what's the problem with MK3I

But it's hard for you to explain why MK2I is more accurate under the same measurement conditions? If there is a significant margin of error with both of MK2 and MK3 in the same time, I have nothing to say. But obviously MK2i is closer to the actual height

jinonsphere said:

But it's hard for you to explain why MK2I is more accurate under the same measurement conditions?

How often have you done this test? Is the MK2 always more accurate?

Actually, I have tested it several times. On the rooftop of the 6th floor, MK2I remained stable at around 20-24 meters each time, while MK3I remained at 62-68 meters. Since both of them use built-in GPS receivers to read alomost same GPS satellites data, why does the more advanced MK3 have greater errors? Is it necessary to perform built-in mathematical model conversion after data reading? Are the calculation formulas for the two types different?

Have you calibrated both watches using a barometer? And then again by GPS ?

Both two watches have been in their original factory condition since  purchased, and i have never used third-party calibration tools. Currently, only the GPS receiver built-in is used to obtain altitude。

I never expected the watch can obtain precise height data. My question is, why the error between the two garmin types is so large？just only using the built-in GPS receiver to receive almost the same data.

jinonsphere said:

Currently, only the GPS receiver built-in is used to obtain altitude。

1) GPS altitude is not the best way.
2) You should calibrate your barometer and altimeter regularly.
3) If you calibrate the altitude via GPS, make sure that you have synchronized the watch with Garmin Express shortly beforehand in order to have the current satellite almanac on the watch. Then start the corresponding activity profile (or altitute calibration) and wait long enough for a good GPS fix.