Add functions for converting to and from binary16 half-precision floating point values. Derived from http://mvn.io7m.com/ieee754b16, of which I am the original author.

3 files changed, 871 insertions, 0 deletions

diff --git a/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary16Test.java b/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary16Test.java
new file mode 100644
index 000000000..d9e84eeab
--- /dev/null
+++ b/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary16Test.java
@@ -0,0 +1,685 @@
+/**
+ * Copyright 2013 JogAmp Community. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without modification, are
+ * permitted provided that the following conditions are met:
+ *
+ *    1. Redistributions of source code must retain the above copyright notice, this list of
+ *       conditions and the following disclaimer.
+ *
+ *    2. Redistributions in binary form must reproduce the above copyright notice, this list
+ *       of conditions and the following disclaimer in the documentation and/or other materials
+ *       provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY JogAmp Community ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
+ * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL JogAmp Community OR
+ * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+ * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+ * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * The views and conclusions contained in the software and documentation are those of the
+ * authors and should not be interpreted as representing official policies, either expressed
+ * or implied, of JogAmp Community.
+ */
+
+package com.jogamp.opengl.test.junit.jogl.math;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+import com.jogamp.opengl.math.Binary16;
+
+public final class Binary16Test
+{
+  /**
+   * Exponents in the range [-15, 16] are encoded and decoded correctly.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testExponentIdentity()
+  {
+    System.out.println("-- Exponent identities");
+    for (int e = -15; e <= 16; ++e) {
+      final char p = Binary16.packSetExponentUnbiasedUnchecked(e);
+      final int u = Binary16.unpackGetExponentUnbiased(p);
+      System.out.println("e: " + e);
+      System.out.println("p: " + Integer.toHexString(p));
+      System.out.println("u: " + u);
+      Assert.assertEquals(e, u);
+    }
+  }
+
+  /**
+   * Infinities are infinite.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testInfinite()
+  {
+    Assert.assertTrue(Binary16.isInfinite(Binary16.POSITIVE_INFINITY));
+    Assert.assertTrue(Binary16.isInfinite(Binary16.NEGATIVE_INFINITY));
+    Assert.assertFalse(Binary16.isInfinite(Binary16.exampleNaN()));
+
+    for (int i = 0; i <= 65535; ++i) {
+      Assert.assertFalse(Binary16.isInfinite(Binary16.packDouble(i)));
+    }
+  }
+
+  /**
+   * The unencoded exponent of infinity is 16.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testInfinityExponent()
+  {
+    Assert.assertEquals(
+      16,
+      Binary16.unpackGetExponentUnbiased(Binary16.POSITIVE_INFINITY));
+  }
+
+  /**
+   * The unencoded exponent of infinity is 16.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeExponent()
+  {
+    Assert.assertEquals(
+      16,
+      Binary16.unpackGetExponentUnbiased(Binary16.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * The sign of negative infinity is 1.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeSign()
+  {
+    Assert
+      .assertEquals(1, Binary16.unpackGetSign(Binary16.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * The significand of infinity is 0.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeSignificand()
+  {
+    Assert.assertEquals(
+      0,
+      Binary16.unpackGetSignificand(Binary16.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * The sign of positive infinity is 0.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testInfinitySign()
+  {
+    Assert
+      .assertEquals(0, Binary16.unpackGetSign(Binary16.POSITIVE_INFINITY));
+  }
+
+  /**
+   * The significand of infinity is 0.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinitySignificand()
+  {
+    Assert.assertEquals(
+      0,
+      Binary16.unpackGetSignificand(Binary16.POSITIVE_INFINITY));
+  }
+
+  /**
+   * NaN is NaN.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testNaN()
+  {
+    final int n =
+      Binary16.packSetExponentUnbiasedUnchecked(16)
+        | Binary16.packSetSignificandUnchecked(1);
+    final char c = (char) n;
+    Assert.assertEquals(16, Binary16.unpackGetExponentUnbiased(c));
