Ethan wrote on Jan 10^th, 2008 at 10:28pm:


Tricky!  For distance something between 5 and 10 yards (or meters!) is probably OK. Just measure out a distance to something that you can hit (barn door - good for me ) and throw while recording.

For the moment - I would hope people would just get out there and do it. When enough see how interesting it is, the standardisation will come. Until then I expect to see some 'optimistic' estimates, but that's the way of it.

Your distance you can measure to probably about 10-20 cm  at best - the uncertain part is locating the point of release. For that reason you need a 'reasonable' distance to get good accuracy. On the other hand, go too far and air drag will have significantly reduced the speed and you will read low. I feel that at least 5m is needed to get a 'reasonable' result. It is possible to make compensations for air drag, but it's probably not what most people want.

The nice thing about timing with a sound recording is that the precision is far higher than any other easily available approach, With care it can be significantly better than 1/1000th second. The difficulty is relating the sounds one records to the events one is timing. The sound of an impact is easy to pick out, less so the sound of the release, at least with my slings.

That was where things like slinging through newspaper came into the earlier proposal. No one at the time (T_S included!) took the bait as far as simplifying it further, although 'informal tests were conducted' .

BTW - all of this stuff can get almost arbitrarily complicated, as an earlier thread suggested. If you split the sites of the microphones in a stereo recording you can easily get doppler shift measurements. Add one or two more channels and you get enough to keep a physics/engineering graduate student happy (or miserable!) for years.