compositionally weaker images in order for them to mess with them later.
However, I think it’s a completely different matter if the photographer crops themselves. I crop almost all the time, as I don’t see a practical difference in “real” focal length and digital cropping (in terms of perspective it’s simply the same). I envision the photo in the viewfinder as well, roughly knowing where I’m going to crop in post. When you’re using a rangefinder camera, it shows you a huge overview of the scene, with overlayed frame lines highlighting the (imprecise) borders of the actual frame. I think planning to crop while taking the photo is more or less the same: You imagine frame lines in your mind for where you want to crop later.
Why would you want to crop? Well, sometimes you might just not have the ideal focal length with you. Digital cropping is equivalent to a digital zoom, giving you more flexibility. You choose a perspective first using your feet, and then you frame using your lens. But if you’ve only got one prime lens with you, this might lead to sub-ideal compositions or sub-ideal perspectives. (BTW I think there’s an error in your paragraph about amateurs with zoom lenses: “A competent photographer shooting with a zoom will pick their perspective first, then use their feet to frame” – it should be the other way around, using your feet alters your perspective, as is also said in the sentences before that). Or you need a longer telephoto lens. And sometimes the right moment is simply more important, so you have to be quick – your composition has to be working in any case of course, but a rough framing can be optimized in post.
Also let’s not forget that cropping was usual with 6×6 medium format cameras. Nobody in their right mind would have suggested this would lead to bad composition back then.
Lastly, you are right about wasting megapixels – but so what? I don’t have any obligation to keep my camera’s megapixels safe or to take care of them. There’s no FairTrade movement for pixels. More megapixels primarily means more freedom: They allow me to print/display bigger if I want that. Or to take framing (at least partly) into post (in that way it’s almost similar to Lytro’s taking focusing into post). As long as the final image is generally what I had in mind when I pressed the shutter, I don’t think the order of steps it took to get there matters. At all.
Cropping to different aspect ratios is unavoidable. But I think your response suggests that you’re happy to compromise rather than challenge yourself and make the most of what you’ve got. Each to his own, I guess.
Ming, I’ve been a reader of your blog for over a year and usually enjoy reading your articles – so when I say ignorance, I strictly mean ignorance in this specific regard. (If this wasn’t clear I’m sorry.) Cropping really is the same as zoom, it’s just applied after the fact rather than before. Many of your articles revolve around composition, so I just don’t understand your point of view: Images with strong composition aren’t created in camera, but in the photographer’s mind. So if you know what you want to achieve when you press the shutter, then I don’t see how this is supposed to mediocrity. And again, this was usual with 6×6 cameras as well.
Apologies if I took it the wrong way – the Internet tends to impersonalize things.
Cropping may technically produce the same results to the composition as zooming, but you’ll also know that I’m not an advocate of zooming to frame, either. It’s more about being able to visualize the composition and perspective than anything else – I think there’s still a difference between 1) you only have 50mm, but visualize a 100mm frame at the time of shooting and compose accordingly and 2) shoot 21mm for everything without making any concessions or adjustments specific to the perspective properties of the focal length and then see if there are interesting crops afterwards. One implies forethought and conscious composition: the other is a lottery. I think it’s pretty clear which approach will consistently yield better results.
So perhaps it’s me who should have been clearer.
Congratulations. Looks like a wonderful exhibit. It’s too bad I won’t be able to see it in person.
Thanks! A lot of the photos are on flickr, fortunately 🙂
I believe you need to distinguish photography for hire, eg., advertising, vs photography for art’s sake. As both a graphic designer and photographer (BFA in both) I believe cropping is an essential part of both applications. First with the camera, then with the computer. I crop for a gallery show much differently than for an ad–different market, different message. Designers and photographers are equally capable of making a bad cropping decisions but cropping is not inherently bad nor good. A good photo will not suffer in the process if handled thoughtfully.
I agree 100%.
Cropping if done thoughtfully can make you a better photographer. When starting out in photography I cropped more than i do now. But it let see what an image was inside the original image. It has helped compose my photos differently now and be more thoughtful.
Its a matter of learning as you go. Use it to your advantage for future image creation.
True, but sadly there’s far too much bad going on…
Jared Polin must be really agree with you..hehe… Nice article Ming,,,
Who’s he?
He is one the earliest photographer i checked out and learn about photography 2 3 years ago, and anyone who have been followed him must know that his famous tagline “Shoot RAW and don’t crop”
His website: http://froknowsphoto.com/
Best reverse take article on framing I’ve ever read. You’ve put some things on their head for me and it makes a lot of sense. I’ve always been anti-cropping, but now at least I know why!
