Landscape Exposures

Barry Keavney

Well-Known Member
Recently, I have been reading different, and sometimes, conflicting ways of maximising the potential for Landscape shots.
My approach was as follows:
I use full Manual Mode, and I'm a fiend for spot metering. I would meter on the foreground and let rip (using the cameras Analogue Exposure Display, I would under-ex by a 1/3 of a stop).
As you would expect, this would over-expose the sky, so I would reverse my approach. Thus, rendering my foreground lifeless and under-ex. I would normally be in excess of f.11, but they were still mediocre at best.
Anyhow, I was reading many techniques and was wondering what you all found worked for you?
Trawling through many, many webpages, I found that alot of photographers take a meter reading of the sky, we'll say 1/160 @ f.11; and another of the foreground, pretend its 1/60 @ f.11...
What they seem to be saying is that one should find a Shutter Speed between the two...1/100? Would this work, is my question, as I haven't had the time to try it.
Also, if this be the case, and I was using Aperture priority; would it be by changing the f stop that I would achieve that number?
Like I say, I usually use Manual mode, and in order for my to find a "common" exposure speed (between dabbling with Shutter and f stop), my Exposure Display would read incorrect because I would be controlling it independently!
Is this making any sense?
Anyhow, what works for you?...

P.s, If I get this nailed, we can throw ND Filters into the mix!;)
 
i think you are over thinking it a bit barry
firstly you have to thinlk what it is you want from the photo and what settings are likely to achive that...
lanscape photography for the most part involves having most of the image, if not all of it, in focus... to do that you need a broad depth of field and so maximum f.11 (dont forget "maximum" in this case means a lower number ... ie the maximum apature of that 50mm i have leant you is f.1.4... you already know this...)
you iso needs to be a low as possible ideally the native iso of the camera... in the case of the d90 i think thats 200)
then the shutter speed should be set from those to things to get correct exposure...
i think that is what you are already saying... right?
i know you are a fiend for manual mode but dont forget that these modern cameras can matrix meter fairly effectivly.. and you can dial in exposure comp
as dave says you dont want burnt high lights ... but you can see if you are loosing highlights by taking the photo and pressing up on the dpad until it shows you lost highlights... by flashing areas of the image at you

trying to average exposure from spot metering is a good way of doing it, but it is just a matter of practice, there is never a hard and fast rule... trial and error, and looking at the highlight warning thing on the camera over time will help it become instinctive...thank god for digital eh

the best way to get an even exposure is to use nd grads.. here i used 2 filters, a less powerfull nd grad over the sea and a more powerfull one over the sky.. then tweaked it on the computer...
in hind sight, this image is a little dark... but you see what i was aiming for ... im no landscape photographer!

52193_1203785227.jpg


... without grads as dave says, fill light in bridge or lightroom can bring back a lot in the lost shadows when you have exposed for the sky... just be carefull, to much and you end up with noise
 
Cheers y'all,
Some very useful info in there. I have taken some "acceptable" landscape shots, but I suppose I'm trying to ensure that I don't miss "that" shot due to a lack of understanding in some vague area; with particular reference to the native iso setting.
Indeed...trial, error & utilising the advice I gain here should help me...and more...to progress to the perfect shot of Autumn!
I've said it before, and I'll say it again; Just because you've read about riding a bicycle, doesn't mean you'll have balance when you get on one!!!
Regards
Barry
 
I believe that part of the issue is taking the photo when the light is too bright. Sunny days are wonderful but a right pain in the arse as the bright sun can wash parts of the photo out. You can compensate to expose the bright areas correctly and correct the areas that are too dark on the computer. There is another option which is to look at the scene and choose the various area that may require differing exposures ie sky background & land foreground. Meter these and then take 2 pictures one metered for the sky, the other for the land. Its a good idea to use a tripod so the camera remains in the same position for both shots. You can then join these two photos together of the computer in photoshop. This is the basis for HDR photography and you can get software that does this automatically but sometimes with differing levels of success. Picturenaut is reasonably good and its free.
 
I'm no expert, but from what I have picked up or tried myself, using a combination of shooting in RAW and understanding layers in Photoshop there are very few problems that can't be solved "later" when you are on the computer.

