Designing a Macro Bracket, pt.2

Continued from Part 1

I'm sort-of coming to the conclusion that it's not worth making the plate myself. There are several straight brackets, some with cold shoes (or even hot shoes, with a PC terminal) for under $10 on Amazon and eBay. I feel like they'd be a lot more secure than the pre-drilled metal braces I've been using, and lighter, for only a few dollars more. I'm still looking for a good inflexible plastic plate, though.

I'm looking at brackets like this: a cheap Opteka and this Nikon. They're designed for a compact camera to use as a plate. I'm guessing that this is the sort of thing I'll be able to get for cheap on eBay, from a Fotodiox or one of the Hong Kong distributors.

Having a shoe on a store-bought bracket is actually not a selling point for me. I don't want the flash to be fixed in one spot, and I've been using a mini-ballhead, with the flash mounted on a short TTL coil cord. I'd like to keep using that, although I don't have a TTL flash. For one thing, it's very adjustable. For another, the connection is much more secure than a PC sync cord, which slips out of the camera's terminal very easily. I'm also hoping to move up to a TTL flash this summer, but I'll post on that separately.

Designing a DIY macro bracket

I'm starting to think about my macro rig for 2010, now the weather's getting nice enough to go outside again. I have a good general idea of what I want to do for a bracket, given my budget constraints and what worked or didn't work last year.

If you're just thinking about a macro bracket for the first time, it's a pretty simple build. It's basically the part of the rig that connects everything: usually a long flat plate that screws into the camera's tripod mount, with some sort of mount for the speedlight on the other end. The value comes from getting the flash away from the lens axis, to add depth and texture to the subject. The trade-off you're trying to minimize is the added weight of an off-axis light, and the hand-shake and fatigue that it introduces.

Over the next few weeks I'll build the thing, but here are my general parameters that I'm starting from:

Ergonomics
It has to have good balance. This is key. Wrist and finger strain are the number one reasons I'll pack it in before I get the shots I want. I picked up a Zeikos battery grip, partly to get my pinky in on the action and partly to move the center of gravity closer to my hand. I think I might get a hand strap as well -- that actually would help a lot, since it would take a lot of work away from the fingers.

Even with good balance, it still has to be relatively lightweight. I think a plastic plate will work better than the stainless steel ones I've been using. I could also drill my own mounting holes very easily. The trick would be finding one that's strong and inflexible.

Another problem with the stainless steel plate is that they scratch the camera body. I've already covered the one from last year with black duct tape -- it looks pretty good.

The plate can't slip. The weight of the speedlight provides a lot of torque when it's moving, and it'll unscrew the plate from the camera and the flash mount. Part of my problem last year was using flat-head screws on plates with beveled mounting holes, or bevel-headed screws on flat mounting holes. This wasn't by choice: I was using pre-drilled plates from the hardware store and had to work with what they had. I tried to counter this with rubber grip tape, but that costs a few turns of the screw into the tripod mount -- again, insecure.

Utility
The bracket should hold the flash in a good position: about 10 o'clock or so. This hasn't been a big problem for me. The mount for the flash should be adjustable; I've been using a mini-ballhead, and may replace it with a cheap geared umbrella adapter for more stability and fine control. It would be nice if the distance from the camera could be adjusted, too, and I'll be thinking about that when I look at plates.

In a perfect world, the plate and flash mount would be so secure that I could shoot in portrait orientation without having anything slip or sag. The challenge here is that the weight of the flash will often turn the mounting screws counter-clockwise...

The rig should include a diffuser for the flash, basically a mini-softbox. I've just built a new one that I have high hopes for, out of a large yogurt container and some translucent vellum paper. This will get its own post once I've really seen how it works, but the key is getting a large apparent size for the light in a design that stays securely in the flash head. If I end up using an umbrella adapter, I may consider building something like a small umbrella.

The flash should be close enough and strong enough to use a low power level - ideally around 1/4 or 1/16 at ISO 100, f/8-f/14. This is to get a very fast flash pulse, which in macro lighting acts as the shutter speed -- the faster the flash, the better it freezes the image. With my 40D, I'm willing to go up to ISO 200, and in fact shot there all last summer.

Value
The DIY rig should be no more expensive than a store-bought solution, and it should be as reliable as one of equivalent cost.

It should be stable and sturdy enough that I won't need to tweak the design or replace parts over the summer.

The parts, especially the plate, should be made of easily available materials. I don't want to get something custom milled at ridiculous cost. Ideally something from the hardware store that I can cut or drill myself. But no false economy, either: I don't want to pay half-price for a half-assed solution.

Part 2: More thoughts on the plate.

WhiBal and color temperature

I picked up a WhiBal color balance card and, though I haven't had the chance to do any really critical (or even really interesting) photography the past few weeks, it's already given me great results with some easy, casual shots. White balance is one of those subtle things that's hard to get right, but really stands out when it's wrong. Fairly or not, when I'm looking a photographer's work, it's the first thing that gives me an impression of their technical skill.