A semi-related question though – I’ve been raised on 35mm slides and I have always felt crop sensors to lack something intangible, which I got back when I bought a D700. Anyone else know that feeling?
Thanks Wouter! Yes and no – could be the lenses, could be the depth of field properties for a given angle of view. I know that after shooting the 6×6 Hasselblad, my 35mm negatives definitely feel like they’re lacking something…
Interesting topic! In 1989 when I was a teenager, I filed out the sides of my 35mm negative holder and was quite firmly anti-cropping. Now, my reflexes are slower, my eyes are worse, my sensibilities have changed… and I have Lightroom and a resolution surplus! 😉
I do still often strive for frames that are presentable exactly as shot, so it does shame me slightly to admit it when a composition can benefit from at least a bit of a trim (among other things) in post. I offset that shame by crediting myself with being sensible enough to take advantage of a readily available solution. (I can’t outrun my 18-year-old self – my only hope is to be more adaptable…)
I guess my recent acceptance of a limited amount of cropping is part of a shift that has come with RAW shooting. I now see how certain adjustments – sometimes even decisions – are better made at a later stage. (Which is not to say there exists a long-term cop-out for sloppy shooting…)
I do feel that these process variations are to a large extent an individual artist’s prerogative. In visual art as well as music, I often say there is no “cheating” – there is only success or failure.
A final thought: if cropping is to be viewed as an evil, it should be mentioned that the tripod is one of our most powerful weapons against it! I just upgraded to a nice geared tripod head and that has had a far greater impact on my work than any new camera body or lens could have possibly had. (I know this doesn’t pertain to all styles, but where the option exists, the benefits can be quite dramatic.)
I wonder how much of edge trimming is because of imperfect (i.e. non-100%) finders. It’s a bigger issue with rangefinders, though I don’t really crop/ trim that much at the edges because the nature of the camera forces you to focus on your primary subject and make it strong enough that the rest doesn’t matter as much.
Success or failure and no cheating may be true, but if your award winning shot can only be printed to A4 because there aren’t enough pixels, then you’ll probably wish you hadn’t cropped…
Completely agree on the good tripod/ head: it engenders precision. I use one for all of my commercial work (and even lugged the big one around Putrajaya last night for an architectural job).
I would consider cropping also an artistic/creativity activity.
True, but it’s I don’t think anybody will argue that an art director or layout person who’s just chopping to fit is foo anything creative…
Dear Ming, Unfortunately this article comes across as “artistic elitism”. When you take a commercial commission, your position is still as an employee. The only solution is to only accept assignments with those clients that agree with you or turn down the work.
Not true. If the client clearly tells you what they want to use the final output for – aspect ratio, leave space here for copy etc – then there’s no problem. There are plenty who have no clue what they want, it’s our job to educate them. If you don’t, then the mediocrity perpetuates.
Agreed, but the client’s artistic understanding, good or bad, can be undermined by the people making the decision. We all now what a child’s swing looks like when designed by a committee. It’s a fact of life. In an ideal situation I would tend to agree with you. Even then, it is truly hard to previsualize each shot perfectly. And, sometimes a better image emerges from a second look.
True. Where possible I try to have the AD/ CD on hand (if we have one) to give feedback straight away. That said, some of the more satisfying assignments I’ve had – both for me and the client – have been when I’ve been given an open brief…
Where the ultimate reproduction of the shot is all about the photograph, I agree with you. But generally it isn’t. As a designer, you don’t crop a shot that is being reproduced as ‘art’, as it were. (Just as you don’t crop a painting unless you are explicitly going for a ‘detail’.) On the editorial battlefield, however, photographs are often used in the context of a layout or editorial concept, which may deserve as much consideration as the aesthetic qualities of your photograph. In other words, compromise may be called for. This does not justify Really Stupid Cropping (which I admit does happen). If you are designing a book or magazine cover to a certain brief, which is more sensible – to crop a bit off the photograph or change the format of the book or magazine for the photographer’s sake? (Designer leaves production manager’s office covered in bruises … ) And another thing (as taxi drivers say), spare a thought for the designer working most of the time with images sloppily taken by a photographer who won’t be concerned to the slightest extent by cropping, insensitive or otherwise, provided that he or she gets acknowledged and paid.
In an ideal world, we’d compose for the output format/ aspect ratio rather than force either one to fit the other. But yes, there are plenty of Really Stupid Photographers and Designers…
Re: Cropping I think it depends on the context. I shot with a prime on the street and sometimes a little bit of cropping is helpful to clean up a composition. I certainly don’t think that there is something inheriently wrong with it. However, I did notice that (seemingly paradoxically) when I made the move away from zooms to fix focal length that I started doing much less cropping. Perhaps because I am thinking through the image more carefully before making the shot.