Bracketing shouldn't be necessary when you can use the RAW file to create (say) three versions at different exposures. Admittedly that is more effort, but it cuts down on memory card usage and (I think) shutter speed time in the field.

Similarly ND Grads - layers could be used to replicate the effect, I think? I only have the circular ND Grads which force the division to be halfway up the picture, which isn't going to be right usually.

What I read somewhere is that you should deliberately overexpose, because it is easier to adjust exposure downwards in PS to bring the detail in the highlights back, but if you underexpose, the dark areas cannot always be adjusted to bring up the detail.
 
Hi Tim,

Certainly combining differently processed raw files can give you a wider dynamic range. And you are right in thinking that verging on the over-exposed provides you with more data to work with than under-exposing. Mind you, over-exposing is probably not quite what is meant though. Many people refer to exposing to the right - don't go too far or there will insufficient data in the upper registers to work with. However, the histogram display is a usually a 'JPEG interpretation' of the exposure so you probably have at least 1 stop more latitude than you might expect from the warning systems. The reason for the 'expose to the right' protocol is summarised here http://www.luminous-landscape.com/tutorials/expose-right.shtml and explained in much better detail in http://www.amazon.co.uk/Real-World-...ef=sr_1_1?ie=UTF8&qid=1282513150&sr=8-1-spell A great book if you want to get the most out of your images and actually very readable.
 
Cheers Pete.

"Exposing to the right" was exactly the concept I was referring to - I'd forgotten the phrase.

Do you think a good alternative to using an ND Grad filter is to combine two copies of the RAW file, one exposed for the sky and one for the rest of the picture? I have a copy of The Essential Guide to Landscape Photography, which has a tutorial using exactly this technique. For a non-straight line horizon, it ought to give a better result.
 
It probably depends on the scene. You have a finite dynamic range in a single shot that is dependent on the sensor so sometimes processing a raw file for different components of a scene will help (or just using the recovery tool). If the range is too great then the true HDR approach will be a better approach but they can start to look a bit unreal IMO. A graduated ND is useful especially if square and mounted in a rotating holder and especially if there is any wind around that might make a composite image tricky (moving clouds and trees etc). As I say, very image dependent but most of the time the duplicate process / recovery route will get you where you want to be.
 
I'm no expert, but from what I have picked up or tried myself, using a combination of shooting in RAW and understanding layers in Photoshop there are very few problems that can't be solved "later" when you are on the computer.

Cameras see as their sensors or films see - and that is not necessarily what the human eye sees. Our eyes and mind are constantly processing the images in real-time, while the camera only reacts to the photons hitting the sensitized surface of the medium. RAW (and/or brackets) and layers are so far the best way to reconcile unrealistic camera images with what matches what your eye saw and soul felt, when you made the exposure.

Using image processing in this way is not covering your mistakes or "cheating", but rather using your skills as a photographer to present an honest interpretation of the image to the viewer. You are not compensating for your own weaknesses, but rather for the limitations of the medium. In photography since Henry Fox Talbot made the first viable negative/positive print in 1836 to the present moment, it has always been thus.

Even if one is mostly working with JPEGs now, please shoot RAW+JPEG if the camera supports it. When I first shot RAW nearly a decade back, I was totally unimpressed. It was not a problem with RAW but with a combination of immature software and my lack of understanding of the potential richness of the format. I shot RAW far less often than I should have.

With years of subsequent experience and understanding, I have been revisiting the early RAW exposures with Photoshop CS5 and drawing image quality out of them I could not even guessed existed within them when first exposed. Alas, if only I could roll back the years and have had the foresight to have shot them all as RAW. RAW contains data representing the state-of-the-art that your camera can capture. Even if you can not fully utilize it at the moment, the time will come that you can and you will be grateful for having done so. For the sake of the future - shoot RAW.

Bracketing shouldn't be necessary when you can use the RAW file to create (say) three versions at different exposures. Admittedly that is more effort, but it cuts down on memory card usage and (I think) shutter speed time in the field.

It all depends upon the dynamic range with which you are dealing, and the answer is always there if your camera can display a histogram. If the graph is piled up against both the right and left vertical axes, you know that you have exceeded the camera's ability to capture the whole dynamic range. At that point, bracket and either use layers and masks or HDR to capture the full range of detail in the final photograph as you process it.

