Anyway, the universe has yet again dealt the camera companies an ironic blow. Just as they come around to the idea that people want higher quality pocket cameras, we come to find out that actually, no one wants a higher quality pocket camera. In fact, in five years the pocket camera will probably be all but irrelevant.

My partner in blogging, peterb, actually called this in his first comment on this post. The truth is that the smart phone (in my case, an iPhone 4) will destroy the pocket camera. There is almost no question that this will happen. I did not believe him at the time because the camera in the iPhone 3GS was actually kind of sucky. I believe him now because the camera in the iPhone 4 is, for the most part, not sucky. I realized this on my most recent trip to Paris where by far my two favorite pocket camera shots were taken with the iPhone. Here is the first one:

The second one is a bit more interesting thing than that, but you’ll have to take my word for it because as a general policy I don’t post family photos to the public Internet. Just a paranoid tick I have.

In any case, here are the four things that make pictures on the iPhone better than all the other cameras:

1. The camera is good enough in terms of image quality, responsiveness and general feature set.

2. In camera HDR. Cannot stress this enough. This one feature makes the camera 100% more usable.

3. Autostitch.

4. The phone is close to the Internet at all times.

We can go over these one by one. The first speaks for itself. The camera is missing nothing that is absolutely critical for about 75% of the photos I ever want to take. By which I really mean 90%. The second is more subtle. What is nice about the HDR is that it compensates for the one major shortcoming of the tiny phone CCD: limited dynamic range. So, instead of black shadows and blown highlights, in most situations you get decent pictures. The pano above takes great advantage of this. I also love love love using Autostitch in the phone. I’ll do panos in my phone in a heartbeat, whereas I hardly ever bother to do panos with my real cameras. This is because doing them in Photoshop sucks and doing them in Autostitch is fun. I can’t explain it. In fact, the picture that I didn’t show you is a 2×2 square HDR pano. It’s the only combined horizontal/vertical panoramic shot I’ve ever actually tried. I would never do this sort of thing with my Nikon, but the fact that I can see the stitch almost immediately in my phone makes me try.

Finally, the fourth one is important because in general we take pictures to share them, and these days the way you share pictures is on the Internet.

I will posit that no current camera-company camera, be it a lowly point and shoot or a $20,000 professional digital back, combines all four of these features in one package. In addition, I (and others who are smarter than me) would argue that the camera companies will never understand how to put a package like this together. This is because they only understand cameras as they were used between 1960 and 1995. I think they still don’t understand how important easy sharing and having a software platform in the camera is to most users. If they did, they wouldn’t have spent the last 10 years in a megapixel pissing match. Instead, they would have given us cameras that are more usable and actually take advantage of things digital can do that film couldn’t.

Anyway, to end, here’s another phone picture I took in Paris which was better than what I got with my big DSLR:

I think this might also be a panorama. But I can’t remember. I can’t remember because I did’t have to sit at my laptop for 20 minutes watching Photoshop chug trying to put it together. See how that works?

1. A good optical viewfinder, with focus confirmation.

2. Reasonable performance.

Only time will tell if the shipped camera actually fulfills these promises. If it does, I guess some of my disposable income will be disposed.

Overnight update:

The photo blogosphere was all aglow with the news of this new machine. In particular, over at TOP you could almost feel the drool over the Internet wires. There were four classes of comments:

1. “WOW, WANT!” Sure, I can agree with this.

2. “Fuji will sell a ton of these because we all know what everyone wants is exactly a niche enthusiast camera!” Well, no. Enthusiast photographers, like all enthusiasts, want something every different from all other users. They tend to just not know it. That’s why no one has done a camera like this before. It won’t sell in large numbers. Just ask Konica about the Hexar. Also, there is little chance that this camera sells for much less than $1000. You don’t sell large numbers of $1000 cameras with fixed lenses and a relatively limited feature set.

3. “Where is the interchangeable bayonet mount?” These people are the same ones who cannot use small cameras because of their huge ape-like fists.

4. “Boy this would have been great as a film camera!” Um. Yeah. Sure.

