Graphics Plus

home *** CD-ROM | disk | FTP | other *** search

/ Graphics Plus / Graphics Plus.iso / msdos / animutil / fastgfx / fg303b / manuals.arj / USER06.DOC < prev next >

Wrap

Text File | 1993-10-02 | 67.2 KB | 1,632 lines

Chapter 6 Graphics Fundamentals 98 Fastgraph User's Guide Overview This chapter describes Fastgraph's fundamental graphics routines, sometimes called graphics primitives. These routines perform such functions as clearing the screen, drawing points, drawing solid and dashed lines, drawing closed shapes (polygons, circles, and ellipses), drawing rectangles (solid, hollow, and dithered), and filling arbitrary regions. Most of these routines have no effect in text video modes, but there are a few exceptions, and they will be noted in the descriptions of those routines. Clearing the Screen The Fastgraph routine fg_erase clears the entire screen in any video mode. In text modes, fg_erase stores a space character (ASCII 32) with a gray foreground attribute in each character cell. In graphics modes, fg_erase sets each pixel to zero. This of course causes each pixel to be displayed its background color. The fg_erase routine has no arguments. The fg_erase routine always sets the entire screen to the background color. Another routine, fg_fillpage, fills the screen with the current color. In text modes, fg_fillpage stores a solid block character (ASCII 219) with the current display attribute (as defined in the most recent call to fg_setcolor or fg_setattr). In graphics modes, fg_fillpage sets each pixel to the current color (as defined in the most recent call to fg_setcolor). Like fg_erase, the fg_fillpage routine has no arguments. Clipping The suppression of graphics outside a pre-defined area is called clipping. Many of Fastgraph's graphics-oriented routines provide clipping, either automatically or through a special version of the routine. Fastgraph includes two routines, fg_setclip and fg_setclipw, to define a rectangular clipping region. The fg_setclip routine defines the clipping region in screen space, while the fg_setclipw routine performs the same function in world space. Each routine takes four arguments: the minimum x, the maximum x, the minimum y, and the maximum y coordinate of the clipping region. The arguments are integer quantities for fg_setclip and floating point quantities for fg_setclipw. For example, the statement fg_setclip(0,159,0,99); would define the upper left quadrant of the screen as the clipping region in a 320 by 200 graphics mode. An implicit clipping region equal to the entire screen is defined as part of the fg_setmode routine's initializations. Clipping is not supported for text modes. Chapter 6: Graphics Fundamentals 99 Points The Fastgraph routine fg_point provides the most basic graphics operation -- setting a pixel to a specified color. The fg_point routine has two integer arguments. The first specifies the pixel's x coordinate, and the second its y coordinate. The pixel is drawn using the current color value, as specified in the most recent call to fg_setcolor. There is also a world space version of this routine, fg_pointw, that uses floating point arguments. Another Fastgraph routine is available for reading a pixel's color value. The fg_getpixel routine has two integer arguments that specify the (x,y) coordinates for the pixel of interest. There is no world space version of the fg_getpixel routine, but you can obtain a pixel's color value in world space by applying the fg_xscreen and fg_yscreen functions to the world space coordinates and passing the resulting values to fg_getpixel. Example 6-1 uses the fg_point routine to draw 100 random points in random colors. It also uses the fg_getpixel routine to insure no two points are adjacent. The program establishes a graphics video mode with the fg_automode and fg_setmode routines. Next, it determines the maximum color value for the selected video mode; note if we used virtual colors (color indices above the maximum color value), some of the colors would be the background color and would thus produce invisible points. The main part of the program is a while loop that first generates a random pair of (x,y) screen coordinates. It then calls the fg_getpixel routine to check the pixels at (x,y) and the eight adjacent positions. If none of these pixels are set, the program generates a random color value and draws a point in that color. After doing this 100 times, the program waits for a keystroke, restores the original video mode and screen attributes, and then returns to DOS. Example 6-1. #include <fastgraf.h> #include <stdlib.h> void main(void); void main() { int area; int color, old_color; int left; int max_color, max_x, max_y; int new_mode, old_mode; 100 Fastgraph User's Guide int x, y; old_mode = fg_getmode(); new_mode = fg_automode(); fg_setmode(new_mode); if (new_mode == 4) max_color = 3; else if (new_mode == 11 || new_mode == 17) max_color = 1; else if (new_mode == 19) max_color = 255; else max_color = 15; left = 100; max_x = fg_getmaxx() - 1; max_y = fg_getmaxy() - 1; while (left > 0) { x = rand() % max_x + 1; y = rand() % max_y + 1; area = fg_getpixel(x-1,y-1)+fg_getpixel(x,y-1)+fg_getpixel(x+1,y-1) + fg_getpixel(x-1,y) +fg_getpixel(x,y) +fg_getpixel(x+1,y) + fg_getpixel(x-1,y+1)+fg_getpixel(x,y+1)+fg_getpixel(x+1,y+1); Chapter 6: Graphics Fundamentals 101 if (area == 0) { color = rand() % max_color + 1; fg_setcolor(color); fg_point(x,y); left--; } } fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Sometimes you may want to display a pixel using an "exclusive or" operation (usually called XOR) to guarantee its visibility. The color of a pixel drawn in XOR mode will be c xor p, where c is the current color and p is the color of the pixel already at that position. This means the pixel's new color will be different from the background as long as the current color is not zero. The Fastgraph routine fg_pointx displays a screen space pixel in XOR mode, while fg_pointxw does the same thing in world space. Their respective parameters are the same as for fg_point and fg_pointw. The Graphics Cursor Many of Fastgraph's graphics routines depend on the position of the graphics cursor as a reference point. For example, Fastgraph includes routines to draw lines from the graphics cursor position to a specified position, and the bit-mapped image display routines discussed in Chapter 10 display or retrieve an image relative to the graphics cursor position. The graphics cursor is not a cursor in the true sense; it is simply a pair of (x,y) coordinates with a special meaning. The fg_setmode routine sets the graphics cursor position to the screen space coordinates (0,0), and the fg_initw routine sets it to the world space coordinates (0.0,0.0). Fastgraph includes four routines for changing the graphics cursor position. The fg_move routine sets it to an absolute screen space position, while the fg_movew routine sets it to an absolute world space position. The fg_moverel routine sets the graphics cursor position to a screen space position relative to its current position. The fg_moverw routine does the same in world space. 