Thursday, September 16, 2010

Landscape Photography - Shooting and Processing

Q: How do I shoot and process my photos? 

People ask me this question at least 20 times per day, seriously!  So, I thought that I better answer it all in a single rather large blog entry so as to have a single point of reference for the future.     I'll include pics with more information in future posts.

I have a feeling that every decent photographer gets this question often, especially from people that are just learning photography.  So what follows is my own take on it.  In a way, it is nothing special.  I don't feel as though I have some secret formula, however I do have a few methods that I follow that are rather ... unusual!  So here it goes.
I have been doing photography since 2006, so I have only used digital cameras.  And I've had no formal training beyond reading every single book I could find on all aspects of photography.  And since I'm also a computer programmer, I have learned how to figure things out on my own, from hand-coding my own website to figuring out all of the software that I use.  So what I do is based purely on personal experimentation, and I'm sure that many experts would excommunicate me from their 'Church of Proper Photographic Techniques' if they saw what I do.  But it works for me and that is what counts.  And that is my advice to anyone that asks.  

It is best to develop your own technique based on what you learn experimentally.  This is very important.  Don't rely on articles that begin with '10 tips to ...' or actually anything referring to 'tips!'  Find out yourself, with the assistance of some foundational knowledge in regards to operating your camera, properly exposing an image, etc.  

Even though you wish to create art, proceed as though you are a scientist.  Scientists begin their education in the lab, learning proper procedures, equations and the like.  But what they are actually learning are the methods of effective discovery.  Scientists learn about the world via experimentation and proof, and photographers can do the same thing.  The big difference is that you are discovering what works for you and your vision of your art while a scientist discovers what works in a verifiable sort of way across all relevant platforms.  Only you know what works for you, yet you can still use the scientific method to discover your artistic self.

Image capture:

Since I like to get really close to the rather violent Pacific Ocean to capture as much drama as possible, I have developed techniques for capture that are very simple and fast.  I'm sure that many photographers do many the following things, but they may not admit it because it does not sound very 'professional.'   To put it simply, I do the following, (in all-manual mode shooting RAW, not jpg!):

1.  I get to my location early and stand there for quite a while, taking it all in.

It is important to soak in the mood and feel of the place that you are about to photograph. Otherwise you can not convey that feeling to your viewer.  The longer you stand there, the better your image will be.  I learned this via trial end error.

2.  I decide what elements strike me the most and imagine them in the image. 

I usually compose the image in my mind before I actually move the camera into position.  I watch the waves and other elements and imagine how to best capture them.  For me, it is best to do this without the camera to hold you back, but do what you feel is best for you.

3.  At a beach, I stand on the shoreline and hand hold the camera with it on the tripod.

Once I have a good idea of what I want to do, I see what the camera can do.  I ask; how wide do I want my view to be?  Is the light good enough to get a short enough exposure to show the proper amount of movement, or do I have to compromise in some way?   Sometimes I have ideas that simply will not work, or perhaps conditions change too quickly to capture it.  So be ready for anything.  Keep an open mind.

4.  I adjust the ND grad filters so that the light is even from top to bottom.

Contrary to what some people say, unless you wish to use HDR, ND grads are absolutely required most of the time when shooting landscapes with dramatic light.  The dynamic range usually just too much unless you are shooting away from the sun.  Even though perhaps you can darken the sky manually in Photoshop with no grad filters, the result is usually ugly or at least difficult to process.  Why make more work for yourself?  Just get the grad filters.  Usually a 0.6 and a 0.9 (or 2 0.9's) will do the job.  Ask yourself,  "Can I get an overall evenly lit exposure with no grads? " I usually use both grads at once

3.  I decide how fast I want the shutter speed to be and adjust the speed.

This is when I decide how much movement I want in the photograph.  Usually on a day with average surf, a 1/8 to 1/4-second exposure shows some movement without it getting chaotic and messy.  Often, a 1/2-second or greater exposure shows the same bits of water moving around in different directions resulting in a messy blob of white cotton.  However, experiment!  Sometimes if the water is flowing smoothly, a 1-second exposure can work.  After enough experimentation, you will get the hang of what works best.  There are many different situations, so experience is the only way to learn.  No book of tips can teach you this.

