When to stop?

[Ian Key]

How do you know when to stop post-processing to get that perfect picture, is it a case of less is more?
Or do you like me keep coming back to a favourite and try to tweak it just one last time?

[John Morton]

I redo favorite images every now and again; but I always start over from scratch, beginning again with the original image and working out from there. That way I can go in a different direction than I did with earlier edits, and have a greater chance of ending up with something 'new' rather than something 'overdone.'

[Mike Buckley]

In most cases (not all), my experience is that the less post-processing that is required, the better the image. That may only apply to me perhaps due to the style of most of my images and my approach to photography.

I rarely return to an image to add more post-processing or to change the initial post-processing. However, if I happen to come across an image that I realize I could have done a better job or used a different approach the first time, I won't hesitate rework the image.

A friend of mine has a daughter who went to art school. When painting a work of art, she was trained to get it to the point that there are no more than six imperfections before moving on to a new piece.

[Donald]

I agree with John (and Mike who posted as I was writing this). I, too, rarely go back to re-work an image. Once I have created the image that I saw in my head when I pressed the shutter, I am reluctant to go back and then create something else, even if that original is full of imperfections (probably a lot more than 6!).

But the maxim that less is more is a good one to strive to implement. But that, of course, is dependent upon having the done the work with the camera to get the best possible raw (intended) material with which to work in post-processing.

I think it's inevitable that when we start out on the digital journey (I know I did anyway) we probably spend far too much time tweaking. That is probably a necessary part of the learning process. As we develop our skill in terms of 'getting-it-right-in-the-camera', but also as our vision grows in terms of what it is we are trying to do in photography, I think we capture the images that more closely fit with what it is we are trying to create. The consequence of this is that we have to do less in post-processing to arrive at the image that we saw in our mind.

EDIT - I would add to what I wrote above:

I believe that another crucial factor in this is dependent on what you are doing in post-processing. If your approach to photography (which is perfectly legitimate) is to take lots of photographs, upload them and then think what you'll do with them, then you probably will do an awful lot more tweaking in post-processing work as you experiment and explore to 'get the look' that you're after.

My approach is to undertake post-processing with a very clear aim in mind - to create the image I 'saw' when I pressed the shutter. That, I believe, is a very different task that you are then undertaking and has implications for your approach to post-processing. So, I am working towards a very clearly understood end point and, through experience, know what the things are that I need to do in post-processing to achieve that end.

[John Morton]

My introduction to digital came by way of a Minolta DiMAGE Scan Elite 5400, and I still remember how often I would scan a 35mm negative or slide, edit it in Photoshop, and exclaim to myself: "THAT'S why I took this photograph!" There was definitely so much more to do with images once they were digitized that my enthusiasm for the whole process knew no bounds.

With a Nikon D700 in my hands, I quickly began to notice that the images coming out of the camera were... well, as reluctant to call them perfect as I was, often I was still hard pressed to decide what else I could be doing to them to "improve" them in post processing. Contrast could be improved (that goes without saying), colors could be tweaked a bit, maybe the color temperature of the white balance could be adjusted subtly; but there wasn't very far to go before that line between 'good enough' and 'too much' was crossed.

Having started out with manual 35mm cameras (anyone remember the first choice of newsrooms, the Pentax K1000?), it was a little difficult to accept that the camera I was using actually did its best at getting great photographs; and that its best was very, very good indeed.

Still though, there is something somewhat elusive about photography that cameras do not quite capture; and it is something as fleeting as a moment because it is in fact the essence of time. Someone here at CiC has commented that all great photographs convey emotion to the viewer, but I would have to agree to disagree on that matter. I would say that what is conveyed is somewhat broader than emotion, although emotion does fit into it: what is being captured by the photograph is the dynamic of time (which is a prerequisite for emotion).

Now I know that sounds counter intuitive: still images capturing the dynamic of time? But to my mind, photographs capture the 'in between' that separates every 'before' from an 'after'; and that is the dynamic from which photography takes its slices of time. That dynamic can be an emotion; it can be a transition, as in the rising or setting of the sun; it can be a fleeting moment when light falls upon a subject 'just so'; or it can be, simply "this specific place at that particular time": the street outside a French cafe at three in the afternoon, or a forest trail when the wind drops off and the sun slips below the branches of the trees.

Deciding where the dynamic of a photograph is to be found is a tentative and tenuous thing, because it is in the framing, and it is in the color of the light, and it is in the density of the shadow. None of that necessarily comes out of any camera as such: those are all things that come out of the photograph because we know they are there and we know how to coax them out to where others can see them.

