## Important Errors

Page Reporter Error
53 Ricky Uy, Sebastian Lipponer Second paragraph, third line, change "area is given by ||v_2|| * h" to "area is given by ||v_1|| * h"
54 Randall Hopper Second paragraph, third line, change "and swapping the remaining two" to ", swapping the remaining two, and negating one."
84 zho The displayed equation should be S(1,1,-1) \dpdu \cross S(1,1,-1) \dpdv = S(-1,-1,1) \dpdu \cross \dpdv = S(-1,-1,1) n = \ne S(1,1,-1) n
98 Mark Bolstad In section 3.2.2, line 4, both \theta_{max} should be \phi_{max}
175 mmp First code fragment: missing space between "class" and "InstancePrimitive"
192 Arek Zimny In figure 4.7, the two instances of "NextCrossing" in the figure and the one in the third line of the caption should be "NextCrossingT".
298 mmp Footnote, line 1, "low-discrepancy and best-candidate samplers" should be "best-candidate sampler"
335 mmp Second fragment, delete "/" after "lensSamples"
353 mmp In Figure 7.37, last line of the caption, change "bricks" to "arches".
416 Stephen Chenney The definition of the hemispherical-directional reflectance should not have that 1/pi term.
430-435 Stephen Chenney The \eta_t in Figure 9.7 and the \omega_t in the specular transmission BTDF should be \eta_o and \omega_o, respectively. Similarly, the SpecularTransmission BxDF implementation should be written to use etao, not etat, Figures 9.9 and 9.10 need to be updated to use \theta_o rather than \theta_t, and various text should be changed to use \theta_o rather than \theta_t.
431 David Wolff In the denominator of the fourth displayed equation, |\cos \theta_i| should be |\cos \theta_t|. (Though N.B. the 430-435 changes above.)
433 Thiago Ize In SpecularTransmission::Sample_f(), (ei*ei)/(et*et) should actually be (et*et)/(ei*ei). (Fortunately, in practice these terms all end up cancelling out, so images shouldn't be changed by this fix.)
442 Arek Zimny Figure 9.15 is drawn incorrectly: \omega_o and \omega_i should make the same angle with respect to \omega_h, \theta_h. As drawn, \omega_o makes that angle with the normal, rather than \omega_h.
448 Stephen Chenney The normalized microfacet distribution actually should be: "\frac{\sqrt{(e_x+2)(e_y+2)}{2\pi} (\wh \cdot \n)^ ..."
449 Stephen Chenney The implementation of Anisotropic::D() should be sqrt((ex+2)*(ey+2)) * INV_TWOPI * powf(...)
497 mmp Figure 11.6: Reflection of ray "r" on left should be drawn with dotted line, incident ray "r'" on right should be drawn with dashed line
523 mmp The last line of code in the third code fragment should be "texel(i-1, 2*s+1, 2*t+1));"
565 mmp Exercise 11.2: Version 1.2 of the OpenEXR file format supportes tiled image files and includes a utility to create tiled files from untiled EXR files. This makes this exercise somewhat easier.
577 Arek Zimny There shouldn't be a \frac{1}{4\pi} term in equation 12.2, the phase function normalization constraint.
581 Vincent Pegoraro In the three term equation on the top of this page, the second term after the first equal sign contains an extra "\frac{1}{4 \pi}" that should be removed. Also, a "\sin \theta" is missing in the third term before the "d \theta".
637 mmp F(x) in the second line from the top should be f(x).
640 mmp Footnote, x ≤ should be x < 0
658 mmp Figure 14.12: Coordinates next to filled dot in figure should be (u, 1-u)
671 Shaohua Fan Change first displayed match to $\beta = E[F] - \int f(x) dx$.
672 Shaohua Fan Third line, O(1/N^-1) should be O(N^-1)
676 "fsf" The multiple importance sampling equation at the bottom of the page is incorrect. It should be (1/n_f) \sum_{i=1}^{n_f} ... + (1/n_g) \sum_{j=1}^{n_g} ...
684 Stephen Chenney "exponent+1.f" in the "Compute PDF from \omega_i from Blinn distribution" fragment should be "exponent+2.f". Also, add a sentence like "Using the definition of the distribution over the hemisphere from Equation 9.9, we have:" before that fragment to make clear where the 2 pi comes from.
