JPH0283670A - Double precision graphic computing system - Google Patents

Abstract

PURPOSE:To improve the computing accuracy of a graphic by generating and storing segment transformation matrices with double precision and single precision corresponding to the input of a deformation element in an internal table, and performing coordinate calculation with the double precision internally corresponding to the designation of the segment transformation matrix with single precision from an application program. CONSTITUTION:A segment transformation matrix generating part 1 generates the segment transformation matrices with double and single precision corresponding to the input of the deformation element and stores them in the internal table 1-1. When the segment transformation matrix designated from the application program at single precision as a parameter coincides with the segment transformation matrix with single precision stored in the internal table 1-1, magnification/compression and movement are performed by applying the segment transformation matrix with double precision stored in the internal table 1-1 and performing an arithmetic processing. Meanwhile, when noncoincidence is obtained, the arithmetic processing is performed by applying the segment transformation matrix with single precision. In such a way, it is possible to perform the coordinate calculation with high accuracy by applying the segment transformation matrix with double precision internally corresponding to the designation of single precision as the parameter from the application program.

Description

[Detailed Description of the Invention] [Summary] Regarding a double-precision figure calculation method that calculates a segment transformation matrix on a figure to enlarge/reduce/move it, etc., double-precision and single-precision segment transformation matrices that correspond to the human power of deformation elements are provided. is created and stored in an internal table, and the coordinates are calculated internally in double precision in response to the specification of a single precision segment transformation matrix from an application program. A segment transformation matrix creation unit that creates precision and single-precision segment transformation matrices and stores them in an internal table; and an S-precision segment transformation matrix in which the segment transformation matrix specified as a parameter in single precision is stored in the internal table. and a segment transformation matrix application unit that applies the double-precision segment transformation matrix stored in this internal table and performs arithmetic processing when the segment transformation matrix matches the internal table, and the segment transformation matrix application unit performs double-precision/single-precision segment transformation. Configure it to apply a matrix and output the calculation result.

[Industrial application field]

The present invention calculates a segment transformation matrix on a figure to enlarge/enlarge it.
This relates to a double-precision graphic arithmetic method that performs reduction, etc.

[Prior art and problems to be solved by the invention] International standards for computer graphics
(Graphical Kernel System)
defines the information (number of parameters, parameter order, parameter data format, etc.) between an application program and a graphics program. Among these, enlarge the shape/
The segment transformation matrix for reduction, rotation, and translation is defined as a real-number type single-precision 2×3 matrix. For this reason, if you use a single-precision segment transformation matrix to perform graphical processing such as enlarging/reducing, even if you try to display the results of structural analysis as a fine graphic, it will not be possible to display the result using the conventional eruption rate. There was a problem.

A detailed explanation will be given below using the conventional processing flow shown in FIG.

In FIG. 6(a), (2) calculates a segment transformation matrix in single precision and sets it as a parameter of the application program. This means creating a 2x3 matrix segment transformation matrix (single precision) corresponding to the specified enlargement/reduction for the segment (figure) to be processed, and setting this as a parameter for the application program. are doing.

In FIG. 6 (b), ■ indicates the segment i to be processed.
Read the figure order from .

(2) Calculates coordinates based on the segment transformation matrix (single precision) and the figure order.

@ displays the calculation result.

(2) determines whether the segment is the end or not.

If YES, it ends. If NO, read out the next graphic order with ■ and repeat @ and @.

As mentioned above, conventionally, the GKS interface was
It was defined as L precision, and correspondingly, the coordinate meter lγ was performed using a single precision segment transformation matrix within the graphics program, so it was not possible to calculate coordinates with double precision.

The present invention creates double-precision and single-precision segment transformation matrices corresponding to the human power of the deformation element, stores them in an internal table, and internally stores them in an internal table in response to the specification of the single-precision segment transformation matrix from an application program. The purpose is to calculate coordinates in double precision.

[, f! means to solve the problem] Means for solving the problem will be explained with reference to FIG.

In FIG. 1, a segment transformation matrix creation section 1 creates double-precision and single-precision segment transformation matrices corresponding to the input of transformation elements and stores them in an internal table.

Segment transformation matrix application unit 2 converts the segment transformation matrix specified in single precision as a parameter to the double precision stored in +δ in this internal table when the 17 columns of segment transformation specified as parameters match the single precision segment transformation matrix stored in the internal table. The segment transformation matrix is applied to perform arithmetic processing.

[Effect]

As shown in FIG. 1, in the present invention, a segment transformation matrix creation unit 1 creates double-precision and single-precision segment transformation matrices corresponding to the human power of deformation elements, stores them in an internal table, and uses the application program to create double-precision and single-precision segment transformation matrices. If the segment transformation matrix specified in single precision as a parameter matches the single precision segment transformation matrix stored in the internal table, the double precision segment transformation matrix stored in this internal table is applied for calculation processing. Then, you can enlarge/reduce, move, etc. On the other hand, if they do not match, arithmetic processing is performed using a single-precision segment transformation matrix.

Therefore, by creating double-precision and single-precision matrices when creating a segment transformation matrix and storing them in an internal table, in response to the fact that single-precision is specified as a parameter from the application program, (B It becomes possible to perform coordinate calculations with high precision by internally applying a high-precision segment transformation matrix.

(Example) First, the overall configuration and operation of the present invention will be explained using FIG.

In FIG. 1, ■ creates a segment (circle). This means that a circle is created as a target figure to be enlarged/reduced, moved, rotated, etc. as shown on the right.