Also, I would like to register my annoyance with the constant use of the term “analog” to describe dial-style controls with click stops and numbers on them. These dials are sending digital signals to a little computer inside the body of the camera just like all other controls on a digital camera wouild. Just get over it. Sheesh.

Anyway, to summarize: WOW. SHINY. WANT.

Most photographers are lousy printers. Most always were. The difference is that in the old days you had to master a substantial skill set in craft before you could demonstrate how lousy you really were. Now anyone can simply buy a technologically superior printer for $500-$1000 and immediately demonstrate their lack of competence.

Read this post (which, IMHO, is a bit of rationalization in an attempt to justify a personal preference, but that’s OK). Here is a bigger link for you to click on, even though I know this sort of thing is bad “style”.. Scan the comments for what Ctein said. As usual, if the post takes itself too seriously, the comments are an order of magnitude worse. Especially mine.

Until around thirty years ago, cameras were primarily mechanical devices. A few had electronic light meters in them, but the user was still expected to set everything by hand. First, you would do it peer into the viewfinder of the camera and turn one knob until your intended subject appeared to be sharp. Then you would stare at the light meter and set the aperture and shutter speed accordingly. Then you would take the picture, confident that it would be perfect because you had spent so much time in thoughtful deliberation over all of these critical settings.

This usage pattern means different things to different people. For the non-enthusiast who just wants a picture, it is complicated tedium. For the professional photographer or photographic artist for whom the image is most important, it is a small bit of technique that is easily mastered and then forgotten about. Finally, for the enthusiastic amateur, it is a secret to be mastered and then cherished, since the actual photograph that he is taking probably isn’t all that interesting.

Not surprisingly, the advent of the automatic camera had a distinct effect on each of these classes of users. For the non-enthusiast, the more that can be automated the better. Automatic exposure, and later, autofocus is a great boon to these people who can now take pictures of higher and higher technical quality while never needing to think about it. Perfect.

Professional photographers also tend to be happy about these features. After all, on the high end cameras the most sophisticated automatic systems are mostly built for them. However, automatic cameras introduce layers of complexity that must first be mastered before a good photographer can take full advantage of them. The late Galen Rowell wrote about this 20 years ago when he compared using an automatic camera to flying a plane on instruments. Rather than interacting with the scene and the settings on the camera directly, you have to observe the scene and then also observe what the automatic system has done for you before you have confidence that the settings are right. Furthermore, when the automatic systems break down, you have to reverse engineer their logic to work out how to make it right. Thus, there is constant fiddling with exposure compensation or autofocus tracking or which focus sensor to use in the viewfinder just to trick the computer into making the settings that you would have dialed in by hand on a simple camera.

So here is the rub: with a manual camera you feel more secure about focus and exposure because you are setting them directly on the machine. With an automatic camera, the interaction is more indirect. You program the camera so that it will automatically do the right thing. This is inevitably difficult at first, but like everything with computers, once the basic abstractions are mastered they can provide a remarkable amount of leverage. Modern autofocus systems are so good that users expect to be able to set them up to accurately track fast and complicated action while still getting every shot sharp. You might have to dig through a dozen different menu settings to get things tuned the way you like, but a modern camera can do it.

In the modern automated world then, it is the poor amateur enthusiast who is left out in the cold. The more indirect way of working with the camera is the least satisfying for this user, because that tactile and emotional connection with the machine was at least half the reason he was using the camera in the first place. These people will complain that they don’t like “giving up control” to the robot brain of the automatic system, or that modern cameras have been “dumbed down” to cater to the uneducated masses. Analogies to automatic transmissions in cars will be made. This is because inevitably the camera enthusiast is inevitably also a car enthusiast, and driving is another hobby where the size of your private parts is reflected by your ability to quickly perform a mostly meaningless feat of marginal manual dexterity (i.e. shift gears).

Luckily, we can ignore the whining and moaning from this particular audience. The truth is that while they are unquestionably more complicated, modern automatic cameras can be mastered and controlled just as effectively as their older manual counterparts. And, while it is true that the someone really needs to take Canon and Nikon out in the back and beat them with a user interface design stick, I’d still much rather be using a modern camera for fast work under marginal conditions. When push comes to shove, the modern camera can hold focus wide open with a 50/1.4 lens at ISO 3200 better than I can. And then it can dial the ISO back to 200 when I go back outside in the sunlight without me babysitting it. That’s just great.

