Pick all that apply

• 1Choice 1.

  Double is IEEE 754 binary floating point, so a decimal fraction like 0.1 that can't be expressed with powers of two is stored as an approximation

• 2Choice 2.

  Switching from Float to Double makes this problem go away

• 3Choice 3.

  For money, the standard practice is to compute in Decimal (decimal floating point) or in integers of the smallest unit (won, cents)

• 4Choice 4.

  Compare Doubles with a tolerance like abs(a - b) < epsilon, but epsilon should scale with the magnitude of the values

• 5Choice 5.

  Decimal is always faster than Double, so it should be the default for all numeric work