AU622542B2 - Method for expediting the rendering of figures in a move-draw language - Google Patents

Description

622 542 COMMONWEALTH OF AUSTRALIA PATENTS ACT 1952 COMPLETE SPECIFICATION NAME ADDRESS OF APPLICANT: Sun Microsystems, Inc.

2550 Garcia Avenue Mountain View California 94043 United States of America NAME(S) OF INVENTOR(S): 4,I Is I It

S.

Nola DONATO Anissa LAM Curtis PRIEM ADDRESS FOR SERVICE: DAVIES COLLISON Patent Attorneys 1 Little Collins Street, Melbourne, 3000.

COMPLETE SECIFICATION FOR THE INVENTION ENTITLED: Method for expediting the rendering of figures in a move-draw language The following statement is a full description of this invention, including the best method of performing it known to me/us:-

A

I

A-j

.I

i BACKGROUND OF THE INVENTION 1. Field of the Invention This invention relates to computer systems and, more particularly, to methods for expediting the operation of graphics accelerators in rendering figures when using a move-draw language.

2. History of the Prior Art There have be a number of examples of the use of move-draw io languages in the prior art. Some dedicated graphics terminals such as the IBM 508U and the Tektronix Graphics Terminal have used move-draw languages.

The SGI Graphics Library includes move-draw processes.

The best known move-draw language is Postscript. Postscript is a computer language which has been developed to enable tho precise and complete description and printing of graphic elements by computers. Postscript is widely used in providing printed graphics through advanced laser printers, For a number of reasons including its facility in providing very accurate graphics output images, implementations and improvements of Postscript and similar languages are being utilized for more general purposes in providing computer applications.

Postscript is a move-draw language, one whose commands direct an output device to either move from an initial point to a second point without drawing between the initial point and the second point, or to draw a straight line or curve segment from an initial point to a second point. Postscript also allows a figure rendered in this manner to be filled or left empty. Thus, by designating 1A 82225.P131 -r r.

move or draw commands and the set of coordinates for each of a number of connected points defining a figure, any desired figure may be rendered.

Normally a process in a move-draw language defining a figure to be s rendered is written as a set of instructions; in Postscript and in this description this set of instruction is called a path. A path begins with the move command, an X coordinate of a first point of the figure, and a Y coordinate of the first point of the figure. The path continues with a draw command, an X coordinate of a second point of the figure, and a Y coordinate of the second point of the figure.

lO The path continues with draw commands each having an associated set of coordinates for a related point until all of the points defining the connected points of the figure have been joined. Disconnected elements to be described are indice'ed by another sequence of move, draw, draw, commands. A path may contain an arbitrary number of such sequences.

It will be appreciated by those skilled in the art that the time required by a computer or by any processor implementing tie proceza will be substantial for complicated figures, the figures for which this language was specifically designed.

Certain processors, for example, co-processors and graphics accelerators implement particular drawing algorithms which are designed to render often used figures more rapidly than they can be rendered by the stepby-step plodding through the complete path process, For example, certain graphics accelerators implement fast algorithms which are designed to rapidly produce rapidly vectors, triangles, and quadrilaterals given only the coordinates of the points defining the particular figures and a definition of the 2 82225,P131 -3figure being rendered. Such graphics accelerators may also provide means for rapidly repeating a figure which is to be rendered a plurality of times whether to produce individual or composite figures. A particular graphics accelerator of this type is described in Australian Patent Application No. 39397/89, entitled APPARATUS AND METHOD FOR LOADING COORDINATE REGISTERS FOR USE WITH A GRAPHICS SUBSYSTEM UTILIZING AN INDEX

REGISTER.

Unfortunately, when these co-processors and graphics accelerators are presented a process written in a move-draw language format such as that described they treat the path process as a list of individual instructions. This occurs because these move-draw languages do not define the figure to be rendered so that the devices may make use of their fast algorithms; consequently, the graphics devices S"slowly traverse through the path process to render the output. Since the figures S 15 themselves are not described by the move-draw languages in a manner to allow 4 4 the use of the fast algorithms, it is axiomatic that the number of times a figure is to be repeated is not described by these languages.