Sometimes the tactile wonder of a mechanical camera still calls out to me once in a while, but it’s mostly false nostalgia. I think the reality is that modern cameras really are better and any reasonably competent photographer should be able to learn how to master one. Of course, if you can’t you can always buy a point and shoot, like a Leica M9.

ISO 12,800

High ISO cameras are all the rage now. In fact, they’ve been all the rage ever since the dawn of the digital SLR era. Somehow we’ve all forgotten that we used to get by with Tri-X (ISO 400) as “fast” film, and pushed Tri-X (ISO 800, maybe more if you tempted fate) as “last ditch desperation” film. These days if a camera sucks at ISO 400 all we can say is that it either has a “tiny crappy sensor” or it was built by incompetent monkeys from Germany.

But here’s the thing. It doesn’t really make your pictures better. I think it encourages people to take lots of bad pictures in dim bad lighting. In particular, it enables that smug “oh I would never shoot with flash” attitude which really means “I don’t really care if the light I use is any good”.

All of this griping might seem hypocritical from a guy who bought a D700 at least partly because of its high ISO performance. But, what can I say. There are times I find it useful. There are times it lets me make a picture I would not otherwise be able to make. But it doesn’t really make the pictures better.

Weather Sealing

Every time I hear someone gripe about not being able to buy a camera body because it is insufficient build quality, and in particular if it is not “sealed against the elements” all I do is snicker and think of this photo.net thread about skydiving with a D-Rebel.

The payoff message is about eight or ten replies in. Trust me, if you aren’t working harder than that guy, you don’t need weather sealing.

That 50/0.9 Lens

See ISO 12,800. I thought about it, but I don’t think I have anything to add.

Lighting Equipment

Of course, the flip side of being obsessed with not using flash is being obsessed with using flash. I swear there are more strobe lit pictures of boring skateboard people, cats, iPods, or half-dressed women with tatoos at that stupid Strobist flickr group than I care to stare at for one second longer. And they all seem to be afflicted with the same one light in front (or behind) and one to the left (or the right) lighting scheme. I guess I should not complain, as I’ve been as guilty of this sort of fooling around as the next guy. After all, those Christmas pictures my dad liked were really well lit, thanks to the kick ass cross-lit living room for Christmas scheme that I stole from, well, the Strobist.

Carrying a tripod

Tripods are a great way to carry around 10 pounds of extra weight so that you can take bad pictures that are really really sharp. Alternatively, they are a great way to carry around 10 pounds of equipment that you never ever use.

Here is the thing. I don’t begrudge anyone their tripod. For certain kinds of work in certain environments, tripods are indispensable. If your goal is to make huge prints and your favorite subjects mostly sit still, then by all means bring the pod along. Or, if you are shooting something that is very far away and requires a huge lens to reach it, you have my blessing. However, spare me all the pious rhetoric about how the only way to get technically acceptable images is to always have your camera on the tripod and the mirror locked up as you breathlessly hold the plunger of the cable release at the ready. For a lot of pictures and in particular a lot of subjects, this would be death.

A camera bag that does not look like a camera bag

Here’s a hint: when you take the camera out of the bag, people will know it’s a camera bag. Get over it.

A Spot Meter

I think Ansel Adams made spot meters popular. His classic treatise on the control of exposure and development in black and white photography unleashed a legion of disciples into the world, each one methodically pointing her spot meter at things to make sure she knows where Zone VIII will fall. I went through a spot meter phase when I shot a lot of slide film. You had to be careful with slide film because if you hit it with a stop or two too much light it tended to just go white on you.

