Computer Science 455
Instructor: R. P. Burton
Third Quiz
October 11-12, 2004
Name _________________________________________ Score ____________/31
1. A transformation matrix with a -1 in (0,0) and (1,1) represents
a. rotation
b. scaling
c. reflection through the origin
d. all of the above
e. none of the above
(d)
2. All of the following reflections can be represented by a transformation matrix EXCEPT
a. reflection across an arbitrary line parallel to a principal axis
b. reflection across an arbitrary line through the origin
c. reflection across an arbitrary line not parallel to a principal axis and not through the origin
d. reflection through an arbitrary point
e. (no exceptions here)
(e)
3. To rotate an arbitrary line segment defined by endpoints (x1, y1) and (x2, y2) so that it is parallel to
a principal axis, it is most efficient to find
a. the sine and cosine of arctan (delta_y/delta_x)
b. the sine and cosine of arctan (delta_x/delta_y)
c. the sine and cosine from the arccosine and arcsine
d. the sine and cosine from delta_x and delta_y
(d)
4. The Cohen-Sutherland endpoint codes for both endpoints of a line segment can be used to determine all of
the following EXCEPT
a. if the line segment can be trivially accepted
b. if the line segment can be trivially rejected
c. which unextended clipping region boundaries are crossed
d. which extended clipping region boundaries are crossed
e. (no exceptions here)
(c)
5. The p's and q's in the Liang and Barsky algorithm can be used to determine all of the following EXCEPT
a. If a line segment is parallel to a clipping region boundary
b. If an extended line segment proceeds from inside to outside, or from outside to inside an extended
clipping region boundary
c. The point where an extended line segment crosses an extended window boundary
d. (no exceptions here)
(d)
6. What is the best characterization of the Nicholl Lee Nicholl algorithm?
a. Clipping is more efficient when Nicholl Lee Nicholl is used in place of either Cohen and Sutherland
or Liang and Barsky if the number of line segments is small.
b. Clipping is more efficient when Nicholl Lee Nicholl is used in place of either Cohen and Sutherland
or Liang and Barsky if the number of line segments is large.
c. Clipping is more efficient when Nicholl Lee Nicholl is used in place of either Cohen and Sutherland
or Liang and Barsky independent of the number of line segments.
d. Nicholl Lee Nicholl cannot be used in place of either Cohen and Sutherland or Liang and Barsky.
(c)
7. In the Nicholl Lee Nicholl algorithm, there are ____ general classes of regions in which the "beginning"
endpoint of a line segment can be placed.
a. two
b. three
c. four
d. nine
(b)
8. The Nicholl Lee Nicholl algorithm compares the slope of a line segment against the slope of at most _____ other
line segment(s) originating from one of the endpoints.
a. one
b. two
c. three
d. four
e. nine
(d)
9. When a clipping region is rectangular but rotated (by other than a multiple of ninety degrees) ______ clipping should
occur before the rotated window is aligned with the coordinate axes.
a. some
b. no
(a)
10. In general, if you clip sets of points, line segments, areas, or text, the clipping algorithm should produce sets of
points, line segments, areas, or text, respectively.
a. true
b. false
(a)
11. The Sutherland and Hodgman clipping algorithm produces _____ intermediate polygon(s) before the final clipped
polygon is produced.
a. no
b. one
c. two
d. three
(d)
12. Relative to Sutherland and Hodgman, Weiler and Atherton is
a. more accurate and easier to code
b. more accurate but harder to code
c. less accurate and harder to code
d. less accurate but easier to code
(b)
13. All-or-nothing string clipping can be accomplished by
a. clipping the major diagonal of the bounding box
b. clipping the minor diagonal of the bounding box
c. clipping a diagonal of the bounding box which intersects the clipping region
d. (any of the above)
(d)
14. Curve clipping is accomplished using
a. Cooprider & Gustafson
b. a point clipping algorithm or a curve-clipping algorithm
c. blanking techniques
(b)
15. Assuming that neither the window nor the viewport is rotated, the window-to-viewport transformation can be
represented by a composition of translation and scaling matrices.
