You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is there real necessity to have two microcontroller classes? In plus microCOM.py and a class inside it don't have the same name. repr method seems to be very strange. Move ORDER_LENGTH and FEEDBACK_LENGTH to consts.
make_message use a very tricky float to int bit conversion (inspired by fast inverse square root I suppose). Why not to use fixed point integer for floats?
The endianness choice must me in constructor.
Note : There is a separate issue for BaseMicro.py
The text was updated successfully, but these errors were encountered:
There is a parent class and two classes inherit from it: MicroProc and Shell as they are two different use of the BaseMicro class.
Indeed __repr__ is strange. Fair point about the constants
make_message does not use this trick, the code on the pico size does (well guessed). As I do not know the range needed for variables, and as I can send floatting point numbers I prefer to ensure that every value can be sent and received
Is there real necessity to have two microcontroller classes? In plus microCOM.py and a class inside it don't have the same name. repr method seems to be very strange. Move ORDER_LENGTH and FEEDBACK_LENGTH to consts.
make_message use a very tricky float to int bit conversion (inspired by fast inverse square root I suppose). Why not to use fixed point integer for floats?
The endianness choice must me in constructor.
Note : There is a separate issue for BaseMicro.py
The text was updated successfully, but these errors were encountered: