Some time ago, I was looking for info on ways to get a reading as accurate as possible for hills I ride. With altimeters, there were two options - one that worked off barometric pressure and one that used GPS satellites. A third option, which I now use, was to get the Delorme Topo USA map software.

Of course, the question with the map was how accurate its figures were. As it happens, I just found a guy to ride with sometimes who has an altimeter as part of his bike computer. This offers an opportunity to make periodic comparisons. So far, it seems his altimeter always gives higher figures, but I don't think he calibrated it properly the first time and I know he has not recalibrated it in a week.

Nevertheless, I can look for the DIFFERENCE in altitude on a climb and see if that matches the map. They come pretty close - probably close enough - but there is one thing I'm wondering about since I don't know the details of the calculation process.

Suppose it is erroneously set to be 100 feet too high. Would this 100 foot error be consistent (allowing for built-in error tolerances) for any altitude? I mean, if I'm at sea level, it would show I am at 100 feet. If I then climed to 1000 feet, would it theoretically show 1100 feet or does the barometric pressure it is based on change different amounts at different altitudes?