JPH03219378A - Graphic plotter - Google Patents

Abstract

PURPOSE:To secure the same degree of graphic plotting quality as Phongshading by using a data holding means shifted by prescribed quantity decided based on a reference coordinate value in a prescribed direction. CONSTITUTION:A (z) gradient DELTAz/DELTAx is calculated by performing division based on the difference DELTAz of a z-coordinate value and the difference DELTAx of an x- coordinate value at both endpoints at a host processor, etc., and the (z) gradient DELTAz/DELTAx is stored in a quotient register 2, and also, a (z) coordinate value zl at one endpoint is stored in an accumulative value register 1 by controlling a selector 4. Also, the (z) gradient DELTAz/DELTAx is supplied from the quotient register 2 to an incremental value shift register 5. The content DELTAz/DELTAx of the incremental value shift register 5 is right-shifted by the complement on 2 of a component xlP less than a decimal point in the x-coordinate value xl sequentially. Also, the selector 4 is set in a state to supply the output value of an adder 3 to the accumulative value register 1 after storing the value zl in the accumulative value register 1. In such a way, plotting with high quality same as that of the Phongshading can be performed.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a graphic drawing device, and more specifically,
The present invention relates to a graphic drawing device typified by glow shading for drawing a graphic based only on integer coordinate values.

<Prior art and problems to be solved by the invention> As a conventional figure drawing method, ■ Line shading (Phong shadi)
ng), a method of drawing based on addresses expressed including data below the decimal point,
and ■ Glow shading (Gouraud sha
ding), a method of drawing based on addresses expressed only by integer values is widely known.

In the figure drawing method (■) above, all addresses are expressed including data below the decimal point, so as shown in Figure 5 (A)
As shown in FIG. 2, it is possible to accurately determine the edge line of a polygon based on the given vertex data, and moreover, it is possible to draw only the pixels located inside the edge conflict line. However, for pixels whose edge lines happen to intersect with the integer grid, only the pixels on the edge line on the specific side are drawn. Therefore, no pixel overlaps with a pixel of an adjacent polygon, and high graphic drawing quality can be achieved. However, this method has the disadvantage that all pixel data must be generated by software, which significantly increases the time required.

In the graphic drawing method (2) above, since all addresses are expressed only by integer values, processing by hardware such as a linear interpolation calculator can be easily accomplished, and the required time can be significantly reduced. However, since all addresses are expressed only as integer values, data below the decimal point must be rounded off, rounded off, etc. in advance, so as shown in Figure 5 (B), Obtaining the edge line of a polygon based on rounded vertex data will make the edge line inaccurate, and furthermore, the pixels on the edge line will also have addresses that have been rounded, so the polygon's edge line will be inaccurate. External pixels will be drawn. Therefore, the pixels of the edge portion overlap with the pixels of the adjacent polygon, and the graphic drawing quality at the edge portion is greatly reduced. In particular, depending on the drawing order, the color of the back polygon may be displayed at the matching portion of the front polygon and the back polygon, resulting in a significant loss of graphic drawing quality.

Furthermore, when performing multi-window display drawing, etc., there may be a strong demand to mix figure drawing with ■ and figure drawing with ■, since the display position of the polygon and the number of pixels within the polygon are different. In reality, mixed display is impossible.

<Object of the invention> This invention was made in view of the above problems,
To provide a new figure drawing device capable of drawing figures by outputting addresses consisting only of integer values from linear interpolation means and achieving figure drawing quality comparable to that of phono shading.

Means for Solving the Problems> In order to achieve the above object, in the graphic drawing device of the present invention, the linear interpolation means includes an increment value holding means for holding increment values to be sequentially added to the cumulative addition value. As such, data holding means is used that is shifted in a predetermined direction by a predetermined amount determined based on reference coordinate values on which interpolation calculations are to be performed. In this case, the shift amount is preferably a two's complement of the decimal value of the reference coordinate value on which the interpolation calculation is to be performed.