102 Fastgraph User's Guide Each of these routines has two arguments that specify the (x,y) coordinates of the new position. For the screen space routines, the arguments are integer quantities. For the world space routines, the arguments are floating point quantities. You can obtain the screen space coordinates of the graphics cursor position with the fg_getxpos and fg_getypos routines. These routines have no arguments and respectively return the x and y coordinates of the graphics cursor position as the function value. To obtain the world space coordinates of the graphics cursor position, you can apply the fg_xworld and fg_yworld functions to the return values of fg_getxpos and fg_getypos. Solid Lines Fastgraph includes eight routines for drawing solid lines. All of them draw lines in the current color value and observe the clipping limits. The fg_draw routine draws a line from the current graphics cursor position to an absolute screen space position, while fg_draww draws a line to an absolute world space position. The fg_drawrel routine draws a line from the current graphics cursor position to a screen space position relative to it. The fg_drawrw routine does the same in world space. You can draw absolute lines in XOR mode with fg_drawx and relative XOR lines with fg_drawrelx, or with their world space counterparts fg_drawxw and fg_drawrxw. XOR lines are often used when drawing cursors (such as cross hairs) or rubberband boxes to insure they are visible against colored backgrounds. Another useful property of XOR lines is that drawing the same lines twice restores what was originally beneath the lines. Each of these routines has two arguments that specify the (x,y) coordinates of the destination position. For the screen space routines, the arguments are integer quantities. For the world space routines, the arguments are floating point quantities. In either case the destination position becomes the new graphics cursor position. This makes it possible to draw connected lines without calling a graphics cursor movement routine between successive calls to a line drawing routine. Examples 6-2 and 6-3 each draw a pair of crossed lines that divide the screen into quadrants. Example 6-2 does this using the fg_move and fg_draw routines, while example 6-3 uses the fg_moverel and fg_drawrel routines. Both examples draw the lines in white, the default for color 15 in all graphics video modes. Example 6-2. Example 6-3. #include <fastgraf.h> #include <fastgraf.h> void main(void); void main(void); void main() void main() { { int max_x, max_y; int max_x, max_y; int mid_x, mid_y; int mid_x, mid_y; int new_mode, old_mode; int new_mode, old_mode; old_mode = fg_getmode(); old_mode = fg_getmode(); new_mode = fg_automode(); new_mode = fg_automode(); Chapter 6: Graphics Fundamentals 103 fg_setmode(new_mode); fg_setmode(new_mode); max_x = fg_getmaxx(); max_x = fg_getmaxx(); max_y = fg_getmaxy(); max_y = fg_getmaxy(); mid_x = max_x / 2; mid_x = max_x / 2; mid_y = max_y / 2; mid_y = max_y / 2; fg_setcolor(15); fg_setcolor(15); fg_move(mid_x,0); fg_move(mid_x,0); fg_draw(mid_x,max_y); fg_drawrel(0,max_y); fg_move(0,mid_y); fg_moverel(-mid_x,-mid_y); fg_draw(max_x,mid_y); fg_drawrel(max_x,0); fg_waitkey(); fg_waitkey(); fg_setmode(old_mode); fg_setmode(old_mode); fg_reset(); fg_reset(); } } Examples 6-4 and 6-5 are variations of example 6-2. Example 6-4 uses world space rather than screen space to draw the crossed lines. Example 6-5 is the same as example 6-2 except it defines a clipping area to restrict drawing to the upper left quadrant of the screen. The clipping suppresses the right half of the horizontal line and the lower half of the vertical line. Example 6-4. Example 6-5. #include <fastgraf.h> #include <fastgraf.h> void main(void); void main(void); void main() void main() { { int new_mode, old_mode; int max_x, max_y; int mid_x, mid_y; old_mode = fg_getmode(); int new_mode, old_mode; new_mode = fg_automode(); fg_setmode(new_mode); old_mode = fg_getmode(); fg_initw(); new_mode = fg_automode(); fg_setworld(-10.0,10.0,-10.0,10.0); fg_setmode(new_mode); 104 Fastgraph User's Guide fg_setcolor(15); max_x = fg_getmaxx(); fg_movew(0.0,10.0); max_y = fg_getmaxy(); fg_draww(0.0,-10.0); mid_x = max_x / 2; fg_movew(-10.0,0.0); mid_y = max_y / 2; fg_draww(10.0,0.0); fg_waitkey(); fg_setclip(0,mid_x,0,mid_y); fg_setmode(old_mode); fg_setcolor(15); fg_reset(); fg_move(mid_x,0); } fg_draw(mid_x,max_y); fg_move(0,mid_y); fg_draw(max_x,mid_y); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Dashed Lines Fastgraph includes four routines for drawing dashed lines. All of them draw lines in the current color value and observe the clipping limits. The fg_dash routine draws a dashed line from the current graphics cursor position to an absolute screen space position, while fg_dashw draws a dashed line to an absolute world space position. The fg_dashrel routine draws a dashed line from the current graphics cursor position to a screen space position relative to it. The fg_dashrw routine does the same in world space. Each of these routines has three arguments. The first two specify the (x,y) coordinates of the destination position. For the screen space routines, these arguments are integer quantities. For the world space routines, these arguments are floating point quantities. The third argument is a 16-bit pattern that defines the appearance of the dashed line. Bits that are set in the pattern produce the visible part of the line, while bits that are reset produce the invisible part. This pattern is repeated as necessary to draw the entire line. For example, the pattern value 3333 hex would produce a dashed line with the first two pixels off, the next two on, Chapter 6: Graphics Fundamentals 105 the next two off, and so forth. Similarly, the pattern value FFFF hex would produce a solid line. The destination position passed to any of the dashed line routines becomes the new graphics cursor position. This makes it possible to draw connected dashed lines without calling a graphics cursor movement routine between successive calls to a line drawing routine. Example 6-6 draws a pair of crossed dashed lines that divide the screen into quadrants. It does this using the fg_move and fg_dash routines and draws the lines in white, the default for color 15 in all graphics video modes. The dash pattern for each line is 3333 hex, which alternates two pixels off and on. Example 6-6. #include <fastgraf.h> void main(void); void main() { int max_x, max_y; int mid_x, mid_y; int new_mode, old_mode; old_mode = fg_getmode(); new_mode = fg_automode(); fg_setmode(new_mode); max_x = fg_getmaxx(); max_y = fg_getmaxy(); mid_x = max_x / 2; mid_y = max_y / 2; fg_setcolor(15); fg_move(mid_x,0); fg_dash(mid_x,max_y,0x3333); fg_move(0,mid_y); fg_dash(max_x,mid_y,0x3333); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Polygons Fastgraph includes routines for drawing filled and unfilled polygons, as well as a routine for determining if a given point is inside