Storage has dropped in price dramatically from my first 64MB card to my present 8GB card, so I don't sweat it. (The 64MB card was over $200Cdn, and one can find 8GB cards for around $20 now, with 8GB cards that can handles the extreme write-speed of my current camera for $60!) We live in wondrous times.

If I bracket for HDR I let the camera shoot all nine exposures at full stop intervals. Once in Adobe Bridge or PhotoMatix, I choose the range that will provide the highest quality image. Much easier to use five out of nine, than to realize you need a couple of brighter or darker exposures and go back half-way across the world to get them. It gives one the option of trying combinations until the very best image quality is achieved.

Similarly ND Grads - layers could be used to replicate the effect, I think? I only have the circular ND Grads which force the division to be halfway up the picture, which isn't going to be right usually.

One can always spot a TV documentary from the late 1970s by the terrible, unrealistic ND grad that every cinématographer thought was the solution to overly bright sky. The filters take exactly the right subject matter and incredible skill to use realistically. However, at the time, they were the only viable solution.

With layers and masks, Adobe Camera RAW (ACR) lets you deal with individual problems within the image - high-contrast sky with low contrast earth, mixed light sources, whatever. In the full version of Photoshop, there is an HSL tab in ACR. Using the blue sliders, one can recover every bit of blue sky that may have been captured in the exposure, and one can even simulate the effect of a polaroid filter, even if one is pointing the lens at 90° to where an actual polariod would work.

What I read somewhere is that you should deliberately overexpose, because it is easier to adjust exposure downwards in PS to bring the detail in the highlights back, but if you underexpose, the dark areas cannot always be adjusted to bring up the detail.

That is as wrong as saying the opposite - that you should do whatever is necessary to protect all highlights. Reflections off chrome trim in full sunlight have no detail whatever, and only look natural if they are pure white - don't sweat it. However, when the white of a building or cloud exceeds the film or sensor's capability of recording it, it is gone - totally - and forever. The key is to expose to protect SIGNIFICANT highlights. Where it is right on the edge, Adobe Camera RAW has a slider called Recover. It can pull the highlights down somewhat with little impact on mid-tones, but it can not recover that which has been totally lost to overexposure.

Ideal exposure shows a histogram where the graph is just barely touching each axis. That guarantees at least a touch of pure black and a touch of pure white which makes the image sparkle in its full dynamic range. There are exceptions, but in general terms one wants the full range. Without a pure white, the image can look dark and muddy. Without a pure black, the image can look weak and washed out. A low dynamic range image is easily stretched in processing. A high-dynamic range image is very difficult to recover to the point of producing acceptable image quality.

The Brightness slider in ACR is a sophisticated version of the traditional gamma or mid-tone adjustment. It leaves pure black and pure white intact, while adjusting the curve of the lightness values in between. By moving it right or left, one can reveal the maximum shadow or highlight detail in a normal dynamic range exposure. ACR allows even finer tuning with the Recovery and Fill light sliders. With care and skill, one can come very close to honestly duplicating what one saw in ones mind at the time the exposure was made.
 
good post there Larry, mirrors much of what i believe about camera RAW and how i go about exposing. I always shoot in the knowledge that I'm not going to capture everything right there and then in my shot. but knowing that with ACR i can usually pull detail out of the shadows and restore a slightly too white sky back to its former glory. typically in ACR you can pull out about a stops worth of useful information in both directions, without having to create the sometimes overused HDR look

It's these small adjustments that help to compensate for the slightly lower dynamic range and exposure latitude that the digital medium loses to its celluloid predecessor.
 
Larry, thanks very much for your informative post.

good post there Larry, mirrors much of what i believe about camera RAW and how i go about exposing. I always shoot in the knowledge that I'm not going to capture everything right there and then in my shot. but knowing that with ACR i can usually pull detail out of the shadows and restore a slightly too white sky back to its former glory. typically in ACR you can pull out about a stops worth of useful information in both directions, without having to create the sometimes overused HDR look

It's these small adjustments that help to compensate for the slightly lower dynamic range and exposure latitude that the digital medium loses to its celluloid predecessor.

I agree with that, David, but you've confused me, and possibly others by saying the following in another thread:

see i dont like making electronic adjustments to photos, so although i see what you're saying. if it can't be done in camera... i wont do it to the shot :)

just dont think it'd proper cricket otherwise :)
followed by
its an interesting way of looking at things. I've always tried to use filters, exposure, natural light and composition to keep things interesting. it seems adjusting digitally, images, isn't as frowned upon as i thought it would be.

I hope this doesn't come across as argumentative or critical, that's not my intention, but your different posts seem to say different things. Is this all a question of extent - are you saying that you like to make small adjustments in ACR, but not larger adjustments, layering, etc?
 
yes absolutely tim... its a question of extent... i dont mind compensating for what are the inherent issues with digital sensors (e.g. sharpness and exposure latitude) but im not so mad keen on using layers, composites, burning areas etc to modify an image.

let me clarify more... if its something i cant do with a small tweak (say 5% here or there) in ACR, then I wont do it. Simple as.

Hope that makes more sense
 
Yes, that makes sense now David.

It's certainly not clear nowadays where the boundaries are between photography skills and PC software skills.

Of course, Photoshop can be used to create a great many images which don't actually start from a photograph in a camera at all, so it is difficult to know where the line is drawn.

If you are old enough, do you remember in the old days people created pictures (eg faces) on computer printout paper (the large stuff with holes and alternate white/green stripes) by writing a program to print out lots of different characters line by line on their line printer? It seemed very clever at the time, but no-one would call it photography. Zoom in 400% to alter pixels, and you are perhaps doing something similar?!

Apologies for taking the thread even more off course!
 
There may be people here who don't really realize what RAW is and is not. First, it is not an image format - like JPEG is an image format. RAW is actually a container more like a ZIP file, and within it, there is the JPEG preview - and actual JPEG file you see when you open the container in ACR; an image thumbnail, file header which typically contains an indicator of the byte-ordering of the file, a file identifier and an offset into the main file data; the actual data as it came off the sensor converted to digital form but otherwise unprocessed by the camera; EXIF data, and camera maker notes, which may help ACR or whatever conversion program understand any special features embedded in the data. The actual contents may vary from camera maker to camera maker.

With a JPEG, the camera applies its settings to the image and the image is done. With RAW no camera settings are applied at all. The image has no white balance, no sharpening, no vivid/natural/monochrome settings or anything else you would expect to be applied to a JPEG image.

When an image is opened, one sees the state of the camera reflected in the preview image - the embedded JPEG - but none of these settings is applied to RAW. If at this point you click on Open Image, it will demosaic the data into a bit-map with the camera settings applied and open it in Photoshop. You don't actually open the data in ACR - you are working on a preview image as you fine tune your settings. Only when you click on open, do the settings actually get applied.

Camera settings are crude - five or seven settings for any parameter for the most part. The whole purpose of RAW in the photographic work flow is that it provides much finer settings than the camera allows. When David says he will only change the default settings by 5% or less, he is not being a purist, just squandering the whole idea behind RAW. RAW puts total control of settings in the hands of the photographer, not the camera. Camera firmware is programmed by anonymous programmers in Tokyo who have no idea what you are photographing, nor the conditions you are working under. Accepting the defaults they have built into the camera is taking yourself out of the photographic process.

RAW has no settings at all. It is just the raw data off the sensor. When you shoot JPEGs, you are actually initially shooting RAW. The difference is that the camera does the processing—not you. The only advantage of RAW is that it allows you to process the same data with a much higher level of control than a camera can provide, and do so at leisure after the shoot. When you process RAW, you are not changing the settings, you are doing the settings! The settings you see in the preview image upon loading RAW are simply arbitrary place-holders.

Since RAW has no white balance of its own, you can use the eye-dropper in conjunction with the colour temperature and tint controls to easily colour balance for a mixed lighting situation even though it may be as drastic as daylight, fluorescent and incandescent in the same image, using layers and layer masks to blend them together. You can also do this once the image is open using Photoshop's colour controls, but then you are doing destructive editing. Any edits done through ACR are totally non-destructive. The same can be done with exposure opening up selected shadows, darkening selected highlights, saturating specific object and muting the colour on others. Whatever it takes to get a good honest representation of what your eye saw and mind perceived at the time of exposure.