a. true
b. false
(a)
16. In general terms, changes to the viewport have _____ effects on the image and changes to the window
have _____ effects on the image.
a. direct, direct
b. direct, inverse
c. inverse, direct
d. inverse, inverse
(b)
17. How is the inverse of a shearing matrix obtained?
a. by complementing the shearing component
b. by inverting the shearing component
c. by transposing the shearing component
d. by inverting and transposing the shearing component
e. (none of the above)
(a)
18. Translation in nD contemplates n translation factors. Scaling in nD contemplates n scaling factors.
Rotation in nD contemplates n sine values and n cosine values.
a. true
b. false
(b)
19. Rotation "about an axis" parallel to a principal axis in 3D contemplates the composition
of _____ 4 x 4 matrices.
a. 2 or 3
b. 4 or 5
c. 6 or 7
d. more than 7
(a)
20. Rotation "about an axis" not parallel to a principal axis but passing through the origin in
3D contemplates the composition of _____ 4 x 4 matrices.
a. 2 or 3
b. 4 or 5
c. 6 or 7
d. more than 7
(b)
21. Rotation "about an axis" not parallel to a principal axis and not passing through the origin
in 3D contemplates the composition of _____ 4 x 4 matrices.
a. 2 or 3
b. 4 or 5
c. 6 or 7
d. more than 7
(c)
22. A transform which superimposes arbitrary 3D Cartesian coordinate frame A onto arbitrary 3D Cartesian
coordinate frame B contemplates the composition of _____ 4 x 4 matrices.
a. 2 or 3
b. 4 or 5
c. 6 or 7
d. more than 7
(c - translate, scale, rotate, rotate, rotate, reverse)
23. An object needs to be transformed relative to local coordinate frame C. Coordinate frame C has a position
and orientation relative to coordinate frame D (which happens to be the coordinate frame for world coordinates).
A transformation which maps coordinate frame ____ onto coordinate frame ____ should be applied to _____ to the data
to facilitate the transformation.
a. C, D, coordinate frame C
b. D, C, coordinate frame D
c. C, D, the object to be transformed
d. D, C, the object to be transformed
(c)
24. A rectangular set of difficult-to-reproduce data has been scan-converted and represented in the frame buffer.
Unfortunately, the data needs to be moved (translated) to a partially overlapping region of the frame buffer using
temporary storage capable of holding only a single scan line.
a. It can’¡Çt be done
b. start with a scan line common to the source rectangle and to the destination rectangle
c. start with a scan line unique to the source rectangle
(b)
25. How many of the following combinations are unsatisfactory: 1) painted and posted, 2) painted and not posted,
3) not painted but posted, 4) not painted and not posted?
a. 1
b. 2
c. 3
d. all four
(b)
26. Suppose a particular design calls for 400 NAND gates, 300 AND gates, and connecting wires. How many
"symbols" need to be defined in master coordinates?
a. none
b. two or three
c. 400
d. 700
e. 700 plus the number of connecting wire segments
(b)
27. Suppose you have a hierarchy consisting of four different kinds of office furniture to be placed in groups
to form departments to form divisions to represent a company. Where does the data describing the (visible)
graphical objects reside?
a. throughout the hierarchy
b. only in the internal nodes
c. only at the leaf level
(c)
28. Suppose four transformation matrices are composed to place the first piece of furniture in the first group
in the first department in the first division in the company. In a properly designed system, how many transformation
matrices need to composed to place the second piece of furniture in the same group?
a. none
b. two
c. four
d. five
(b)
29. Of the following, which is the worst place to clip?
a. in master coordinates
b. in world coordinates
c. in screen coordinates
(b)
30. An LED display is viable because, relative to incandescent light sources,
a. LEDs can be manufactured in all the needed colors
b. LEDs virtually never burn out
c. LEDs can be turned on and off very quickly
(c)
31. How many of the following produce true 3D output: 1) a "vibrating" mirror system, 2) a head-mounted
display, 3) a holographic display, 4) a rotating LED panel, 5) a "stereo box" with no moving parts, and
6) a "Viewmaster"
a. none
b. one or two
c. three or four
d. five or six
(a)