Tidying up edges is one thing, changing from 24mm to 85mm is quite another. As for zooms vs primes – I think you’re starting to visualize the perspective of your lens, which is a good thing 🙂
Interesting and thought provoking article. I’ve been thinking for a while to do a project where I just shoot squares on a D800; I tend to think framing in terms of the format, but with these pixel counts I can afford to throw something away and frequently a better composition is available in another format than 3:2. It just requires a mental effort (for me) to compose with the intent to cut off the sides or top and bottom, although I’ve become fairly proficient at stitching.
One practical point you didn’t bring up against cropping is convenience of composing and focusing: when cropping away a lot, one needs to compose with a much smaller part of the viewfinder/LCD, something which is not as pleasant as using the full size. Frequently this leads to problems; focus errors, clutter that was not noticed and poor visualization of how the image would scale. Or at least that’s how I feel; I don’t do heavy crops. Leica users seem to complain that using tele lenses is hard because the image in the viewfinder is small and I thinkg the same effect is going on here.
Naturally, if one composes for 21 mm and wants to crop to 35 mm, I would say that one needs to be extremely lucky that the composition is strong in both cases. But if one is prepared to compromise quality then much more is possible…
I often feel 3:2 is a bit of a compromise too – most of the time my intuitive compositions are 16:9 or square. However, Nikon provides viewfinder masking for 3:2 DX and 5:4 only…sigh. As for Leica users – it’s not so much framing as focus precision; the RF patch covers a large portion of the final scene, but is very small in the finder making it difficult to see exactly what you’re focusing on.
I am a Creative Director and as the the title suggests, I direct the creative. From my point of view when I brief a photographer I always explain exactly how I want the image to work, and the photographer understands how its going to be used. And it is my priority to get that shot. Once we get it and time permitting we can explore other options.
There are occasions where we have a general idea of the images we want for a client and this is where a good art director and photographer can really explore the possibilities within the brief In these cases what we are producing is an image library for a client to use when needed. When creating an image library ou are never really sure how they will be finally used, but what I find is that a good image is always a good image.
When photographers are asked to leave room for cropping all that tells me is that the art director is not confident in what he wants, this puts undue pressure on the photographer to deliver something that he is not sure about, and if things don’t work in the end everyone blames the photographer, when really there should have been more up front planning. My art directors no never to brief a photographer with a half baked idea.
Finally, i do love the first image, for me the children’s expression tells me that the roller coaster is really big, in the end I don’t need to see it.
Vito – I completely agree, but sadly we rarely have the luxury of an art director, or worse, have an art director who doesn’t know their job and tells you to leave room to crop…that is the unfortunate state of the creative industry in this country.
Hi – and thanks for this article – your thoughts about photography really help me to grow.
The following quote makes me wonder if I have misunderstood something:
“… then use their feet to frame – perhaps tweaking very slightly with the zoom to finesse things, but not by more than a few millimetres – any more, and the perspective deviates noticeably from what was intended”.
I always thought that the only thing that determines perspective is the distance and zooming is nothing more than cropping/framing (keeping perspective the same).
When you say ‘perspective’ do you mean ‘angle of view’ or ‘point of view’?
(After writing this question I think I only translated it wrong!?)
Thanks
Chris
I think you’re confusing angle of view (plus the impact it has on verticals, converging lines, relative sizes, etc) and physical position. Zooming isn’t cropping. You can shoot the same subject and frame it to be the same size, but if you do this with 12mm and 200mm lenses you’ll notice the surrounding objects are very different relative sizes to your main object – that’s the impact of a change in perspective.
Thanks for trying to answer may very ‘basic’ question.
Quote: “You can shoot the same subject and frame it to be the same size, but if you do this with 12mm and 200mm lenses you’ll notice the surrounding objects are very different relative sizes to your main object – that’s the impact of a change in perspective.”
Yes, but the reason is because you are changing the distance to the subject – or am I wrong?
What I mean is: If I shoot a subject from 20 meters away with a 200mm lens and then change to a 35mm lens and take a second picture from the same distance -> and after that I crop it to get the same frame as with the 200mm lens -> the picture will look the same (with regard to perspective). So this would lead me to the conclusion that it makes no difference (!) with regard to perspective (!) whether I change the lense or do some cropping. (You can find an explanation of what I mean hear: http://www.mhohner.de/essays/myths.php#focalper ).