+    Assert.assertEquals(1, Binary16.unpackGetSignificand(c));
+    Assert.assertEquals(
+      16,
+      Binary16.unpackGetExponentUnbiased(Binary16.exampleNaN()));
+    Assert.assertEquals(
+      1,
+      Binary16.unpackGetSignificand(Binary16.exampleNaN()));
+    Assert.assertTrue(Binary16.isNaN(c));
+    Assert.assertTrue(Binary16.isNaN(Binary16.exampleNaN()));
+  }
+
+  /**
+   * Packing NaN results in NaN.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testPackDoubleNaN()
+  {
+    final double k = Double.NaN;
+    final char r = Binary16.packDouble(k);
+    Assert.assertTrue(Binary16.isNaN(r));
+  }
+
+  /**
+   * Packing negative infinity results in negative infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackDoubleNegativeInfinity()
+  {
+    Assert.assertTrue(Binary16.NEGATIVE_INFINITY == Binary16
+      .packDouble(Double.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * Packing negative zero results in negative zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackDoubleNegativeZero()
+  {
+    Assert.assertTrue(Binary16.NEGATIVE_ZERO == Binary16.packDouble(-0.0));
+  }
+
+  /**
+   * Packing positive infinity results in positive infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackDoublePositiveInfinity()
+  {
+    Assert.assertTrue(Binary16.POSITIVE_INFINITY == Binary16
+      .packDouble(Double.POSITIVE_INFINITY));
+  }
+
+  /**
+   * Packing positive zero results in positive zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackDoublePositiveZero()
+  {
+    Assert.assertTrue(Binary16.POSITIVE_ZERO == Binary16.packDouble(0.0));
+  }
+
+  /**
+   * Integers in the range [0, 65520] should be representable.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testPackDoubleUnpackFloat()
+  {
+    for (int i = 0; i <= 65536; ++i) {
+      final double in = i;
+      final char packed = Binary16.packDouble(in);
+      final float r = Binary16.unpackFloat(packed);
+      System.out.println(String.format(
+        "packed: 0x%04x 0b%s in: %f unpacked: %f",
+        (int) packed,
+        Binary16.toRawBinaryString(packed),
+        in,
+        r));
+
+      if (i <= 2048) {
+        Assert.assertEquals(in, r, 0.0);
+      }
+      if ((i > 2048) && (i <= 4096)) {
+        Assert.assertTrue((r % 2) == 0);
+      }
+      if ((i > 4096) && (i <= 8192)) {
+        Assert.assertTrue((r % 4) == 0);
+      }
+      if ((i > 8192) && (i <= 16384)) {
+        Assert.assertTrue((r % 8) == 0);
+      }
+      if ((i > 16384) && (i <= 32768)) {
+        Assert.assertTrue((r % 16) == 0);
+      }
+      if ((i > 32768) && (i < 65536)) {
+        Assert.assertTrue((r % 32) == 0);
+      }
+      if (i == 65536) {
+        Assert.assertTrue(Double.isInfinite(r));
+      }
+    }
+  }
+
+  /**
+   * Integers in the range [0, 65520] should be representable.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testPackFloatDoubleEquivalent()
+  {
+    for (int i = 0; i <= 65536; ++i) {
+      final float f_in = i;
+      final double d_in = i;
+      final char pf = Binary16.packFloat(f_in);
+      final char pd = Binary16.packDouble(d_in);
+
+      System.out.println("i: " + i);
+      System.out.println(String.format(
+        "pack_f: 0x%04x 0b%s",
+        (int) pf,
+        Binary16.toRawBinaryString(pf)));
+      System.out.println(String.format(
+        "pack_d: 0x%04x 0b%s",
+        (int) pd,
+        Binary16.toRawBinaryString(pd)));
+
+      Assert.assertEquals(pf, pd);
+    }
+  }
+
+  /**
+   * Packing NaN results in NaN.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testPackFloatNaN()
+  {
+    final float k = Float.NaN;
+    final char r = Binary16.packFloat(k);
+    Assert.assertTrue(Binary16.isNaN(r));
+  }
+
+  /**
+   * Packing negative infinity results in negative infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackFloatNegativeInfinity()
+  {
+    Assert.assertTrue(Binary16.NEGATIVE_INFINITY == Binary16
+      .packFloat(Float.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * Packing negative zero results in negative zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackFloatNegativeZero()
+  {
+    Assert.assertTrue(Binary16.NEGATIVE_ZERO == Binary16.packFloat(-0.0f));
+  }
+
+  /**
+   * Packing positive infinity results in positive infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackFloatPositiveInfinity()
+  {
+    Assert.assertTrue(Binary16.POSITIVE_INFINITY == Binary16
+      .packFloat(Float.POSITIVE_INFINITY));
+  }
+
+  /**
+   * Packing positive zero results in positive zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testPackFloatPositiveZero()
+  {
