SLRs

I think the best thing about SLRs isn’t their elimination (well, exponential reduction) of shutter lag, nor the support for high ISOs, or even advanced exposure and metering modes. It’s that even at relative high apertures (f/5.6), you can keep a shallow depth of field. Consider this photograph:

Something-flower

(Does anyone know what type of flower this is, BTW?) The photo wouldn’t be half as good if everything were in focus, as a normal camera would have rendered it. But by throwing the distracting (and ugly!) background out of focus, the shot comes out a lot better. I don’t entirely love the depth of field on this one; I wish you could see a little more of the plant clearly (which would have required that I stop the lens down a bit more), but I also wish the background were even further out of focus (which would have required that I open up the lens a bit more). BTW, a little bit of HDR going on here, as it wasn’t the best lighting.

mIMG_2571

There’s another example. Too shallow, or at least, I should have manually selected the autofocus sensor to use one on the left, so that all the caterpillars were in focus. But the background (green and purple bushes) are pleasantly blurred, keeping your attention on the tree.

m-IMG_2593

Here I totally disregarded the rule of thirds. I like it anyway. The other leaves were pretty nearby, so they’re only slightly out of focus. But again, it draws your attention in closer.

m-IMG_2596

There’s the best example. The trees in the background were across the street, and thus extremely out of focus. The camera focused on the leaves, which are tack sharp.

And now, I’m going to go finish mowing the lawn. There were just too many photo opportunities I noticed… 😉

Although I’m attending a Fishercats game tonight… It’ll be my first time with an SLR there. Let’s see how that goes.

Central NH Photos

I joined my family today in central/Northern New Hampshire. It was opening day at Clark’s Trading Post, a favorite of my brother. My brother encouraged me to go with them, and the forecast called for a decent day, so I figured I’d tag along. The area’s always been quite photogenic, and I figured it’d be a good chance to continue my exploration into HDR.

Climax Locomotive

That’s the train at Clark’s. This is what I like to think of as a halfway-decent HDR shot: a close inspection will reveal some technical flaws, but it’s normally a difficult subject to shoot well. It’s largely a black train, glossy in parts, matte in others, but it also has shiny metal highlights, plus the sky. This isn’t to say that it can’t be photographed, just that the results are ordinarily less than stunning. What I like about this shot is that your first thought isn’t, “What type of surrealist artwork is this?!” As seems to be typical of my shots, the sky looks kind of wonky, but overall, I’m happy with the shot.

Clark's

There’s an example of the type of stuff I’m still on the fence about. It’s just kind of jarring in a way, as the colors, while “correct” in a sense, are unnatural. Rather than correcting for the fact that the camera can’t capture the whole scene the same way the human eye might, it goes further and does something even our eyes can’t. It’s a little surreal, but the style is growing on me. (Trivia: look closely and see how many things you can spot wrong. When stitching together multiple photos in which people are moving, things are bound to not quite match up right. They’re fairly subtle in this photo.)

Clark's

There’s another building, again showing the more “legitimate” aspects of HDR photography in my mind. (Besides the ghostly half-man.) This would ordinarily be a lighting disaster. The building was receiving what was almost direct sunlight, while a dark shadow existed. And the sky was somewhere in between. In this case, I think the blended exposures work perfectly. The same goes true for this shot:

Tuttle House

I didn’t know how it’d turn out at first… It looks like a simple shot, but it wasn’t! My typical method is to set my camera up for auto-bracketing, taking three shots in a row, one properly exposed, one too dark, and one too bright, thus increasing the odds that there’s a good shot in there somewhere. Often the main one looks good, but I know that some of the details from the others will boost it when converting to HDR. In this case, though, none of the three worked. If the bottles looked good, the wooden headstones were washed out. If the wall looked good, the grass was far too bright. I had a nicely-bracketed set of three bad photos. Fortunately, Photomatix worked it magic and produced a good shot.

We eventually tired of Clark’s and went exploring the area. Lost River was nearby… I pondered what lens to take, since it wasn’t practical to carry all of them, and made the right choice to bring my wide-angle 18-50mm lens. And here I realized that shooting for HDR, much like switching to using two computer monitors, starting to carry a cell phone, or buying high-thread-count sheets, is a habit that rapidly becomes very hard to break. I’d take a “normal” shot, but realize that parts were under- or over-exposed, so I’d retake the shot as a bracketed set of three, and spent some time in the car ride back home on my laptop merging them.