It’s not clear to me that all of this is needed any more. If you have sufficient experience with your camera and are good at reading light, you can mostly trust your camera’s “multi-segment” meter to get pretty close to the exposure that you want. Sure, you might have to compensate for extreme tones once in a while, but that’s easily done. Finally, you can easily find out if you missed badly by quickly scanning the histogram of the picture on the back of your camera. Oh wait, that would be chimping…

Not Chimping

People like to dump on chimping. I think this is dumb. I think if we are going to carry these complicated digital cameras around we should exploit every advantage that they afford us. And, to my mind chimping is one of the biggest. You can chimp to get a good meter reading while nothing is happening, and then shoot away when the going gets good. You can chimp to set up your flash on manual and then take an hour of flash pictures without worrying about the TTL metering going nuts. You can chimp to make sure you didn’t blow out that highlight. You can chimp to make sure your mom’s eyes are in focus.

I don’t completely understand the hate, but it probably stems from the sort of “I get everything in one shot” dick waving that is popular among most male photographers. Only the weak, the thinking goes, would need to use something as dumbed down as the chimping screen to check their work. Real men get it right the first time and every time. Well, I guess if you really are perfect then more power to you. I’ll use my dumbed down chimping screen, thanks.

A Bigger Camera

For most people, bigger cameras are mostly just heavier. Aside from providing you with a good workout, all that extra weight doesn’t amount to much of an advantage in my opinion. Of course, some big cameras have unique advantages that are not in smaller cameras. But, if you are reading this weblog, chances are you did not need those things. So, carry smaller cameras. Your back will thank you.

Of course, if you are one of those genetic freaks whose hands are just SO HUGE that you can’t even pick up a Canon S90 point and shoot because your fingers are the size of footballs, then by all means get that 1D mark IV. Who am I to stand in your way.

A Smaller Camera

Smaller cameras are smaller. This means you carry them more, which is a great advantage. However, I am tired of people claiming that smaller cameras are “less intrusive” or make the fact that you are taking pictures “less obvious.” I occasionally test this theory by trying to take hip shot candids of people with my small point and shoot. Here is what I learned:

1. The pictures are always out of focus.

2. The people point and yell at me anyway.

3. I never get anything better than the time I pointed a Nikon 8008s film SLR with a 24mm lens right into the face of that lady at the bus stop and snapped 5 frames.

The truth is that the big camera brings out more anxiety in you than in your subject. If you project a relaxed state of mind and have a rapport with your subject, no one will care if you are pointing that 24-70/2.8 zoom in their face. The way to disappear with a camera is to get people used to the fact that you will always be pointing a camera at them. Then they forget you are there.

There. I made it to ten. One parting thought: reading long winded blog posts about photography also won’t make your pictures better. Happy new decade. Go shoot some stuff.

I would be on shaky ground if I tried to present you with a purely technical discussion of what makes pictures look sharp. I don’t have the background in optics or the various engineering fields necessary to talk about the subject at that level. But in my experience in taking photographs, I do know that there are a collection of factors that you, the photographer can control, that lead to a general impression of sharpness, or not. I will list these here for your reference.

Focus

Every picture as a point of best focus. When you are good, and sometimes when you are lucky, you manage to put the most important part of your subject at the point of best focus in the picture. This makes the picture look sharper.

Depth of Field

As you move away from the point of best focus things get progressively blurrier. The concept of depth of field refers to how far to the front or back of the focus point you can be and still “appear” sharp. Depth of field is a function of the focal length of the lens, the focus distance, and the aperture that you are using. These are all numbers can you can read off. But, depth of field is also a function of how blurry you are willing to allow the picture to be. This last parameter is called the “circle of confusion”, and all the DOF charts or graphs all depend on first fixing how large you allow this circle to be.

Generally you get more depth of field by taking pictures of things that are far away with short focal length lenses at small apertures like F8 or F11. Of course, things that are far away are often tiny and boring and you can’t shoot everything with a short lens. Also, stopping down too much can introduce diffraction and other problems. In addition, having too much depth of field can cause its own issues, as anyone who has found herself in a photograph with a telephone pole growing out the side of her head can attest. So, working with depth of field, like everything in photography, is an exercise in compromise.

Image Size

The more you magnify a picture the more blurry it looks. I think this fact alone would discourage people from having absolute discussions of sharpness, but it doesn’t. I have a modest example. I took this picture in France, and it looks great at flickr sizes:

.

When you print it 12×18 and stare at it, you realize that I didn’t put the focus where I should have and the right half of the shot is mostly out of focus. So is the picture sharp? Yes and no. There is no single answer.

