diff --git a/Nuclex.Support (x86).csproj b/Nuclex.Support (x86).csproj
index dcc9a8d..0a0e8ab 100644
--- a/Nuclex.Support (x86).csproj	
+++ b/Nuclex.Support (x86).csproj	
@@ -108,6 +108,10 @@
     <Compile Include="Source\Collections\TransformingReadOnlyCollection.Interfaces.cs">
       <DependentUpon>TransformingReadOnlyCollection.cs</DependentUpon>
     </Compile>
+    <Compile Include="Source\FloatHelper.cs" />
+    <Compile Include="Source\FloatHelper.Test.cs">
+      <DependentUpon>FloatHelper.cs</DependentUpon>
+    </Compile>
     <Compile Include="Source\Licensing\LicenseKey.cs" />
     <Compile Include="Source\Licensing\LicenseKey.Test.cs">
       <DependentUpon>LicenseKey.cs</DependentUpon>
diff --git a/Source/FloatHelper.Test.cs b/Source/FloatHelper.Test.cs
new file mode 100644
index 0000000..d84e752
--- /dev/null
+++ b/Source/FloatHelper.Test.cs
@@ -0,0 +1,66 @@
+using System;
+using System.Collections.Generic;
+
+#if UNITTEST
+
+using NUnit.Framework;
+
+namespace Nuclex.Support {
+
+  /// <summary>Unit Test for the FloatHelper class</summary>
+  [TestFixture]
+  public class FloatHelperTest {
+
+    /// <summary>Tests the floating point value comparison helper</summary>
+    [Test]
+    public void TestFloatComparison() {
+      Assert.IsTrue(
+        FloatHelper.AreAlmostEqual(0.00000001f, 0.0000000100000008f, 1),
+        "Minimal difference between very small floating point numbers is considered equal"
+      );
+      Assert.IsFalse(
+        FloatHelper.AreAlmostEqual(0.00000001f, 0.0000000100000017f, 1),
+        "Larger difference between very small floating point numbers is not considered equal"
+      );
+
+      Assert.IsTrue(
+        FloatHelper.AreAlmostEqual(1000000.00f, 1000000.06f, 1),
+        "Minimal difference between very large floating point numbers is considered equal"
+      );
+      Assert.IsFalse(
+        FloatHelper.AreAlmostEqual(1000000.00f, 1000000.13f, 1),
+        "Larger difference between very large floating point numbers is not considered equal"
+      );
+    }
+
+    /// <summary>Tests the double precision floating point value comparison helper</summary>
+    [Test]
+    public void TestDoubleComparison() {
+      Assert.IsTrue(
+        FloatHelper.AreAlmostEqual(0.00000001, 0.000000010000000000000002, 1),
+        "Minimal difference between very small double precision floating point " +
+        "numbers is considered equal"
+      );
+      Assert.IsFalse(
+        FloatHelper.AreAlmostEqual(0.00000001, 0.000000010000000000000004, 1),
+        "Larger difference between very small double precision floating point " +
+        "numbers is not considered equal"
+      );
+
+      Assert.IsTrue(
+        FloatHelper.AreAlmostEqual(1000000.00, 1000000.0000000001, 1),
+        "Minimal difference between very large double precision floating point " +
+        "numbers is considered equal"
+      );
+      Assert.IsFalse(
+        FloatHelper.AreAlmostEqual(1000000.00, 1000000.0000000002, 1),
+        "Larger difference between very large double precision floating point " +
+        "numbers is not considered equal"
+      );
+    }
+
+  }
+
+} // namespace Nuclex.Support
+
+#endif // UNITTEST
\ No newline at end of file
diff --git a/Source/FloatHelper.cs b/Source/FloatHelper.cs
new file mode 100644
index 0000000..412ac98
--- /dev/null
+++ b/Source/FloatHelper.cs
@@ -0,0 +1,148 @@
+using System;
+using System.Collections.Generic;
+using System.Runtime.InteropServices;
+
+namespace Nuclex.Support {
+
+  /// <summary>Helper routines for working with floating point numbers</summary>
+  /// <remarks>
+  ///   The floating point comparison code is based on this excellent article:
+  ///   http://www.cygnus-software.com/papers/comparingfloats/comparingfloats.htm
+  /// </remarks>
+  public static class FloatHelper {
+
+    #region struct FloatIntUnion
+
+    /// <summary>Union of a floating point variable and an integer</summary>
+    [StructLayout(LayoutKind.Explicit)]
+    private struct FloatIntUnion {
+
+      /// <summary>The union's value as a floating point variable</summary>
+      [FieldOffset(0)]
+      public float Float;
+
+      /// <summary>The union's value as an integer</summary>
+      [FieldOffset(0)]
+      public int Int;
+
+      /// <summary>The union's value as an unsigned integer</summary>
+      [FieldOffset(0)]
+      public uint UInt;
+
+    }
+
+    #endregion // struct FloatIntUnion
+
+    #region struct DoubleLongUnion
+
+    /// <summary>Union of a double precision floating point variable and a long</summary>
+    [StructLayout(LayoutKind.Explicit)]
+    private struct DoubleLongUnion {
+
+      /// <summary>The union's value as a double precision floating point variable</summary>
+      [FieldOffset(0)]
+      public double Double;
+
+      /// <summary>The union's value as a long</summary>
+      [FieldOffset(0)]
+      public long Long;
+
+      /// <summary>The union's value as an unsigned long</summary>
+      [FieldOffset(0)]
+      public ulong ULong;
+
+    }
+
+    #endregion // struct DoubleLongUnion
+
+    /// <summary>Compares two floating point values for equality</summary>
+    /// <param name="left">First floating point value to be compared</param>
+    /// <param name="right">Second floating point value t be compared</param>
+    /// <param name="maxUlps">
+    ///   Maximum number of representable floating point values that are allowed to
+    ///   be between the left and the right floating point values
+    /// </param>
+    /// <returns>True if both numbers are equal or close to being equal</returns>
+    /// <remarks>
+    ///   <para>
+    ///     Floating point values can only represent a limited series of natural numbers.