Trees & Stuff

This is at the Lost River section; hardly the best shot, but a quick-and-dirty example of an ‘acceptable’ use of HDR. The trees are well-lit, and so is the sky. Difficult to pull off with one exposure, but a piece of cake with three and HDR!

As we went along, I noticed various parts of running water. (I thought this was a lost river… Pretty easy to find.) I’d never actually taken the stereotypical long-exposure moving-water shot, so this served as a good opportunity. I set the camera to ISO 100 (pretty insensitive to light, very low noise, but meaning slower shutter speeds), and stopped the lens down to f/22, which gave me about a one-second shutter speed. I set the camera down on a railing and pressed the shutter. Viola!

But I was curious… How would my newfound obsession with HDR play into this? Could you “bracket” that type of shot, and merge them with any success?

Waterfall

I actually didn’t expect this to work, but it ended up being one of my favorite shots from the day. The shots were something like 1/3 second, 1 second, and 2 seconds, so I expected that the water / person would have moved too much. As luck would have it, they didn’t, and the result was that shot. (A nice side-effect is that I rarely remember to stop the lens down for landscapes, but I necessarily did here… At f/22, everything, in theory, is in focus. Although if you look closely, you’ll notice that some of the photo is kind of soft where things got matched up slightly off-kilter.

We then went to the Indian Head Resort, where, in a welcome break from $15 admission tickets, we paid 50 cents to climb the tower. (In hindsight, they should have paid me to climb that thing!)

Indian Head Tower

Doesn’t it look big and scary? Nevermind that I used a wide-angle lens feet away from the base to distort the perspective, nor that I made it an HDR exposure to boost the ominous dark clouds that really weren’t that ominous or dark.

Indian Head Rock

That’s the Indian Head. For those easily confused like me, the Indian Head, and the Old Man of the Mountain are two separate things. I initially remarked, “It kind of still looks like a face,” before realizing that it was the Old Man that came tumbling down, not the Indian Head.

This was yet another one of those shots that was pretty tricky. I pulled out a polarizing filter for this one to try to boost contrast and get the sky to not look so gloomy; you wouldn’t know from the picture, but it helped. You also wouldn’t know from the picture, but this, too, is an HDR shot.

Facing the other way, I decided to take a series of shots holding the camera vertical, intended to be stitched together into a panorama. Since I have under 500MB free on my hard drive (?!) and since I couldn’t find PanoTools or any of its ilk, I ended up trying Windows Live Photo Gallery, which I installed at Mr. T’s suggestion but never got around to using much. (I also brought the image into Photoshop, where I cropped it and tweaked it.)

I’d like to give it good reviews, as it was very easy and quite intuitive. The problem is that I’m fairly certain this isn’t actually how things looked. The pond looks right, but I’m fairly certain that there were more ‘humps’ off to the left. I’m really not sure what happened, but the end result looks good, so I’m happy.

HDR

I brought my brother to organ lessons today, which are held in an old (18th century, I believe) church in Amherst. So I brought the camera along in anticipation of some more HDR fun.

Church Interior, HDR

This is a so-so interior shot. Note that it’s kind of noisy (grainy), I’ll touch on that in a bit. Also note, just as an interesting tidbit, the green cast on things: it was sunny outside and there were lots of green trees right outside the window. This shot, in my mind, is more about what HDR should be used for: getting ‘appropriate’ results where the camera normally wouldn’t hold up. To get detail in the pews and yet not have the upper third of the image ruined by the windows would ordinarily be difficult.

Church Exterior, 1 Church Exterior, 2

Here, side by side (I hope) are two versions of the same thing. They were both taken from a set of three bracketed shots. That is, I have the camera take three shots back-to-back, the first “normal,” the second overexposed (too bright), and the last underexposed (too dark). The underexposed one gets the detail in the sky and clouds well, but the foreground is really dark. The overexposed one fills in all of the details that would otherwise be too dark, such as the leaves on the tree.

The problem is that, after seeing both, I don’t like either. The “exposure blending” one looks natural, but very washed out and dull. The HDR one, by contrast, looks absurdly unreal, but is very contrasty, colorful, and preserves all the detail.

Burnt

That’s a building right next door to the church that burned down. (And has been sitting there in ruins for months.) It’s weird how the tree in the upper left seems almost like it was Photoshopped in because the lighting doesn’t match at all. This is actually a case where HDR helped more than it hurt, though: getting any details in the charred part of the building and the white part of the building, with the sun reflecting off of it, was impossible. Here, HDR helps both co-exist. Of course, the sky looks bizarre (why does it change colors like that?), the purple tree at far right looks unsettling, and the trunk of the car in the lower right of the image exhibits noticeable “ghosting” (or is that a reflection?).