4.  I adjust the aperture to get the exposure needle right into the middle.

I usually do not have time to get a precise light reading or even think about something like a zone-system.  Fortunately, a digital exposure is nearly free so go for it!

5.  I go in, plant the tripod in my planned place, take a shot and run.

If you are properly prepared and have composed the image ahead of time, you can get in there, plant the tripod, shoot and run in those few seconds in between wave impacts.  And it is surprising how many shots turn out really well.  Still, if you like the scene and you want a good rendition of it, you may have to do 5, 10 or even more variations.  Timing is everything, so watch the waves (or whatever is dynamic) for a while so you can catch it at the most dramatic moment.  Even then, you may need 10 tries!  Don't settle for the first few even if you like them.  It is amazing how randomness can make shot #10 be the best.  If I had a policy of only taking 3 photos of a composition, I'd probably only have half of the good images I have today!

Since I usually use a very wide angle of less than 20mm on a 35mm sensor, I have to get very close to the foreground.  Always find good and substantial foreground to anchor your photo.  That is how a viewer can be transported to your location!

6.  I look at the back of the camera to see what I got and magnify it to 100%.

First I look at the overall picture to see if I have light detail in all areas, then I zoom in to see if it is sharp from foreground to background.  Take a moment here and concentrate.  You do not want to return home thinking that you made the shot of a lifetime only to find out that the foreground is out of focus!  On the widest angle photos, I often focus to infinity and everything from 3ft-infinity is razor sharp on my Canon 17-40L lens with the aperture between F10 and F14.  F22 can make everything soft, so avoid it!  Again, experiment in front of your home with focal sharpness so that you are prepared in the field!  Do not rely on focal charts or advice, even from me.  Find out yourself what works with your gear.  Again, excommunication from the 'Church of Proper Hyperfocal Distance' will probably be your fate, but try it out!  It simplifies things dramatically on super-wide shots.  I have no time to mess around with focus when a 12-footer is bearing down on me.  So I focus manually ahead of time.  When you zoom in, you must be more careful, so experiment with all focal lengths ahead of time.

7.  I look at the histogram.

We are working with digital files now and there are differences versus film.  (Experienced digital photographers know the following information, but I'll go over it anyway.)  With film, you needed to get the most perfect exposure possible on the film.  Sometimes you needed to underexpose in order to get the rich, saturated color of a sunset with detail.  But a RAW digital file is really just a data file.  So, the most important thing is to capture as much data as possible.  And that means getting as close to overexposing it as possible.  That is because each pixel contains a value from 0-255.  0 is dark and 255 is pure white.  So rather than trying to get as many pixels as possible to around 128, try to get them closer to 180 or more. For example, lets say you have a properly exposed picture where the dark areas are 5, the average is 128 and the bright areas are 220.  Most of the picture looks good, but the dark areas (the 5's) are really dark even though you can see detail in the dark areas with your own eyes.  

Because the camera cannot capture the dynamic range of brightness that your eyes can, you need a little help from the data file!  So, what I do is make a picture that is as bright as I can get it without the white areas becoming blown out and overexposed.  Even if it looks a little too bright in the back of the camera, the dark areas will have the detail I need when I process the shot.  So now, you may have the darkest areas at 20 and the brightest at 240.  Just don't let the bright areas get too close to 255 or you will lose detail!  Again, experiment for yourself.  It is the only way to know for sure.

I'll go into processing below, but the main point is to experiment and get as much data into your picture file as possible.  Then you have more leeway when processing.  You also get more leeway by shooting in RAW mode vs. jpg mode.  This is very important because there is more data in a RAW file vs. a jpg file.  Also, a jpg file is compressed even at a 12 setting.  Each time you save, a little bit of detail is lost, so avoid jpgs except as final output.

