Everything Numbers

Types of Numbers

• The Standard Roster
  • int
  • long
  • long long
  • usigned (number type)
  • float
  • double
  • long double
• Extra Guys from stdint.h
  • uint(2/8/16/32/64)_t
  • int(2/8/16/32/64)_t

Annoyances of Floating Point Numbers

Precision

Epsilon ($\epsilon$)

Since there is an underlying precision issue in floating point numbers, people set a $\epsilon$ as a boundary of sorts. If $x<\epsilon$ then $x$ is essentially “0”

Arithmetic

Those in math.h

| Function | Description | | ———– | ————————— | | pow(x, y) | $x^y$, returns double | | sin(x) | $\sin x$ | | cos(x) | $\cos x$ | | tan(x) | $\tan x$ | | sqrt(x) | $\sqrt{n},n>=0$ | | fabs(x) | $|x|$ | | exp(x) | $e^x$ | | log(x) | $\log_e x$ | | log10(x) | $\log_{10} x$ | | log2(x) | $\log_2 x$ |

Those in stdlib.h

| Function | Description | | ——– | ————————— | | rand() | $x^y$, returns double | | sin(x) | $\sin x$ | | | |

frand() for a random number between 0 and 1 srand() (unsigned) seed for rand()

#C #memory