a convex polygon. All the polygon routines observe the clipping limits. The fg_polygon routine draws an unfilled polygon in screen space. Fg_polygon requires an array of integer x coordinates as its first argument, and an array of integer y coordinates as its second argument. Each (x,y) coordinate pair from the two arrays is treated as a polygon vertex. In other 106 Fastgraph User's Guide words, the x coordinate of the first polygon vertex is the first element of the x coordinate array, and the y coordinate of the first vertex is the first element of the y coordinate array. Similarly, the second elements of each array define the second vertex, and so forth. The third argument for fg_polygon is an integer quantity that specifies the number of elements in the two coordinate arrays (that is, the number of polygon vertices). Another routine, fg_polygonw, draws an unfilled polygon in world space. The fg_polygonw routine is the same as the fg_polygon routine, except its x and y coordinate arrays must contain floating point values instead of integers. The polygon drawing begins at the first vertex specified in the coordinate arrays. The polygon routines then draw a solid line to the second vertex, then to the third vertex, and continue in this manner through the last vertex. If necessary, they then close the polygon by drawing a line connecting the last vertex and the first vertex. Example 6-7 illustrates unfilled polygon drawing using the fg_polygon routine in the EGA monochrome or enhanced modes (modes 15 and 16). The program exits if neither of these video modes are available. Example 6-7. #include <fastgraf.h> #include <stdio.h> #include <stdlib.h> void main(void); #define VERTICES 10 int x[] = {200,300,400,400,300,240,160,160,200,210}; int y[] = {100, 80,100,220,320,320,240,200,160,150}; void main() { int old_mode; old_mode = fg_getmode(); if (fg_testmode(16,1)) fg_setmode(16); else if (fg_testmode(15,1)) fg_setmode(15); else { printf("This program requires a 640 x 350 "); printf("EGA graphics mode.\n"); exit(1); } fg_setcolor(1); fg_polygon(x,y,VERTICES); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Chapter 6: Graphics Fundamentals 107 As shown in this example, the fg_polygon routine expects the x and y components defining the polygon vertices to be in separate arrays. Another routine, fg_polyline, draws an unfilled polygon from polygon vertices in one integer array. With fg_polyline, the first array element is the x component of the first vertex, the second element is the y component of the first vertex, the third element is the x component of the second vertex, and so forth. The first fg_polyline argument is the vertex coordinate array of alternating x and y components, and the second argument specifies the number of vertices. In other respects, fg_polyline is identical to fg_polygon. An additional feature available with fg_polyline but not fg_polygon is the ability to specify vertex offsets. The fg_polyoff routine's two integer arguments respectively define offset values that are added to each fg_polyline vertex. This makes it possible to define polygon vertices as relative values, that when combined with the offsets, determine the polygon's absolute position. Perhaps the most important polygon display routine is fg_polyfill, which displays a filled convex polygon in screen space (the polygon is filled with pixels of the current color). Its first argument is a vertex array in the same format used by fg_polyline. The next argument is a work array used internally. The size of the work array must be at least twice the polygon height if defined as an integer array, or four times the polygon height if defined as a byte array. The final argument specifies the number of polygon vertices. Example 6-8 illustrates the use of the fg_polyline, fg_polyoff, and fg_polyfill routines. The program first draws an unfilled polygon, centered within the left half of the screen. It then draws a filled version of the same polygon, centered within the right half of the screen. In each case, the centering is accomplished by passing appropriate values to fg_polyoff. After waiting for a keystroke, the program establishes a clipping region in the lower right corner and redraws the filled polygon at the same position but in a different color. This of course results in only a portion of the filled polygon being drawn. Like example 6-7, this example requires a 640 by 350 graphics mode. Example 6-8. #include <fastgraf.h> #include <stdio.h> #include <stdlib.h> void main(void); #define VERTICES 10 int xy[] = {200,100, 300, 80, 400,100, 400,220, 300,320, 240,320, 160,240, 160,200, 200,160, 210,150}; int work_array[700]; void main() { int old_mode; old_mode = fg_getmode(); if (fg_testmode(16,1)) fg_setmode(16); 108 Fastgraph User's Guide else if (fg_testmode(15,1)) fg_setmode(15); else { printf("This program requires a 640 x 350 "); printf("EGA graphics mode.\n"); exit(1); } fg_setcolor(1); fg_polyoff(-120,-25); fg_polyline(xy,VERTICES); fg_polyoff(200,-25); fg_polyfill(xy,work_array,VERTICES); fg_waitkey(); fg_setcolor(2); fg_setclip(480,639,175,349); fg_polyfill(xy,work_array,VERTICES); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } The fg_polyfill routine fills convex polygons. From Fastgraph's perspective, a polygon is convex if any horizontal line drawn through the polygon crosses the left edge exactly once and the right edge exactly once (excluding horizontal and zero-length edge segments). Note that this definition includes shapes that are not convex in the traditional sense. In addition, any non-convex polygon can be decomposed into two or more convex polygons. The filled convex polygon is a fundamental tool of three-dimensional computer graphics. A common practice is to build an image or object from several adjacent or connecting polygons. Such polygons typically overlap at one or more edges. For instance, the coordinates defining the right edge of one polygon may also define the left edge of another polygon immediately to its right. For an overall image to appear correct, its component polygons must fit together correctly. The fg_polyfill routine applies the following rules to handle overlapping polygon edges: Points located exactly on non-horizontal edges are drawn only if the polygon's interior is directly to the right. Points located exactly on horizontal edges are drawn only if the polygon's interior is directly below them. A vertex is drawn only if all lines ending at that point meet the above two conditions. These three rules ensure that no pixel is drawn more than once when filling adjacent polygons. However, they may not be suitable for displaying polygons that are not adjacent because some of the pixels on the polygon's edge will not be included. If this is an issue, first draw the filled polygon with fg_polyfill, then draw an unfilled polygon having the same vertices with fg_polyline. Chapter 6: Graphics Fundamentals 109 The final Fastgraph routine relating to polygons is fg_inside, which checks if a specified point lies inside a convex polygon. Its