8.  I adjust the exposure if necessary and try again until it looks good.
Yes, just keep trying until you are satisfied.  

That's it.  I know that this all sounds basic but this is what works for me.  And after doing this for a while, it is really just four steps.  Set it up, take a shot, make adjustments and shoot again.  Don't make things more complicated than they have to be!  I suppose that none of this is revolutionary.  Many of you may do the same thing, I'm not sure.  The important thing is to develop your own way of getting the most data that you can.  

Light = Data!

Again, a digital file from a camera, especially a RAW file, is really a data file.  So you should try to get as much data as you can into your file.  The more light you capture, the more data you have.  This often means slightly overexposing the image while making sure that no areas become truly overexposed.  Often you end up with a picture in the back of the camera that looks a bit overexposed and sometimes downright ugly!  But that is fine because when you darken the image in Photoshop or your image editor of choice, colors and details are there and often can look as good as it did to your own eyes without having to crank up the color saturation slider.  And that is the objective.

It is easier to darken a bright image than to brighten a dark image.  Let's say you have a dark area of the image with pixel values from 2/256 - 4/256 (where 0 is black and 256 is white.)  Values of 2 or 4 are very dark.  If you have a value of 2 and you wish to brighten it up, the next value is 3.  On a percentage basis, going from 2 to 3 is a 50% increase in brightness.
There is no middle ground between 2 and 3, or the resulting 20 and 30 if you brighten it!  So, when you have dark regions and you try to brighten them, you end up with ugly blotchy patches of 'noise' and you lose detail.  Now, if you have dark areas with values around 20 for example and you wish to brighten them a bit, you can go from 20 to 21.  That is only a 5% increase.  So there is a smoother path for brightening vs. from 2 to 3 which is a 50% increase.  Big percentage changes between pixels mean lots of noise in an image.  What is important is not whether you know the numbers, but rather that you know to get as much light (and therefore data) into your image as you can.  The camera can not capture all the dynamic range of light values that you can see, so get as much as you can in the camera.

Another advantage of slightly overexposing an image is that you do not have to saturate it in post-processing to get back the color you saw with your own eyes.  Often times, a 'properly' exposed image looks flat in the back of the camera compared to what you see before you.  Then you get home and try to bring back the color and you get lots of ugly noise.  That is because there is not enough data to recreate the scene you just photographed.  When you you have a lot of data, you can darken the picture by adjusting the levels.  Darkening via levels brings back the color without increasing the noise.  Then it looks more accurate compared to what you originally saw. I spend a LOT of time getting into a place with dramatic light, so I want to make sure that I can bring it home with me.  I am not a Photoshop guru!

Remember, no amount of playing around in Photoshop will make up for the glaring bright light of the majority of most days.

An essential tool for landscape photography are neutral density (ND) graduated filters as I mentioned above.  The sky is usually very bright compared to the land.  Even though your eye can adjust to the dynamic range of bright and dark areas, the camera can not record it.  (My appologies for the repetition to those who use them all the time!)  So you place the darker half of the filter over the brightest areas of the scene before you.  Slide them up and down in the filter holder until the light looks even.  Take a test shot or two.  Don't skip quickly by this step or the image may be unsalvageable.  A long discussion of ND grads is beyond the scope of a blog entry so read up on them if you are unfamiliar.  Sometimes people think that ND grads are not necessary because they can go into Photoshop and darken the sky. Usually this does not work very well because the sky pixels will be so close to 255 that darkening them does nothing at all and can get really ugly!  Darkening near 255 works the same as brightening near 0, though not as dramatically.

The proper use of ND grad filters allows you to get the maximum amount of data possible into your picture's data file (in the darker areas) so you can more easily process it when you get home.  I must admit that I am lazy when it comes to Photoshop.  I don't want to spend much time adjusting images and I don't know many tricks.  I'd rather put all my time into being there when the light is best.  And actually taking an extra hour to get the best shot can save an hour in Photoshop.  Where would you rather be, in nature or on your computer processing a picture?