HDR-IMG_1978_79_80

Finally, we have this image. Despite the fact that I was basically shooting into a window with sunlight streaming in, it was quite dark. I shot at ISO 800 and f/3.5, and got 1/50, 1/100, and 1/25 shutter speeds when bracketing. The ISO was fairly high at 800, but the XTi performs well and has negligible noise. UNLESS you try to pull lots of detail out of the shadows, as Photomatix apparently does. At a quick glance, this is an OK shot, but look at the walls and their graininess. If you view it larger, it’s even more prominent. Further, the image came out a bit soft, so I’d like to have applied some sharpening in Photoshop, but that just accentuates all the noise.

So I’ll concede that HDR has some benefits, but that it’s easy to overdo it. And, above all, shoot at a low ISO if you want good results!

HDR

I’ve posted before about High-Dynamic Range photography, with, err, lackluster photos to demonstrate the concept. I’ve been trying my hand with Photomatix, which has given be slightly better results:

HDR Front Yard

It’s worth noting that those were dark, ominous clouds. I shot this with auto bracketing on in my camera, meaning it took one “normal” shot, one overexposed +2 EV, and one underexposed -2 EV. And then, in very simple terms, Photomatix ‘merges’ them into one shot, taking the ‘best’ parts of each.

So I Can Close the Tab

I came across Ken Rockwell’s site the other day, and, as I perused a lot, I came across his interesting mention of the Casio EX-F1. I’ve “graduated” from integrated point-and-shoots to digital SLRs, although this camera costs more than my digital SLR and three lenses put together.

Photographically, it’s mediocre. 6 megapixels. Except you don’t buy this thing for its resolution. You crave it because:

  • 60 frames per second at 6 megapixels. (Note that most movies are shot at 24 fps.)
  • “[S]tereo HDTV movies,” although I confess that I’m not quite sure what that means.
  • Continuous shooting mode, where it’s just constantly shooting at 60fps, and, when you hit the shutter, saves the ones around that time. Thus, you can actually get shots from before you click the shutter.
  • A maximum shutter speed of 1/40,000 second. That is not a typo.
  • 60 frames per second is ridiculous. But if you can take a cut in resolution, you can go further, all the way to 1,200 frames per second at a pitiful 336×96.

Actually, 336×96 isn’t just tiny, it’s a really weird size. I’ve resized (and cropped) a random photo of mine down to 336×96:

336×96 Pixels

In conclusion… 6 megapixel camera, with a long zoom lens equivalent to 36-432mm. And it’s an HDTV video camera. And it’s got the crazy bonus of letting you use shutter speeds of 1/40,000 of a second, and capture low-res video at 1,200 frames/second. I wouldn’t carry it as my main camera (though it would probably be entirely usable for that), but I’d love one of these in my bag for video and such.

I also wonder about the “trickle-down” effect. Although really, more like the “trickle-out effect.” Nikon’s D3 will give pretty clean shots at ISO 6400, something no other camera even tries to offer. It goes up to ISO 25,600. Canon and Nikon are very close when it comes to the frames-per-second rate of their high-end digital SLRs; 7-9 frames/second. (Hint: get rid of the shutter, which is useless on a digital camera where you can just “read” the sensor for a given period of time.) Companies keep focusing on packing more and more megapixels into smaller and smaller sensors. As I’ve said before, I have a 20×30″ print from my 6-megapixel camera. (Cropped a bit, too, actually.) I only “upgraded” to my 10-megapixel XTi because the old one broke and you can’t buy a 6-megapixel SLR anymore. Maybe, just maybe, we’ve seen an end to the megapixel arms race. We exceeded the resolution you could squeeze out of film a long time ago, and now we’re giving medium format a run for its money. When I go to buy a new SLR in maybe five years, I don’t want it to be more than 10 megapixels. But I hope that it goes a lot further than 6 megapixels. And if a “prosumer” point-and-shoot camera does 60 frames per second at full resolution, all of a sudden 3 frames per second on an SLR looks pathetic. Similarly, I’m unaware of any still camera (aside from maybe weird scientific-engineered stuff) that will take a 1/40,000-second exposure, or any flash that’s capable of running at 7 frames per second.