These days with digital cameras, people tend to hold their pictures to a ludicrous standard in terms of sharpness versus image size. You open the file up in Photoshop and crank it up to 100% and proclaim that your camera can’t focus, your lenses are crap and you botched the depth of field yet again. Never mind that you’d have to print the shot at 60 by 40 inches to really notice. Don’t be that guy. At this point I will note that I’m the only one I know who is bothered by the focus problems in the above picture from France.

Subject Motion

Aside from focus problems this is probably the number one reason pictures don’t look sharp. If what you are taking a picture of moves while the shutter is open, it will be blurry. The solution is to either nail the thing in place or use a faster shutter speed. Probably half of all example images proclaiming that a camera or lens doesn’t work are actually unsharp because of subject motion.

Camera Motion

The flip side of subject motion is camera motion. If your camera shakes up and down while the shutter is open then it’s just like your subject is jumping up and down in front of you while the shutter is open. What you get in the end is a blurry picture. Probably half of all example images proclaiming that a camera or lens doesn’t work are actually unsharp because of camera motion.

Sometimes it’s fun to play around with using subject and camera motion on purpose… but it usually doesn’t work:

Subject Texture

Here is an intersting wrinkle. Some subjects just look sharper than others. Rocks look sharp:

Water and fog and mist, not so much:

Filling your pictures with things that have a lot of fine detail or clear sharp edges tends to make people think they are sharper. One of the reasons for this is the special importance of edges in the psychology of perceived sharpness. Edge sharpness is so important that the most used filter in Photoshop is the so called unsharp mask which, in its simplest form, is nothing more than a computation that increases contrast around the edges to make them look sharper.

Lighting

Another often overlooked aspect of sharper pictures is the lighting. In general directional light that emphasizes textures and edges makes pictures seem sharper. Also, juxtaposing high contrast light against low contrast light is a good way to make your subject “pop” out of the picture. Head over to the Strobist for more lessons about this. The principles are the same for either flash or natural light. Here is a lame example:

If you want your pictures to look flat and boring and completely lacking in texture, find light that is head on and very uniform:

Flat light from the front makes it impossible to find the edges in the picture, and sharp looking edges are psychologically important.

Resolution

I put this last because out of all of the factors I have listed it probably has the least to do with perceived sharpness. Resolution is nothing but a measure of the smallest detail that your imaging system can show as an independent entity in a picture. It can effect sharpness in that if you are taking a picture of something with a lot of fine detail, that detail will look sharper if it is better resolved. But there are plenty of pictures that are relatively low resolution that look sharp anyway. The best examples are HD television and projected films, both of which we think of as sharp even though the raw resolution is pretty low. Very small but sharp looking web photos are another example.

I think the rise of digital cameras have made people overly conscious of resolution and they tend to latch on to it in the hope of finding a single number that means “good image quality.” I think history will show that hardly anyone needs more than a few megapixels, and of those people, hardly any really needed more than six.

Related to resolution, you will also notice that I didn’t mention anything about lenses. I don’t have much to say about lenses for two reasons. First, I think the other aspects of image sharpness are more important than the raw performance of the lens. Second, I have never felt that I am “lens limited” with respect to how my pictures look. If there is something wrong with them, there is almost always something else to blame, and it’s usually my technique. If I were a real photographer, I would feel more confident about proclaiming grand truths about lenses. But as it is, the reality is that all the lenses you can buy are probably good enough. And if they are not, you’ll probably find out before me.

To sum up, in order to take sharper pictures you should:

1. Find sharper subject matter.

2. Keep the subject still or your camera still or both.

3. Use good light.

4. Learn to focus.

Seems simple enough. Maybe I’ll figure it out in another 20,000 pictures.

But here is a line from the review that confuses me: “Doing serious photography with a camera at arms length just isn’t my thing.”

You hear this complaint over and over again from the people in the world who just can’t live without an eye-level viewfinder. I understand the shortcomings of the LCD screen, but this complaint still baffles me. The implication is that the only way to use the LCD on the back of the camera to compose pictures is to lock yer arms straight out in front of your face like a mentally crippled robot and then point the camera in the right direction. So here is my question: do serious photographers not have elbows? Can any of you out in the Intertube-land enlighten me? Do you have elbows?