So for me, having other people say "Wow, that's a great photograph" and saying myself "Yes, that's what it was about - that place, at that time" are sometimes two different things; and getting those great photographs out of the camera can be a lot easier than bringing out the temporal dynamics in a photograph that define the essence of a moment in time.

Admittedly, getting the photograph 'right' in the camera during the image capture is the only way to go and it is a level of technical expertise we should all strive toward as photographers; but post processing isn't necessarily about 'fixing' photographs that have technical imperfections. With the power and capabilities we find in today's photo editing software, post processing is also a journey of discovery where every photograph has the potential to express itself in ways that are not immediately obvious.

So, I think that there is a lot to be said for post processing and for re-processing old images through new edits: not because we have to, but because we want to (and certainly, because we can)... because it is fun!

[Donald]

A wonderfully insightful, thoughtful and thought-provoking commentary, John. One that invites us to think deeply about our photography.

Given what you write in your second last paragraph, I hope my comments did not convey the impression that I consider post-processing to be just about fixing the imperfections that cannot be addressed in the camera. Not at all. I do think of post-processing as an integral part of the process of creating the image that I envisaged making. I believe that pressing the shutter is the end of the first part of making an image. The second part starts when you begin processing in the darkroom (if you're shooting film) or on computer (if you're shooting digitally).

When I write about my own reluctance to return to an image, perhaps reluctance is not the best word. What I find is that my soul and passion is not engaged in the same way as they were when I first made the image. When I first work on a picture, then I cannot be disturbed (as my partner knows). My whole passion and energy is focused solely on working with that image and creating what I think, at that time, is the best image I can make. I lock into that image and am oblivious to all that is around me.

Your statement, in relation to the dynamic of the image, that, "None of that necessarily comes out of any camera as such: those are all things that come out of the photograph because we know they are there and we know how to coax them out to where others can see them.", is, I think, so true and the end result that you set out is what I hope I am striving to do at every step in the process of making an image.

In conclusion, I hope that others both now and in the future can find your post above and learn from it. I will certainly save the link to it and refer aspiring photographers to it, inviting them to reflect on its contents as they proceed on their journey of learning.

[Mike Buckley]

Donald's point about having a clear vision of an image before releasing the shutter has profound impact on your approach to postprocessing. Though I agree with him that there is nothing wrong with snapping a lot of pictures and deciding later how to postprocess them, I do believe that the photographer who uses the other approach is likely to consistently get the most number of top quality images in the least amount of time.

It just occured to me that the answer to Ian's questions could and perhaps should be different depending on where a person is in the postprocessing learning curve and whether there have been any major developments in the postprocessing software being used. I'll explain.

The person who is relatively new to postprocessing will surely learn about a particular technique a year from now that could be applied to images made years earlier. However, the more one studies the great tutorials at CiC from the very beginning, the less likely that is going to happen. Even so, there is a LOT to learn and sometimes it takes awhile for the proverbial light bulb to go on. So, be prepared for and excited about the idea of learning something really great even after you have become comfortable with your postprocessing skills.

Sometimes a software developer will release a new tool that does something that could not have been done previously. A pro photographer that I used to be in touch with was so pleased with two new tools made available in CS5 that she is revisiting most of her photographs. (She feels that CS5 has made more advances than any previous release.) I, on the other hand, primarily use postprocessing software that hasn't had an upgrade since June 2008, but that's a whole other story.

[Mike Buckley]

Great post, John. Lots to consider and reconsider.

[FrankMi]

How do you know when to stop post-processing to get that perfect picture, is it a case of less is more?
Or do you like me keep coming back to a favourite and try to tweak it just one last time?

There are a lot of variables at work that can be significantly different for different photographers, images, and experience levels.

If you are trying to get the image to look as closely to what you saw when you pressed the shutter, I find that usually less post processing is better but that can vary wildly with your skill in both taking and post processing the image.

When I was just starting to learn post processing, I found myself trying a lot of different shooting and post processing approaches to learn the best technique for a particular image. To continue exploring new techniques I still fall back on this procedure on a frequent basis. As I became more skilled in the various post processing techniques I also became more skilled in getting the image captured correctly. Both learning aspects contributed to produce a better image with far less post processing effort and a much better sense of when to stop making adjustments.

For some, reality isn't the goal but rather to produce a pleasing art form that may have very little in common with the original image so where to stop post processing depends on your artistic vision and what you are trying to produce.

Either way, the best time to stop post processing an image is when YOU are satisfied with the results. Just be aware that your vision of the result may not be widely accepted by others.

If you are trying to produce a result that has wide appeal, you may need to go back repeatedly and apply the feedback you receive in post processing (or reshoot the image - or apply what you have learned to your next image). Respectfully understand that you can't satisfy everyone's vision of what is right so in the end, you will still most likely stop post processing an image when YOU are satisfied with the results.