With ‘perspective’ I mean the effect how fast elements that are further away get smaller.
So does zooming (without changing the distance) change the perspective?
To quote the source above:
“Perspective has nothing to do with focal length”
“Perspective is only influenced by your position, and field of view is only influenced by focal length. ”
thanks again
chris
I’m thinking in terms of angle of view: how much of your frame does your subject occupy? I think we may be talking across purposed – your example isn’t quite comparing the same thing. If you’re cropping a 35mm frame to 200mm, it will have the same angle of view but not the same depth of field properties.
This was a very interesting article and you had me fooled by the headline, I thought the article would take a different direction.
I would really dislike it if other people cropped my photos, specially if the didn’t understand my idea behind the composition and the image itself. I am not a pro photographer myself but to some extent I have experienced the same myself in my profession as a software developer.
Personally I mainly crop my photos for 3 reasons:
– Adjust for a tilting horizon or a similar user error. Maybe the most frequent reason but at least it is only a small amount that gets cropped away.
– Wrong format, when I think the photo will work better in another format like 4:3, 5:4 etc
– Zoom in post. I only use prime lenses and sometimes it is just not physically possible to get the framing exactly right.
Of the reasons you have, format/ aspect ratio is the only one that’s really valid. You can always compose to include more with a prime. Out of curiosity, what did direction did you expect from the title of the article? 🙂
I agree that it is the one that is truly valid but sometimes there are limitations like a water shore or a wall etc. “You can always compose to include more with a prime.”, I actually think that it is the opposite that is the major problem, that is to include a little less without messing to much with the composition.
I connected the title with dogma-style photography and the articles I read about it but at I also recalled one of your recent post about using a wide-screen format (something like 1:4.*) so maybe I wasn’t to worried.
Ah, I suppose it depends on the situation to some extent. Yes, if you’re stuck with something too long then you’re out of luck – stitch, I suppose. I generally go wider with primes (28/85, or lately 21/50) so this isn’t so much of an issue. I’d rather include more context than have too much isolation. Are you referring to this article on 2.4:1 widescreen?
Oh, to add – I’ve also encountered cases as yours where the client does the cropping. Let’s take food photography, I took a photograph of a dessert just fine, however the person handling the menu design wasn’t a photographer. So I when I saw the final result, I was appalled to see that a perfectly fine photograph is now composed as such that the focus was on the bowl than the content of the bowl! 😦 Had they used the photo as it is, it would’ve been just nice. Even if they cropped it – the least they could do is maintain the composition to focus on the subject.
I absolutely hate that when it happens. I’ve dumped some clients because of it; they then attribute the images to you and it makes you just look incompetent.
“The final reason not to crop has to do with reasons of image quality: why pay for 36 million (or however many) pixels, and throw away half of them?”
Simple – if it makes the picture better, crop it! No reason to be calculative over the sensor’s pixels. You pay for the camera technology as a whole. 36 MP gives a lot more flexibility than ever before. Even if you took just 1/4 of it – if it is a good picture, it is a good picture.
However when I speak to those new in photography, I will still advise them to frame the image as best as they can. This trains them to be observant, it trains them to be careful in getting the image RIGHT! I think this is the right way to do, I’m not against cropping if it improves the picture but I make it a point to avoid cropping whenever possible. I find that if new comers are very reliant on cropping, they’ll end up being too dependent on post-processing.
Even for my commercial work – I compose as best as I could. Cropping just wastes a lot of time. I will frame it right and align the lines as best as I could to minimize cropping / adjustment work.
If you’re in a situation where you have control over all of the elements as is usually the case for commercial work, then you shouldn’t need to crop at all…
Best not to work for a newspaper then – they are the worst. I once shot a bride and groom at a fancy wedding venue. The pose was vertical dancing pose, full body, against a simply gorgeous floor to ceiling paned window and blue sky and a garden outside in the distance. The bride was perfect and the dress draped beautifully. When the newspaper editor for the lifestyle section of the paper got hold of the image, he used the heads.
Yep. I hate editorial work. Mind you, we tend to leave a lot extra in commercial too…
Ming, a question – if one of the main reasons that cropping is bad is that it signifies lack of previsualization on the part of the photographer, does this really matter to the viewer? I mean, the final image should stand on its own, without attached tags saying it was uncropped (or unprocessed, for that matter) – in this case, who, besides the photographer him- or herself, can say whether it was cropped or not? If the image is strong, does it really matter (to anyone but the photographer) whether it was framed correctly on the spot or framed correctly in Lightroom?