+    Assert.assertTrue(Binary16.POSITIVE_ZERO == Binary16.packFloat(0.0f));
+  }
+
+  /**
+   * Integers in the range [0, 65520] should be representable.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testPackFloatUnpackDouble()
+  {
+    for (int i = 0; i <= 65536; ++i) {
+      final float in = i;
+      final char packed = Binary16.packFloat(in);
+      final double r = Binary16.unpackDouble(packed);
+      System.out.println(String.format(
+        "packed: 0x%04x 0b%s in: %f unpacked: %f",
+        (int) packed,
+        Binary16.toRawBinaryString(packed),
+        in,
+        r));
+
+      if (i <= 2048) {
+        Assert.assertEquals(in, r, 0.0);
+      }
+      if ((i > 2048) && (i <= 4096)) {
+        Assert.assertTrue((r % 2) == 0);
+      }
+      if ((i > 4096) && (i <= 8192)) {
+        Assert.assertTrue((r % 4) == 0);
+      }
+      if ((i > 8192) && (i <= 16384)) {
+        Assert.assertTrue((r % 8) == 0);
+      }
+      if ((i > 16384) && (i <= 32768)) {
+        Assert.assertTrue((r % 16) == 0);
+      }
+      if ((i > 32768) && (i < 65536)) {
+        Assert.assertTrue((r % 32) == 0);
+      }
+      if (i == 65536) {
+        Assert.assertTrue(Double.isInfinite(r));
+      }
+    }
+  }
+
+  /**
+   * Integers in the range [0, 65520] should be representable.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testPackUnpackDouble()
+  {
+    for (int i = 0; i <= 65536; ++i) {
+      final double in = i;
+      final char packed = Binary16.packDouble(in);
+      final double r = Binary16.unpackDouble(packed);
+      System.out.println(String.format(
+        "packed: 0x%04x 0b%s in: %f unpacked: %f",
+        (int) packed,
+        Binary16.toRawBinaryString(packed),
+        in,
+        r));
+
+      if (i <= 2048) {
+        Assert.assertEquals(in, r, 0.0);
+      }
+      if ((i > 2048) && (i <= 4096)) {
+        Assert.assertTrue((r % 2) == 0);
+      }
+      if ((i > 4096) && (i <= 8192)) {
+        Assert.assertTrue((r % 4) == 0);
+      }
+      if ((i > 8192) && (i <= 16384)) {
+        Assert.assertTrue((r % 8) == 0);
+      }
+      if ((i > 16384) && (i <= 32768)) {
+        Assert.assertTrue((r % 16) == 0);
+      }
+      if ((i > 32768) && (i < 65536)) {
+        Assert.assertTrue((r % 32) == 0);
+      }
+      if (i == 65536) {
+        Assert.assertTrue(Double.isInfinite(r));
+      }
+    }
+  }
+
+  /**
+   * Integers in the range [0, 65520] should be representable.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testPackUnpackFloat()
+  {
+    for (int i = 0; i <= 65536; ++i) {
+      final float in = i;
+      final char packed = Binary16.packFloat(in);
+      final float r = Binary16.unpackFloat(packed);
+      System.out.println(String.format(
+        "packed: 0x%04x 0b%s in: %f unpacked: %f",
+        (int) packed,
+        Binary16.toRawBinaryString(packed),
+        in,
+        r));
+      if (i <= 2048) {
+        Assert.assertEquals(in, r, 0.0);
+      }
+      if ((i > 2048) && (i <= 4096)) {
+        Assert.assertTrue((r % 2) == 0);
+      }
+      if ((i > 4096) && (i <= 8192)) {
+        Assert.assertTrue((r % 4) == 0);
+      }
+      if ((i > 8192) && (i <= 16384)) {
+        Assert.assertTrue((r % 8) == 0);
+      }
+      if ((i > 16384) && (i <= 32768)) {
+        Assert.assertTrue((r % 16) == 0);
+      }
+      if ((i > 32768) && (i < 65536)) {
+        Assert.assertTrue((r % 32) == 0);
+      }
+      if (i == 65536) {
+        Assert.assertTrue(Float.isInfinite(r));
+      }
+    }
+  }
+
+  /**
+   * Signs in the range [0, 1] are encoded and decoded correctly.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testSignIdentity()
+  {
+    System.out.println("-- Sign identities");
+    for (int e = 0; e <= 1; ++e) {
+      final char p = Binary16.packSetSignUnchecked(e);
+      final int u = Binary16.unpackGetSign(p);
+      System.out.println("e: " + e);
+      System.out.println("p: " + Integer.toHexString(p));
+      System.out.println("u: " + u);
+      Assert.assertEquals(e, u);
+    }
+  }
+
+  /**
+   * Significands in the range [0, 1023] are encoded and decoded correctly.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testSignificandIdentity()
+  {
+    System.out.println("-- Significand identities");
+    for (int e = 0; e <= 1023; ++e) {
+      final char p = Binary16.packSetSignificandUnchecked(e);
+      final int u = Binary16.unpackGetSignificand(p);
+      System.out.println("e: " + e);
+      System.out.println("p: " + Integer.toHexString(p));
+      System.out.println("u: " + u);
+      Assert.assertEquals(e, u);
+    }
+  }
+
+  /**
