JPS6117189A - Graphic processor - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a graphic processing device that creates graphic data.
In particular, the present invention relates to graphic processing suitable for high-speed transfer processing of graphic data within a display memory.

[Background of the invention]

There are two methods for controlling data such as characters and figures output to a display device such as a CRT screen or a printing device, based on the difference in data stored in readable/writable display memory.
There is a street.

(1) Method for storing character codes In the display memory, characters and graphic information are encoded and stored in elementary form, and the code read out in synchronization with the display timing is used to generate a character generator, etc. Access pattern memory and obtain display data.

However, although this method has the advantage that the display memory requires a relatively small capacity, it has the problem that it is not suitable for displaying fine graphics due to the memory capacity.

(2) Method for storing pixel data as is The display memory stores display data for each pixel. In this method, pixel data for the display screen is stored.
Although it requires a large capacity display memory, it has the advantage of being able to display arbitrary graphics and highly flexible character display.

As the price of memory decreases, the use of the latter method tends to increase, but since the latter method processes large amounts of data, processing performance becomes an issue. In particular, data transfer processing for two-dimensional areas is an important issue.

A known device for developing (writing) the contents of a surface pattern table into a display memory (or frame memory) is disclosed in Japanese Patent Laid-Open No. 178285/1985.

However, in the device disclosed in the above-mentioned publication, the information that can be developed is limited by the size of the surface pattern table, and a function of transferring an arbitrary area on the display memory to another area cannot be realized.

In addition, since the specified conversion of the coordinate information of the expansion point is performed using a unique circuit configuration, the size of the pattern is limited to a unit of a power of 2, and the size of the pattern is limited to a power of 2. Information cannot be expanded to arbitrary pixel positions.

[Purpose of the invention]

The present invention was made to improve the above-mentioned drawbacks, and its purpose is to store data of multiple pixels in one word of memory, and to transfer pixel information to another pixel location. It is an object of the present invention to provide a graphic processing device which can calculate pixel positions of a transfer source (source side) and a transfer destination (destination side) according to a transfer mode and transfer graphic information at high speed.

[Summary of the invention]

The features of the present invention for achieving the above object include means for calculating a shift amount from information specifying pixel positions within one word of a transfer source and a transfer destination, and means for shifting multiple bits at a time,
Furthermore, it is supplied with a command register that stores a command code included in a command given from the outside, and part of the information of the command code stored in the command register, which is decoded to determine the current pixel coordinate position. a code decoder that generates the type and code of the operation to be performed on the code; a code register that stores the output of the code decoder; and the contents of the code register and other part of the command code. a control device that decodes a predetermined arithmetic control signal to generate a predetermined arithmetic control signal; and a control device that is supplied with the arithmetic control signal output from the control device and parameters included in a command given from the outside, and performs either addition or subtraction based on these. According to the above steps, an arithmetic device for calculating pixel drawing position coordinates on the image memory and a means for specifying a position on the image memory based on the pixel drawing position coordinates calculated by the arithmetic device are provided. At the point.

[Embodiments of the invention]

Preferred embodiments of the present invention will be described in detail below based on the drawings.

FIG. 2 shows the configuration of an arithmetic unit 1200 in a graphic processing device embodying the present invention. Conventionally, pixel data was transferred using software such as a microprocessor, but in the present invention, pixel data is transferred at high speed using a unique hardware configuration.

That is, source address register 1210, destination address register 1211, address selector 1212, reducer 81213, barrel shifter 1214
, transfer mode register 1215, address decoder 12
16, logical operation unit 1217, data buffer 1218
, display memory 16, bit mode register 1220,
Consists of.

The display memory 16 has a structure in which one word has 16 pits, and addresses are sequentially assigned. Source address register 1z1
0 and destination address register 121
1 has a 24-bit configuration, with the upper 20 bits specifying a display memory address and the lower 4 bits specifying a bit position within one memory word.

The source address register 1210 specifies the pixel location of the transfer source, and the destination address register 121
1 specifies the pixel position of the transfer destination. Address selector 1
212 selects an address to be sent to the display memory.

The subtracter 1213 is a source address register 1210 and a destination address register 1211.
Subtract the bit and calculate the shift amount of the barrel shifter.