Sometimes, HDR (High Dynamic Range) processing can take the place of ND grad filters.  Especially in cities or canyons with jagged horizons.  In HDR photography, you take several pictures ranging from severely underexposed to severely overexposed and the HDR software averages them out to make a composite which can look quite realistic if processed subtly.  Sometimes people push HDR processing to make a dramatic and surreal image.  Good processing of this type can look very artistic, but for this article, I'm focusing on making natural and artistic landscapes from a single RAW file exposure.

Processing - Let me see it and feel it again...

When I get back to my computer, I want to recreate and re-experience the feeling that I had when I first saw the scene with my own eyes.  And for me, that is the essence of landscape photography as art.  I want to see it again, and feel it again.  And if I can create a TIFF file that brings me back to that place and time again and again, that is 'art' for me.  Sometimes it may not be art for other people, but that is okay.  Not everyone will relate to every picture.  So rather than worry about whether others will like it, I just try to make sure that I like it.  And hopefully others will too.

When I review the photos, I use Capture One RAW processing software by Phase One.   There are many other RAW processing programs that do a fine job however.  When I browse through them, I end up liking about 10% of the photos I made.  And I do not take many shots on a single sunrise or sunset outing.  On average I take about 20-30 shots and like 2 or 3.  Try to be a ruthless editor, but don't delete any for a few days.  You may have a change of heart!  Even though I try really hard to make every exposure a fine work of art, most simply do not turn out.  The exposure and focus are usually fine, but somehow that feeling of being there is just not... there.  The process of review is rarely what I expect.  Sometimes I think I captured some amazing images and I might even show them around in order to do a bit of chimping (going ooh ooh oooh, aah aah aaah.)  Then, I look at them later and think to myself; "What was I thinking?  This is rather flat and boring!"  Sometimes I almost delete photos that later I like a lot.  In short when you are reviewing your shots, never believe everything you think!

If I have 2 nice images that I like to look at again and again, I am happy.  Often I'll wait a day or two before processing them in order to see if I still like them the next day.  Before making my final decisions, I look at the ones I rejected just to see if I may like them now.  You never know sometimes. 

In my RAW editor, I first make the basic brightness, contrast and color adjustments for the entire image.  Then I convert the RAW file into a 16-bit TIFF file.  16-bit files are bigger and the color is better and more smoothly blended than with jpgs or 8-bit TIFFs.  There are many large volumes written about the advantages of RAW files, so if you wish to go into more detail, read a book or browse the internet.  Suffice to say, shoot RAW and convert to a 16-bit TIFF!

When I am ready to process the TIFF image, I look at it in Photoshop (There are other good TIFF editors too) and compare it to my own memory and sometimes to the back of the camera.  Yes, I actually hold the camera up to the screen to compare!  It seems as though every picture must be adjusted in some way because the camera cannot capture the range of light and color that our eyes see.  That is true even though I use neutral density graduated filters to even out the light in the sky with the light on the land.  Often times the brightest areas are too bright while the darkest areas are too dark, even though with my eyes I could see detail in all areas.  Often, right after I take the shot, I'll hold up the viewfinder to the scene before me and compare, just so I will have an idea on how to process it.  Chimping at this point will guarantee failure just as in tennis when you admire a great shot that you just made.  Chimping in tennis will get you a mouth full of yellow fuzz!

At this point I must say that I rarely use layers, even though the experts think I should.  I tend to use one single layer and make all adjustments to that layer.  I suppose that this heresy will get me excommunicated from the 'Church of Proper Processing', but it works for me.  If you use layers, just continue using them.  There are some advantages such as undoing a change that you made last week because it did not print out right, but the added complexity is just not worth it to me. I also rarely subclass in programming for the same reason.  I just go back to the base object and create a new one.  It is easier.