first argument is a vertex array in the same format used by fg_polyline and fg_polyfill. The second argument is the number of vertices in the polygon, while the last two arguments specify the screen space x and y coordinates of the point being tested. The fg_inside function returns 1 if the point lies inside the polygon and 0 if not. If the vertex array does not define a convex polygon, the return value is undefined. Circles and Ellipses Fastgraph includes routines for drawing filled and unfilled circles and ellipses. Both screen space and world space versions of these routines are available, and all of them observe the clipping limits. The fg_circle routine draws an unfilled circle in screen space, centered at the graphics cursor position, using the current color. The routine's only argument specifies the circle's radius in horizontal screen space units. Another routine, fg_circlew, draws an unfilled circle where the radius is measured in horizontal world space units. The analogous routines for drawing filled circles are fg_circlef and fg_circlefw. All four circle drawing routines leave the graphics cursor position unchanged. The fg_ellipse routine draws an unfilled ellipse in screen space, centered at the graphics cursor position, using the current color. The routine requires two arguments that respectively specify the length of its horizontal and vertical semi-axes. In other words, the first argument is the absolute distance from the center of the ellipse to its horizontal extremity, and the second argument is the absolute distance from the center to the vertical extremity. Another routine, fg_ellipsew, draws an unfilled ellipse in world space. The analogous routines for drawing filled ellipses are fg_ellipsef and fg_ellipsfw. All four ellipse drawing routines leave the graphics cursor position unchanged. Example 6-9 illustrates the use of the fg_circlew and fg_ellipsew routines. The program first uses the fg_automode routine to propose a graphics video mode and then uses the fg_setmode routine to select that video mode. It then makes color 15 the current color, which by default is white in all color graphics modes and "on" in the monochrome graphics modes. Next, it establishes a 200-unit by 200-unit world space coordinate system. The program then uses fg_ellipsew to draw an ellipse and fg_circlew to draw a circle, both centered at the middle of the screen (which is the origin of our world space coordinate system). The circle has a radius of 1/16 the width of the screen (12.5 horizontal world space units), and the ellipse is horizontally inscribed within the circle. Example 6-10 illustrates the use of the fg_circle and fg_ellipse routines. It is functionally identical to example 6-9, but it uses screen space rather than world space coordinates to draw the circle and ellipse. Note the arguments to the fg_circle and fg_ellipse routines are dependent on the maximum x and y coordinates of the selected video mode. If we didn't compute these arguments in this manner, the actual size of the circle and ellipse would be proportional to the pixel resolution of the video mode. No such dependency exists when using world space, but we pay a price for this feature in slightly slower execution speed. 110 Fastgraph User's Guide Example 6-9. Example 6-10. #include <fastgraf.h> #include <fastgraf.h> void main(void); void main(void); void main() void main() { { int old_mode; int mid_x, mid_y; int old_mode; old_mode = fg_getmode(); int x,y; fg_setmode(fg_automode()); fg_setcolor(15); old_mode = fg_getmode(); fg_setmode(fg_automode()); fg_initw(); fg_setcolor(15); fg_setworld(-100.0,100.0,-100.0,100.0); mid_x = fg_getmaxx() / 2; fg_movew(0.0,0.0); mid_y = fg_getmaxy() / 2; fg_ellipsew(12.5,12.5); x = mid_x / 8; fg_circlew(12.5); y = mid_y / 8; fg_waitkey(); fg_move(mid_x,mid_y); fg_setmode(old_mode); fg_ellipse(x,y); fg_reset(); fg_circle(x); } fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Chapter 6: Graphics Fundamentals 111 Solid Rectangles Fastgraph includes four routines for drawing solid rectangles, two for screen space and two for world space, with and without clipping. None of these routines affect the graphics cursor position. The fg_rect routine draws a solid rectangle in screen space, without regard to the clipping limits, using the current color. Fg_rect requires four integer arguments that respectively define the minimum x, maximum x, minimum y, and maximum y screen space coordinates of the rectangle. The minimum coordinates must be less than or equal to the maximum coordinates, or else the results are unpredictable. The fg_clprect routine is identical in all respects to the fg_rect routine, except it observes the clipping limits. The world space versions of the solid rectangle drawing routines are fg_rectw and fg_clprectw. Like fg_rect and fg_clprect, they require four arguments that define the extremes of the rectangle, but the arguments are floating point world space coordinates. You also can use fg_rect in text modes. When used in a text mode, fg_rect expects its four arguments to be expressed in character space (that is, rows and columns) rather than screen space. This means the four arguments respectively specify the minimum column, maximum column, minimum row, and maximum row of the rectangle. Fastgraph constructs the rectangle by storing the solid block character (ASCII decimal value 219) in each character cell comprising the rectangle. The rectangle is drawn using the current character attribute, but because the solid block character occupies the entire character cell, the background component of the attribute is essentially meaningless. Example 6-11 demonstrates the use of the fg_rect routine by drawing 200 random-size rectangles in random colors. The program first uses the fg_automode routine to propose a graphics video mode and then uses the fg_setmode routine to select that video mode. Next, it determines the horizontal and vertical screen resolution for the selected video mode, using the fg_getmaxx and fg_getmaxy routines. The main part of the program is a for loop that generates a random rectangle in each iteration. Inside the loop, the C library function rand is used to generate the extremes of the rectangle. If necessary, the program then exchanges the coordinates to make the minimum coordinates less than or equal to the maximum coordinates. Finally, it again uses rand to generate a random color number between 0 and 15, and then draws the rectangle in that color. After drawing all 200 rectangles, the program restores the original video mode and screen attributes before returning to DOS. Example 6-11. #include <fastgraf.h> void main(void); #define RECTANGLES 200 #define SWAP(a,b,temp) { temp = a; a = b; b = temp; } 112 Fastgraph User's Guide void main() { int i; int minx, maxx, miny, maxy; int old_mode; int temp; int xres, yres; old_mode = fg_getmode(); fg_setmode(fg_automode()); xres = fg_getmaxx() + 1; yres = fg_getmaxy() + 1; for (i = 0; i < RECTANGLES; i++) { minx = rand() % xres; maxx = rand() % xres; miny = rand() % yres; maxy = rand() % yres; if (minx > maxx) SWAP(minx,maxx,temp); if (miny > maxy) SWAP(miny,maxy,temp); fg_setcolor(rand()%16); fg_rect(minx,maxx,miny,maxy); } fg_setmode(old_mode); fg_reset(); } Unfilled Rectangles Fastgraph includes a routine for drawing unfilled rectangles. The fg_box routine draws an unfilled rectangle in screen space, with regard to the clipping limits, using the current color. The arguments to fg_box are the same as those for fg_rect. The depth of the rectangle's edges is one pixel by default, but you can change the depth by calling fg_boxdepth. The fg_boxdepth routine expects two arguments. The first argument is the width of the rectangle's left and right sides, while the second is the height of its top and bottom sides. Once you call fg_boxdepth, fg_box draws all unfilled rectangles using the depth values specified in the most recent call to fg_boxdepth. Unlike fg_rect, the fg_box routine has no effect in text video modes. The analogous world space routine is fg_boxw. Example 6-12 is the same as example 6-11, but it draws unfilled instead of solid rectangles. The program uses fg_box to draw the each rectangle and fg_boxdepth to define the rectangle depth at three pixels in each direction. Note that you don't need to call fg_boxdepth for each rectangle if you want all of them to have the same depth. Example 6-12. #include <fastgraf.h> void main(void); Chapter 6: Graphics Fundamentals 113 #define RECTANGLES 200 #define SWAP(a,b,temp) { temp = a; a = b; b = temp; } void main() { int i; int minx, maxx, miny, maxy; int old_mode; int temp; int xres, yres; old_mode = fg_getmode(); fg_setmode(fg_automode()); fg_boxdepth(3,3); xres = fg_getmaxx() + 1; yres = fg_getmaxy() + 1; for (i = 0; i < RECTANGLES; i++) { minx = rand() % xres; maxx = rand() % xres; miny = rand() % yres; maxy = rand() % yres; if (minx > maxx) SWAP(minx,maxx,temp); if (miny > maxy) SWAP(miny,maxy,temp); fg_setcolor(rand()%16); fg_box(minx,maxx,miny,maxy); } fg_setmode(old_mode); fg_reset(); } The fg_boxx and fg_boxxw routines are screen space and world space "exclusive or" versions of the box drawing routines. They simplify displaying rubberband boxes (hollow rectangles that move in response to keystrokes or mouse movement) because drawing an object in XOR mode makes it appear, but drawing it again in the same position and in the same color restores what was under the object originally. Given this useful property, here is an outline of what's necessary to draw a rubberband box: 1. Draw the box in its initial position with fg_boxx or fg_boxxw. 2. Wait for a user response that signifies box movement. 3. Use fg_boxx or fg_boxxw to redraw the box in the same position as in step 1. This restores what was there originally. 4. Draw the box in its new position with fg_boxx or fg_boxxw. 5. Repeat steps 2, 3, and 4 until the box is no longer needed. 114 Fastgraph User's Guide Example 6-13 shows a simple use of fg_boxx in the 320 by 200 MCGA/VGA 256-color mode (mode 19). The program fills the left and right halves of the screen with different colors and then displays an XOR box the overlaps each half. Even though the box is drawn in the same color as the right half, it is still visible because drawing something in an XOR mode (except in color 0) guarantees that it will be visible against its background. After a keystroke, the program redraws the same box, which of course restores what was there in the first place. Example 6-13. #include <fastgraf.h> void main(void); void main() { int old_mode; old_mode = fg_getmode(); fg_setmode(19); fg_setcolor(9); fg_rect(0,159,0,199); fg_setcolor(15); fg_rect(160,319,0,199); fg_waitkey(); fg_boxx(80,239,50,149); fg_waitkey(); fg_boxx(80,239,50,149); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Dithered Rectangles The process of alternating different color pixels across a region of the display area is called dithering. This technique is especially useful in the graphics modes with few colors, such as CGA and Hercules modes, because you can simulate additional colors through effective uses of dithering. Fastgraph includes two routines for drawing dithered rectangles, one for screen space and one for world space. Neither routine observes the clipping limits, nor do they affect the graphics cursor position. The fg_drect routine draws a dithered rectangle in screen space. Like the fg_rect routine, fg_drect requires four integer arguments that respectively define the minimum x, maximum x, minimum y, and maximum y screen space coordinates of the rectangle. The minimum coordinates must be less than or equal to the maximum coordinates, or else the results are unpredictable. However, fg_drect also requires a fifth argument that defines the dithering matrix, which in turn determines the pixel pattern used to build the dithered rectangle. The size and format of the dithering matrix are dependent on the video mode. Chapter 6: Graphics Fundamentals 115 The world space version of the dithered rectangle drawing routine is fg_drectw. Like fg_drect, it requires four arguments that define the extremes of the rectangle, and a fifth argument that defines the dithering matrix. As mentioned earlier, the size and format of the dithering matrix are dependent on the video mode. The dithering matrix is a four byte array in all video modes except the 256 color graphics modes (modes 19 through 27), where it is an eight byte array. This array contains a pixel pattern that fg_drect or fg_drectw replicates across the rectangle's area. The structure of the dithering matrix closely mimics the structure of video memory in each graphics mode. The remainder of this section will present some simple mode-specific examples to illustrate the structure of the dithering matrix in the different graphics modes. Suppose we would like to produce a "checkerboard" of light blue and white pixels. That is, in a given row of a rectangle, consecutive pixels will alternate between these two colors. Additionally, the pattern for adjacent rows will be shifted such that there will always be a white pixel above and below a light blue pixel, and vice versa. Hence this pixel pattern would look something like B W B W W B W B B W B W W B W B where each B represents a light blue pixel, and each W represents a white pixel. The following examples describe the dithering matrix that could be used to produce such a pixel pattern in each graphics mode. CGA Four-Color Graphics Modes The CGA four-color graphics modes (modes 4 and 5) use a four-byte dithering matrix that Fastgraph treats as a four-row by one-column array. Since each pixel in these modes requires two bits of video memory, each byte of the dithering matrix holds four pixels. Thus, the pixel representation of the dithering matrix would appear as shown below on the left; its translation to numeric values appears on the right. [3] B W B W [3] 01 11 01 11 [2] W B W B [2] 11 01 11 01 [1] B W B W [1] 01 11 01 11 [0] W B W B [0] 11 01 11 01 Because these modes do not offer a light blue color, we've used light cyan (color value 1 in palette 1) to approximate