The barrel shifter 1214 has a function of shifting 16-bit data by an arbitrary bit length at a time, and the amount of shift is controlled by the calculation result of the subtracter 1213.

Transfer mode register 1215 stores various transfer modes. The address decoder 1216 decodes the lower four bits of the destination address register 1211, performs decoding corresponding to the specifications of the transfer mode register 1215 and the bit mode register 1220, and outputs mask information specifying the operation bit position within one word. Output. The logical operator 1217 selectively performs all operations only on the bits designated by the mask information. The data buffer 71218 temporarily stores data transferred to and from the display memory 16.

Here, although not shown, the source address register 1210, destination address register 1211, transfer mode register 1215, and bit mode register 1220 are written with data and controlled by a central processing unit (CPU) or other dedicated control device. has been done.

FIG. 3 is a diagram for explaining the operation of the embodiment of FIG. 2, and shows the case of two transfer modes designated by the transfer mode register 1215. FIG. 5A shows a one-pixel transfer mode in which only one pixel of data is transferred at a time. Address selector 1212 first selects source address register 1.
210 is selected, one word of data including the source pixel is read from the display memory 16, and the data buffer 12 is read out.
18 to the barrel shifter 1214. On the other hand, a subtracter 1213 subtracts the lower 4 bits of the source address and destination address, and a barrel shifter 1214 performs a shift operation of a plurality of bits.

Next, the address selector 1212 selects the destination address register 1211, reads out data 1@ including the destination pixel position, and sends it to the logical operator 1217 via the data buffer 1218. On the other hand, the lower 4 destination addresses
The bits are decoded by address decoder 1216, and mask information specifying the destination pixel location is output. The logical operator 1217 performs a 1-replacement operation on the destination. The result of this calculation is stored at the destination address of the display memory via the data buffer 1218. By repeating this one-pixel transfer process while sequentially updating the source address and destination address, a large amount of data can be transferred at high speed regardless of memory word boundaries.

FIG. 3(b) explains the operation in the multiple pixel transfer mode. In this case, the address decoder 16 sets "1" in a plurality of bit positions according to the designation of the transfer mode register 15. Therefore, when transferring a plurality of horizontally consecutive bits, the speed can be further increased.

In this way, according to this embodiment, even when data for multiple pixels is stored in one word of the display memory, one or more pixels can be stored in three memory accesses: source read, destination read, and destination write. Pixel data can be transferred to any pixel position, and high-speed transfer is possible.

Furthermore, in the embodiment shown in FIG. 2, the pit mode register 1220 has a function that allows efficient processing even when one pixel data is expressed by multiple pits (color or multi-tone processing). Five different operating modes can be selected according to the specifications. FIG. 4 shows the structure of one word of the display memory in each mode.

(a) 1 bit/pixel mode This mode is used when expressing one pixel with one bit, such as in a black and white image.
Store pixel data.

(b) 2-bit/pixel mode This mode expresses one pixel with two pits and is used for displaying up to four colors or four gradations. One word in the display memory contains consecutive 8
Stores pixel data.

(C) 4 bits/pixel 7 to 1 pixel is expressed by 4 pits, and 1 pixel of display memory is used.
A word stores data of four consecutive pixels.

(d) 8-bit/pixel mode One pixel is represented by 8 pits, and one word in the display memory has 2 bits.
Stores data for pixels.

(e) x6 bit/pixel mode One pixel is expressed by 16 bits, and one word in the display memory corresponds to one pixel data.

FIG. 5 illustrates the transfer of one pixel data using the 4-bit/pixel mode as an example. It is necessary to read one word of data containing a source pixel and embed only the source pixel data in the destination pixel position.

FIG. 6 shows the flow of one pixel transfer process in the 4-bit/pixel mode. First, one word of the display memory 16 containing the source pixel is read out and temporarily stored in the data buffer 1218. On the other hand, the lower 4 pits of address information specifying the source pixel and the destination 7
The lower four bits of the address information specifying the green pixel are subtracted.

This value represents the difference in pit position between the source pixel and the destination pixel. The source read data is shifted by barrel shifter 1214 and source pixel (C,)i
-1: Aligned to the destination pixel position.

Next, one word included in the destination pixel (Ca) is read out, and the logical operator 1217 reads out the word included in the destination pixel (Ca).
) is calculated.

