This is a listing of the errors found in the Third Edition of the OpenGL Programming Guide (Woo, Neider, Davis, Shreiner: ISBN 0-201-60458-2, published in 1999) and its source code examples.
NOTE: Silicon Graphics has moved the location of the source code examples. They are no longer on sgigate.sgi.com. Instead, the source code files have been moved to ftp://ftp.sgi.com/opengl
This is a link to the list of bugs in the Second Edition (Woo, Neider, Davis: ISBN 0-201-46138-2, published in 1997).
There is no comprehensive list of bugs in the First Edition, published in 1992.
If you have either the Second or First Edition and find problems, please check to see if the problems are repeated in the Third Edition before reporting them. If you find unreported bugs in the Third Edition, please submit them by e-mail.
stencil.c: add glFlush()
at the end of display() (Reported by Alan Hanson)
page 24, Figure 1-3
shows five pictures of squares, rotating clockwise. To match code
example 1-3, they should be rotating counter-clockwise. (Reported by Roman
Baeriswyl)
page 25, double.c:
The downloaded version of double.c works fine, but the book doesn't match
it. The switch statement in the mouse function should have three cases:
GLUT_LEFT_BUTTON, GLUT_MIDDLE_BUTTON, and GLUT_RIGHT_BUTTON. The example
on page 25 misses the GLUT_RIGHT_BUTTON case.
page 36, Example 2-1
typo: remove "Special characters" from first line of code.
(Reported by Jichun Zhu)
page 36, 2nd paragraph,
1st sentence: remove the word "to" from "The internals of GLUT will pass
this to function two arguments..." (Reported by Scott Dahl)
page 54/55: at the
beginning of the display() function, add glLineWidth(1.0f) so that redraw
will restore the line width. To get code example 2-5 to match figure 2-9,
add glLineWidth(1.0) to the very last code on page 55. (Reported by Kevin
Foad)
page 63: in the sentence
above Figure 2-12, it says: "In Figure 2-12, the dashed lines represent
added edges." However, there are no dashed lines in Figure 2-12.
The new image we replaced in the 3rd edition lacks the dotted lines that
should be in the interior of the polygon. (Reported by several people)
page 89, Figure 2-17:
The model on the left isn´t an icosahedron, because it has too much
faces. Only ten faces should be seen, and only five triangles should be
connected to each vertex. (Reported by Christian Vetter)
page 114,
Figure 3-8: The triangle in the figure is not centered at the origin, but
in the source code of the programming example, the triangle's center is
almost at the origin. This strongly affects the rotation example,
causing a mismatch between the figure and the code. Also, the amount
of rotation in the code and diagram do not match. The orientation
of the original and the rotated triangle in the figure just look like the
ones that are being shown by the compiled program. To get the results the
figure shows, the triangle must be rotated about 30 degrees clockwise about
a ray parallel to the z-axis and that goes through the lower right vertex
of the triangle. But in the program and also in the code given in the book
the triangle is rotated +90 degrees, if the same projection like in the
figure is chosen, counterclockwise. (Reported by Manuel Hewitt and David
Holland)
page 119,
3rd line from the bottom, typo: the line of code should be gluLookAt(...)
instead
of gluLookat(...). (Reported by Mike Soukup)
page 183: In the 1st paragraph
in the section entitled 'Select a Lighting Model,' the second sentence
is: "In Example 5-1, the only element of the lighting model that's defined
explicitly is the global ambient light." However, the global ambient light
is not defined at all in Example 5-1. (Reported by Jan Vlug)
page 196: In Example 5-7,
the first line in the display callback reads:
glClear( GL_COLOR_BUFFER_MASK | GL_DEPTH_BUFFER_MASK );
"MASK" should be "BIT"; that is:
glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT );
(Reported by Michael Anderson)
page 210, the last
sentence reads, "Remove all the glMaterialfv() calls, and use the more
efficient glColorMaterial() calls to achieve the same lightning."
It should read, "Remove most of the glMaterialfv() calls...", because it
is not possible to change GL_SHININESS with glColorMaterial(). (Reported
by Roman Baeriswyl)
page 220: The
sentence that starts "None of these techniques is hard to use-- in fact..."
is wrong. It should say "are" instead of "is" (Reported by John Alway)
chapter 7; pages 257,
264, and 274: code examples 7-1, 7-2, and 7-5, should not include
the header files <GL/gl.h> and <GL/glu.h>. At best,
inclusion of these files is redundant, because <GL/glut.h>
also includes them. On MS Windows, <windows.h> must be
included before either gl.h or glu.h, to ensure the proper
definition of internal macros. See pages 15 and 16 for more detail about
the proper inclusion of header files.
page 270: In Example
7-3, each call to glVertex3f should be changed to call glVertex3fv
(Reported by Jose Antonio Robles)
page 309: The
second illustration from top, in the box, reads: "Short (byte 0) | Short
(byte 1) ---> Short (byte 0) | Short (byte 0)"
It should read: "Short (byte 0) | Short (byte 1) ---> Short (byte 1)
| Short (byte 0)" (Reported by John Alway)
page 332: In Figure 8-15:
the formula for calulating P'(i,j) has a typo. In the formula, P(m+1, n+j)
should have been P(m+i, n+j). (Reported by Sung Kang)
page 338: The next to
last paragraph reads: "The convolutions of pixels at the left and
top edges of an image..." It should read, "The convolution of pixels
at the edges of an image..." because the left and top edges are not handled
differently to the right and bottom edges. (Reported by Dave Baldwin)
page 339: The 3rd paragraph
reads: "GL_REPLICATE_BORDER computes the convolution in the same way as
in GL_CONSTANT_BORDER mode, except that the last row of column of pixels
is used for the pixels that lie outside..."
It should read, "GL_REPLICATE_BORDER computes the convolution in the
same way as in GL_CONSTANT_BORDER mode, except for the treatment of pixel
locations where the convolution filter extends beyond the edge of the source
image. For these locations, it is as if the outermost one-pixel border
of the source image was replicated." (Reported by Dave Baldwin)
page 421: section entitled
"The q-Coordinate"
Should read "See Plate 28 for an example..." not "See Plate 27 for
an example..."
page 452: In the 2nd
full paragraph, it should read "using 1.0/16.0, 2.0/16.0,..." instead of
"using 16.0/1.0, 16.0/2.0,..." (Reported by David Bourguignon)
page 453: Example 10-2
is not directly in the file accpersp.c. It belongs to the jitter.h
file, which is included in accpersp.c (Reported by David Bourguignon)
page 461, in dof.c
there is a missing declaration:
page 486: the general
quadratic equation has the term a5yx. It should be a5yz.
(Reported by Tsai Pi Hsi and Gerry Murphy)
page 529: In Example
12.8 (trim.c):
page 658: the function
page 665: the resulting
cross product (near the bottom of the page)
page 674: the orthographic
projection matrix R is wrong. Each element in the far right column (matrix
elements with index number 12 through 15) of the matrix R should have a
minus sign in front of them. (Reported by Sung Kang)
Color Plate 28:
Credits should say "Image courtesy of Stefan Brabec and Wolfgang Heidrich,
Max-Planck-Institute for Computer Science, Saarbrücken, Germany."
Color Plate 30 typo:
"virtual" is misspelled. (Reported by Stephane Raimbault)
To
the www.opengl.org web pageLast Modified: September 1, 2002