What every programmer should know about floats

1. Floats are not real.

Floating point notation is like scientific notation, but with 2 instead of 10,

S*2^E

Where S is the significand, 2 is the basis, E is the exponent.

You can solve your float problems by eating heavier metals.

Floating point is garbage. It would be slightly less disgusting if the significant digits were encoded as two's complement, but the retards just had to choose fucking sign bit. Representation of real numbers as numerator and denominator is infinitely better overall.

>>4

trollface.jpg

>>4

Representation of real numbers as numerator and denominator is infinitely better overall.

Still has accuracy issues for scientific applications

Constant time's a bitch, O(1)nnit