Computer Science 455
Instructor: R. P. Burton
Seventh Quiz

April 14-15, 2003

1. Ray-tracing is more efficient when rays are traced
   • forward from the light source
   • backward from the eye
   • (neither approach has any significant effect on the efficiency)
   (b)


2. What happens in ray tracing when a ray strikes a partially transparent object?
   • it passes through the object
   • it reflects off the object
   • (both (a) and (b))
   (c)


3. When a volume is represented using the octree approach, normals to the surface
   • always align with the principal axes and, hence, are generally meaningless
   • cannot be determined in any meaningful way
   • can be determined in a meaningful way
   (c)


4. When a 2D "fractalized" surface (for graphic presentation) results from random displacement (to an appropriate degree), normals
   • cannot be determined since the fractal surface is infinitely detailed
   • can be determined without any special, additional steps
   • can be determined, but require complex calculations not usually needed to determine normals
   (b)


5. If, for the purpose of antialiasing, rays are sent through pixel corners rather than through pixel centers, the number of rays (for a typical display)
   • decreases dramatically
   • stays about the same
   • increases dramatically
   (b)


6. In a radiosity context consisting of 1000 one-sided patches, about how many form factors need to be computed?
   • about 500
   • about 1000
   • about 500 x 1000
   • about 1000 x 1000 /2
   • about 1000 x 1000
   (d)


7. The eye can respond to about 400,000 distinct frequencies. What is the maximum number of different colors from which Gap could choose for this year's shirts?
   • 400,000
   • 400,000 x 400,000
   • a lot more than (b)
   (c)


8. Which of the following is the least pure color of light?
   • white
   • pink
   • red
   (a)


9. The standard primaries are
   • R, G, & B from the RGB color model
   • "Extremes" from the chromaticity diagram
   • Frequencies at which the eye has peak sensitivity
   • Imaginary color distributions
   (d)


10. Considering the dimensionality of the chromaticity diagram, how many color points maximize the ratio of colors captured to color points? (a color point is a point on the chromaticity diagram and "colors captured" represents the length/area/volume/… of the geometric figure determined by the color points.)
    • two
    • three
    • four
    • the more the better
    (b)


11. Identify the LEAST correct statement:
    • Tint the red by adding just white.
    • Tone the red by adding just black.
    • Shade the red by adding just black.
    (b)


12. Suppose each pixel consists of a red, green, and blue component, each of which can be either on or off. How many colors can be represented by pixels which are 2 x 3?
    • 2 x 3
    • 2 x 3 x 8
    • 6 x 6 x 6
    • 7 x 7 x 7
    (d)


13. The HSV color model can be derived geometrically most easily from
    • the RGB color model
    • the CMY color model
    • (its derivation from either (a) or (b) is equivalently easy)
    (c)


14. Where are the pure hues in the HSV hexacone? Assume the table is at the top and the point is at the bottom.
    • on the table
    • around the perimeter of the table
    • down the central axis
    • along the outside
    • in the interior
    • at the point
    (b)


15. Where are the shades in the HLS color model?
    • at the points
    • on the "table" between the two cones
    • on the outside
    • in the interior
    (c)


16. Which of the following techniques always intersects even small objects and large, far away objects?
    • supersampling
    • adaptive supersampling
    • stochastic ray-tracing
    • statistical supersampling
    • (none of the above)
    (e)


17. Consider a 500 x 500 grid with one ray/grid cell, and a scene consisting of 100 objects. How many ray segments could be produced in raytracing this scene? Assume no transparent objects.
    • 500 x 500
    • 500 x 500 x 100
    • a lot more than 500 x 500 x 100
    (c)


18. In the context of raytracing, mathematical elegance and computational efficiency go hand in hand.
    • true
    • false
    (b)


19. To find the intersection of a ray and a sphere, the ray is represented _____ and the sphere is represented _____.
    • explicitly, explicitly
    • explicitly, implicitly
    • implicitly, explicitly
    • parametrically, explicitly
    • parametrically, implicitly
    • explicitly, parametrically
    (e)


20. When solving for the potential intersection of a ray and a sphere, the discriminant can indicate
    • if there is no intersection point
    • how many intersection points there are
    • both (a) and (b)
    (c)