Once I have the TIFF file open, the first thing I do is select the darkest areas and brighten them up a bit (if necessary) using the 'levels' adjustment in Photoshop.  Then I darken the brightest areas just a bit too if necessary.  If you adjust things too extremely, you will end up with blotchy areas with lots of noise, so be careful.  Sometimes the darkest areas can not be 'saved', so leave them alone.  In short, you can't get something out of nothing.  If the light was not there to begin with, you cannot invent it when you are processing.  (Well, you can but when attempting to record reality, it is not a good idea!)  If you 'invent' light, that is manipulating an image.  You are putting something into the picture that was not originally there in the scene.  Other than dust spots or something very small, I avoid manipulating an image.  It ruins my memory of the place if it does not look/feel as it did when I took the shot.  Slightly brightening up a spot is not manipulation because you are restoring the image to the way it really looked to your eye.  Long exposures can be regarded as in-camera manipulation but those images often accurately record the feel of the place.  People know they are long exposures so there is no deception there.  Manipulation is an issue of honesty more than anything.  If you say that you manipulated an image, fine with me!

When I select an area for brightening or darkening, I will feather it to soften out the edges of the change.  Then the natural look is maintained.  Experiment with this.  To my eye, brightness usually changes gradually as I look from one spot to the next, so I want my photos to look the same way even though the camera often can not record it properly.

Once the file looks properly exposed in all areas, I make sure that the colors are strong and natural looking, just as they looked in the back of the camera and to my memory.  If an area has too much color and has lost detail, such as with a very red sky at sunset, I'll desaturate the area a bit.   If the color is flatter than reality, I may saturate it a few percent.  I do this according to my comparison on the scene between the back of the camera and what I saw with my own eyes.  If the back of the camera looks dull compared to the scene before me, I make a mental note that I must resaturate the photo to get it back to reality.  Over-saturating a colorless scene will introduce ugly blotchy noise and you will lose detail, so be careful with saturation!  I'd rather have detail.  Sometimes you lose mass-appeal by desaturation.  I've seen people admire images with hyper-red blown-out skies that have no detail whatsoever because the photographer cranked up the saturation slider as far as it could go.  And sometimes a hyper-saturated and blotchy image will get admirers to ooh and ahh.  But try to maintain your integrity!

I must admit that many professional photographers look at me as though I've lost my mind when I talk about holding out the camera and comparing it to the scene before me, or comparing the back of the camera to my computer monitor or print.  But the back of the camera is my best link between the original scene and my output, so I'll just let them think I'm crazy!

Then I sharpen each area of a photo one by one.  I'll select some rocks, or the ocean, or a mountain, feather it, and sharpen each thing separately.  Over sharpening can induce some really nasty noise so be careful.  I sharpen each thing at different strengths and radii.  Again, experiment with what works best.  I tend to leave the sky alone, but the ocean often looks good with a smart sharpening radius of 2.0-4.0 while rocks may be sharpened anywhere between 0.2 and 4.0.  Sometimes I'll give places two passes of sharpening, one at around 4.0 and another at a much smaller radius.  You can never be too careful about inducing noise.  Even noise reduction will not save an over sharpened shot. All that work only takes a couple of minutes but it is worth it in a big print.

Once every area is very sharp, but not too sharp, I may reduce the noise in the sky or other smooth areas.  Dark areas of the sky can be the noisiest parts of the photo even at ISO 50, so look at these areas carefully.  Then I'll look at all areas of the photo once again at 100% magnification and look for the any problems of any sort.  Look out for salt spray or dust spots.

That is about it.  In short, I try to get it right in the camera because I really don't want to spend much time in Photoshop.



My fine-art website
Learn from the old masters with my 325-page e-book!
Hi-res downloads and framed photo prints



  1. Your approach to learning photography mirrors my own in SO many ways...only I started with film cameras. Thanks for sharing the results from some of your experiments that have turned out successfully for you. I found them very helpful and I look forward to reading more of your blog posts.

  2. Great blog! I like your approach, honesty, and your amazing results:-).

    One question I have... You mention "feathering" a lot. I know there are a lot of ways to do this in Photoshop, what's your method?

  3. Excellent - very interesting process. Thanks for sharing

  4. Thank you so much for this, I loved reading about how you go about creating such amazing images