I had the same type of experience with the RS200sd. Even after trying to calibrate it on the track (several times!), it would be off on the very next lap. Worst of all is, as you note, it's inconsistent and there's no accounting for whether it's reading too much or too little sometimes. Minor changes to form and footstrike seem to be able to throw the pace off completely.

My solution was to go get the Garmin Forerunner 301--it works quite nicely and I find it to be very accurate in terms of overall pace/distance ("instantaneous" pace/distance are a little less accurate).

I've been looking at getting the rs200sd...have these results been typical for most people or not, cause if I'm going to drop $200 on a watch like that I'd expect it to be **** near perfect.

Just ran Philly with the RS200sd. Watch showed 26.32 miles, although it stopped working for about 30 seconds during the middle of the race. Can't explain it. Not sure if it affected the mileage count. Watch said I averaged 7:50...chip time said I averaged 7:51. I calibrated the watch two weeks before on a track.

I'd say it's pretty accurate!

vel26, you are complaining about a difference of 0.26 in 7 mi. You real measurement for that course (using only the watch) is 7.06 +/- 0.13, or 7.06 +/- 1.9% Granted, that is a little more than is advertised, but < 2% error is pretty good..

Pommer... you aren't going to get much better that what this gadget does. There was a VERY long thread (here[/URL" target="_blank">) about comparing accelerometer (he had a Fitsense) to GPS. A guy had both and wore both for all his runs. Found the accelerometer to be more accurate, for pace especially.

------------------
Craig[/URL" target="_blank">

quote:<HR>Originally posted by vel26:
Actually, 0.26 was the variation on only 3 runs of that course. I have ran it a few more times and have measurements ranging from 6.84 to 7.23 miles. This is based on 9 runs. The range gets a little bigger the more times I run this course. You can call that consistent if you want but I'll just leave the foot pod at home and call it 7 miles every time. Maybe not accurate but extremely consistent!<HR>

okay, 7.035 +/- 2.8% given your new data.

THIS IS BARELY SIGNIFICANT. YOU ARE ONLY GOING TO DO BETTER WITH A MEASUREMENT WHEEL.

Calling it 7 mi is fine, but why did you buy the thing if that's good enough for you?

The real benifit of the accelerometer distance measurement is very good instantaneous speed measurement. Did you look at KudzuRunner 's experiment?


------------------
Craig[/URL" target="_blank">

Since my name's been invoked, I'll weigh in. I know nothing about the Polar, but since it's pod-based, it sounds like it's in the same family as the Fitsense.

Ah, you gotta love these devices.

Here's what I suspect is going on: variations in stride rate.

The key thing with pod-based units--and something that distinguishes them radically from GPS units such as the Garmin--is that you've got to maintain an extremely regular stride rate. You've got to pick a stride rate that you can hew to, calibrate the device over an accurately measured course (and do NOT trust car odometers!!; I use a 10-speed bike with a computer odometer), and then, every time you run, hit the same stride rate. When you run faster, you do so not by increasing stride rate, but bu increasing stride-length. Or at least this is what you should do if you expect your pod-based unit to give consistent measurements.

Most of us vary our stride rate a bit. Most of us start off for the first quarter to half mile with a slightly slower stride, and many of us, if we're tired at the end of a run, slow down a bit--particuarly the end of a long run.

That will throw off your measurement.

I've measured out one mile, in quarter-mile increments, in a local park. At least once a week, when I'm using my Fitsense, I run that mile--sometimes midrun--to see if I'm still calibrated. Weird stuff happens. When I recently changed from Brooks Adrenaline 5 to Adrenaline 6, I had to increase the CalVal from 76 to 82 in order to stay calibrated.

Perfection--even near perfection--is pretty hard to achieve. And lets not forget that even on certified courses, the actual path you run may be slightly longer.

Still, you're right to expect more than you've been getting. Don't give up quite yet. First, measure out exactly one mile on a stretch of easily accessible park trail or road. (Not with a car.) Then calibrate as I'm suggesting, and practice keeping to that stride rate no matter how fast or slow you're running. I think you'll see more consistent results.

Adam

Why should stride rate matter? I thought that the footpod measured acceleration three dimensions, and therefore could calculate with pretty fair accuracy how far one has travelled. Theoretically, one should be able to pirouette down the road and the footpod would be able to track the distance travelled.

Does anyone know how these things work and why changes in stride should matter?