(That said, I’m having a hard time figuring out when you’d need a 1/40,000-second exposure. I only hit my camera’s 1/2,000-second limit when I’m too lazy to stop the lens down…)

Photography

I’ve had yosigo in my Flickr contacts list for a while now, as I’m a fan of their work.

Check out this photo, for example. It’s a very simple photo, and any of us could have taken it. He didn’t do a ten-minute exposure with a special lens, and he didn’t use any fancy filters. Contrast hasn’t been bumped up in Photoshop. A five-year old with a camera could take this shot.

And yet, no one else did. While we’re all out there hunting down elaborate shots of obscure subjects in perfect lighting, he’s taking shots like these. Incredibly simple; that seems to be his trademark.

In learning photography, we all learned the Rule of Thirds. Never center something in the photo. Imagine a 3×3 grid, and “center” on one of those. Tell him that. He keeps taking ultra-simplistic shots, and they keep being amazing.

Perhaps most “frustrating” as a photographer, though, is that I feel that it’d be very easy to imitate his work, and yet almost impossible to maintain the “Wow” factor of his photos. If I took a photo of a plate with a piece of bread on it, it’d probably wind up in my “delete” pile.

Hugin

Between Andrew’s amazing panorama created with Hugin, and Garrett almost simultaneously sending me a link to a Lifehacker* post about Hugin, I figured it was only fitting that I try it out. It’s essentially a software app (OpenSource no less) to stitch together a series of shots and create a panorama.

Since I don’t like reading instructions, I downloaded the software and worked on setting ‘control points’ for a set of five photos I recently took, thinking they’d be good for creating a panorama.

Hugin without reading the directions

I think I need to read the directions.

* It’s interesting that Garrett sent me a link to Lifehacker. Kyle’s recently become a big fan, too. I discovered the site at least a year ago, and always considered it a niche website that no one I knew would have heard about. I’m glad to be proven wrong!

A Little Competition ;)

Colored Amp

Okay, so I much prefer Andrew’s. But I woke up this morning and Kyle’s guitar amp was the first thing I saw. “That’d make a good photo,” I thought.

I don’t know what to make of it, though. I always envisioned it as something that would be a good picture at a concert, lit by various stage lighting. In this case, it was lit by light coming in through the window, and the shot would essentially have been black and white.

So I wonder what to make of my modifications. If I told you that I set up two slave flashes, one with a blue gel over it, and one with a red filter over it, and positioned them on opposite sides of the amp, you’d think I was a really advanced photographer, going to great lengths to create a good shot. But I’m a college student with no external flash, much less two radio slaves with colored gels I can slip in. So I used Photoshop to add in some colored lighting effects, and dialed the opacity down to make it more subtle. I also applied some very “broad” sharpening, because the “Jay” was initially thrown too far out of focus. It’s still not nearly as good as I’d like, but the outline’s better.

So the net result, I think, is that it conceivably looks like a shot taken at a concert, or something else interesting, versus something taken of my roommate’s guitar amp on the floor on a sunny morning. But part of me thinks it’s “wrong” to go that far in Photoshop.

Quick Photoshop Tutorial

As a photographer, and an obsessive-compulsive nut, I spend a lot of time working on tweaking photos in Photoshop. Inspired by a recent project, here’s my workflow for one image. Here’s what I started with:

It’s an aerial photograph, and certainly not a bad photograph. But it can be better!

The first thing I tend to address is how much detail is “lost” in the shadows and highlights. Often you’ll find a photo that looks washed out. (Less often, you’ll find one that’s excessively dark.) The good news is that, within reason, you can fix this up. The classic way is the Levels tool in Photoshop (Ctrl+L). But newer versions have a Shadows & Highlights wizard, something I’ve taken to using before I even get to Levels. The Shadows & Highlight tool (Image / Adjustments / Shadow & Highlight) gives you two sliders, one to bring out detail in the shadows (the dark regions), and one to bring out detail in the highlights (the exceedingly light regions). Here’s how it looks after a quick adjustment:

It’s definitely looking better. But it’s still not quite there. You have to be careful with the Shadows & Highlight tool, because applying too much results in an image that looks entirely artificial. You can adjust it so it starts to look like real life, but take it much further and you’re quickly at, “Wow, someone got carried away in Photoshop…” So I had to stop when the image was still slightly washed out. Take a look at the shadows cast by the buildings, which are the darkest part of the photo. The darkest part should be black, not gray. This is where Levels (Ctrl + L, or Image / Adjustments / Levels) comes in:

A few things to note. The graph is called a histogram, and it’s basically a graph of how much of the photo is each shade of gray, ranging from pure black on the left to pure white on the right. Short of just hitting the ‘Auto’ button (which often produces results I don’t quite like), the easiest way to tweak the levels is to drag the sliders appropriately. Pull the black arrow (underneath the graph, above the “0” text box) until it points to where the graph is just starting to slope upwards. (Or, more accurately, to where the graph is non-zero.) You could do the same with the whites, although in this case, it’s properly-adjusted.