The card is a small, neutral piece of plastic that you use to determine an accurate white balance, either with a custom white balance before shooting, or as a batch setting in post-production. It comes in several sizes; I got one about the size of a business card, because I do a lot of macro and wanted a smallish one, but for larger scenes Amazon carries a 3.5"x6" Studio size, and there's a wider variety directly from the manufacturer. It comes with a stand, a lanyard, and a thin, cheap-feeling case; they also make a really small one that attaches to your keychain. One nice aspect of the card is that the color comes from the material, not a printed surface, so any scratches won't affect its effectiveness. There's a sticker on one side that's used for setting white and black points in post, and it has an autofocus target as well.

The WhiBal looks like a typical Kodak gray card used for exposure (in that both are gray rectangles), but the two are used for different purposes. A gray card has a standard 18% reflectivity, so your camera's light meter gives you a consistent exposure with them. But they're not entirely neutral in color, and they aren't even the same color in different types of light -- so using a gray card for white balance will give you the wrong color temperature, and it won't even be a predictable amount of wrong in different situations. On the other hand, the WhiBal isn't 18% gray, and using it for exposure will be a bit off; I'm not sure by how much, or how consistently, but I'll bet you could experiment with exposure compensation to meter off the white and black sticker.

In a RAW workflow, you use the WhiBal by placing it in the scene you're shooting, adjust the angle so you don't see any glare on the stickers, and taking a shot of it. Then you go on shooting as normal. If the lighting conditions change, take another shot. Once you're in post-production, you can use the "click white balance" tool of your digital darkroom on the card, and then copy that setting across all the images in those lighting conditions. All the software I've used has a batch-copy tool to make this simple. It's supremely easy and consistent -- it takes about five seconds to shoot the card, and another five to set the entire shoot's white balance.

By the way, in case you're interested, I've measured the color temperature of my Vivitar 285HV flash as 5750 degrees, with a tint of -2 in LR. My 40D's on-camera flash measures 6400 degrees, with no tint adjustment.

The image of the robin above was balanced using the WhiBal card, and for the photography I do -- which is very nature and animal oriented -- it's a great example of the card's impact. Auto white balance was totally off. The Daylight setting I was shooting in was just slightly too warm, and like many images I've taken in the fall and winter, the dried leaves just looked subtly wrong. It's not terribly visible in the comparison image, but it's something that's bothered me for a long time, and I picked this specific image because the WhiBal's setting made the colors of the dead foliage feel "right" for the first time.

So the verdict is good. It's easy to use, tremendously helpful, and not terribly expensive, even for a low-budget photographer like myself. I wish I'd gotten one earlier.

Image straightening, DPP wishes, and kill Zoombrowser

The latest version of Canon's Digital Photo Professional software (v3.8) is due to be released soon, and it's bringing the ability to straighten images -- a feature whose absence has often added steps to my own workflow. It probably won't be as easy as Lightroom's straighten tool, where you basically just draw a line across the image and LR rotates the image until it's straight, but however they implement it will be a lot easier than moving the whole image into a second editor.

DPP is pretty handy. I wish the Edit Image and Main windows were better integrated, and until it lets you edit metadata it's going to be basically useless for most serious workflows, but it's not hard to use and the image quality is fantastic. There are a few things I would really, really like to see though.

First: Smarter sharpening tools. I don't mind if they're a bit slower than Lightroom's fantastic mask-based sharpening tool or a high-pass filter overlay in Photoshop, but they do need to give similar control. Right now the sharpening tool is a blunt instrument that beats your image softness into submission, instead of a subtle precision tool.

Second: Better batch tools. There's no easy way to copy individual settings across multiple images. You can copy and paste a 'recipe', as Canon calls it, from one to several, but you can't tell it what that recipe includes or excludes. I may want a consistent white balance across a whole shoot, but groups of photos with different exposure and sharpness needs. This should be an easy feature to add, so perhaps I'm just missing it.

Third: Ditch Zoombrowser and integrate metadata editing into DPP. That's the only thing ZB can do that DPP and EOS Utility can't. Zoombrowser might have a place for point-and-shoot users who aren't shooting RAW and just need basic editing and library tools, but there's really no need for it to be part of the EOS software suite, except for metadata. And honestly, given the alternatives out there, Zoombrowser looks and feels like legacy junkware: iPhoto, Picasa, and the built-in Windows media tools, all handle basic adjustments, library organization, and metadata better than ZB does. If it's easier for me to deal with keywording after I've uploaded my images than it is to use the bundled software that handles keywording, then there's something wrong with the bundled software. DPP should have a metadata tab in the Edit Image window, and it should be easy to batch edit multiple images.

(I know that one was a bit of a sidetrack, but as long as Canon thinks of Zoombrowser as a useful part of the EOS software suite, DPP's not going to get decent metadata tools.)

My last and biggest wish for DPP is smart noise reduction. I'm actually really happy with the NR tools in general, but there's an opportunity here for Canon to make them even easier and faster. Here's my thinking: Canon knows how my 40D's sensor generates noise in response to ISO and exposure adjustments. It knows how that's different from the sensor in a 1D or 1000D. What I want DPP to do is look at those factors in each RAW image and figure out how much luminance and chrominance noise reduction to apply. Give us defaults to tell it what baseline to go for, like "light", "average", or "heavy", and then DPP can set the NR sliders for us. It would take some cleverness to add this feature, but the hard part -- understanding the sensors and the RAW data -- is already done. It'd be taking advantage of Canon's main strength in the post-processing arena, which is why you'd use DPP in the first place.