At this time, since '1#' is generated only at the destination pixel position as mask information, only one destination pixel is updated and write data is obtained. The type of logical operation can be replacement, logical operation, etc. In cases other than the 4-bit/pixel mode, exactly the same calculations are performed, only the format of the mask information is different.

As described above, according to this embodiment, even when data of one pixel is expressed by a plurality of pits, source reading, destination reading, destination writing,
This has the advantage that pixel data can be transferred to any pixel position with three memory accesses.

Further, FIG. 9 shows a pointer movement direction (8D) of a rectangular area transfer command as an application example of the present invention. Various pointer scans are possible to perform pixel-by-pixel processing. <
Eight movement directions, a> to (h), are defined and can be specified independently for the source area and destination area. Therefore, various combinations of transfer methods are possible. FIG. 8 shows a transfer example in which 8D=1 is specified for the source area and 5D=5 is specified for the destination area.

Next, methods for performing these various pointer scans will be explained in detail. That is, the operation of each part of the graphic processing apparatus shown in FIG. 1 will be explained using a rectangular area transfer command as an example.

FIG. 1 shows a schematic configuration of the entire graphic processing apparatus according to an embodiment of the present invention. As is clear from the figure, the graphic processing device of this embodiment includes a command register 10, a control device 1
1, an arithmetic control unit 1200, a code decoder 13, a code register 14, a mode decoder 15, and a display memory 16.

The command register 10 temporarily stores a command code COM among commands transferred from the outside.

Part of the information in the command code COM is transferred to the control device 11, decoded, and arithmetic control is performed according to a predetermined processing algorithm.

The arithmetic unit 1200 executes coordinate calculations for calculating drawing addresses of figures and processing of figure data, and its detailed configuration is shown in FIG. The code decoder 13 receives commands and
Predetermined code data according to the present invention is generated based on part of the information of the code COM and information provided from the arithmetic unit 1200.

The code register 14 temporarily stores code data generated by the code decoder 13. The mode decoder 15 decodes the processing mode field of the command to control arithmetic processing. The display memory 16 stores graphic information or data in units of pixels.

Table 1 shows a list of commands interpreted and executed by the graphic processing device of this embodiment. Each command consists of a command code of 1 word (16 bits) followed by several words of parameters, but some commands do not require any parameters at all.

Commands are classified into the following three types.

(1) Register access commands This is a group of commands that control reading and writing to registers inside the graphic processing device.

<2) Data transfer commands A group of commands for controlling data transfer to display memory or reading from display memory.
1(3)
Graphic commands are commands that control the generation of various graphics. This table shows 23 types, including straight lines, rectangles, groups of straight lines, polygons, circles, circular arcs, ellipses, elliptical arcs, and filled shapes.

Since these commands are well known to those skilled in the art, a detailed description of them will be omitted.

Next, the operation of the embodiment shown in FIG. 1 will be explained in detail by taking a typical copy command as an example.

Figure 7 shows the format of the copy command.As is clear from this figure, the copy command consists of a 1-word (16-bit) command code followed by 4-word parameters 1 to 4. Become.

The copy command is a temporary command that transfers data in a rectangular area (source area) in the display memory 16 to another area (destination area).
By setting the code and parameters, various pointer scanning directions can be selected during transfer.

FIG. 8 shows a conceptual diagram of an example of the operation of the copy command. Parameters Xs and Ys define the starting point coordinates of the source (transfer source) area 168.

On the other hand, parameters DX and DY define the direction and size of the area. That is, based on the starting point coordinates,
If DX is positive, move to the right; if DX is negative, move to the left.
Also, if DY is positive, the area is upward, and if DY is negative, the area is downward.9, DX.

Its size is specified by the absolute value of DY.

The S bit of the command code (FIG. 2) is a bit that defines the order of source area scanning. If S is O”,
Scanning is given priority to the horizontal direction, and scanning is performed in the vertical direction every time one line in the horizontal direction is completed. When S is '1n, on the other hand, scanning is performed with priority given to the vertical direction.

Therefore, the S bit of the command code and the parameter DX
, DY, the source region 168 can be scanned in eight ways, as shown in FIG.

That is, FIG. 9 (→ indicates S=O, DX) O.