light blue. The B pixels thus translate to color value 1, or 01 binary. White is available as color value 3 in palette 1, so the W pixels translate to color value 3, or 11 binary. The hexadecimal equivalent of the binary value 11011101 (for array elements [0] and [2]) is DD, and the hexadecimal equivalent of the binary value 01110111 (for array elements [1] and [3]) is 77. As shown in example 6-12, 116 Fastgraph User's Guide these are precisely the values assigned to the elements of the dithering matrix. Example 6-14 uses mode 4 to display a 50-pixel by 50-pixel dithered rectangle in the upper left corner of the screen. The dithering matrix represents the blue and white checkerboard pattern discussed in the preceding paragraph. Example 6-14. #include <fastgraf.h> void main(void); void main() { char matrix[4]; int old_mode; old_mode = fg_getmode(); fg_setmode(4); matrix[0] = matrix[2] = 0xDD; matrix[1] = matrix[3] = 0x77; fg_drect(0,49,0,49,matrix); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } CGA Two-Color Graphics Mode The CGA two-color graphics mode (mode 6) uses a four-byte dithering matrix that Fastgraph treats as a four-row by one-column array, as in the other four-color CGA modes. However, each pixel in this mode only requires one bit of video memory, so each byte of the dithering matrix holds eight pixels. Thus, the pixel representation of the dithering matrix would appear as shown below on the left; its translation to numeric values appears on the right. [3] B W B W B W B W [3] 0 1 0 1 0 1 0 1 [2] W B W B W B W B [2] 1 0 1 0 1 0 1 0 [1] B W B W B W B W [1] 0 1 0 1 0 1 0 1 [0] W B W B W B W B [0] 1 0 1 0 1 0 1 0 Mode 6 obviously does not offer a light blue color, so we've used black (color value 0) in its place. The B pixels thus translate to color value 0. White is available as color value 1, so the W pixels translate to color value 1. The hexadecimal equivalent of the binary value 10101010 (for array elements [0] and [2]) is AA, and the hexadecimal equivalent of the binary value 01010101 (for array elements [1] and [3]) is 55. Thus, to make example Chapter 6: Graphics Fundamentals 117 6-14 run in mode 6, we only need to change the fg_setmode argument from 4 to 6 and change the dithering matrix values as shown below. matrix[0] = matrix[2] = 0xAA; matrix[1] = matrix[3] = 0x55; Tandy/PCjr 16-Color Graphics Mode The Tandy/PCjr 16-color graphics mode (mode 9) also uses a four-byte dithering matrix that Fastgraph treats as a four-row by one-column array. Each pixel in this mode requires four bits of video memory, so each byte of the dithering matrix only holds two pixels. Thus, the pixel representation of the dithering matrix would appear as shown below on the left; its translation to numeric values appears on the right. [3] B W [3] 1001 1111 [2] W B [2] 1111 1001 [1] B W [1] 1001 1111 [0] W B [0] 1111 1001 The B pixels translate to color value 9 (light blue), or 1001 binary, and the W pixels translate to color value 15 (white), or 1111 binary. The hexadecimal equivalent of the binary value 11111001 (for array elements [0] and [2]) is F9, and the hexadecimal equivalent of the binary value 10011111 (for array elements [1] and [3]) is 9F. Thus, to make example 6-14 run in mode 9, we only need to change the fg_setmode argument from 4 to 9 and change the dithering matrix values as shown below. matrix[0] = matrix[2] = 0xF9; matrix[1] = matrix[3] = 0x9F; Hercules Graphics Mode The size and format of the dithering matrix in the Hercules graphics mode (mode 11) are the same as in the CGA two-color mode (mode 6). Hercules Low-Resolution Graphics Mode The size and format of the dithering matrix in the Hercules low- resolution graphics mode (mode 12) are the same as in the CGA four-color modes (modes 4 and 5). As far as our checkerboard example goes, we'll use black (color value 0) in place of light blue, and bold (color value 3) instead of white. Thus, the B pixels translate to 00 binary, while the W pixels translate to 11 binary. The hexadecimal equivalent of the binary value 11001100 (for array elements [0] and [2]) is CC, and the hexadecimal equivalent of the binary value 00110011 (for array elements [1] and [3]) is 33. Thus, to make example 6-14 run in mode 12, we only need to change the 118 Fastgraph User's Guide fg_setmode argument from 4 to 12 and change the dithering matrix values as shown below. matrix[0] = matrix[2] = 0xCC; matrix[1] = matrix[3] = 0x33; EGA/VGA/SVGA 16-Color Graphics Modes The EGA/VGA/SVGA 16-color graphics modes (modes 13 through 18, 28, and 29) use a four-byte dithering matrix that Fastgraph treats as a four-row by one-column array. Unlike the other graphics modes, which allow you to store pixels of several colors in the dithering matrix, these modes treat the dithering matrix as a bit map for a specific color. Since each color in the dither pattern must be stored in a separate bit map (that is, in a separate dithering matrix), you must call fg_drect once for each color. Furthermore, you must use the fg_setcolor routine before each call to fg_drect to define the color used with the dithering matrix. In all EGA/VGA/SVGA graphics modes, each byte of the dithering matrix is a bit map that represents eight pixels. Using our familiar checkerboard example, the pixel representation of the dithering matrix would appear as shown below. [3] B W B W B W B W [2] W B W B W B W B [1] B W B W B W B W [0] W B W B W B W B Translating this pattern to numeric values is simple. Just construct one dithering matrix for each color in the pattern (there are two colors in this example), where pixels of the current color translate to 1, and other pixels translate to 0. Following our example, the translation for the B pixels appears below on the left, while the translation for the W pixels appears on the right. [3] 1 0 1 0 1 0 1 0 [3] 0 1 0 1 0 1 0 1 [2] 0 1 0 1 0 1 0 1 [2] 1 0 1 0 1 0 1 0 [1] 1 0 1 0 1 0 1 0 [1] 0 1 0 1 0 1 0 1 [0] 0 1 0 1 0 1 0 1 [0] 1 0 1 0 1 0 1 0 The hexadecimal equivalent of the binary value 01010101 is 55, and the hexadecimal equivalent of the binary value 10101010 is AA. As shown in example 6-15, these are precisely the values assigned to the elements of the dithering matrices. Chapter 6: Graphics Fundamentals 119 Example 6-15 uses mode 13 to display our light blue and white checkerboard pattern. Note you must call fg_drect twice -- once for the light blue pixels (color value 9), and again for the white pixels (color value 15). Note also how fg_setcolor is used before each call to fg_drect to define the color of the pixels fg_drect will display. Example 6-15. #include <fastgraf.h> void main(void); void main() { char matrix[4]; int old_mode; old_mode = fg_getmode(); fg_setmode(13); matrix[0] = matrix[2] = 0x55; matrix[1] = matrix[3] = 0xAA; fg_setcolor(9); fg_drect(0,49,0,49,matrix); matrix[0] = matrix[2] = 0xAA; matrix[1] = matrix[3] = 0x55; fg_setcolor(15); fg_drect(0,49,0,49,matrix); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } 256-Color Graphics Modes The 256-color graphics modes (modes 19 through 27) use