I don't honestly need DPP to connect to online services or handle other post-export file handling, though I wouldn't necessarily mind it as long as my other wishes come true first. I don't really need brush adjustments either. There are things that Lightroom and Aperture will always do better than DPP, and for the price I don't mind going over to GiMP or some other third-party program to do them. The fine line, for me, is where the third-party programs sit in the workflow. If I'm moving an image to GiMP, I don't want to spend more time waiting for it to open than I spend actually doing whatever it is I need to do. So picking out dust spots is fine, but sharpening is not.

Portfolio

Smugmug added a new feature this week that it's calling Smart Galleries, essentially a way to place virtual copies of an image into more than one gallery. It's something like Flickr's sets and galleries in one.

Anyway, I've made a "best of" gallery as a portfolio for my nature photography, so when I point someone to my website, I can give them a single link that shows off what I'm most proud of.

Check it out here!

Back-button Autofocus

If you're shooting with a camera that supports it, look into using back-button autofocus, which separates the autofocus system from the shutter button.  Then, the shutter button starts and locks the light meter, and the AF button, pressed with your thumb, starts the autofocus magic. 

Mid-range and pro Canon DSLRs will often have a dedicated button for this ("AF-ON"); the Rebels can use the AE-Lock ("*") button for it.  They all have a custom function that can let you select, basically, which of all these buttons do what -- there are actually several ways to map different functions to these buttons besides the one I'm talking about here, one of which might be even more perfect for your work style.  I can't speak to Nikons from personal experience but I know they have this ability, and I assume other brands do as well -- my EOS A2 can do this, so it's not a new idea.

It takes a little bit of effort to retrain your thumb and forefinger, but not much.  And for a lot of photographers, it's worth it.  If you think about it, having one button control the focus, the light meter, and the shutter all at the same time doesn't make a lot of sense to begin with.  A lot of times, the subject of your image is going to need be in focus and exposed properly, sure -- but that's no harder with back-button AF, whereas photos where that's not the case are going to be much harder to manage without it.

Imagine trying to shoot a person against a bright, colorful sunset with fluffy clouds, and mackerel sky.  You're visualizing a mid-distance portrait with her silhouetted against the background, so you switch over to partial metering, because you don't want the light meter to find some ugly middle ground by looking at everything: this scene is about contrast and color.   So in aperture-priority mode* for shallow DoF, you meter off of the sky to bring her exposure down by two stops, full black, and AE-Lock it.  Then you autofocus on the person, but the light meter wakes up again: your shutter speed drops to bring her exposure up higher than you want, the sky gets overexposed, and everything is either noisy or blown out.  With back-button AF focus, you could have held the exposure with the shutter button, then focused with your thumb, without the light meter knowing or caring.

*Shooting in manual mode would also avoid this sort of problem, except of course that the light is fading fast in this hypothetical situation, and somebody's got to keep an eye on the light meter, whether it's you or the camera.

Another situation where this comes in useful is when you're shooting in low but shifting light, where the AF drive has to work hard to find its subject in the first place.  Imagine shooting a street corner on a cloudy day, with the street signs in focus and readable, and moving cars and pedestrians around it.  You get the street sign in focus easily enough, but clicking the shutter button will send the AF hunting back and forth for whatever it thinks you want.

Sure, you can switch the lens AF off, but that's is a workaround, and it's easy enough to miss a shot because you forgot to switch it back on again.  Back-button AF means you don't have to think about working around slightly complicated, everyday focus/metering issues like this because they just don't come up.  It's no harder than having the shutter button do all the work, and in many cases it's just a lot easier.

2:1 magnification

I picked up a set of super inexpensive Pro Optic Extension Tubes. They feel pretty cheap and I doubt they'll last for years and years, but they come with electrical connections at half the price of the Opteka, a third the price of the Kenko, and a hell of a lot cheaper than Canon's.  If I could afford the Kenko tubes, I'd probably have gone with them, since they'll last a lot longer, but as long as they keep the lens on the camera and pass through information to stop down the aperture, extension tubes don't need to be fancy.  There's no glass in them: they work by moving the lens farther from the sensor, so it projects a larger image.  It's a lot like a film or slide projector, where the farther it is from the wall, the larger the picture.  But dimmer too: you have a fixed amount of light making up the image, and the larger that image, the 'thinner' it's going to be.

We're covered in snow and slush here in Boston, so my macro expeditions have been limited to the far corners of the kitchen, where I found the exotic Chlorophytum comosum: the spider plant.  With the 100mm f/2.8 USM macro lens, the full set of tubes gives about 2:1 magnification.  At this point, you really need buckets of light in order to get an aperture with a reasonable depth of field: my flash was set to 1/4 power and was only a few inches away.  Unlike bugs, spider plants don't run away from the camera, so I used a tripod to make these a bit easier.