Collision Detection - Latest Comments in collision detection: Study explores how a fly avoids a swatter

Re: collision detection: Study explores how a fly avoids a swatter

Clive — Tue, 09 Sep 2008 14:40:39 -0000

Electrically-charged devices are cool, but what I really want is a steam powered automatic fly killer ... made of like brass and wood and stuff.

Re: collision detection: Study explores how a fly avoids a swatter

turmon — Tue, 09 Sep 2008 00:44:36 -0000

The question of the "pressure wave" versus "see it coming" theories seems to still be left open.

This is worth knowing, because it would have consequences for informed swatter design. Transparent swatters versus sparse mesh. Windex works in either case.

Re: collision detection: Study explores how a fly avoids a swatter

Bram Cohen — Mon, 08 Sep 2008 18:18:26 -0000

Electric fly swatters work really, really well. They have a bunch of parallel metal bars which get statically charged alternating positive and negative, and when a fly touches them it shorts the circuit and zzzzap, it's dead. Seems to completely and totally evade all mechanisms the fly has for detecting an incoming object, presumably because air can get through it just fine.

Unfortunately many electric fly swatter owners never come to appreciate how effective they are at killing flied, because they're purchased for ... other uses, and the fly eradication potential is overlooked.

Re: collision detection: Study explores how a fly avoids a swatter

Clive — Mon, 08 Sep 2008 13:32:17 -0000

James, that is a breaktaking comprehensive guide to flyswatting!

Re: collision detection: Study explores how a fly avoids a swatter

James Sherrett — Mon, 08 Sep 2008 12:51:14 -0000

In a former life, prior to my engagement with geekiness, I spent summers guiding fishermen in northern Ontario. One of the necessary tasks of the guide is to clean the fish. Nothing, in my experience, attracts flies like fish guts. Flies of all types and wasps. For sport (and because many of my guests were not so interesting), I'd see how many of the flies I could kill.

I could usually kill 20 or 30 flies in a day. So that taught me a few things.

1. Flies have an optimal size for swat avoidance. The deer fly is the most elusive I ever found. Bigger flies (horse flies and blue-ass houseflies) can take a swat and still live, but they're slower and easier to hit. Small flies are harder to hit but they don't tend to move too fast, either in the air of taking off.

2. Flies circling in the air are hard to take with your hand but easy to knock down with something larger. The larger thing you swing ought to be flat, like a cutting board or hard-backed book, and light enough to get a quick, concussive swing. You're not trying to actually kill the fly with the blow, just knock it down and stun it, then kill it.

3. Flies crawling on fish guts (or a counter or plate) will take off very fast. Rob's comment, "The one certainty is that a fly will always fly *up*. Slowly move your hand up behind the fly until you're about 3-4 inches behind. Then swipe the air about 1-2 inches above the fly. Works 90% of the time." is bang on from my experience. This is the money technique.

The exception to this is when flies are buzzing against a window, which is really the only time I've found a fly swatter the most useful tool.

Hope that helps you rid the world of a few flies. I don't count on it, but I hope. Happy hand washing!

Re: collision detection: Study explores how a fly avoids a swatter

Andrew Kember — Mon, 08 Sep 2008 11:11:28 -0000

Whilst a "hipster PDA" might be a worthy iPod touch replacement for me, I feel that a rolled-up newspaper for fly-swatting might make me break a sweat. For next summer, I'm seriously considering the purchase of an Executioner (http://www.theexecutioner.c..., a battery-powered high-voltage hand-held insect electrocuter. Apparently it sparks when the insect hits it, which sounds quite satisfying.

Re: collision detection: Study explores how a fly avoids a swatter

Clive — Sun, 07 Sep 2008 20:00:18 -0000

Rob, xchmp: Fly hacking! I like it!

sdn: The Windex concept is an even more obviously superior hack, though I confess I'm sort of emotionally hooked into the idea that I can kill flies *with my own bare hands*. It's like using a trebuchet or something ... I feel more old-skool.

Re: collision detection: Study explores how a fly avoids a swatter

Rob Hyndman — Sun, 07 Sep 2008 11:13:44 -0000

I just catch them with my hand. The one certainty is that a fly will always fly *up*. Slowly move your hand up behind the fly until you're about 3-4 inches behind. Then swipe the air about 1-2 inches above the fly. Works 90% of the time.

Re: collision detection: Study explores how a fly avoids a swatter

sdn — Sun, 07 Sep 2008 11:08:21 -0000

swatters are pointless. the best way to kill flies: spray them with windex. this is how i de-infested my house last summer.

Re: collision detection: Study explores how a fly avoids a swatter

xchmp — Sun, 07 Sep 2008 09:33:37 -0000

The trick is to distract the fly. Hold a hand out somewhere to the side of it, not so close that it takes off, but close enough that it can see it. Waggle your fingers. Slowly move your swatter behind it, until it's slightly closer than your other hand, then swat as fast as possible.

I'd guess what's happening is that the fly's swat-reaction system gets overloaded with information. It has to deal with five moving objects (your fingers and thumbs). And then another much more slowly moving object comes along (the swatter). Since its attention is on the more obvious threat, it isn't able to integrate the data from the rapidly approaching swatter until it's too late. Interestingly, this doesn't usually squash the fly - the swatter tends to knock the fly out of the air as it takes off. Which allows you to remove it from the area without having to deal with fly innards.

Re: collision detection: Study explores how a fly avoids a swatter

Clive — Sun, 07 Sep 2008 09:20:13 -0000

Testing to see if these comments work! (This is Clive.)