686 Stephen Chenney Given the aforementioned Anisotropic changes on 448 and 449, the 2.f * M_PI term in the computation of anisotropic_pdf should be deleted.
689 Stephen Chenney The definition of the hemispherical-directional reflectance should not have that 1/pi term. Also, the M_PI term in the definition of BxDF::rho() should be removed, and the \pi in the equation for the estimator at the very bottom of the page should be deleted.
697 Stephen Chenney The last sentence on the bottom of the page should be "So p(\theta) = \sin \theta/(1 - \cos \thetamax) and p(\omega) = 1 / (2 \pi (1 - \cos \thetamax))."
703 Arek Zimny In the last line of code in AreaLight::Sample_L(), "ps" should be the first argument to L(), not "p".
706 Stephen Chenney Figure 15.11 would be better if the radius line "r" connected the middle of the sphere to the point where the tangent line intersects the sphere.
713 Kevin Egan, Trevor Thompson Not only is the equation for sampling the Henyey-Greenstein phase function incorrect, but the code implementation is wrong as well. (For more details, see the bug report in the pbrt bug database.)
748 Wei-Wei Xu and Shaohua Fan The numerator and denominator of the fraction of the first displayed equation should be flipped--i.e. it should be "p_A = p_\w \frac{|\cos \theta_i|}{|| \p[i] - \p[i+1] ||^2}".
777 Matthias Zwicker The displayed equation toward the bottom of the page should have |\cos \w_o| in the numerator.
850 mmp Second fragment, second line, delete "/" before "Block(u)"
869 mmp First paragraph, 4th line, delete "octree"

## Minor Errors

Page Reporter Error
72, 74 Leonhard Gruenschloss Missing cross reference in mini-indices to Matrix::Transpose() (page 852)
colophon mmp Third paragraph: "that appears" should be "that appear"
notation mmp Add to index of notation, t(\p, \omega) ray casting function (defined on p. 734).
readme.htm David Gould In the Windows section, change cd c:\pbrt\examples to cd c:\pbrt\scenes and change imageview.exe pbrt.exr to imageview.exe simple.exr. Also, this line: exrtotiff pbrt.exr pbrt.tif doesn't work because "exrtotiff" isn't included in the standard install so isn't in the \bin\win32 directory.
xxxi Cliff Wooley "have generously in suppliyed" should read "have generously supplied".
11 Francis Huynh In 5th line of pseudo-code at bottom of page, swap "light_vector" and "eye_vector" to be consistent with usage of \omega_o and \omega_i.
18 Larry Lai In second paragraph, line 4, "which is be the focus" should be "which is the focus"
54 Leonhard Gruenschloss Second paragraph, change "zeroing one of the two components" to "zeroing one of the components".
70 Randall Hopper Section 2.7.4, first equation set, second line, delete comma between z_1 and z_2.
73 Leonhard Gruenschloss First equation, "p = a * cos(alpha) = a * (v * a)" should be "p = a * ||v|| * cos(alpha) = a * (v * a)"
98 mmp In paragraph for section 3.2.2, change "will use" in line 2 to "uses", and change "won't" in line 4 to "doesn't".
120 Arek Zimny A comma is missing in the expression for one of the hyperboloid's partial derivatives; specifically the right hand side of the one for \partial^2 p/\partial u \partial v, before the zero.
142 mmp Second paragraph, second line, wrong font is used for ']' in 'f[0]'.
204 Arek Zimny In second to last paragraph, "- 1" after "prims.size()" should also be in code font.
210 Brian Budge At the bottom, "sort() depends on the fact that a < b and b < a is only true if a == b" should actually be "sort() depends on the fact that the only time a < b == false and b < a == false will be when a == b".
217,218 Nils Thuerey It would be more clear to name the variable "tplane" "tsplit" in order to match usage in the text and figure 4.15.
221 mmp 5th paragraph, insert "and" after (1987), before "Amanatides"
242 mmp In equation 5.2, there shouldn't be 'prime' symbols after the theta or domega symbols.