Hope this helps!

[Polar01]

Ian: I am going to add my fifty cents worth, John has an excellent post, what is it all about. However to get to your question of when is enough, enough. We all try ot get the image to look as close to as what we saw, sometimes though what we take a image of is not what we actually see but what can be seen. Ok I will explain, one of my best images, if seen before I processed it, would have been what the heck did he see in that, it not a very interesting image, ok ugly is what it was. However it was what I saw unseen that was there, I could bring out the best in the clouds, the hills in the background, the old buildings on the ridge, and the foreground. So against the orginal image yes it was over processed as it looks nothing like it, now it is an image that people walk pass, stop, backup and look at. It is what John talked about ""Yes, that's what it was about-that place, at that time"" then you know that you have the right amount of processing.

Cheers:

Allan

[jiro]

Using my own experience to answer the question of the poster, what I noticed is that I used to play around, tweaking everything on most of my images because I haven't found my "style" yet. On my first 6 months in digital photography, I started applying heavy contrast, vivid colors, a lot of sharpening, and a bit of extra exposure to my images. Then, seeing the work of some exceptional photographers that I admire, I noticed that their images are quite the opposite of what I am producing. Then narrowing down to only two photographers that I admire and study their work, I noticed that their images does not have that extra sharpening needed, their colors are not overly saturated, and they both apply vignettes on their images. I did not copy their style per se but rather tried to emulate them and from there I was able to find my own "style". Finding mine, if it took me 15 to 30 minutes and even an hour before to tweak one image, now, it only takes me about 3 - 5 minutes to process my image from RAW and I'm done. If my computer is faster, I might even say that I can process an image 1 -2 minutes max and that's it. It took me about 15 months to find out what I like or what my style is. It varies from person to person so there's no rule for that. I also agree with the idea that if you got the image you want from the camera's output, you'll spend less time and effort adjusting it on the post-processing. I hope you'll find your style soon, Ian and good luck with your photography.

[Manfred M]

Simple answer - when it is good enough.

In the real world, creative endeavours (not just photography) are limited by time (deadlines) and budget. These are real constraints that do help dictate how to define “good enough”.

In my professional life, I manage a design engineering group, and this like photography is a creative endeavour. I have had the opportunity (and challenge) of working with recent engineering graduates on projects. One of the biggest challenges that recent grads seem to have is knowing when to stop. Part of the issue with someone new to any field is that there is a learning curve because “we don’t know what we don’t know”. With experience (especially with the tools we use), we get better and faster.

I’ve learned a few “tricks” that apply equally to engineering and photography:

1. Planning – whether one is taking a picture or designing a product one has to have a plan as to how to achieve that goal. In photography, that starts with having a clear idea as to what you want in the image. This applies to both shooting and post-production.

2. Creativity is an iterative process – Nobody designs a product or business process in a single go at it and the same thing goes for an image. We have to work a scene with our camera and get multiple different shots to get good coverage. The same thing goes in post-processing; we try things to see how they work, and keep the things that we like and discard the things that we don’t like.

3. Make your mistakes early in the process – these mistakes are easiest and fastest to correct. The engineering example is to find your mistakes in the design stage; it is far faster and cheaper to correct them there than in the production stage.

The same thing applies in photography. If you take a picture that is really good, you will have a lot less work in post-production. I find that there is about a 10 or 20:1 ratio of effort difference in getting it right in camera versus fixing things in post-production. What I am trying to say, an extra minute of work behind the camera will save me 10 – 20 minutes of Photoshop work.

4. Know when to stop – amateur photographers are a bit like recent engineering grads. They want to keep on improving but don’t know when things are good enough. Here too, I have a couple of rules of thumb that help:

a. If you keep flipping back and forth with the same or similar approach in post-production you should stop because you are no longer making any tangible improvement to your work; and

b. If you have spent more than 5 or 10 minutes trying to decide what to do next, you have likely the “good enough” point in your work.

The only caveat I will add here is dealing with “image editor’s block”, where you can’t figure out exactly what you don’t like about the image. In that case go away from your screen for 20 or 30 minutes and then get back at it. If you still can’t figure out what to do, then you have hit the “good enough” point.

[Donald]

The only caveat I will add here is dealing with “image editor’s block”, where you can’t figure out exactly what you don’t like about the image. In that case go away from your screen for 20 or 30 minutes and then get back at it. If you still can’t figure out what to do, then you have hit the “good enough” point.

......... and if you feel that it's at the , "This is as good as it's going to be" point and you're still unsure if it's good enough, hit the delete button.

[Mike Buckley]

Great post, Manfred. I sell printing services pertaining to the design documents that engineers and architects create. Your post about their collective and individual work processes makes so much sense, especially when explained in context of the photographic process as you did.