It doesn’t matter to the viewer insofar as final output quality is affected by the reduced resolution available for printing. However, what does matter is if the composition is weaker because the properties of a give perspective aren’t fully utilized…
Cropping can affect the output quality for reasons other than resolution. One example of this is a landscape shot I saw last year printed quite large.
The right portion of the image seemed strangely out of focus compared to the left side. At first I could see no reason for this then I realized what I was seeing was the fact that the wide angle lens used obviously wasn’t great at the edges. This meant that when the image had been cropped to change the composition the focus of the image was not at all uniform even though the subject was distant and thus relatively. This drop in image quality was visible on both the subject and the sky above it.
Similar non uniformity could result from other properties of lenses if not corrected in post processing vignetting or CA.
That’s a good point actually – you’d have to be cropping fairly heavily to see this with good lenses, though. But non-uniformity is definitely an issue.
Absolutely, I agree re the ill trained art directors, etc…..and alluded to this very issue in your ‘stereotypes’ post a week back. if a photographer has given the composition adequate thought….then cropping should be minimal, at best. sometimes, though, you must admit…we all reach for shots at times. generally, iI find myself doing this in the street or fast moving sports/ social gatherings….and something weird is in the frame that either I didn’t see because i was so intent on getting the ‘essence’ of the scene. Thanks, Ming, as always….
No problem. Sometimes it pays to just take a moment to scan the frame, not just the center; or try to see things as a decomposed abstract of form. Then you just move the camera around until the frame balances itself out. More on this in a future article…
i could not agree more with you MT. its really bugging me a lot as a “creative businessman” where clients want to have highest quality as possible with a very tight budget and ridiculous deadline (also add serious competition in the industry, ready to take your prospective jobs). but for recent year i do make approach to educate client about this issue so along time at least we can hope we can reach the same ground with them.
2nd point about zooming change the perspective, i think the main culprit is because how our industry change a lot these days.. more affordable gear so people can make progress really fast where at some point the gear exceed their knowledge(basic tech stuff) skipping more fundamental aspect of photography itself. (personal experience).
“skipping more fundamental aspect of photography itself. (personal experience).”
Agree as well here. When I had a DSLR kit I had 3 zooms. All quality Canon L lenses so good grade. Since selling out of that and into a small system I have not even thought of buying into a zoom lens again. I prefer to have one less thing to think on the camera and be more thoughtful in my composition now. Not to say I wasn’t thoughtful but it was easy to be a bit lazy in the mind when out and about.
It’s just another aspect to creating an image vs. taking an photo.
The first point resonates with me because I’ve had clients hire me based on my portfolio; they want similar images (of course). When I deliver those, their art department complains there isn’t enough room for cropping…what they don’t realize or refuse to acknowledge is that by delivering a larger image and cropping, you’re changing the composition and overall impact of the image significantly.
Education – both for photographer and client – seems to be relatively ignored these days. We have to do our bit for both.
Ming,
Wonderful topic. Perspective is key. Best Wishes – Eric
Thanks Eric!
I like all of the non-cropped pictures the best with the exception of the first. The expressions on the kids faces are priceless.
Haha, it would be interesting to hear why you don’t like the first one?
In the other pictures I strongly prefer the non-cropped. However in the first one the tightest crop shows the facial expressions better. The facial expressions are amazing and there is still some context and leading lines etc… and it made me smile and laugh. 🙂
That makes sense. I too prefer the non-cropped – as that was the way it was initially composed – but I just wanted to show that there were options…
I agree regarding throwing away pixels. We should be trying to become better photographers than post processing gurus when it comes to composition (but also applies to getting the light correct in camera).
I crop to take edges off if required, or change the size and aspect of the image. I have been shooting more with the 1:1 format on the OMD and experiementing with it. I am starting to use this menue more so I can move beteen shots and formats on the run, once I feel for how i’d like the image. We are very lucky given the technology today. So sometimes I may crop a 3:2 or 4:3 into this if I feel it improves an image. Probably about 1 in every 20 images I’m guessing.
A question for you though Ming……if we go to 1:1 does this reduce the pixel count in the OMD or retain and condense them? Sorry if it’s a silly question!
It reduces the total number of pixels because you’re basically throwing away the unused bits.
So throwing them away (by reducing the sensor size again) before the image is even taken? Is that correct?
It basically crops before capture, yes. So unless your sensor is multi aspect – not many cameras have this – then you might as well save all of the info and crop afterwards.
Thanks Ming…interesting. May change my take on experimenting with this being the case!! Maybe just buy Blad…!!!
Blads don’t work held vertically – had to try once with the CFV back, because it shoots 645…