246 sfan "The conversion of spherical angles to a direction can be found by" -> "The conversion of a direction $(x,y,z)$ to spherical angles can be found by"
250 mmp Equation 5.5, top of rightmost term, "d" should be roman font
275 mmp Delete space after "::" in "Environment Camera Method Definitions" fragment at the top of the page. Then delete one space in each of the following two lines such that the "float"s continue to line up with "const" in the first line.
286 Benjamin Wüthrich In the displayed equation at the bottom of the page, \Pi_T(x) should actually be \Pi_T(\omega).
320 Leonhard Gruenschloss In "first hundred points of the 2D Hammersley sequence", change "Hammersley" to "Halton", and then in the next sentence, change "Halton" to "Hammersley".
322 Leonhard Gruenschloss In second paragraph, sentence starting with "Plots of the first...", exchange "Hammersley-Zaremba" and "Halton-Zaremba".
426 mmp Figure 9.5: thinner lines should be used to denote angles theta', theta''. (Also p.245, F5.11)
427 Nuttapong Chentanez In the displayed equation at the bottom of the page, the final delta function should just be $\delta(\phi_i - \phi_r)$.
431 Thiago Ize In the displayed equation after equation 9.5, the \omega_t on the left-hand side should be \omega_o. Then, on the right hand side, T(\omega_i, \n) should be T(\omega_o, \n). In the remainder of the sentence after the equation, the same change should be made to the T term and the \omega_i at the end should also be changed to \omega_o.
438 nirwan In the third line of the "OrenNayar Public Methods" chunk, "OrenNayarR" should be "R".
492 Werner Jainek In figure 11.4, point p' should be at the location given by the sum of the two vectors du*dpdu+dv*dpdv. (Currently it isn't!)
499 Arek Zimny In 4th line of 2nd paragraph, change "dsty" to "dsdy"
502 erikdendekker The assignment in the last line of the fragment at the middle of the page should be "*dtdy = ...", not "*dtdx = ..."
539 Arek Zimny Below bottom displayed math change "average value of the step function in one dimension" to "average value of the step function in two dimensions".
550 Wojciech Jarosz "dots are is randomly placed" should be "dots are randomly placed"
587 Arek Zimny Should add "in closed form" to the end of "The DensityRegion cannot implement the VolumeRegion::Tau() method" and change the next sentence ("Therefore....implementations") to something like "However, it can be implemented with Monte Carlo, as is done in Section 15.7.2."
609 mmp Third fragment, 4th line, insert 3 spaces before "screenY1"
647 mmp Second equation from bottom, change \sin^{-1} to \arcsin (be consistent with usage elsewhere.)
676, 678 mmp Pull in spacing on "f" subscripts (many)
681 Stephen Chenney Delete the first sentence of the last paragraph. Then insert "Recall that Blinn's microfacet distribution function is" at the start of the second sentence and add the clause ", where \cos \theta_h..." from the deleted first sentence to the end of the second sentence.
682 Kevin Egan The last line of the "Compute sampled half-angle vector..." code fragment should be "if (!SameHemisphere(wo, H)) H.z *= -1.f;"
714 Volodymyr Kachurovskyi At the second line from the bottom, "N d" should be "N / d".
725 Keith Jeffery In the "Allocate and request samples for sampling one light" fragment, the line "bsdfComponentOffset[0] = sample->Add2D(1);" should instead be "bsdfComponentOffset[0] = sample->Add1D(1):"
772 mmp In the denomenator of the top equation on the page, the left paren ( is incorrectly set as a subscript of p. Furthermore, the first 'p' in the denomenator should be italicized.
805 mmp Figure 17.2 L_0 should be letter 'o' not zero
853 mmp Second fragment, delete "Matrix4x4::"
857 mmp In the second sentence of the first full paragraph, change "The code to the random number generator" to "The code that implements the random number generator".
921 humper C.3.4 - In the input files, use "PixelFilter", not "Filter"
927 Jim Price The type of the kd-tree "emptybonus" parameter should be "float" and not "integer"
935 Jim Price "volumegrid" is missing from the list of supplied volume plugins
1005 Craig Kolb Under "ray tracing", "early, phototypical example" should be "early, prototypical example".
1018 mmp In the third bullet list on the page, "export" should be in typewriter font (2x)