In the graphic drawing device described above, the incremental value holding means of the linear interpolation means is a data holding means that is shifted in a predetermined direction by a predetermined amount determined based on the reference coordinate values on which interpolation calculations are to be performed. Even if the reference side coordinate value of one end point of the line segment contains data below the decimal point, other coordinate values corresponding to the reference side coordinate value of the grid point closest to the preset side are obtained (hereinafter referred to as the reference side coordinate value). Obtaining coordinate values when the side coordinate values are integers is called compensation), and then other coordinate values corresponding to the reference side coordinate values of grid points located sequentially in a predetermined direction can be sequentially moved forward. can. In other words, if the address of the end point of a line segment contains data below the decimal point, and this line segment is an edge line of a polygon, there is the disadvantage that the edge line shifts like in conventional glow shading. will be resolved. When polygons are broken down into scan lines, the shift direction of the data holding means is set in the direction in which the end points of the scan lines are included in the polygons, allowing figure drawing similar to phono shading to be performed. It can be achieved. If the shift amount is a two's complement of the decimal value of the reference coordinate value on which interpolation is to be performed, the shift amount can be easily set. Furthermore, since the display position of the polygon and the number of pixels within the polygon are the same as when drawing using the phono shading method, when performing multi-window display drawing, etc. Can be mixed with drawing.

The case where the correction is performed on the line segment PI P2 shown in FIG. 1, assuming that the X coordinate value is the reference side coordinate value and the second coordinate value is another coordinate value, will be described in more detail.

However, as a compensation method, there is a method of compensating to an integer X coordinate value depending on the increasing direction of the X coordinate from one end point Pi,
Since there is a method of compensating to an integer X coordinate value in the decreasing direction of the X coordinate from the other end point P2, only the former method will be described below.

First, if we define the integer component of the X coordinate value x1 of the end point Pi as xll, the component below the decimal point as xlP, the 2nd coordinate value as 21, and the 2nd slope of the line segment Pi P2 as △2/△X (the 1), the X coordinate value of the integer is x Ll+1, and the 2 coordinate value 2 corresponding to the X coordinate value xll+1 is z −z l + (1−x IP) (Δ2/ΔX
) is obtained.

That is, to the X coordinate value z1 of the end point, add the value obtained by multiplying the 2's complement value of the decimal part of the X coordinate value by 2 slope Δ2/ΔX,
After that, normal interpolation calculations can be performed, pixel data similar to that in the case of line shading can be obtained at high speed, and high-quality and high-speed graphic drawing can be achieved. Note that, as is clear from the above equation, if xlP-0, the endpoint pixel will not be drawn.

<Example> Hereinafter, embodiments will be described in detail with reference to the accompanying drawings showing examples.

FIG. 2 is a block diagram showing an embodiment of the essential parts of the graphic drawing device of the present invention, in which, like the conventional linear interpolation device, an accumulated value register (1) that holds an initial value or a cumulative addition value;
a quotient register (2) that holds a quotient obtained by dividing the difference between one coordinate value by the difference between the other coordinate values; an adder (3);
It also has an increment value shift register (5) which can be shifted to the right in place of the increment value register of the conventional linear interpolation device. Furthermore, (SEL)
is a signal for controlling the selector (4), and (CK
) is the clock signal fed to the accumulated value register (1), (WE) is the write 9 enable signal for the incremental value shift register (5), and (MODE) is the right-shifted and unshifted states. This is a signal for selecting the state. Also, since this device is shown as a device for interpolating two coordinate values, the cumulative value register (1)
The output value from is supplied to an adder (3) and also to a depth buffer (6) for hidden surface processing.

The operation of the linear interpolation device having the above configuration is as follows.

Once the line segment to be drawn is determined as shown in Figure 1, a high-level processor (not shown) performs division based on the difference △Z between the two coordinate values and the difference △X between the Calculate △2/ΔX, and calculate Z slope △2/△
While storing X in the quotient register (2), selector (
4) to store the two coordinate values z1 of one end point in the cumulative value register (1).

Note that the two gradients Δ2/ΔX are supplied from the quotient register (2) to the increment value shift register (5). and·
The content Δ2/ΔX of this incremental value shift register (5) is the component xlP below the decimal point of the X coordinate value x1 of one end point.
are sequentially shifted to the right by the two's complement of .

In addition, after the selector (4) stores zl in the accumulated value register (1), it is immediately controlled in the opposite direction and becomes in a state of supplying the output value from the adder (3) to the accumulated value register (1). .