The D700, of course, is Nikon’s “smaller” and “more affordable” model of “full frame” digital camera. That is, the sensor in the camera is the same size as a piece of 35mm film used to be. Nikon thinks this is so important they even gave a new name to the full frame cameras, calling them “FX” cameras. The older digital SLRs with the smaller sensors are now called “DX”.

It took me a long time to convince myself that I wanted to pony up the extra money for an FX camera. After all, there is nothing sacred about 35m frame size. It’s just a historical coincidence. The supposed advantages of the FX cameras are easy to summarize and refute:

1. Better image quality per pixel. Nikon’s pro cameras all have 12mp sensors. This means the pixels in the FX cameras are larger, which in theory means various things about the image will be better on a per-pixel basis. But realisitically, the 12mp DX sensor that is in the D300 and D90 is already almost unimaginably excellent. Just ask peterb.

2. Your old lenses have the same field of view and depth of field as they used to. This is no big deal, you can either get used to it or get different lenses. The Nikon DX lens line is pretty well filled out, covering very wide to very telephoto. Just suck it up.

3. Nicer viewfinder, and some handling advantages. It’s true that the FX viewfinders are larger and easier to use. On the other hand, if you shoot with autofocus most of the time who cares? Also, in the D300 the AF sensors cover the whole frame which is awesome!

FX, it seemed to me, had a thin set of advantages at a relatively high price. So, why did I end up caving? Three reasons:

1. The sensor is insanely good.

2. My old lenses work right again.

3. The viewfinder is really nice.

I have a modest example what the sensor can do here:

This picture was taken by room light at ISO 1250. As a point of comparison, I think it looks as good as what my D200 would do at ISO 200, maybe 400 if I was lucky. But that’s just the noise characteristics. Pictures out of the D700 just look sharp to me. When you hit focus correctly and you control the camera shake, all of the detail that you imagine will be in the picture is there. This didn’t always happen with the D200 no matter how careful I was. The D700 lets you get cleaner, sharper pictures in more situations than I ever imagined possible. That doesn’t mean the pictures are better, but it does make the process easier.

Also contributing to this ease of operation is that my 35/2 is finally a 35/2 again. Let me explain. I like the point of view provided by a 35mm lens used on a 35mm film body. The 35mm F2 lens was always was one my favorites on my film cameras. But, on a DX camera, the field of view is cropped. To get the wider field of view, you have to use a 24mm lens. But the 24mm lens has two problems

1. Everything looks like you used a 24mm lens.

2. The 24mm lens is a 2.8, not F2.

These don’t seem like they should be insurmountable problems, but I never could get comfortable using the 24 to get my 35mm point of view. In addition, the D200 was never great at focusing the lens. So, I would just give up and use my zoom lenses. There is nothing wrong with zoom lenses, but they are bigger, heavier, and slower (F4).

The D700 brings back the joy of the 35/2. The field of view, perspective, and depth of field are finally right. The autofocus is faster and more accurate than with the D200. And as a bonus you can take pictures of things you can barely see, lit only by the slightest light, and they look like you were working in daylight.

The viewfinder is really great too.

In addition to resurrecting my old prime lenses, I also picked up the Nikon 24-85mm f/3.5-4.5 AF-S G zoom lens. This is the zoom I used to use on my D100 and stupidly sold with the D100 to get a D70 with the 18-70 kit lens. On the D700 it’s a bit wider than the 18-70 was on the DX bodies, and as a bonus the lens is a bit smaller and lighter too. I plan to be happy with it. At some point I’ll get a 20mm lens to get just that much wider, and then maybe a longer telephoto prime too. I’m not in any hurry.

Overall the D700 feels like a machine that I have no right to be using. The thing is a bit too big, a bit too fast and most of all a bit too good for what I’ll do with it. Realistically all I will do with it is family snaps, vacation pictures maybe the odd bit of personal cityscape work around Pittsburgh. Still, I’m lucky enough have this camera available to me at this point and time, so I decided I’d take advantage of it. It may be too much camera, but it will be a hell of a lot of fun to use.