In the example of DY)O, the starting point (Xs,
Ys), it horizontally scans in the right direction by a distance of DX as shown by the arrow, then moves up one line and performs the same horizontal scan. Then, after repeating horizontal scanning for DY lines, the end point (indicated by a circle in the figure) is reached and the process ends.

FIG. 9(b) shows that S=0. In this example, DX>O, DY (O), from the starting point (Xg, Ya), horizontally scan the distance DX in the right direction as shown by the arrow, and then horizontally scan one line below in the same way. , the end point is reached after horizontal scanning is repeated for DY lines.

Fig. 9(C) is an example of S=O, DX<O, DY>0, where horizontal scanning is performed from the starting point (Xs, Ys) by a distance of DX to the left as shown by the arrow, and then Move up one line and perform horizontal scanning in the same way. Then, the end point is reached after the horizontal scanning has been repeated by DY lines.

In Fig. 9(d), S = O, DX (0, DY (in the example of 0, reed), from the starting point (Xs, Ys), horizontally scan a distance of DX in the left direction as shown by the arrow, and then Horizontally scan the lower side of 12 ins in the same way. Then, after repeating the horizontal scan for DY lines, the end point is reached.

FIG. 9(e) shows that S=1. DX>0. In the example of DY>0, from the reed starting point (X[l, Ys), vertical scanning is performed upward by a distance of DY as shown by the arrow, and then it is moved one line to the right and vertical scanning is performed in the same way. Then, the end point is reached after repeating the vertical scan by DX lines.

FIG. 9(f) shows S=1, DX)0. DY (in the example of 0, reed, from the starting point (Xs, Ys), vertically scan downward by a distance of DY as shown by the arrow, then vertically scan one line to the right in the same way. Then, vertically scan by DX line. The end point is reached when the scan is repeated.

In Fig. 9(g), S = l, DX (0, DY, >00 example), from the starting point (Xs, Ys), vertically scan upward by a distance of DY as shown by the arrow, and then The left side of one line is vertically scanned in the same way.The end point is reached when the vertical scanning is repeated by DX line.

FIG. 9(h) shows that S=1. DX<0. In the example of DY<0, from the starting point (Xs, Y-), vertical scanning is performed downward by a distance of DY as shown by the arrow, and then one line to the left is similarly vertically scanned. Then, the end point is reached after repeating vertical scanning for DX lines.

On the other hand, in the destination area, the starting point is the current drawing point coordinates (X, Y), and there are eight scanning directions as shown in Figure 10, depending on the 3-bit value of the DSD quield in the command code. selected.

(a) in the same figure is when the DSD is ``000'', and the same scanning as in FIG. ””
In the case of 100"'101"'110"'111", the same scanning as in (b) to (h) of FIG. 9 is performed.

In the embodiment of FIG. 1, the first word in the command transferred from the outside is gmd as a command code,
The number is placed in the command register 10.

Control device 1 according to the top 4 pits of the command code.
Control of copy processing is started at 1. The S bit and DSD field in the command code (see Figure 7) are:
The signal is sent to the code register 14 via the code decoder 13.

The lower processing mode field of the command code is decoded by the mode decoder 15.

Parameters 1 to 4 are sequentially sent to a primary register (not shown) inside arithmetic device 1200. Current drawing point coordinates (
Although the registers for storing the values (X, Y) are not shown,
It is provided inside the arithmetic device 1200.

FIG. 11 is a layout diagram showing an example of the configuration of the code register 14 according to the present invention. This sign register holds the next 10 bits of information. In this figure, the information stored in each register section is indicated by an arrow, taking the case of a copy command as an example.

(1) Ql This is the first bit for switching the coordinate registers X and Y. The copy command is used to completely determine the priority order of scanning in the X direction and Y direction in the source area 168. For this reason, 1 as Ql is the command code S
Bit is set.

When Ql-0, select the X register, select the X register as specified, and when Ql = 1, select the X register when X is specified,
When X is specified, each X register is selected.

(2) Q2 This is the second bit that switches the coordinate registers X and Y. In the copy command, destination area 1
It is used to switch between the X direction and Y direction of scanning in 6D. For this purpose, bit 2 of the DSD field (FIG. 2) of the command code is set as Q2.

When Q2=0, the X register and the X register are selected as designated, and when Q2=1, the registers are selected by inverting the designations of X and Y.

