Closed
Description
AbstractTxBuffer_t<> could benefit from the 24-bit floating point format used in the Model 4811 meter; if limiting uplinks to 11 bytes, it gets you from 2 values maximum to 3 max, plus a few extra bytes. Since very few measurements have even 17 bits of precision, a natural reduction of IEEE-754 onto 24 bits is quite useful.