Still an issue with Calculated Sensor

[PeterGraz]

Dear Bertrand, on October 2, in response to my reference to errors in the calculation with Calculated Sensors (#4585), you convinced me by means of the following examples that the calculations are now correct (your screenshots).

These particular examples are still calculated correctly today (1.5.18). However, if you adjust the decimals (e.g. to 1 or 3) for the telemetry value used in your examples (Tension RX), the result changes by a factor of a power of ten and is then incorrect. Your examples are therefore only correct (by chance) for two decimal places. Furthermore, in your 2nd example, if you change Tension RX from 2 to 1 decimal place, the constant 0.02 is changed to 0.2! Regards, Peter

[bsongis-frsky]

It's time to use float there internally for all calculations, and store the result as needed.

[PeterGraz]

Very good decision!