It is possible to translate from the move-draw language format to a format 4 t 20 which the particular graphics device can understand in utilizing its internal t4 t algorithms to obtain accelerated rendering of figures. For example, such a translation might cause the device to count the number of vertices of a figure to determine whether a particular algorithm might be used, However, this translation in itself is normally sufficiently time consuming that no overall increase in speed is obtained, Moreover, such translations may well be 920124,krcpO2,53915,90,3 -r I individual to the particular graphics devices so that individual processes would need to be prepared for each graphics device to be utilized.

82225.P131 SUMMARY OF THE INVENTION In accordance with the present invention there is provided a method of implementing a move-draw format to control the operation of a graphics accelerator of a computer system comprising the steps of providing a path process including the individual coordinates of a shape to be rendered, providing a definition of the shape to be rendered, referring to the definition of the shape to determine whether the graphics accelerator is able to implement an algorithm for such type of shape, and selecting the individual coordinates from the path to render the shape while ignoring any move-draw elements of the path process if the graphics accelerator is capable of implementing the algorithm.

According to the present invention there is also provided a method of implementing a move-draw format to control the operation of a graphics accelerator of a computer system comprising the steps of providing a definition of a shape to be rendered, providing a command and individual coordinates for each point defining the shape to be rendered, referring to the definition of the shape to determine whether the graphics accelerator is able to implement an algorithm for such type of shape, and selecting the individual coordinates to render the shape V 20 while ignoring the commands of the process if thee graphics accelerator is capable of implementing the algorithm.

A preferred embodiment of the invention is described in detail hereinafter, with reference to the accompanying drawings in which like elements are referred to by like designations throughout the drawings.

W 92o0123,krsp.2,53915S9,5 ~r o<iS~ l--r-CI- BRIEF DESCRIPTION OF THE DRAWINGS FIGURE 1 is a diagram representing a particular figure which may be rendered as the output of a computer device.

FIGURE 2 illustrates a particular process used in a move-draw language for representing a pair of composite figures defined by four triangles.

FIGURE 3 illustrates a pair of composite figures defined by triangles useful in describing the operation of the present invention.

FIGURE 4 illustrates a particular process used in a move-draw language for representing the composite figures of Figure 3 in accordance with the present invention.

FIGURE 5 illustrates in more detail a process used in a move-draw language for representing a pair of triangles in accoraance with the present invention.

FIGURE 6 illustrates a pair of triangles useful in describing the operation of the present invention.

Ij 82225.P131 To, P4. O~ T ,I NOTATION AND NOMENCLATURE Some portiorns of the detailed descriptions which follow are presented in terms of algorithms and symbolic representations of operations on data bits s within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art.

An algorithm is here, and generally, conceived to be a self-consistent 1o sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like. It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities.

Further, the manipulations performed are often referred to in terms, such as adding or comparing, which are commonly associated with mental operations performed by a human operator. No such capability of a human operator is necessary or desirable in most cases in any of the operations described herein which form part of the present invention; the operations are machine operations. Useful machines for performing the operations of the present invention include general purpose digital computers or other similar devices. In all cases the distinction between the method operations in 4r 82225.P131 '1 4A 7- I- i; i 1 III-- u operating a computer and the method of computation itself should be borne in mind. The present invention relates to method steps for operating a computer in processing electrical or other mechanical, chemical) physical signals to generate other desired physical signals.

82225.P131 tS '~t

N>

DETAILED DESCRIPTION OF THE INVENTION Referring now to Figure 1 there is illustrated a figure 10. The polygon represented by a number of points P1 through P8. In order to render polygon 10 using a move-draw language, a process such as the following might be written: Move P1 Draw P2 Draw P3 Draw P4 Draw Draw P6 Draw P7 Draw P8 Draw P1 In any particular language, the wording of the commands might vary, but, in general terms, these are the commands given, in the first step of the process, the position of the cursor, pen,or other output device is moved to the coordinates of point P1 from whatever its position may be. In the second step, a straight line is rendered from !he initial point P1 to the coordinates of the point P2. In the third and each succeeding step, a .straight line is rendered from point at which the output device lies at the end of the preceding step to the coordinates of next point P3 through P8 and back to point P1. In this manner, the complete polygon illustrated in Figure 1 is described at the output.

KM 82225.P131 0'* 1: LI- rg It will be recognized by those skilled the art, that each step of the path process described requires that the particular command to be utilized be recovered from memory and that the command be interpreted and executed to the particular point described. Moreover, even though a command may be repeated a number of times in succession (the draw command, for example), the command must be retrieved from memory each time it is executed. Running the process requires some given particular time to execute all of its steps when run on any particular computer system.