In this case, this is all we need. Press “OK” and you’re done. (A more advanced technique that’s sometimes called for involves using the eyedroppers above the “Preview” checkbox. Click the black eyedropper, and then click on the photo to select what you want to define as 100% black. Do the same with the white eyedropper, selecting what should be pure white in the photo. The gray one is more tricky, as you’re setting color balance, but you essentially want to find a “neutral gray,” something like pavement. You really just click around until you find something that looks good. But in this case, none of this is necessary.)

After clicking OK, here’s what you get:

The shadows are now black. More generally, and perhaps more importantly, the photo no longer looks washed out. It’s properly exposed.

In this case it’s not bad, but the next step is to look at the color cast of the image. It’s more pronounced in other photos, which often have various color tints that they shouldn’t.

Here’s the Variations menu (Image / Adjustments / Variations):

Make a habit of clicking on the photo under “Original,” which will reset it. When you first start otherwise, it’ll save whatever settings you have from last time. Unless things are insanely bad, I tend to slide the slider over from the center to where it’s shown in the photo. Do this mostly just “to taste.” My end result is subtle:

Now, we’re basically there. One last thing. I like really “sharp” images. It’s not blurry per se, but it doesn’t have that “tack sharp” status. Photoshop makes it very easy to sharpen images. I tend to use the “Smart Sharpen” available in newer Photoshop versions, though the older “Unsharp Mask” remains popular and effective. Here’s the Smart Sharpen (Filters / Sharpen / Smart Sharpen…) dialog:

You set two things: the amount of sharpening to apply, and the radius of the sharpening. Here, I used settled on an amount of 66%, and a radius of 0.5 pixels. This sort of blend is common. Another alternative is the “haze” sharpen, which is something like 20/20 (a low 20%, but over a huge 20 pixels). Applying the 66%, 0.5-pixel sharpening gives us this image:

So here we are. We’ve made lots of subtle changes, but all of them are quick. But then jump back to where we started:

One additional tip: I didn’t need it here, but adjusting Saturation can help with some more dull-looking photos. Ctrl+U is the shortcut to memorize, but Image / Adjustments / Hue & Saturation is the menu entry. Just don’t go too overboard.

Digital Photo Recovery

I just discovered PhotoRec, a tool for recovering digital camera images.

For the non-geeks, a quick basic background…. When you save a file, it writes it to various blocks on the disk. Then it makes an entry in the File Allocation Table, pointing to where on the disk the file is. When you delete a file, the entry is removed from the File Allocation Table. That’s really all that happens. The data is still there, but there’s nothing pointing to where on the disk it is. This has two implications. The first is that, with appropriate tools and a little luck, you can still retrieve a file that you’ve deleted. (Whether this is comforting or distressing depends on your perspective…) The second is that, with no entry in the File Allocation Table, it’s seen as “free space,” so new files saved to the disk may well end up getting that block. It’s technically possible to recover stuff even after it’s been overwritten, but at that point it’s much more complex and much more luck is involved.

Last night we went out to dinner… We took lots of photos, but some were deleted. So I figured PhotoRec might recover them. So I gave it a try.

The filesystem shows 163 photos. After running PhotoRec, I have 246 photos. What’s odd is what photos I have. It’s not the ones from last night. They’re scattered from various events, and several are from almost two months ago.

This does leave us with an important tip, though: if you delete an essential photo, stop. Each subsequent thing you do to the disk increases the odds of something overwriting it. In a camera, just turn it off. Taking more photos seriously jeopardizes your ability to recover anything.

In my case, I didn’t have anything really important… I just wondered how it would work. And I got strange results for recovered files. (Which has me wondering a lot about how its files get written out to disk, actually.) But it’s good knowledge for the future. (By the way, PhotoRec runs under not just Linux, but also, apparently, Windows, and most any other OS you can imagine.)