Monthly Archives: July 2009

One Other Thing

Posted on by

… about the new camera. It’s not just a (high speed) video camera, it also takes stills, though it doesn’t do all the things that an SLR does.

damselfly

This is a damselfly, which are related to dragonflies. The main difference to my eye (not being a biologist) is wings open vs. closed when resting. There are other differences, too, though rectal tracheal gills was a phrase I didn’t need to see. Oops. Spoiler alert.

A great, or at least pretty good, blue heron.

blueheron

A monarch butterfly and caterpillar

before-and-after

Some bug on Queen Anne’s Lace
bug-1224

Deconstructing BJ Thomas

Posted on by

Myth of raindrop formation exploded

“Myth” may be overstating it, which is par for the course. But someone actually observed what happens to large raindrops, to see why we get the distribution of raindrop sizes.

The pair got the idea from the unusual but well known transformations of fuel droplets travelling at high speed in diesel engines. As they travel, drops flatten from a sphere into a pancake-like disc; this catches passing air and inflates like a liquid parachute that eventually explodes in a shower of smaller droplets.

Vendor Gifts as Proxy for Economic Recovery

Posted on by

I buy things for the lab (which may not be a true statement in the near future; we shall see), and some vendors like to thank me for my business by throwing in some item that I didn’t buy but might find useful or enticing. I have gotten t-shirts, beanbags for juggling (or possibly hackey-sack), the ever-useful thumb drive that is 3 generations smaller than what’s on the market (128 MB. Oh, joy), and , of course, office supplies like pens, highlighters and post-it notes emblazoned with the company logo. Often, however, the gift is food of some sort.

I’ve noticed that in the recent tougher economic climate that the companies engaging in this practice had been scaling back or discontinuing their give-away advertising efforts. I’m happy to report that the “lab snacks” assortment we get from one optics vendor, which had been downsized for much of the past year (or at least ours had been), have returned to the full cornucopia. Full recovery can’t be too far down the road.

High-Speed Thoughts

Posted on by

I’ve been having way too much fun with my new camera, and (as any regular reader knows) have posted several videos. Many more are still unedited and unuploaded (is that a word?), and even more will just sit as a collection of magnetic orientations on my hard drive because they ended up being not particularly interesting, but I don’t like throwing things out if there’s any chance they might be useful at some point.

Lessons learned, in no particular order:

Some things look interesting in slow motion, some things don’t.

Obviously, slow-moving (or nonmoving) things are a waste of a shot. If you can discern details with your eyes, slowing things down won’t help much. A fan filmed in slow-motion looks just like a fan moving at a slower setting. A medium-sized bird flying along looks a lot like a bigger bird flapping more slowly. What you need is some nonlinearity — some kind of inflection — for these situations to become interesting. A bird taking off, for instance, is a lot more interesting (to me) than one just flying along. But you often do it anyway, just to see if slowing it down reveals any details.

Some things are really frikkin’ fast, and slowing them down by a factor of 14 (30 fps to 420 fps) doesn’t really put a dent in it. The sweet spot is around 100 Hz. At 10 Hz or lower you can see what’s going on without assistance, and when you get above several hundred Hz you aren’t sampling the phenomenon often enough to truly see the detail — Nyquist’s theorem in action.

These are movies — things move

This is a no-brainer, but for someone used to taking stills, it’s a problem of adapting. Zooming in too far means that when the action happens, the target quickly moves out of frame, and it’s really hard to follow and keep the target in-frame. Zoom out too far, though, and there isn’t enough detail to make the shot worthwhile. For a regular video, I’d say pick your poison. For slow-motion, I think you can err a little on the side of zooming in, because you get more frames, and still have the chance to capture something interesting.

The flip side of this is that my motion has a smaller effect on the shot. The camera has a steady-cam setting, and any of my motion is “damped” by the frame rate. Physical jitter isn’t as annoying. I use a tripod when I can, but it’s not a must unless you need both hands for the shot itself.

Lighting is really important

Another no-brainer, but it still takes getting used to. You can’t really use a flash, and you might not realize how bright flashes are. Even though the camera films at lower resolution (and I assume this is to group pixels together to gather more light) the camera still doesn’t do as well in ambient light as a still photo. 420 fps is only about two milliseconds of exposure time, and you have to gather enough light so that noise isn’t an obstacle, which limits you. I’ve purchased a couple of LED desk lights for indoor shooting, as well as outfitting a traditional el-cheapo-desk lamp with a CFL (the fixture is limited to 60 Watts, but a 100 W-equivalent-brightness CFL draws only about 20 W). The drawback to this is that cheap desk lighting gives you a 60Hz flicker. I film near a window when I can, to get sunlight involved.

A corollary to this is that the aperture is wide open, which reduces depth-of-field. The camera has an occasional tendency to auto-focus on the wrong item, and the shot ends up being blurry. But that’s not a disaster, because …

Film is free

As with all digital photography, this is another habit one has to shake off if one is used to film and developing — the spectre of not having enough film, and letting that dictate whether you take the shot. Digital cards are cheap, and you can delete bad pictures right away if necessary. You can also decide if you want to re-shoot, (if circumstances permit) if you know that a shot was bad. If I’m doing a set-up shot with consumables, like match-lighting or balloon-busting, I prepare several samples.

Regular video isn’t all the great.

Even with the steady-cam on, there is a lot of jitter when holding the camera. I haven’t tried Rhett’s water-bottle hack to add mass, because I haven’t been in many situations where I want regular-speed video.

New Tool for the Stealth Paparazzi

Posted on by

Invisible flash takes photos without the glare

Although the dark flash gives a crisp image without disturbing those in the picture, the results have an odd colour balance that looks like a view through a night-vision scope.

To give the pictures more normal hues, Krishnan and Fergus used colour information from a brief, flash-free photograph of the same scene taken quickly after the dark flash image.

The second image is dim and blurry, and so it lacks some of the fine detail of its dark flash twin. However, the researchers use software to combine the sharp detail from the first image with the natural colours from the second image, resulting in a remarkably natural end result.