Other than achieving proper exposure and focus, the camera contributes nothing. You can shoot under incandescent bulbs with a daylight balance, and achieve a perfect white balance in processing. You have complete control over the response curve of the image, with no need to accept the programmer's response curve. You can fine tune it to the needs of the image. Using the HSL tab, you have fine control over not just red, green and blue channels, but eight different bands of the spectrum. RAW puts you in control - not the Japanese programmer.

Not to blow my own horn, but last year I was hired by Focal Press, the top publisher of photography books to vet a manuscript by a couple of very well known writers for a book on RAW. I know RAW.
 
Certainly. It is my time of life to share any experience and knowledge I have picked up along the way with the next wave of shooters. I have gained so much from those who went before me, that it is now my time to give to the community that has made my life terrific. Feel free to pass along anything I write and/or the URL to my web-site.
 
Great post Larry.

However, although technically you are right, the way I look at it is this; all cameras require that you give them a set of settings (the arbitrary placeholders) which, if you request a file format other than RAW, it will use to store the image, and which it will also use for the in-camera display screen and (I assume) the embedded thumbnail in the RAW file.

Traditionally photographers were stuck with those settings and had to do any further manipulations in the darkroom to adjust them. With RAW you can replace them on the PC. For some, this takes away the skill they perceive in deciding on settings "blind" as in traditional photography, based on judgement, experience or just luck. For others it is called progress!

If you don't change the camera settings, most software goes ahead and applies those settings when moving on from the RAW.

So, with any non-RAW format, you have to make any changes on top of the camera settings, as in the darkroom. If you store in RAW, you don't, you can start afresh.

Is this a valid way of looking at it?
 
Great post Larry.

However, although technically you are right, the way I look at it is this; all cameras require that you give them a set of settings (the arbitrary placeholders) which, if you request a file format other than RAW, it will use to store the image, and which it will also use for the in-camera display screen and (I assume) the embedded thumbnail in the RAW file.

Correct.

Traditionally photographers were stuck with those settings and had to do any further manipulations in the darkroom to adjust them. With RAW you can replace them on the PC. For some, this takes away the skill they perceive in deciding on settings "blind" as in traditional photography, based on judgement, experience or just luck. For others it is called progress!

You are still replacing them in the darkroom. I have read of RAW files being referred to as "Digital Negatives". Just like with a film camera, you need to be precise in setting the exposure, focusing the lens, having the optimum balance between aperture, shutter speed and ISO setting. Like with a film camera shooting negatives, once in the fume-room, you can set the white balance and lightness/darkness of the print. You can dodge and burn digitally as well as physically, though in the digital domain you can actually see what you are doing. I use Photoshop as I did my fume-room, except decades have passed and the tools have evolved as has everything else in photography. However, Photoshop serves exactly the same purpose.


If you don't change the camera settings, most software goes ahead and applies those settings when moving on from the RAW.

So, with any non-RAW format, you have to make any changes on top of the camera settings, as in the darkroom. If you store in RAW, you don't, you can start afresh.

Is this a valid way of looking at it?

JPEG can be processed in much the same way as RAW, but every edit is to some degree destructive. In processing RAW, you open a fresh instance of the image each time and blend it with the other previous instances. JPEG is 8-bits per channel, giving you 256 discrete steps between black and white in each of the red, green and blue channels. RAW is at minimum 12-bits per channel making the discrete steps much smaller - with 4096 steps. I shoot 14-bit RAW with 16384 steps, which I process in a non-destructive way within a 16-bit per channel workspace.

So yes, when you process JPEG you are altering the image as the camera produced it. It will take a reasonable amount of editing without beginning to show rounding errors, banding in the sky or other gradients. And with RAW you are always starting with the original, unchanged raw data off the sensor. However, most RAW processing programs either use a built-in database or an XML "sidecar" file that chronicles the changes you made. Thus the next time you open it, the settings will still be there. Since the settings are kept externally, nothing has changed with the raw data. It is just a convenience so you don't lose the fruit of your earlier labours.
 
Last edited:
Back
Top