As pointed out above, certain processors, for example, graphics coprocessors and graphics accelerators are designed to implement particular drawing algorithms more rapidly than they are able to traverse through the normal step-by-step rendering of a path process written in a move-draw language. For example, certain graphics accelerators implement exceptionally fast algorithms which are designed to produce vectors, triangles, ahd quadrilaterals when furnished only the coordinates of the particular figure and a definition of the figure being rendered. Such graphics accelerators may also provide means for rapidly rendering a group of figures of the same kind or for rendering a number of line segments at a time if they have prior knowledge of how many connected line segments follow.

Unfortunately, when presented a process written in a move-draw language format as has been described, these co-processors and graphics accelerators simply treat the process as a list of individual instructions and slowly traverse through the process to render the output. This occurs because these move-draw languages neither explain the figure to be rendered nor the 34 82225.P131

"$A

B :I i IILLI)-- I~ II~-~ number of times a figure of this type is to be rendered so that the graphics devices may make use of their fast algorithms.

Figure 2 illustrates the data structure of a particular path process for s drawing four triangles which are illustrated in Figure 0. The data structure or listing shown in Figure 2 includes a type command and a pair of coordinates for each particular point of the triangles; this listing is simply a more formal arrangement of the particular process defined above for a move-draw language. It will be understood by those skilled in the art that the particular elements of each step of the process must be carried out in order so that the particular triangles illustrated in Figure 3 may be rendered, A review of the pseudocode representation of the process listed in Figure 2 illustrates that no indication is contained in the path of shapes which are to be rendered or of the number of times each such shape is to be rendered. Consequently, a process defined as in Figure 2 cannot utilize the fast drawing algorithms implemented in various graphics accelerators and other pieces of hardware utilized in providing graphic output images.

It will also be understood by those skilled in the art that the particular representations of each of the elements of each step of the process shown in Fgt1ur 2 are pseudocode representations of the actual code which might be usod for the particular steps. In fact, the structure for each of the commands is such that thirty-two bits of storage are used for each of the lines of code in a preferred embodiment of the invention. This data structure is used because most graphics accelerators and other systems operate more rapidly if they are able to handle information which Is divided on il ntger boundaries, In a particulir preferred embodirent of the Invention, the actual division takes 82225.P131

II

;A

Vpace on boundaries which include four bytes. For this reason, thirty-two bits of storage are used for each individual line of code.

It has been determined that the actual memory space required for each of the "type" commands is much less than the thirty-two bits that are provided by the data structure in order to allow the smallest elements tc be on integer boundaries. However, the increase in speed attained by arranging the data structure to divide on integer boundaries is sufficiently great that the waste of memory involved has been found a reasonable trade-off. The present i o invention makes use of the excess space provided In the type designations of the data structure to further increase the speed of operation in running such a move-draw process, up to three times that of the graphics accelerator described in the co-pending patent application referred to above, while allowing the process to run without modification on devices which do not include the fast algorithm facility.

i The present invention accomplishes the foregoing by designating in the otherwise unused "type" space the particular figure which the lines of code are rendering and the number of times the figure is to be rendered. Figure 4 illustrates a path modified in accordance with the invention to render the four triangles shown in Figure 3. Again the representations are pseudocode representations in order to better convey the meaning of each element of the process. At each of the lines of instruction in the path process in which the type designation is given, the shape to be drawn is also included and a count is given.

82225.P131 For example, in Figure 4 the first instruction at line 0 of the path process includes a four to indicate that four shapes of the same type are to be rendered.

The "tri" indicates that the shape to be rendered is a triangle. The next two lines of code (lines 4 and 8) indicate the X and Y coordinates of the initial point P1 of the first triangle. The line numbered 12 provides the next type designation. It should be noted that since this type designation includes only the "line" command (the command for drawing a straight line) since it is not necessary within a shape to indicate again the count or the shape being rendered. The line of code numbered 36 begns the second triangle; and, again, the count (which is now three) and the shape are included. It should be noted that the decreasing count for each triangle allows the process t' witch out to some other process and yet return without having to go back to the original point of the process, It should also be noted that in the particular movedraw code illustrated, it is not necessary for the last side of the triangle to be designated by an indiidual command; the shape merely closes upon itself.

Figure 5 illustrates in more detail, using command language from the preferred embodiment of the invention, the path for rendering a pair of triangles which are illustrated in Figure 6. In the path shown in Figure 5, the coded language has the following meaning: L-at the beginning of code means draw a shape, Brelers to the FRACT code format, F-refers to floating point code format, 2D-refers to two dimensional shapes having but two coordinates.

82225,P131 7 Other coded entries use the usual pseudocode representations for a move-draw language. Thus the first line of code illustrated in Figure indicates that a single figure is to be drawn and the shape of the figure is a triangle. LF2D signifies that the coordinates are coded as floating point s numbers, that the particular command is to draw the triangle, and that the shape is two dimensional so that only two coordinates are expected for each point.

The MOVE command is the normal pseudocode representation of the command given in a move-draw language to accomplish a move. The remaining lines defining the first triangle are self explanatory.

The tenth line of code 36 begins the second triangle which for the purpose of explanation of this invention has its coordinates defined in the FRACT number format. Because a number format is used which is different than the format used for the first triangle, a one is placed in the count position to designate to the graphics accelerator hardware that this is the only one of these shapes to be defined. Again, the code defines the shape as a triangle, and LB2D explains that a draw command is being executed for a two dimensional triangular shape with FRACT coordinates.

The code illustrated in Figure 5 is used as follows when run on the graphics accelerator referred to in the co-pending patent application. The hardware looks to the first line of the geonmetric designation, determines that one triangle is to be drawn in the floating point format, then igncwA. all but the three additional sets of coordinates necessary to provide the three vertices of the triangle. If the count had been higher, the program would have continued through all of the triangles until all of the triangles had been rendered without the need to obtain any of the move-draw commands from memory, Thus, the 82225.P131

-I

Ki:~fl S. c 11 i; hardware need not call down the commands for move and draw, it merely looks for the expected coordinates and draws the shape using its exceptionally fast algorithm for the described figure.

This invention offers a substantial number of advantages. It accomplishes its purposes without requiring more storage than that required to implement the process in the ncrmal move-draw path structure. It allows a process to run much faster on graphics devices having algorithms for accelerating the rendering of particular shapes. It is entirely compatible with other versions of the process coded using the normal path form. It is able to be used for rendering both polygons and vectors, Although the invention has been described in a preferred embodiment, it will be clear to those skilled in the art that many modifications may be made to the invention described without departing from the spirit and scope of the invention. It will, therefore, be appreciated that the invention should be considered to be particularly defined by the claims which follow.

1 82225.P131

Claims (4)

1. A method of implementing a move-draw format to control the operation of a graphics accelerator of a computer system comprising the steps of providing a path process including the individual coordinates of a shape to be rendered, providing a definition of the shape to be rendered, referring to the definition of the shape to determine whether the graphics accelerator is able to implement an algorithm for such type of shape, and selecting the individual coordinates from the path to render the shape while ignoring any move-draw elements of the path process if the graphics accelerator is capable of implementing the algorithm.

2. A method of implementing a move-draw format to control the operation of go 9, a graphics accelerator of a computer system as claimed in Claim 1 in which the step of providing a definition of the shape to be rendered comprises providing a 15 designation of a particular element of the shape and providing a designation of the it number of times the particular element is to be repeated.

3. A method of implementing a move-draw format to control the uperation of a graphics accelerator of a computer system comprising the steps of providing a 20 definition of a shape to be rendered, providing a command and individual coordinates for each point defining the shape to be rendered, referring to the definition of the shape to determine whether the graphics accelerator is able to implement an algorithm for such type of shape, and selecting the individual coordinates to render the shape while ignoring the commands of the process if thee graphics accelerator is capable of implementing the algorithm, t y 920124,rcpe,002,53915,90,16 I iii L I -I

4. A method of implementing a move-draw format to control the operation of a graphics accelerator of a computer system as claimed in Claim 3 further comprising the steps of providing a designation of the number of times the shape is to be repeated, and repeating the shape the number of times designated.

82225.P131 iA4iE~ -o ~tu -4 18 A method of implementing a move-draw format substantially as hereinbefore described with reference to the drawings. DATED this TWENTY FIFTH day of JANUARY 1992 SUN MICROSYSTEMS, INC. by its Patent Attorneys DAVIES COLLISON CAVE 920124,kxcspcXOO2,5391S.90,j8