Therefore, first, the content z1 of the accumulated value register (1) and the content 2-j (△2/Δx) of the incremental value shift register (5) shifted to the right by a predetermined number of bits (j is an integer)
By supplying this to the adder (3), we can obtain the 2-coordinate value z-zl+2-j (Δ2/ΔX) corresponding to the point where the X-coordinate value has changed slightly in the increasing direction of the X-coordinate value from the end point Pi. I can do it. After that, by sequentially shifting the contents of the incremental value shift register (5) to the right corresponding to the bits whose two's complement value of xlP is "1" and adding them to the contents of the cumulative value register (1), Finally, the two coordinate values z3-zl + (1-xlP) (△2/
ΔX) can be obtained. After that, the two slopes Δ2/ΔX held in quotient register 0 are shifted again by the increment value.
It is sufficient to supply it to the register (5) and hold it as it is without shifting it to the right. Using the above two coordinate value z3 as a reference value, the two gradients Δ2/ΔX are sequentially cumulatively added to obtain the accurate value corresponding to the integer X coordinate value. Two coordinate values can be obtained sequentially.

Specific example> The above coordinate value x1. zl and two gradients Δ2/ΔX are respectively "0011.0101" and "00110.
10110""00001.01101',
If the coordinate value x3 is "0100.0000" in binary, the distance Jx in the scan line direction between the point with the coordinate value Xl and the point x3 is the decimal part xlP of Xl - "0.01
“1” in the least significant bit of the reciprocal “0.1010°” of “01”
(take two's complement), J
x-“0°1011”. Therefore, the value “0000.101101”00°00101 obtained by right-shifting the 2 slope Δ2/ΔX corresponding to the digit where the bit of JX is “1”
By adding 101"0.000101101" to the two-coordinate value z1, it is possible to obtain the two-coordinate value z3-"00111.101001111" corresponding to the point whose X coordinate value is x3.

After that, the X coordinate value increases by "1", so the increment value shift register (5) is filled with 2 slopes Δ2/ΔX-
00001,01101″ is kept as is and the above 2
By sequentially cumulatively adding the coordinate value z3, two coordinate values corresponding to the integer X coordinate value can be sequentially obtained.

Note that in the above embodiment, two coordinate values are linearly interpolated based on the X coordinate value, but it is of course applicable to linear interpolation of the X coordinate value, luminance value, etc. based on the X coordinate value. It is.

FIG. 3 is a schematic diagram showing the state in which three adjacent polygons are drawn by the figure drawing device of this embodiment. Since pixel sharing is reliably eliminated in any edge line, the figure drawing quality is improved. has improved significantly. In addition, in the edge line between the front polygon and the back polygon shown in Figure 4, only the pixels that coincidentally match the integer grid are drawn in the color of the back polygon, and the remaining pixels are drawn in the color of the front polygon. Since it is drawn in color, even if the back polygon is drawn later, the color of the back polygon is hardly drawn, and the graphic drawing quality can be significantly improved.

In the above embodiment, the incremental value shift register (5)
is described as being capable of only being shifted to the right, but if it is also capable of being shifted to the left, it can be used to solve the linear equation for determining the clip/knob area using the clip window.

<Effects of the Invention> As described above, the present invention is capable of achieving high-quality drawing comparable to that of the phono shading method while performing drawing using the glow shading method, and moreover, it is capable of achieving a high quality drawing comparable to that of the phono shading method, and is also superior to drawing using the glow or shading method. This has the unique effect of being able to coexist with drawing based on the engineering method.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram for explaining the graphic drawing operation according to the present invention, FIG. 2 is a block diagram showing an embodiment of the main part of the graphic drawing device of the present invention, and FIG. 3 is an embodiment of the diagram shown in FIG. A schematic diagram showing a state in which three adjacent polygons are drawn by a graphic drawing device, FIG. 4 is a diagram illustrating a state in which a front polygon and a back polygon are matched, and FIG. 5 (A) explains phono shading. FIG. 5(B) is a schematic diagram explaining glow shading. (5)...Incremental value shift register as data holding means

Claims (1)

[Claims] 1. In a graphic drawing device that sequentially generates addresses using linear interpolation means and writes pixel data to an image memory based on the generated addresses, the linear interpolation means holds increment values to be sequentially added to cumulative addition values. A graphic drawing device characterized in that the data holding means (5) is shifted in a predetermined direction by a predetermined amount determined based on a reference coordinate value on which an interpolation calculation is to be performed as an increment value holding means.