+   * Unpacking NaN results in NaN.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testUnpackDoubleNaN()
+  {
+    final double k = Binary16.unpackDouble(Binary16.exampleNaN());
+    Assert.assertTrue(Double.isNaN(k));
+  }
+
+  /**
+   * Unpacking negative infinity results in negative infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testUnpackDoubleNegativeInfinity()
+  {
+    Assert.assertTrue(Double.NEGATIVE_INFINITY == Binary16
+      .unpackDouble(Binary16.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * Unpacking negative zero results in negative zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testUnpackDoubleNegativeZero()
+  {
+    Assert.assertTrue(-0.0 == Binary16.unpackDouble(Binary16.NEGATIVE_ZERO));
+  }
+
+  /**
+   * Unpacking 1.0 results in 1.0.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testUnpackDoubleOne()
+  {
+    final char one = 0x3C00;
+    final double r = Binary16.unpackDouble(one);
+    System.out.println(String.format("0x%04x -> %f", (int) one, r));
+    Assert.assertEquals(r, 1.0, 0.0);
+  }
+
+  /**
+   * Unpacking -1.0 results in -1.0.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testUnpackDoubleOneNegative()
+  {
+    final char one = 0xBC00;
+    final double r = Binary16.unpackDouble(one);
+    System.out.println(String.format("0x%04x -> %f", (int) one, r));
+    Assert.assertEquals(r, -1.0, 0.0);
+  }
+
+  /**
+   * Unpacking positive infinity results in positive infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testUnpackDoublePositiveInfinity()
+  {
+    Assert.assertTrue(Double.POSITIVE_INFINITY == Binary16
+      .unpackDouble(Binary16.POSITIVE_INFINITY));
+  }
+
+  /**
+   * Unpacking positive zero results in positive zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testUnpackDoublePositiveZero()
+  {
+    Assert.assertTrue(0.0 == Binary16.unpackDouble(Binary16.POSITIVE_ZERO));
+  }
+
+  /**
+   * Unpacking 2.0 results in 2.0.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testUnpackDoubleTwo()
+  {
+    final char one = 0x4000;
+    final double r = Binary16.unpackDouble(one);
+    System.out.println(String.format("%04x -> %f", (int) one, r));
+    Assert.assertEquals(r, 2.0, 0.0);
+  }
+
+  /**
+   * Unpacking -2.0 results in -2.0.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testUnpackDoubleTwoNegative()
+  {
+    final char one = 0xC000;
+    final double r = Binary16.unpackDouble(one);
+    System.out.println(String.format("%04x -> %f", (int) one, r));
+    Assert.assertEquals(r, -2.0, 0.0);
+  }
+
+  /**
+   * Unpacking NaN results in NaN.
+   */
+
+  @SuppressWarnings("static-method") @Test public void testUnpackFloatNaN()
+  {
+    final float k = Binary16.unpackFloat(Binary16.exampleNaN());
+    Assert.assertTrue(Float.isNaN(k));
+  }
+
+  /**
+   * Unpacking negative infinity results in negative infinity.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testUnpackFloatNegativeInfinity()
+  {
+    Assert.assertTrue(Float.NEGATIVE_INFINITY == Binary16
+      .unpackFloat(Binary16.NEGATIVE_INFINITY));
+  }
+
+  /**
+   * Unpacking negative zero results in negative zero.
+   */
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testUnpackFloatNegativeZero()
+  {
+    Assert.assertTrue(-0.0 == Binary16.unpackFloat(Binary16.NEGATIVE_ZERO));
+  }
+
+  /**
+   * Unpacking 1.0 results in 1.0.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testUnpackFloatOne()
+  {
+    final char one = 0x3C00;
+    final float r = Binary16.unpackFloat(one);
+    System.out.println(String.format("0x%04x -> %f", (int) one, r));
+    Assert.assertEquals(r, 1.0, 0.0);
+  }
+
+  /**
+   * Unpacking -1.0 results in -1.0.
+   */
+
+  @SuppressWarnings({ "static-method", "boxing" }) @Test public
+    void
+    testUnpackFloatOneNegative()
+  {
+    final char one = 0xBC00;
+    final float r = Binary16.unpackFloat(one);
+    System.out.println(String.format("0x%04x -> %f", (int) one, r));
+    Assert.assertEquals(r, -1.0, 0.0);
+  }
+}
diff --git a/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary32Test.java b/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary32Test.java
new file mode 100644
index 000000000..42e7f08a1
--- /dev/null
+++ b/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary32Test.java
@@ -0,0 +1,93 @@
+/**
+ * Copyright 2013 JogAmp Community. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without modification, are
+ * permitted provided that the following conditions are met:
+ *
+ *    1. Redistributions of source code must retain the above copyright notice, this list of
+ *       conditions and the following disclaimer.
+ *
+ *    2. Redistributions in binary form must reproduce the above copyright notice, this list
+ *       of conditions and the following disclaimer in the documentation and/or other materials
+ *       provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY JogAmp Community ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
+ * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL JogAmp Community OR
+ * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+ * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+ * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * The views and conclusions contained in the software and documentation are those of the
+ * authors and should not be interpreted as representing official policies, either expressed
+ * or implied, of JogAmp Community.
+ */
+
+package com.jogamp.opengl.test.junit.jogl.math;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+import com.jogamp.opengl.math.Binary32;
+
+public class Binary32Test
+{
+  @SuppressWarnings("static-method") @Test public void testInfinityExponent()
+  {
+    Assert.assertEquals(
+      128,
+      Binary32.unpackGetExponentUnbiased(Float.POSITIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeExponent()
+  {
+    Assert.assertEquals(
+      128,
+      Binary32.unpackGetExponentUnbiased(Float.NEGATIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeSign()
+  {
+    Assert.assertEquals(1, Binary32.unpackGetSign(Float.NEGATIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeSignificand()
+  {
+    Assert.assertEquals(
+      0,
+      Binary32.unpackGetSignificand(Float.NEGATIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public void testInfinitySign()
+  {
+    Assert.assertEquals(0, Binary32.unpackGetSign(Float.POSITIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinitySignificand()
+  {
+    Assert.assertEquals(
+      0,
+      Binary32.unpackGetSignificand(Float.POSITIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public void testNaNExponent()
+  {
+    Assert.assertEquals(128, Binary32.unpackGetExponentUnbiased(Float.NaN));
+  }
+
+  @SuppressWarnings("static-method") @Test public void testNaNSignificand()
+  {
+    Assert.assertTrue(Binary32.unpackGetSignificand(Float.NaN) > 0);
+  }
+}
diff --git a/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary64Test.java b/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary64Test.java
new file mode 100644
index 000000000..78806985f
--- /dev/null
+++ b/src/test/com/jogamp/opengl/test/junit/jogl/math/Binary64Test.java
@@ -0,0 +1,93 @@
+/**
+ * Copyright 2013 JogAmp Community. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without modification, are
+ * permitted provided that the following conditions are met:
+ *
+ *    1. Redistributions of source code must retain the above copyright notice, this list of
+ *       conditions and the following disclaimer.
+ *
+ *    2. Redistributions in binary form must reproduce the above copyright notice, this list
+ *       of conditions and the following disclaimer in the documentation and/or other materials
+ *       provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY JogAmp Community ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
+ * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL JogAmp Community OR
+ * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+ * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+ * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * The views and conclusions contained in the software and documentation are those of the
+ * authors and should not be interpreted as representing official policies, either expressed
+ * or implied, of JogAmp Community.
+ */
+
+package com.jogamp.opengl.test.junit.jogl.math;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+import com.jogamp.opengl.math.Binary64;
+
+public class Binary64Test
+{
+  @SuppressWarnings("static-method") @Test public void testInfinityExponent()
+  {
+    Assert.assertEquals(
+      1024,
+      Binary64.unpackGetExponentUnbiased(Double.POSITIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeExponent()
+  {
+    Assert.assertEquals(
+      1024,
+      Binary64.unpackGetExponentUnbiased(Double.NEGATIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeSign()
+  {
+    Assert.assertEquals(1, Binary64.unpackGetSign(Double.NEGATIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinityNegativeSignificand()
+  {
+    Assert.assertEquals(
+      0,
+      Binary64.unpackGetSignificand(Double.NEGATIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public void testInfinitySign()
+  {
+    Assert.assertEquals(0, Binary64.unpackGetSign(Double.POSITIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public
+    void
+    testInfinitySignificand()
+  {
+    Assert.assertEquals(
+      0,
+      Binary64.unpackGetSignificand(Double.POSITIVE_INFINITY));
+  }
+
+  @SuppressWarnings("static-method") @Test public void testNaNExponent()
+  {
+    Assert.assertEquals(1024, Binary64.unpackGetExponentUnbiased(Double.NaN));
+  }
+
+  @SuppressWarnings("static-method") @Test public void testNaNSignificand()
+  {
+    Assert.assertTrue(Binary64.unpackGetSignificand(Double.NaN) > 0);
+  }
+}