This makes sense only if the computer does not sense every pass of the magnet at higher speeds. It would then correct for this by throwing out very short term low speed readings in the middle of sustained high speed readings. Is this true?

Also, place the magnet on your stiffest spoke so the attractive force between the magnet and switch don't knock your wheel out of true.

Also, place the magnet on a spoke leading to the valve because that'll be opposite the rim joint which is usually the heaviest part (I don't know if this applies to CF rims).

Also, I found my circumference measurement didn't give me accurate readings on the local measured 3-mile route so I did a simple adjustment... now I match the route sheets and everything really well.

A few years ago I found that I have a "measured mile" on a road near my house. I use the bike computer's documentation as a guide to get the measurement close. I then take a ride with the appropriate tire pressure and then I ride the measured mile (there is also a sign along the way for the 1 KM mark). Last time I was within 2%. I noted setting that was used on my biek computer and adjusted it 2%. Next time I rode by I was within 1%... good enough for me. There is quite a difference between tires that are listed as the same size, but of different manufacturers or different models.

I tried the measure on the floor thing, but forgot to add my weight to the bike so I was off about 10% the first time I did that. It does work, but it was a bit of a pain.

Happy riding,
André

LOL - I actually made the image in 2005. It was for a GPS-specific thread here on BF. I just happened to remember it was still on my ISP storage site.

This is necessary for basic accuracy, nothing to do with a gnat's ass.

My VDO instructions say to get the magnet on the spoke to pass the sensor something like 1-3 cm. I've got disc brakes, so I'm fairly limited as to where on the fork blade the sensor can be. On mine, the magnet is close enough to almost touch when it passes by. Could be further away and still work fine, but I enjoy tight tolerances in all things.

Mine is pretty accurate, especially when compared with its' own data. I.e. I don't really care if I'm going exactly 14 mph, I just want to know what my speed is relative to yesterday, to my average, to the last hill, etc. Am I getting faster all the time, and how far did I ride? Really all I care about...

Cadence, the only thing I really care about is cadence.

I'm not sure exactly what you mean. My understanding of decent computers is that they'll integrate over a window of a few seconds, calculating distance as (#rotations x circumference( and time as the difference between the recorded time of the first and last measurement. If the computer is even reasonably decent and correctly set up, it won't be dropping measurements. That sort of behavior makes your speed look very glitchy.

As I understand it, there are two competing effects:

*As you go faster, accuracy *increases* because you get more measurements (ie, rotations) over the averaging window.
*As you go faster, accuracy *decreases* because it eventually gets hard for the electronics to determine exactly when the magnet is right over the sensor. Note this is also a function of placement of the magnet on the wheel.

What I don't know is how those two effects play out, and which dominates. I expect it varies with the computer. There also is probably a "sweet spot" where it gets the benefits of averaging before error starts increasing significantly.

I know you posted this to Scheherezade, but you have to understand that some of us just feel the need for speed once in a while.

My average computer speed (with starts and stops) is around 11-12 in the winter time. Actual cruising speed is more like 13-14. And, since my computer measures up to .1mph increments, I would have to be up to at least 15mph to see a .2mph increase (which I do when I'm pushing a little harder/tailwind/downhill).

I know you were joking, but I'm just clarifying.

On a wireless computer, I place the sensor as high up on the fork
where it will match the magnet for a signal.That way I can
slide the computer an inch along the handle bar to make room
for the light battery I strap on the stem.
I had to get a computer so I knew when to slow down and
stop getting speeding tickets in a 30 mph zone.(cheech).

I bet those GPS distances are based on map info which is far from perfect and 2% is pretty close. I think GPS is only accurate to a few feet so it'd be pretty difficult for it to decipher your exact travel distance without using map information.

High-end GPS units are pretty accurate. The consumer devices - stuff found in cell phones or those things people use on golf courses - less so, because the chips are smaller and less powerful. I've heard something like accurate to a 100' circle for that type. That said, I'm sure it's all getting better and more powerful by the day...

I've got an old Garmin handheld that was generally good to about 15' or so, unless it could only get a few satellites. Also, the error would usually go down if I stayed still a while.

The only time I noticed it be significantly off when it thought it was good was when I'd be near a tall building, putting all the satellites it could see on one side of the sky.

You really don't want to think about the technology being used here, verses the true value. Just don't go there.

It's only accurate to a few feet at any one place, but since it's traveling along a path, the error only matters at the start and stop, so a max of 11 feet at both ends.

What maps? My GPS doesn't use maps, it's a hiking/outdoor GPS. You know, REAL GPS from before those cagers *******ized it. Why someone needs GPS to find their way on a road I'll never know.

Actually, civilian GPS receivers are about as accurate as they will ever be. A military GPS unit can pinpoint your exact location, within a foot or so. The GPS units that are manufactured for the public are programmed to be off by at least a few yards. They use a "randomizing" program that throws them off a bit.

I think if you had the patience, you could really dial in a computer.

The hardest part would be finding an accurate stretch of road mileage (the so-called measured mile) to use as a benchmark.

I got so lost in a hilly west Austin neighborhood on Monday that I had to use Google Maps a turn at a time to find my way out. Granted, I have zero sense of direction; but it's a confusing 'hood. Were it not for GPS, I would have been either riding forever or calling my wife, who is much better with directions than I am.

Anyway, I was pretty happy to have GPS and Google Maps on my phone...

This is not true. Civilian GPSs are only inaccurate due to atmospheric interference, and a nearby WAAS station can correct a lot of that out. Military GPS has the ability to average readings over the course of many minutes or even hours, so if you're setting up an observation station or an artillery position, you can get very accurate readings.

It's true that the GPS system DOES have the ABILITY to intentionally incur inaccuracies; the system is called "selective availability". The whole system is shifted around up to 100 meters, though they can make it be off by as much as they want. Along with the signal, they transmit encrypted corrections for the intentional inaccuracies, and military units are able to decrypt and apply the corrections.

Selective availability was turned off on May 1 2000 at the order of Bill Clinton. If SA were still on, a whole lot of civilian GPS use wouldn't really be practical.

The only remaining inaccuracies are variable signal delays caused by ionospheric effects.

One other thing that military GPS does that AFAIK no reasonably-priced consumer GPS unit does right now is carrier phase positioning. Time signal positioning gets you to within 10 feet or so (my handheld GPS often reads 11 or 12 feet error). Carrier phase positioning has the promise of getting as close as a few centimeters. However, it's significantly more difficult, and most people don't really care that much if they're 5 or 10 feet off.

Ignore please

Have you actually ever looked at the napkin size multi-lingual directions Einstein?(so be nice)

I use a Garmin GPS and a Cateye cyclocomp that I calibrated using the roll out method with me on bike.

They match almost always to within .05 miles (The Garmin always a few hundredths less) for a wide range of ride distances (mostly flat)

As to repeatability - the Cateye always reads to the .01 mi for my commute. When it increases by .02 (without any subtle route changes) I know that I need to put air in my front tire.

Al

I'm using a Garmin too and have found it to be fairly accurate and would agree with the accuracy by nosiebeam.

And it in fact seems to adjust for elevation based on topography rather then a known mileage.

You have got to be kidding - that GPS makes it's calculations, like 3 times a second. It's not averaging, or cutting corners, on anything.

Other than changes due to national (and military) security, it's as accurate as things can be!

Suuuuure... We know that you were actually at work. Don't worry... we get bored at work too!