(3) S1x This is 2-bit information that holds the arithmetic code in the first X direction. Generally, the upper bit of the two bits is used to select addition or subtraction, and the lower bit is used to perform the addition or subtraction (the bit is only '1'). This is to select whether the bit is 0'').

In the copy command, the sign of the parameter DX is set in the upper bit, and l'' is set in the lower bit. That is, the upper bit is used as information specifying the operation sign in the X direction of the source area 168.

(4) siy 2-bit information that holds the operation code in the first Y direction,
This is for selecting an operation like the above-mentioned 81x.

In the copy command, the sign (upper bit) and '1'' (lower bit) of the parameter DY are set, respectively, to specify the operation sign in the Y direction of the source area 16S.

(5) 82x This is 2-bit information that holds the second X-direction arithmetic code, and specifies the X-direction arithmetic code of the destination area 16D in the copy command. Bit 1 of the DSD field of the command is set in the upper part, and bit 1'' is set in the lower part.

(6) S2y This is 2-pit information that holds the second Y-direction arithmetic code, and the copy command specifies the Y-direction arithmetic code of the destination area 16D. Bit 1 of the DSD field of the command is set in the upper part, and '1' is set in the lower part.

To summarize the above, 5IXI sly, S2X.

82y can take four different states, and 00"
and '10', add or subtract '0', respectively (in other words, do not perform any processing), '01#', add '1', and '11' subtract @ln. It turns out.

Note that, as described above, in the copy command, each register section S lx + 81)' + 82x of the code register 14
, 82y are all set to 1'', but these bits may be controlled and changed in commands for other processing.

FIG. 12 shows the processing flow of a copy command. The contents of the expression specifying each register are as follows.

(1) Xs (Ql) Specifies the Xs register when Ql-0 and the Ys register when Ql-1. That is, the first
Or coordinate values in the priority scanning direction.

(2) Ys (Ql) When Ql-0, set Ys register, and when Ql-1, set Xg register.
Specify a register. That is, it is the coordinate value of the source region 168 in the second scanning direction.

(3) X (Q2) Q2=O specifies the X register, and Q2=1 specifies the X register. That is, it is the coordinate value of the destination area 16D in the first or priority scanning direction.

(4) Y(Q2) Specify the X register with Q2-0 and the X register with Q2-1. That is, it is the coordinate value of the destination area 16D in the second scanning direction.

(5) 5lx(Ql) Select Six when Q1=0 and select Sly when Q1=1. That is, it is the code of the first (priority) scanning direction of the source region 168.

(6) 5ly (Ql) Select Sly when Q1=O and select SIX when Qt=1. That is, it is the code of the source region 168 in the second scanning direction.

(7) S 2 x (Q 2 ) Select 82x when Q2=0 and select S2y when Q2=1.・
This is the code of the first (priority) scanning direction of the destination area 16D.

(8) S2y (Q2) Select S2y with Q2-0 and 82X with Q2=1. This is the code in the second scanning direction of the destination area 16D.

Generalizing the above, in the process shown in Figure 12, (i) Xi (Qj) specifies the (i) register when Qj is 0'', and specifies the Yl register when Qj is 61''. (2) Y i (Q j ) specifies the Yi register when Qj is '0'', and specifies the Xi register when Qj is 1#, (3) S i x (Q j ) selects the Six register when Qj is '0', and selects the Six register when Qj is '61', and (4)8iy(Qj) selects the Six register when Qj is '0'.
This means that all registers are selected, and when Qj is '1'', Six registers are selected.

When a copy command is input, the control device 11 sequentially outputs control information according to the flow shown in FIG.

The control information is stored in a suitable memory (not shown) inside the control device 11, for example as a microprogram.

Parameters 1 to 4 following the command code in the copy command, namely Xs and Ys.

DX and DYk are input sequentially and stored in a register inside the arithmetic unit 1200 (step 81).

Then, processing for one line in the first (priority) scanning direction begins. For this, Xs(Ql) and X(Q2),
That is, the start coordinate values of the source area 16S and destination area 16D in the first (priority) scanning direction are saved (temporarily stored) in another register (step 82).

Next, pixel information k of the coordinate point specified by (Xs, Ys) is transferred to the coordinate point specified by (X, Y) (
Step 83). This process of transferring one pixel is the same as that already described for the arithmetic unit 1200.

