Programming Languages and Computer Graphics — Mock Test 7

C++ and Python allow a class to inherit from several classes directly. Java and VB.NET allow multiple inheritance only of interfaces.

A union is a user-defined data type; a function is a subroutine; an interactive 3D environment is virtual reality; a shadow mask belongs to colour CRT output devices.

DDA and Bresenham are the classic line-drawing algorithms.

Note: The midpoint line algorithm is also a line-drawing algorithm; the official key is A and B.

An 8-bit signed (two's complement) value ranges from −128 to 127; unsigned from 0 to 255. The least values are −128 and 0.

A class with a pure virtual function is abstract and cannot be instantiated. Virtual functions implement run-time polymorphism through base pointers, and a pure virtual function may still be given a body.

Hue is the dominant wavelength (colour) of light; saturation is purity and luminance is brightness.

The line from C(0,0,−10) through P1(1,2,0) is (t, 2t, −10 + 10t); at t = 3 it passes through P2(3,6,20), so P1 (nearer) obscures P2. P3(2,4,6) is not on it (t = 2 gives z = 10), so P3 does not obscure P1 and P2 does not obscure P3. A, C and D are correct.

Bresenham draws lines; Cohen–Sutherland clips lines; Sutherland–Hodgeman clips polygons; Z-buffer removes hidden surfaces.

Bezier curves are widely used in painting, drawing and CAD packages; splines are controlled by parameters such as bias and tension (beta-splines); sweep representations build 3D objects with translational or rotational symmetry; natural objects are modelled with fractals.

Sutherland–Hodgeman clips polygons against a convex clip window and handles convex polygons correctly; concave ones may give extra edges.

arrfptr is an array of 3 pointers to functions, each returning a pointer to an array of 10 (int). So x is an array of three function pointers.

Clipping removes the objects and line parts that lie outside the clip window.

Multiply x by 2 and y by 3: (0,0) → (0,0), (0,2) → (0,6), (3,0) → (6,0), (3,2) → (6,6).

Scaling changes an object's size; translation moves it, rotation turns it and shearing slants it.

The value 1 is arr[2][3][1], which in pointer form is *(*(*(arr + 2) + 3) + 1). arr[3][4][2] and arr[2][3][2] are out of bounds.

string s1; (default), string s2(s1); (copy), string s4{'a'}; and string s5{7, 'a'}; (initializer lists of chars) are listed as valid. Following the official key (after Stroustrup's examples), string s3{7}; is treated as an error, since there is no string(int) constructor.

A camera captures 2-D array input; an LCD is a transmissive display; inkjet printers output binary (dot) images; a flat-bed scanner captures input one line (1-D array) at a time.

Signed and unsigned types of the same size have the same number of values; signed types simply shift half of them to negatives, so the signed positive range is smaller. C is wrong.

The keyword explicit appears only on the declaration inside the class; the out-of-class definition is just Date::Date(int dd) { … }.

x′ = 0.5x halves the horizontal component; y′ = 1.5y increases the vertical by a factor of three halves.

If classes differ only by a parameter, use templates; if actual types are known only at run time, use a class hierarchy; inline operations with templates give run-time efficiency; reaching derived-class specifics through a base needs explicit casting.

The loop variable is initialized first; a do-while executes its body before testing the condition.

A raster-scan CRT draws the picture as rows of dots, line by line, exactly as a television does, so it can use a TV monitor. R explains A.

The painter's (depth-sort) algorithm draws polygons back to front, so nearer ones overwrite farther ones — the simplest way to remove hidden surfaces in a minimal 3D pipeline.

A full 4 × 4 matrix in homogeneous coordinates, with a non-trivial bottom row (e, f, g, h), is a general projective transformation — a homography.