an eight-byte dithering matrix that Fastgraph treats as a four-row by two-column array. Each pixel in these modes requires eight bits of video memory, so each byte of the dithering matrix only holds a single pixel. We therefore need the two column dithering matrix to produce a meaningful dither pattern. The pixel representation of the dithering matrix would appear as shown below on the left; its translation to numeric values appears on the right. [6] B W [7] [6] 9 15 [7] [4] W B [5] [4] 15 9 [5] [2] B W [3] [2] 9 15 [3] [0] W B [1] [0] 15 9 [1] 120 Fastgraph User's Guide The B pixels translate to color value 9 (light blue), and the W pixels translate to color value 15 (white). Example 6-16 uses mode 19 to draw our light blue and white checkerboard pattern. Example 6-16. #include <fastgraf.h> void main(void); void main() { char matrix[8]; int old_mode; old_mode = fg_getmode(); fg_setmode(19); matrix[0] = matrix[3] = matrix[4] = matrix[7] = 15; matrix[1] = matrix[2] = matrix[5] = matrix[6] = 9; fg_drect(0,49,0,49,matrix); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Closing Remarks There are two other important items pertaining to the fg_drect and fg_drectw routines. These items apply regardless of which graphics video mode is being used. First, the dithering matrix may not contain virtual color values. That is, the pixel color values stored in the dithering matrix must be between 0 and the maximum color value for the current video mode. If any color value is redefined using the fg_defcolor routine, Fastgraph always ignores the redefinition and instead uses the actual color value. Note this does not apply to palette registers or video DAC registers, because in these cases we are redefining the color associated with a color value and not the color value itself. Second, Fastgraph aligns the dithering matrix to specific pixel rows. Fastgraph draws the dithered rectangle starting with the pixel row specified by the rectangle's lower limit (the maximum y coordinate for fg_drect, or the minimum y coordinate for fg_drectw) and proceeds upward until reaching the rectangle's upper limit. In all cases the dithering matrix used by fg_drect and fg_drectw contains four rows. If we let r represent the pixel row corresponding to the rectangle's lower limit, then the first row used in the dithering matrix is r modulo 4 (assuming the dithering matrix rows are numbered 0 to 3). This alignment enables you to use the same dithering matrix in multiple calls to fg_drect and fg_drectw for building an object of adjacent dithered rectangles (for example, an L-shaped area) and still have the dither pattern match where the rectangles intersect. Chapter 6: Graphics Fundamentals 121 Region Fill Fastgraph includes routines for filling arbitrary regions. The fg_flood and fg_paint routines fill a region with the current color value by specifying a screen space point in the region's interior. The fg_floodw and fg_paintw routine also fill a region, but they require the interior point to be expressed in world space. All region fill routines have two arguments that specify the (x,y) coordinates of the interior point. For fg_flood and fg_paint, the arguments are integer quantities. For fg_floodw and fg_paintw, they are floating point quantities. None of them change the graphics cursor position. The difference between the "flood" and "paint" versions of these routines is simple: fg_flood and fg_floodw will not extend the region fill beyond the clipping limits, while fg_paint and fg_paintw ignore the clipping limits. As a result, fg_paint and fg_paintw are significantly faster. The region being filled must be a closed polygon whose boundary color is different from that of the specified interior point. The region may contain holes (interior areas that will not be filled). Fastgraph fills the region by changing every interior pixel whose color is the same as the specified interior point, to the current color. If the interior point is already the current color, the region fill routines do nothing. It is important to note fg_paint and fg_paintw do not treat the screen edges as polygon boundaries. Filling an open polygon will cause these routines to behave unpredictably. This is not the case with fg_flood and fg_floodw as long as the clipping limits are not beyond the screen edges. Example 6-17 illustrates a simple use of the fg_paint routine in a 320 by 200 graphics mode. The program uses fg_bestmode to select an available video mode (if no 320 by 200 graphics mode is available, the program exits). After establishing the selected video mode, the program uses fg_move and fg_drawrel to draw a hollow rectangle in color 10 and a hollow diamond in color 9. The diamond is drawn in the middle of the rectangle, thus making it a hole with respect to the rectangle. The program leaves these shapes on the screen until a key is pressed. At that time, it calls fg_paint to fill that part of the rectangle outside the diamond with color 10. After waiting for another keystroke, the program again uses fg_paint to fill the interior of the diamond with color 15. Finally, the program waits for another keystroke, restores the original video mode and screen attributes, and returns to DOS. Example 6-17. #include <fastgraf.h> #include <stdio.h> #include <stdlib.h> void main(void); void main() { int old_mode, new_mode; new_mode = fg_bestmode(320,200,1); if (new_mode < 0) { printf("This program requires a 320 x 200 "); printf("graphics mode.\n"); exit(1); } 122 Fastgraph User's Guide old_mode = fg_getmode(); fg_setmode(new_mode); fg_setcolor(10); fg_move(100,50); fg_drawrel(120,0); fg_drawrel(0,100); fg_drawrel(-120,0); fg_drawrel(0,-100); fg_setcolor(9); fg_move(160,80); fg_drawrel(30,20); fg_drawrel(-30,20); fg_drawrel(-30,-20); fg_drawrel(30,-20); fg_waitkey(); fg_setcolor(10); fg_paint(160,70); fg_waitkey(); fg_setcolor(15); fg_paint(160,100); fg_waitkey(); fg_setmode(old_mode); fg_reset(); } Summary of Fundamental Graphics Routines This section summarizes the functional descriptions of the Fastgraph routines presented in this chapter. More detailed information about these routines, including their arguments and return values, may be found in the Fastgraph Reference Manual. FG_BOX draws an unfilled rectangle in screen space, with respect to the clipping region. The width of the rectangle's edges is one pixel unless changed with the fg_boxdepth routine. FG_BOXDEPTH defines the depth of rectangles drawn with the box display routines. The fg_setmode routine initializes the box depth to one pixel. FG_BOXW draws an unfilled rectangle in world space, with respect to the clipping region. The width of the rectangle's edges is one pixel unless changed with the fg_boxdepth routine. FG_BOXX draws an unfilled rectangle in screen space, with respect to the clipping region, in "exclusive or" mode. The width of the rectangle's edges is one pixel unless changed with the fg_boxdepth routine. Chapter 6: Graphics Fundamentals 123 FG_BOXXW draws an unfilled rectangle in world space, with respect to the clipping region, in "exclusive