Next, coordinate values X5 (Ql) and X(
Q2), code values S 1 x (Ql ), S 2x
(Q2) is added. That is, the specified coordinate points of each area are each moved by one pixel in the first scanning direction (step 84).

The processes of steps S3 and S4 are repeated until the specified coordinate point reaches the end point of one line (step S5
).

When the processing for one line is completed and the determination in step S5 is established, X s (Ql), X (Q2
) in step S6), the second scanning direction coordinate value Y
[1 (Ql) and Y (Q2) each have a code value Sl
y(Ql)S2y(Q2) is added to set the starting point coordinates of the second line (step 87).

All data in the source area 168 can be transferred by repeating the processes in steps 82 to S7 described above until all line processing in the second scanning direction is completed - until the determination in step S8 is satisfied.

As mentioned above, in this embodiment, various pointer scanning modes during area data transfer can be expressed in a single processing flow shown in FIG. It can be simplified.

Furthermore, as another example of the command, a pattern command will be described below.

FIG. 13 shows the format of the pattern command.

As is clear from the figure, this command consists of 1 word, 16
It consists of a pit command code and a one-word parameter.

The ``p-y'' command executes the pattern processing inside the graphics processing device.
This is a command to expand pattern information stored in memory onto display memory. In selecting the mode of operation of the command, the pointer can be caused to perform various scans.

Parameter 1 (Sy, sx) in this case specifies the size of the area in which the pattern is developed, and is the lower byte).
(SX) and upper byte (Sy) define the number of times the pointer is repeated in the first (priority) scanning direction and the second scanning direction, respectively.

Figures 14 and 15 show the scanning direction of the pointer.
A total of 4 pits (bits) can be set, and 16 scanning modes can be selected.

That is, Fig. 14 (a) to (h) shows the case where sL is '0'' and 8D changes from ooo'' to '111''.
Figure 15 (a) - 8 different scanning directions of the pointer
(h) is 5L75E"l" and is the scanning direction of the pointer in eight directions when sD changes from o o o" to '111".

In FIGS. 14 and 15, the black dots indicate the starting points of scanning, and the white circles indicate the ending points of scanning.

FIG. 16 shows the configuration of the - sign register 14 and the values set in the sign register in the pattern command.

In the case of a pattern command, bits 11 to 8 of the command code (that is, a total of 4 bits of SL and SD described above) are decoded in the code decoder 13 based on a predetermined logical formula, and the result is the 11th bit. As shown in the figure, each code bit position Q 11 Q of the code register 14
21 S I X IS 13', S 2x,
82 YK Cef Tosa t'L.

Table 2 shows a list of the values of the code register 14 in various cases of scanning mode pits. Note that the * mark in the same table indicates that either "0" or "Pa1" may be used.

Each sign bit in FIG. 11 and Table 2 has the following meaning in the notch turn command.

Table 2 (1) Ql, Q2 Since they are not used in pattern commands, e'0'' is always set for each.

(2) Slx, S1y serve to control the increment of the X and Y coordinate values in the first scanning direction.

(3) S2x, S2y Controls the increment of the X and Y coordinate values in the second scanning direction.

FIG. 17 shows the processing flow of pattern commands.

First, the parameter 1 - that is, S)'+SX following the command code is input and stored in the register inside the arithmetic unit 12 (step 511).

Next, processing for one line in the first scanning direction begins.

First, the starting coordinate values (X, Y) of the pointer are saved in another register (Stella 7'812).

Then, as a drawing process for one pixel, the information (pattern bit) in the pattern memory is used to draw the image in the display memory.
Drawing is performed at the pixel position specified by the start coordinate values (X, Y) (step 513).

Next, code values slX and Sly are added to the coordinate values X and Y of the coordinate pointer, respectively, and the results are used as new X and Y coordinate values. By this process, according to the contents of Slx, 81)', the
The display position will move by one pixel (step 51).
4).

For example, 51x-"01" (1 addition), 5iy='''
00" (O addition), the coordinate point moves to the right. Also, 5IX - "11" (1 subtraction), 51y = "'01
” (add 1), the coordinate point will move toward the upper left.

