Question about the 'Normalized integer format' of the MEASUREMENTS class.

[michpro]

I see an inconsistency in the description of the measurement encoding: It says: “This integer is always signed and given as a two's complement number.”, whereby in the example the exponent 0x85 is decoded as [-5] => normalizer = 10^(-5) exponent and treating this as a number in the two's complement code it should be decoded as [-123] => normalizer = 10^(-123). In microcontrollers, numbers with a sign treated as a 'normal' code of complements to two are computationally simple, and the proposed solution forces the creation of an additional function for conversion.

[michpro]

I misread: integer (mantissa) is encoded in the code of complements to two, so there is no inconsistency,... but the construction of the normalizer is more complicated than it could be :)