or" mode. The width of the rectangle's edges is one pixel unless changed with the fg_boxdepth routine. FG_CIRCLE draws an unfilled circle in screen space. The circle is centered at the graphics cursor position. FG_CIRCLEF draws a filled circle in screen space. The circle is centered at the graphics cursor position. FG_CIRCLEFW draws a filled circle in world space. The circle is centered at the graphics cursor position. FG_CIRCLEW draws an unfilled circle in world space. The circle is centered at the graphics cursor position. FG_CLPRECT draws a solid (filled) rectangle in screen space, with respect to the clipping region. FG_CLPRECTW draws a solid (filled) rectangle in world space, with respect to the clipping region. FG_DASH draws a dashed line from the graphics cursor position to an absolute screen space position. It also makes the destination position the new graphics cursor position. FG_DASHREL draws a dashed line from the graphics cursor position to a screen space position relative to it. It also makes the destination position the new graphics cursor position. FG_DASHRW draws a dashed line from the graphics cursor position to a world space position relative to it. It also makes the destination position the new graphics cursor position. FG_DASHW draws a dashed line from the graphics cursor position to an absolute world space position. It also makes the destination position the new graphics cursor position. FG_DRAW draws a solid line from the graphics cursor position to an absolute screen space position. It also makes the destination position the new graphics cursor position. FG_DRAWREL draws a solid line from the graphics cursor position to a screen space position relative to it. It also makes the destination position the new graphics cursor position. FG_DRAWRELX draws a solid line in "exclusive or" mode from the graphics cursor position to a screen space position relative to it. It also makes the destination position the new graphics cursor position. FG_DRAWRW draws a solid line from the graphics cursor position to a world space position relative to it. It also makes the destination position the new graphics cursor position. FG_DRAWRXW draws a solid line in "exclusive or" mode from the graphics cursor position to a world space position relative to it. It also makes the destination position the new graphics cursor position. 124 Fastgraph User's Guide FG_DRAWW draws a solid line from the graphics cursor position to an absolute world space position. It also makes the destination position the new graphics cursor position. FG_DRAWX draws a solid line in "exclusive or" mode from the graphics cursor position to an absolute screen space position. It also makes the destination position the new graphics cursor position. FG_DRAWXW draws a solid line in "exclusive or" mode from the graphics cursor position to an absolute world space position. It also makes the destination position the new graphics cursor position. FG_DRECT draws a dithered rectangle in screen space, without regard to the clipping region. The rectangle's dither pattern is defined through a dithering matrix whose format is dependent on the video mode being used. FG_DRECTW draws a dithered rectangle in world space, without regard to the clipping region. The rectangle's dither pattern is defined through a dithering matrix whose format is dependent on the video mode being used. FG_ELLIPSE draws an unfilled ellipse in screen space. The ellipse is centered at the graphics cursor position, and its size is determined by the specified lengths of the semi-axes. FG_ELLIPSEF draws a filled ellipse in screen space. The ellipse is centered at the graphics cursor position, and its size is determined by the specified lengths of the semi-axes. FG_ELLIPSEW draws an unfilled ellipse in world space. The ellipse is centered at the graphics cursor position, and its size is determined by the specified lengths of the semi-axes. FG_ELLIPSFW draws a filled ellipse in world space. The ellipse is centered at the graphics cursor position, and its size is determined by the specified lengths of the semi-axes. FG_ERASE clears the screen in either text or graphics modes. FG_FILLPAGE fills the screen with the current color in either text or graphics modes. FG_FLOOD fills an arbitrary closed region with the current color value, with respect to the clipping limits. The region is defined by specifying a screen space point within its interior. FG_FLOODW fills an arbitrary closed region with the current color value, with respect to the clipping limits. The region is defined by specifying a world space point within its interior. FG_GETPIXEL returns the color value of a specified pixel. FG_GETXPOS returns the screen space x coordinate of the graphics cursor position. FG_GETYPOS returns the screen space y coordinate of the graphics cursor position. Chapter 6: Graphics Fundamentals 125 FG_INSIDE determines if the specified point is inside a given convex polygon. FG_MOVE establishes the graphics cursor position at an absolute screen space point. FG_MOVEREL establishes the graphics cursor position at a screen space point relative to the current position. FG_MOVERW establishes the graphics cursor position at a world space point relative to the current position. FG_MOVEW establishes the graphics cursor position at an absolute world space point. FG_PAINT fills an arbitrary closed region with the current color value, without respect to the clipping limits. The region is defined by specifying a screen space point within its interior. FG_PAINTW fills an arbitrary closed region with the current color value, without respect to the clipping limits. The region is defined by specifying a world space point within its interior. FG_POINT draws a point (that is, displays a pixel) in screen space. FG_POINTW draws a point in world space. FG_POINTX draws a point in "exclusive or" mode in screen space. FG_POINTXW draws a point in "exclusive or" mode in world space. FG_POLYFILL draws a filled convex polygon in screen space. The polygon is closed if necessary. FG_POLYGON draws an unfilled polygon in screen space, using two coordinate arrays to define the polygon vertices. The polygon is closed if necessary. FG_POLYGONW draws an unfilled polygon in world space. It is the same as the fg_polygon routine, except the coordinate arrays contain world space values. FG_POLYLINE draws an unfilled polygon in screen space, using a single coordinate array to define the polygon vertices. The polygon is closed if necessary. FG_POLYOFF defines the screen space offset applied to each vertex for polygons drawn with fg_polyfill or fg_polyline. FG_RECT draws a solid (filled) rectangle in screen space or character space, without regard to the clipping region. FG_RECTW draws a solid (filled) rectangle in world space, without regard to the clipping region. 126 Fastgraph User's Guide FG_SETCLIP defines the clipping region in screen space. The clipping region is a rectangular area outside of which graphics are suppressed. FG_SETCLIPW defines the clipping region in world space.