The processing in steps 813 and 814 is repeated for one line (the number of times specified by the lower byte of the parameter). When the processing for one line is repeated, the line end determination at step 815 is established.

Thereafter, in step 816, step 81
The coordinate values X and Y saved in step 2 are restored and added to the code values S2x and 82Y', respectively.

As a result, the coordinate point is moved by one pixel in the second scanning direction, and is set as the starting point coordinate of the second line (step 517).

The processing of steps 812 to S17 is performed in step 818.
Desired pattern information can be drawn in a predetermined area by repeating the process until the completion determination is satisfied (in other words, the number of times specified by the upper byte Sy of the parameter).

In other various commands as well, the operation register and operation contents are specified in accordance with the contents of the code register, thereby facilitating pointer scanning operations.

Those skilled in the art will readily understand that the use of sign registers also facilitates the description of algorithms in the case of commands for generating straight lines and curves such as circles.

〔Effect of the invention〕

As described in detail above, according to the present invention, coordinate values for various calculation modes can be easily calculated with a small amount of control information, and when data of multiple pixels is stored in one word of display memory. This also has the advantage that high-speed data transfer within the display memory is possible.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, FIG. 2 is a diagram showing the internal configuration of the arithmetic unit shown in FIG. 1, and FIG.
The figure is an explanatory diagram of data transfer with respect to Fig. 7, Fig. 4 is an explanatory diagram of pit/1iii element mode, Fig. 5 is an explanatory diagram of pixel data transfer, and Fig. 6 is an explanatory diagram of pixel data transfer. , Figure 7 shows the format of the copy command.
, FIG. 8 is a conceptual diagram showing an example of the operation of a copy command, FIGS. 9 and 10 are conceptual diagrams for explaining the scanning direction of the pointer in the source area and destination area, respectively, and FIG. FIG. 1 is a layout diagram showing an example of the internal configuration of the code register, FIG. 12 is a conceptual diagram for explaining the copy command in an embodiment of the present invention, and FIG. 15 is a conceptual diagram for explaining the scanning direction of a pointer for a pattern command. , FIG. 16 is a diagram showing an example of the configuration of the code register and information set in the code register when a pattern command is executed, and FIG. 17 is a processing flow diagram of the pattern command. 1210...Source address register, 1211...
・Destination address register, 1213・
... Subtractor, 1214 ... Barrel shifter, 1216.
...Address decoder, 1217...Logic operator, 1
0... Command register, 11... Control device, 12
00... Arithmetic device, 13... Code decoder, 14.
...Sign register, 15th mouth 3rd mouth 40 1st thought〒゛-ta〒〒゛-ta〒〒〒゛-ta〒〒〒〒〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉〒〉 -ri 4 bY bracket 80.6 謬1 is 謬 110! 4 130th bit b1t command code 11011sD Processing mode parameter I
SY sX fan 1
20 凋 1t + mouth foil IS Kasumi full'7 mouth

Claims (1)

[Claims] 1. A command code included in a command given from the outside in a graphics processing device that controls the generation and transfer of graphics data for each pixel onto an image memory that stores graphics data in units of pixels. a first means that is supplied with some information and decodes it to generate a type and code of an operation to be performed on the current pixel coordinate position; and an output of the first means and other information of the command code. provided some information,
A control means that decodes this to generate a predetermined arithmetic control signal, and is supplied with the arithmetic control signal outputted from the control means and the parameters included in the command given from the outside, and based on these, the information is stored in the image memory. 1. A graphic processing device comprising: a calculation means for calculating pixel drawing position coordinates; and means for specifying an address on an image memory based on the pixel drawing position coordinates calculated by the calculation means. 2. The graphic processing device according to claim 1, wherein the operation is either addition or subtraction. 3. The first means is supplied with a command register that stores a command code included in a command given from the outside, and information on a part of the command code stored in the command register, and decodes the command register. Claim 1 or claim 1 further comprising: a code decoder that generates the type and code of an operation to be performed on the current pixel coordinate position; and a code register that stores the output of the code decoder. The graphic processing device according to any one of Item 2. 4. The code register is a register that stores at least a reference pixel coordinate position on the image memory, a current pixel coordinate position, the type of operation to be performed on the current pixel coordinate position, and whether or not the operation can be executed. A graphic processing apparatus according to any one of claims 1 to 3, characterized in that: