1Version 1.0.3
2        * Added support for enabling/disabling large stack optimisations, used in
3          halfFunction.h.
4          (Piotr Stanczyk)
5	* Added ImathNoise.(h/cpp) files. Initializes Perlin noise to match the
6	  Renderman implmenetation.
7	  (Pixar Contribution)
8	* Fixed a number of missing includes to comply with stricter
9	  enforcement by gnu compilers.
10	  (Piotr Stanczyk)
11	* Depracated compiler flag: -Wno-long-double since it is no longer
12	  supported under OS X.
13	  (Piotr Stanczyk)
14	* A minor API change to Imath::Frustum has been made:  the functions
15	  'near' and 'far' have been renamed to 'nearPlane' and 'farPlane' due
16	  to conflicts with certain windows headers.  The former alternate
17	  accessor names for these values on windows ('hither' and 'yon')
18	  remain, though should be considered deprecated.
19	  (David Lenihan)
20	* Added SVD, eigenvalue solver, and procrustes fit calculations
21	  to ImathMatrixAlgo.
22	  (Chris Twigg, Ji Hun Yu)
23	* Added Imath::FrustumTest for frustum visibility testing.
24	  (Eric Johnston)
25	* Fixed a stack corruption in the matrix minorOf functions.
26	  (Nick Rasmussen)
27	* Visual studio 2008 project files have been added to the vc/vc9
28	  directory, and several minor visual studio compile fixes have
29	  been applied.
30	  (Nick Rasmussen)
31	* Updated the so verison to 7.
32	  (Piotr Stanczyk)
33	* Depracated the MacCode_Warrior and Shake submodules.
34	  (Piotr Stanczyk)
35
36Version 1.0.2
37	* Added support for targetting builds on 64bit Windows and minimising
38	  number of compiler warnings on Windows. Thanks to Ger Hobbelt for his
39	  contributions to CreateDLL.
40	  (Ji Hun Yu)
41	* Removed a spurious restrict qualifier in the matrix manipulation code
42	  that was causing the 64-bit MS compiler to generate code in release
43	  mode that caused incorrect results.
44	  (Ji Hun Yu)
45	* Added patches for improving universal binaries on OS X. Thanks to
46	  Paul Schneider for the contribution
47	  (Piotr Stanczyk)
48	* Imath::Box optimization: remove loops from methods by partially
49	  specializing the class, for boxes of two and three dimensions.
50	  (Piotr Stanczyk)
51	* Added explicit copy constructors to Imath::Matrix33<T> and
52	  ImathMatrix44<T> to make conversions between float and double
53	  matrices more convenient.
54	  (Florian Kainz)
55	* Added slerpShortestArc() and euclideanInnerProduct() functions
56	  to Imath::Quat<T>.
57	  (Nick Porcino)
58	* Added 4D vector class template Imath::Vec4<T>.
59	  (Nick Porcino)
60	* Copy constructors and assignment operators for Matrix33<T>
61	  and Matrix44<T> are up to 25% faster.  Added matrix constructors
62	  that do not initialize the matrix (this is faster in cases where
63	  the initial value of the matrix is immediately overwritten anyway).
64	  (Nick Porcino)
65	* Rewrote function closestPointOnBox(point,box).  Shortened
66	  the code, improved numerical accuracy, fixed a bug where
67	  closestPointOnBox(box.center(),box) would return the center
68	  of the +Z side of the box, even if the +/-X or +/-Y sides
69	  were closer.
70	  (Florian Kainz)
71	* Rewrote function findEntryAndExitPoints() in ImathBoxAlgo.h.
72	  Results are now consistent with those from intersect(), also
73	  in ImathBoxAlgo.h.
74	  (Florian Kainz)
75	* Made Vec2<T>::length() and Vec3<T>::length() more accurate for
76	  vectors whose length is less than sqrt(limits<T>::smallest());
77	  (Florian Kainz)
78	* Made Quat<T>::angle() more accurate for small angles.
79	  (Don Hatch)
80
81Version 1.0.1:
82	* Removed Windows .suo files from distribution.
83	  (Eric Wimmer)
84
85Version 1.0.0:
86	* Bumped DSO version number to 6.0
87	  (Florian Kainz)
88	* Rounding during float-to-half conversion now implements
89	  "round to nearest even" mode: if the original float value
90	  is exactly in the middle between the two closest half values
91	  then rounding chooses the half value whose least significant
92	  bit is zero.
93	  (Florian Kainz)
94	* Installation Tuning:
95	  - Corrected version number on dso's (libtool) - now 5.0
96	  - Separated ILMBASE_LDFLAGS and ILMBASE_LIBS so that test programs
97	    in configure scripts of packages dependent on IlmBase can link
98	    with static libraries properly
99	  - eliminated some warning messages during install
100	  (Andrew Kunz)
101
102Version 0.9.0:
103	* Initial release of this code as a separate library.
104	  Previously the libraries contained were part of
105	  version 1.4.0 of OpenEXR
106	* New build scripts for Linux/Unix
107	  (Andrew Kunz)
108	* New Windows project files and build scripts
109	  (Kimball Thurston)
110