[John Morton]

Perhaps then, with reference to Manfred's insightful post, we must note that "when to stop" can often be "when to start doing something else"; and this shift in emphasis can easily be the result of how the images are going to be used. Something that is going through a web press onto newsprint might need an idiot-proof edit (or perhaps "gremlin-proof" would be more applicable): not to bring everything out but to keep things from being smeared together; in which case one is definitely editing with the aim of keeping things easily recognizable, rather than tweaking to bring out subtle nuances.

Sometimes, there is such a thing as "too good"; and strong contrast may in such cases win out over subtle gradients, for instance.

I believe the early ink jets demanded somewhat of a similar approach; although I was lucky enough to get my hands on a refurbished QMS/Minolta color laser printer so I didn't really experience that aspect of home printing myself (thank goodness).

So not only the realities of business deadlines but also the practicalities of production become a factor; and perhaps some of the most delicate juggling of editing strategies emerge within that nether region of colour management between what we see on screen and what we can get in print.

But that can be more a question of "when can I finally be done with this" than "when should I stop".

Perhaps the purest answer to the original question, "When to stop," is: 'when you have a finished print in your hand that you want to make more copies of rather than trying to get a better version.'

[Steve S]

I mainly do wildlife photography, with a few landscapes here and there. The way I approached editing, is first i defined the things "I" like in a photograph.

For me, i like............... clean/ very sharp/very contrasty/accurate colors (neutral to slightly on the cool side for the most part)

After you find the things you like in an image, you need to design a 'workflow' the gives you consistant / 'repeatable' results.

A good workflow will put limits on the basic things all photos need, and keep you from overediting.

[Mike Buckley]

A good workflow will put limits on the basic things all photos need, and keep you from overediting.

That's an excellent point that I had never thought of even though I now realize that adhering to a precise workflow has helped limit my overediting. I initially documented every step of my workflow so I wouldn't forget what to do when and so my editing would be consistent. As I continued to learn, I modified my workflow document from time to time. Though I haven't changed it in a long time, I still refer to it to ensure that I am reminded of the available processes within the postprocessing toolbag and the order of using them.

[jprzybyla]

I like Steve who posted in this thread above, shoot mainly nature and birds. My main concerns taking the image are exposure and focus with an eye to what the background will look like. I watch for blown highlights and deep shadows on the subject while setting the exposure, with either of those the image is most times unrecoverable. I want the focus to be precise on the bird because only then are the fine details in the feathers shown. I admit to being a chimper carrying a Hoodman Loupe to examine the image on the LCD screen of the camera. Most times when I have done this correctly post processing is minimal. My workflow always is the same. I begin by cropping the image for the best composition, the move on to tweaking the exposure and setting the white and black clipping points. Rarely do I change the white balance. The automatic white balance in the Nikon D7000 most times is excellent. If I do change the white balance mostly it is to warm an image that appears too cool. I do this in the Basic and Tone Control Modules of Lightroom 4. Next the image is sharpened, once again minimally which is only done to overcome the effects of the image being converted from analog to digital and effect of the Bayer Filter. Then I apply the lens correction for whatever particular lens I am using. Depending on the image I may apply a slight amount of post crop vignetting. The image is then copied from the DNG to a TIFF and opened in Photoshop Elements. If there is noise in the image I use Neat Image as a plug-in to control that. If the image needs some cleaning up I use the Spot Healing Brush and the Clone Stamp in Photoshop Elements. The final step in Photoshop Elements is to apply Local Contrast Enhancement as required.

I do this on every image in the order listed. On some images it is necessary to use the Adjustment Brush in Lightroom to correct or sharpen areas of an image. Whereas the main expsoure controls affect the whole image (globally) the Adjustment Brush only affects the areas that you brush over and mask (locally). I find that if the image is excellent to begin with the risk of doing too much doesn't exist. I believe that doing too much post processing comes about from trying to make a good image excellent, which in my opinion is just about impossible. I think the computer axiom of... garbage in, garbage out holds true.

[Scott Stephen]

I just need to go out amd make some magenta friends. Then I could stop worrying and working in Post to properly capture what I saw when I shot the picture..

[John Morton]

Good point, Scott.

Now according to Dan Margulis (author of "Professional Photoshop: the classic guide to color correction"), one can in fact edit color accurately simply by 'working with the numbers' and watching the balance of red, green, and blue (or cyan, magenta, yellow and black) for any identifiable flesh tone as those statistics appear in the "Info" palette in Photoshop... provided one already knows the balance for flesh tones.

As true as this advice is, I've found that profiling my monitor and laser printer using a ColorMunki device is consistently a much more helpful approach.