21. Which of the following does not lead potentially to the "midstream" termination of the calculation of the intersection of a ray and a sphere?
    • determining that the ray's origin is inside the sphere
    • determining that the ray's origin is outside the sphere
    • determining that the ray points away from the sphere
    • determining the ray's closest approach to the sphere
    • (all lead potentially to "midstream" termination)
    (a)


22. Spherical inverse mapping consists of
    • computing a ray reflected from the inside of a sphere rather than from the outside of a sphere
    • mapping a location on a sphere onto a location in a 2-dimensional grid
    • mapping a location in a 2-dimensional grid onto a sphere
    • unfolding a sphere and placing it in a 2-dimensional plane
    (b)


23. When finding the intersection of a ray and a plane, the ray is represented in ____ form and the plane is represented in ____ form.
    • explicit, explicit
    • explicit, implicit
    • implicit, explicit
    • parametric, explicit
    • parametric, implicit
    • explicit, parametric
    (e)


24. (as discussed in class) After a ray has been determined to intersect a plane, the ray is determined to intersect a polygon in the plane by
    • an inside-outside test in 3D
    • an inside-outside test in 2D
    • a winding number test in 3D
    • a winding number test in 2D
    (b)


25. For ray/box intersection, suppose the values of t where the ray intersects the x slab are -4.59 and 13.8. For the y slab, they are 4.59 and 2.29. For the z slab they are -1.15 and 1.15. What can you conclude?
    • the ray is inside the box for t between -4.59 and 13.8.
    • the ray is inside the box for t between -1.15 and 1.15.
    • the ray is inside the box for t between 0 and 1.15.
    • the ray is never inside the box.
    (d)


26. When finding the intersection of a ray and a quadric, the ray is represented in ____ form and the quadric is represented in ____ form.
    • explicit, explicit
    • explicit, implicit
    • implicit, explicit
    • parametric, explicit
    • parametric, implicit
    • explicit, parametric
    (e)


27. Fractal geometry dates from
    • the time of Euclid
    • the time of Euler
    • shortly before the time when most of the members of the class were born
    (c)


28. The applications of fractal geometry are limited to semi-realistic pictures of natural phenomena and exciting pictures, but with virtually no applications in the sciences or engineering.
    • true
    • false
    (b)


29. How long is a typical fractal curve, including the Koch snowflake?
    • up to twice the length of its sides
    • up to three times the length of its sides
    • up to ten times the length of its sides
    • infinitely long
    (d)


30. Why are fractal curves generated with functions in the complex plane?
    • because fractals are imaginary, and require an imaginary component
    • because the complex plane is needed to represent the complexities of fractals
    • to separate the x and y components
    (c)


31. Which of the following is most useful for generating fractals?
    • orbits which diverge to infinity
    • orbits which converge to the origin
    • orbits which neither diverge nor converge
    (c)


32. When a point on an edge is displaced using the random displacement method
    • the point is most likely to be at the middle of the edge
    • the displacement is proportional to the length of the edge
    • both (a) and (b)
    (c)


33. What is the fractal dimension of the Koch snowflake?
    • 1
    • between 1 and 2
    • 2
    • between 2 and 3
    • 3 or higher
    • (it has no such dimension)
    (b)


34. A large scale view of a fractal is always sufficient to predict exactly the details of a magnified view.
    • true
    • false
    (b)


35. Suppose you go up in the mountains and dig a hole. You arrange all the rocks from the hole in order of decreasing size. You expect to find
    • about the same number of each size of rocks
    • more big rocks than little rocks
    • more little rocks than big rocks
    • no correlation between size and quantity
    (c)


36. What is the source of 1/fbeta noise?
    • solar radiation
    • Brownian motion of molecules
    • Oceanic tides
    • Nobody knows
    (d)


37. 1/fbeta noise can NOT be observed in
    • electronic components
    • ocean flows
    • automobiles flowing down a freeway
    • church hymns
    • (no exclusions here)
    (e)


38. When computing the Julia set by inverse iteration, all of the following are either recommended or acceptable EXCEPT
    • using polar coordinates
    • taking the square root of the radius
    • taking as the new angle either the preceding angle or the preceding angle plus pi
    • (no exceptions here)
    (c)


39. About how many lines of code are required to compute the Julia set by inverse iteration?
    • less than 50
    • more than 50, but less than 150
    • more than 150, but less than 500
    • more than 500
    (a)