And the viewfinder is really nice too.

With digital processing techniques, I generally don’t need a log book like this. I can just open the file again in Lightroom or Photoshop and the software will have kept all my settings for me. But, once in a while this is not enough.

A couple of years ago I made what is now a favorite picture in Paris. The one with the boat on the river at sunset:

I made this picture by combining two pictures. One was very dark, but kept detail in the sky.

One was very light, but kept detail in the boat.

I combined the two pictures together as layers in Photoshop. The bright one was on the bottom. The dark one was on top. Then I made a mask that hid the dark parts of the dark picture that I didn’t want and let the bright parts show through. The mask looks like this:

I had all this information in the Photoshop file that I kept with this picture in it. So I knew at least in principle how I had done it. But when I tried to apply the same technique on some recent pictures I kept failing over and over again. I realized that I could not remember how I had made the mask.

There were two aspects of this mask that vexed me. First, it is in almost perfect alignment with the elements in the pictures that it is meant to mask off. Second, the mask itself seems to be a composite of the light and dark pictures. It’s almost as if I combined the pictures together to make the mask so I could combine the pictures together.

After playing around with this for most of a weekend, I finally think I know what I had done originally. The first, and most important step, is to the use the Photo Merge feature in Photoshop to merge as a first step in combining the shots. It turned out that I was shooting these pictures hand held, and so even the parts of the picture that did not move are not in perfect alignment because I moved the camera between shots. My first 15 attempts at masking were foiled because the normal hand alignment of the bits was not precise enough for the final picture to look clean.

As a side note, if you look at the composite carefully you can see that it in fact is not nearly perfectly aligned but it’s a lot better than you can get by hand. I can now see these flaws in the print, and it will bug me a lot because I have a disease.

OK. With the layers aligned, making the mask is easy. First, add an all white mask to the dark layer. There are a few different ways to do this depending on which Photoshop you are running. Recall that this mask is a 8-bit grayscale layer. You can edit this layer by opening the Channels pane in the Layers and Channels inspector and then clicking the eyeballs on and off until all you see is the alpha channel that the mask lives in. You can edit the mask using all of the bitmap tools that Photoshop provides including the drawing tools, copy and paste, and so on. Areas of the mask that are white or light colored reveal the picture underneath them. The parts of the mask that are black or dark block those bits of the picture from being visible. We have started with an all white mask, so right now all you can see is the top layer, which has the dark picture in it.

Having created the empty mask, use the magic wand tool to select all of the sky, following the outline of the trees. This will work well because of the high contrast between the tree and the sky. Copy those bits out of the light picture and paste them in the mask. This will make the sky part of the mask mostly white and very light gray. In this form, the mask will mix the dark clouds in the top layer with some of the light colors from the light layer.

Next, go back to the dark layer. Use Load Selection to load the selection from the layer mask. Then invert the selection and copy the dark bits from the foreground of the dark picture. Paste those in the mask. Now adjust the darkness of the dark part of the mask so that it blocks out what you want it to block out. Voila you are done. The resulting mask reveals the sky part of the dark picture but blocks most of the dark bits.

Note that this will only work well if you have a large expanse of white sky that is easy to select. In a sense, I got lucky because the contrast in the original scene was so high.

Of course, these days you can just buy some off-the-shelf HDR software to do all of this work for you. You push a button, adjust some color settings and get this:

Note how the automatic merge is also better than mine. It brought some of the warm light from the foreground of the dark shot into the final picture, which I did not do. It also aligned the mask with the trees much better than I did. Here is my work blown up to 100%:

Here is what the computer did:

The software I used is called Photomatix and it seems to be a bit more intuitive to use than the built-in HDR merging in Photoshop. I have not been able to figure out how to do the final color corrections in Photoshop so the picture looks “normal” instead of freaked out. But Photomatix seems to do a good job.

So there you have it. In the end, I found both my original answer and a better solution to the original problem. If any of you Photoshop wizards out there have an even better scheme for doing this, let me know. Next time I try this I want to be more prepared. Or, maybe I’ll just learn how to use ND grad filters.