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
I guss currently the Bias is calculated according to the IEEE 754 Standard: Bias: 2^(NoOfExpBits-1)-1. So with 8bit exponent a bias of 127, or with 5bit a bias of 15.
I guss currently the Bias is calculated according to the IEEE 754 Standard: Bias: 2^(NoOfExpBits-1)-1. So with 8bit exponent a bias of 127, or with 5bit a bias of 15.
It would be nice to set a custom bias akin similar to what mentioned in https://en.wikipedia.org/wiki/Minifloat , a bias of 2.
Otherwise this is all great and useful for education.
The text was updated successfully, but these errors were encountered: