diff --git a/fixedpoint/fixedpoint.py b/fixedpoint/fixedpoint.py
index 6f30c20..23d04d3 100644
--- a/fixedpoint/fixedpoint.py
+++ b/fixedpoint/fixedpoint.py
@@ -676,10 +676,6 @@ def __unsupported(*args: Any, **kwargs: Any) -> NotImplemented:
         """Unsupported FixedPoint method."""
         return NotImplemented
 
-    # For future reference, in case division is needed:
-    # https://courses.cs.washington.edu/courses/cse467/08au/labs/l5/fp.pdf
-    __floordiv__ = __rfloordiv__ = __ifloordiv__ = __unsupported
-    __truediv__ = __rtruediv__ = __itruediv__ = __unsupported
     __matmul__ = __rmatmul__ = __imatmul__ = __unsupported
     __mod__ = __rmod__ = __imod__ = __unsupported
     __rlshift__ = __rrshift__ = __unsupported
@@ -814,6 +810,58 @@ def __imul__(self: FixedPointType, multiplier: Numeric) -> FixedPointType:
         self._bits, self._signed, self._m, self._n = self.__mul(other)
         return self
 
+    def __floordiv(numerator: FixedPointType,
+                   denominator: FixedPointType) -> AttrReturn:
+        """Perform division and return attributes of the result.
+        This is the
+        """
+        m: int = numerator._m + denominator._n + 1
+        n: int = numerator._n + denominator._m
+        signed: bool = bool(numerator._signed or denominator._signed)
+
+        bits = int((numerator._signedint << (denominator._n + denominator._m)) // denominator._signedint)
+        return bits & (2**(m + n) - 1), signed, m, n
+
+    def __floordiv__(self: FixedPointType, other: Numeric) -> FixedPointType:
+        denominator, props = self.__to_FixedPoint_resolved(other)
+        return self.__class__.__new(*self.__floordiv(denominator), **props)
+
+    def __ifloordiv__(self: FixedPointType, denominator: Numeric) -> FixedPointType:
+        other = self.__to_FixedPoint(denominator)
+        self._bits, self._signed, self._m, self._n = self.__floordiv(other)
+        return self
+
+    def __rfloordiv__(self: FixedPointType, numerator: Numeric) -> FixedPointType:
+        other = self.__to_FixedPoint(numerator, self._signed)
+        return self.__class__.__new(*other.__floordiv(self, self.overflow,
+                                                      self._owarn),
+                                    self.overflow, self.rounding, self.str_base,
+                                    self.overflow_alert,
+                                    self.implicit_cast_alert,
+                                    self.mismatch_alert)
+
+    # Both division operators use the maximum precision possible - it's up to the user to truncate
+    # the result how they see fit.
+    __truediv = __floordiv
+
+    def __truediv__(self: FixedPointType, other: Numeric) -> FixedPointType:
+        denominator, props = self.__to_FixedPoint_resolved(other)
+        return self.__class__.__new(*self.__truediv(denominator), **props)
+
+    def __itruediv__(self: FixedPointType, denominator: Numeric) -> FixedPointType:
+        other = self.__to_FixedPoint(denominator)
+        self._bits, self._signed, self._m, self._n = self.__truediv(other)
+        return self
+
+    def __rtruediv__(self: FixedPointType, numerator: Numeric) -> FixedPointType:
+        other = self.__to_FixedPoint(numerator, self._signed)
+        return self.__class__.__new(*other.__truediv(self, self.overflow,
+                                                      self._owarn),
+                                    self.overflow, self.rounding, self.str_base,
+                                    self.overflow_alert,
+                                    self.implicit_cast_alert,
+                                    self.mismatch_alert)
+
     def __pow(self: FixedPointType, exponent: int) -> AttrReturn:
         """Perform exponentiation and return attributes of the result."""
         if not (isinstance(exponent, int) and exponent > 0):
diff --git a/tests/test_operators/test_division.py b/tests/test_operators/test_division.py
new file mode 100644
index 0000000..7105fad
--- /dev/null
+++ b/tests/test_operators/test_division.py
@@ -0,0 +1,37 @@
+#!/usr/bin/env python3
+# This test was not written by the original authors of the fixedpoint library and does not conform
+# to its quality standards.
+
+import pathlib
+import sys
+sys.path.insert(1, str(pathlib.Path('./../..').resolve()));
+import fixedpoint as uut
+
+from random import *
+import math
+
+# fuzz test divide
+for i in range(1000):
+    x = uut.FixedPoint(0, m=int(random() * 100)+1, n=int(random() * 100)+1, signed=1)
+    y = uut.FixedPoint(0, m=int(random() * 100)+1, n=int(random() * 100)+1, signed=1)
+
+    # because we are choosing only numbers that are precisely representable in fixed-point,
+    # no error is introduced at these steps.
+    x._bits = getrandbits(x._m + x._n + int(x._signed))
+    y._bits = getrandbits(y._m + y._n + int(y._signed))
+    if (y._bits == 0): y._bits = 1
+    d = x / y
+
+    xf = float(x)
+    yf = float(y)
+    df = xf / yf
+
+    # TODO: establish actual error bounds
+    MANTISSABITS = 52
+    exp = math.frexp(df)[1]
+    tol = (2**(exp - MANTISSABITS))
+    err = df - float(d)
+    if (math.fabs(err) > tol):
+        print(f"division of {x.qformat} / {y.qformat} has error {err}")
+        print(f"d = {float(d)}, df = {xf / yf}, tol = {tol}")
+        print()