+    ///     For example, the values 2.00000000 and 2.00000024 can be stored in a float,
+    ///     but nothing inbetween them.
+    ///   </para>
+    ///   <para>
+    ///     This comparison will count how many possible floating point values are between
+    ///     the left and the right number. If the number of possible values between both
+    ///     numbers is less than or equal to maxUlps, then the numbers are considered as
+    ///     being equal.
+    ///   </para>
+    ///   <para>
+    ///     Implementation partially follows the code outlined here:
+    ///     http://www.anttirt.net/2007/08/19/proper-floating-point-comparisons/
+    ///   </para>
+    /// </remarks>
+    public static bool AreAlmostEqual(float left, float right, int maxUlps) {
+      FloatIntUnion leftUnion = new FloatIntUnion();
+      FloatIntUnion rightUnion = new FloatIntUnion();
+
+      leftUnion.Float = left;
+      rightUnion.Float = right;
+
+      uint leftSignMask = (leftUnion.UInt >> 31);
+      uint rightSignMask = (rightUnion.UInt >> 31);
+
+      uint leftTemp = ((0x80000000 - leftUnion.UInt) & leftSignMask);
+      leftUnion.UInt = leftTemp | (leftUnion.UInt & ~leftSignMask);
+
+      uint rightTemp = ((0x80000000 - rightUnion.UInt) & rightSignMask);
+      rightUnion.UInt = rightTemp | (rightUnion.UInt & ~rightSignMask);
+
+      return (Math.Abs(leftUnion.Int - rightUnion.Int) <= maxUlps);
+    }
+
+    /// <summary>Compares two double precision floating point values for equality</summary>
+    /// <param name="left">First double precision floating point value to be compared</param>
+    /// <param name="right">Second double precision floating point value t be compared</param>
+    /// <param name="maxUlps">
+    ///   Maximum number of representable double precision floating point values that are
+    ///   allowed to be between the left and the right double precision floating point values
+    /// </param>
+    /// <returns>True if both numbers are equal or close to being equal</returns>
+    /// <remarks>
+    ///   <para>
+    ///     Double precision floating point values can only represent a limited series of
+    ///     natural numbers. For example, the values 2.0000000000000000 and 2.0000000000000004
+    ///     can be stored in a double, but nothing inbetween them.
+    ///   </para>
+    ///   <para>
+    ///     This comparison will count how many possible double precision floating point
+    ///     values are between the left and the right number. If the number of possible
+    ///     values between both numbers is less than or equal to maxUlps, then the numbers
+    ///     are considered as being equal.
+    ///   </para>
+    ///   <para>
+    ///     Implementation partially follows the code outlined here:
+    ///     http://www.anttirt.net/2007/08/19/proper-floating-point-comparisons/
+    ///   </para>
+    /// </remarks>
+    public static bool AreAlmostEqual(double left, double right, long maxUlps) {
+      DoubleLongUnion leftUnion = new DoubleLongUnion();
+      DoubleLongUnion rightUnion = new DoubleLongUnion();
+
+      leftUnion.Double = left;
+      rightUnion.Double = right;
+
+      ulong leftSignMask = (leftUnion.ULong >> 63);
+      ulong rightSignMask = (rightUnion.ULong >> 63);
+
+      ulong leftTemp = ((0x8000000000000000 - leftUnion.ULong) & leftSignMask);
+      leftUnion.ULong = leftTemp | (leftUnion.ULong & ~leftSignMask);
+
+      ulong rightTemp = ((0x8000000000000000 - rightUnion.ULong) & rightSignMask);
+      rightUnion.ULong = rightTemp | (rightUnion.ULong & ~rightSignMask);
+
+      return (Math.Abs(leftUnion.Long - rightUnion.Long) <= maxUlps);
+    }
+
+  }
+
+} // namespace Nuclex.Support