■ Inputs the transformation elements (scaling, rotation, soft). This means inputting scale (enlargement/reduction), rotation, and shift (movement) as shown in the dotted line frame on the right.

(2) Creates a segment transformation matrix. This means that double-precision and single-precision segment transformation matrices consisting of 2×3 matrices are created and stored in an internal table, as shown on the right.

(2) applies a segment transformation matrix. This is a double-precision segment transformation matrix that is applied when the segment transformation matrix application unit 2 determines that the internal table stores a segment transformation matrix that matches the Φ product segment transformation matrix specified by the application program as a parameter. Perform all coordinate calculations by applying to the segments, for example, generate coordinate values of each segment of the deformed circle with high precision (double precision) as shown on the right end 4
I am trying to do that. - Force, segment transformation (When the negative matrix application unit 2 determines that the segment transformation matrix is not stored in the internal table, coordinate calculation is performed by applying a single-precision segment transformation matrix to the segment.

↓'11 determines whether the figure is complete or not. If yes, end. If no, repeat steps after ■.

Next, in FIG. 2, which describes in detail the operation of the segment transformation matrix creation unit 1 in FIG. 1 using the flowchart in FIG. 2, (2) calculates a segment transformation matrix in double precision and sets it in an internal table. This means that the 2×3 matrix segment transformation matrix (see FIG. 4) corresponding to the deformation element manually inputted in FIG. 1 (2) is obtained in double precision.

(2) converts the obtained double precision value to single precision and sets it in the internal table.

(2) Set the segment transformation matrix reduced to single precision as a parameter of the application program. This is explained in the third section below.
This is to apply the double-precision segment transformation matrix in Figures ■ and ■.

As described above, double-precision and single-precision segment conversion matrices are set in the internal table.

The operation of the segment transformation matrix application unit 2 in FIG. 1 will be explained in detail using the flowchart in FIG. 3.

In FIG. 3, ■ determines whether the specified segment transformation matrix and the single-precision segment transformation matrix in the internal table match. If yes, set the double-precision segment transformation matrix of the internal table to a variable using ■. If no, set the segment transformation matrix specified by ■ to the variable.

■Reads the figure order from the segment.

[Phase] calculates coordinates based on the segment transformation matrix (double precision or Qj-+R degrees) substituted into variables and the graphic order.

[Phase] displays the calculation result.

(2) determines whether it is the end of the segment.

If yes, end. If no, repeat the steps after [phase].

As a result, if a single-precision/double-precision segment conversion matrix is stored in the internal table in Figure 2, even if single-precision is specified from the application program, the double-precision segment conversion matrix will be internally stored. By applying 7, it is possible to perform coordinate calculations in double precision. On the other hand, in the internal table f
If an IL precision segment transformation matrix is not stored, coordinate calculation is performed in QL precision by applying the specified single precision segment transformation matrix.

FIG. 4 shows an example of a segment transformation matrix creation routine.

Here, as input parameters, as shown in the figure, a 2×3 matrix unit matrix as a segment transformation matrix, a fixed point, a shift level, a rotation angle (radian), a scaling factor (X direction, Y direction), and a coordinate switch are given. . As an output parameter, a 2×3 matrix of illustrated segment transformation matrices is obtained. Each element M11, M12, M21, and M22 of this segment transformation matrix represents an amount of enlargement, reduction, and rotational movement as shown in the figure. Each element M13, M2
S represents the amount of parallel movement as shown.

FIG. 5 shows an example of application of the segment transformation matrix.

Here, as shown in the figure, a segment name and a segment transformation matrix (single precision) are specified as input parameters.

Correspondingly, as mentioned above, if the internal table stores a segment transformation matrix that matches the segment transformation matrix (single precision), the double precision segment transformation matrix in the internal table is applied. Calculates coordinates with precision and displays the results on the screen. On the other hand, if a negative result is not stored, a single-precision segment conversion table is applied to calculate coordinates in single-precision, and the results are displayed on the screen.

〔Effect of the invention〕

As explained above, according to the present invention, when creating a segment transformation matrix, double precision and single precision are created and stored in an internal table, and the segment transformation matrix specified in single precision as a parameter from the application program is generated. If a matrix is stored in an internal table, a double-precision segment transformation matrix is applied to calculate the coordinates.
Corresponding to the specification of single precision according to the standard GKS interface, a double precision segment transformation matrix is internally applied to perform coordinate calculations in double precision, thereby improving the single precision of graphics.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram of the configuration and operation of one embodiment of the present invention, and FIG.
The figure shows a flowchart for creating a segment transformation matrix, Figure 3 shows a flowchart for applying a segment transformation matrix, Figure 4 shows an example of a segment transformation matrix creation routine, Figure 5 shows an example of applying a segment transformation matrix, and Figure 6 shows an explanatory diagram of the prior art. . In the figure, ■ represents a segment transformation matrix creation unit, 1-1 represents a segment l-conversion table, and 2 represents a segment transformation matrix application unit.

Claims (1)

[Claims] In a graphic operation method that calculates a segment transformation matrix on a figure to enlarge/reduce/move it, etc., double-precision and single-precision segment transformation matrices corresponding to the input of transformation elements are created and an internal table is created. When the segment transformation matrix specified in single precision as a parameter matches the single precision segment transformation matrix stored in the above internal table, the segment transformation matrix is stored in this internal table. The segment transformation matrix application unit (2) performs calculation processing by applying a double-precision segment transformation matrix.
A double-precision graphic calculation method characterized by being configured to output calculation results by applying a single-precision segment transformation matrix.