JPS59181387A - Graphic generator - 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] [Technical Field] The present invention relates to a graphic generation device, and more specifically, to a device that displays or prints graphics such as characters and images in a raster scanning format, in which a font memory is arranged in a bank configuration to provide multifunctionality. This invention relates to a graphic generation device that enables display or printing.

[Prior art]

Conventionally, in devices that display or print figures including characters, images, etc. in a dot matrix, it is common to display or print only that figure for one piece of figure information. In other words, each dot pattern stored in the font memory is read and output when the corresponding code information is input, and the shapes that are not stored in the font memory are read out and output. Multiple dot patterns are input from the outside, stored once in R and AM, and then output. Therefore, various figures, for example, composite characters such as rAJ (Angstrom), which combines the character rAJ and the symbol "0", characters with black and white inverted, 90
If you want to display or print characters rotated by 180° or 180°, characters enlarged or reduced, or figures that are not used elsewhere, you must design each time and store the don't φ pattern in the font memory. Alternatively, these graphics must be manually stored in RAM, requiring replacement of font memory or large-capacity bit map memory in page units.

〔the purpose〕

The purpose of the present invention is to solve such conventional problems, and to solve the problems of the prior art,
Another object of the present invention is to provide a graphic generation device that can optionally display or print new graphics, etc., and can expand its functions.

〔composition〕

Hereinafter, the configuration of the present invention will be explained using examples.

FIG. 1 is an overall block diagram of a printer controller showing an embodiment of the present invention.

The printer controller includes a CPU 3 that performs overall sequence control, a program memory (ROM) 4, a work memory (R, AM) 5, and a font memory 10 that stores character patterns corresponding to character codes. and,
There are two types of overlapping graphics (including text and images), such as the interface that connects the face 2 and the plotter 8 in printing units of the plotter 8.
One is character composition (multi-strike), which is a function that prints a character on paper and prints another character in the same position to create a new character that the printer does not own as a graphic. The other method is 7-om synthesis, in which the same information (e.g., ruled lines, company name, date, title, mark, etc.) is fixedly printed on a series of printing paper, and the information is printed continuously. This is a function that holds this state and repeatedly presses the button until printing is completed.

In the printer controller shown in FIG. 1, the figure code to be overlapped with the base figure code can be stored in one page group memory 12, so that it can be used for both the above-mentioned character composition and form composition. If the plotter 8 in FIG. It prints images on sheets in rask scanning format, and its resolution is main scanning,
Both sub-scans are "/300'. Rask scan is performed with a laser beam on the photoreceptor, and the laser beam is controlled to be on and off by a modulator. The modulation signal is sent from the printer controller, etc. A modulated laser beam is supplied as a serial image signal to form an electrostatic latent image on the photoreceptor.Toner is attached to the electrostatic latent image and transferred to printing paper, thereby printing the image. be done.

The printer controller shown in FIG. 1 is located between the plotter 8 and an external device (host machine, computer, etc.) 1, and is physically coupled to the external device 1 at an external connection interface portion. The external device 1 sends character codes and control codes (for example,
ASCII code or JIS code), the printer controller selects the character bataso in the font ψ memory 15 in the controller while controlling the printing position of the character code to be printed based on the control code,
It is sent to the plotter 8 as an image signal and printed.

The printer controller shown in FIG. 1 is characterized by the following points. (1) In order to perform high-speed printing, a buffer memory (page memory) for two pages is provided in the controller, and while one is used as an output buffer to the plotter 8, the other is By using it as an input buffer from the host machine 1 and alternately switching between two pages of buffer memory, printing can be performed without falling behind the printing speed of the plotter 8, which is 12 PPM. (it
) In order to perform both portrait/landscape printing, the order in which character codes are stored in the page memory 12 and the order in which character patterns are read out in the font memory 15 are switched for boat rate, portrait, and landscape printing, respectively. Can be controlled on a page-by-page basis.

Note that the boat rate is a page printed in the direction of a portrait, etc. shown in Figure 2(d), and the landscape is a page printed in the direction of a landscape painting, etc., shown in Figure 2(e). . 1ii) Due to the selection of 2 infeeds, the line feed width has basic modes of 1/8' (8LPI) and 1/' (6LPI). 1/g'L F (line feed) is 38 dots/line, 115'L
F is composed of 50 dots/line or 48 dots/line. 17B 'L F in page memory 12
A control flag for selecting 115' LP is provided, and the reading range of the font memory 15 is controlled based on this flag. Note that 1/s' L
Printing position control (PLU
: Partial Line Up, PLD
: Partial Line Down) is possible,
This is used for superscript printing and subscript printing. Also, "/4'LF (4LPI)" is 115
' Using the basic mode of LP, select Printer 1.
l-1 Automatically within the controller 1/6'//4"L
2 Controlled as F. ”/$' LP, 1/6
'Lp, 1/4' L FPLD control is set in each page memory 12 as a control flag for each character.

Ov) Character pitch control page - Memory mapping is formed with a space width of "/60' and a line width of 1 line (8LPI, 6LPI), and a character width of /6o' (v=i, 2, 3).・・・・・・
) can be controlled. Therefore, the character width is 1/i o'
t 1/+ 2″, 1/15″+ 1/20″, P
, S (Proportional 5pace), etc. can be mixed.

(■) Font selection The printer controller can store character patterns of up to 5 fonts in units of 128 characters/font, for a total of 640 characters.

128 characters/font is called one font bank, and each font can be freely selected from the host machine 1 using a control code, and each font can be switched character by character. Also, in the 5 font bank, "/ 't"/ '10
Any font size of 12''/'9'/20', P, or S can be arbitrarily set, and normally a font memory 15 using ROM is set in each font bank.

(Yl) Character synthesis, page synthesis The structure is such that the buffer memory for two pages can each store character codes for two characters, and normally only one character is used, but character synthesis ( For example, A+ -A, a
+ = a, etc.) to create a new character, store two characters in page memory 12 and font memory 15.
can be accessed twice to synthesize a character pattern and output it as an image signal.

The part that stores character codes that are normally used is called the main character area, and the part that stores character codes that are used during composition is called the subcharacter area. In other words, character synthesis is called multi-strike, and the graphic code is stored in the page memory 12,
This is a function that stores another figure code in the same location and outputs it at the same time when outputting to create a new figure.

Page compositing, also called form overlay, is a process in which the same information (e.g., ruled lines, company name, date, title, etc.) is retained on a series of printing sheets in a semi-fixed manner until the series of printing is completed, and then used repeatedly. It is a function.

(Two types of underline synthesis underline and double underline are available.) Each page memory 12 can be set as a control flag, and when reading the font memory 15, the underline is synthesized using this control flag. do.

- Font download and font conversion By providing RAM as the font memory 15 in place of the font memory 15 using ROM in five font banks, character codes and their character patterns can be read from the host machine 1 as data. , AM, and after the setting is completed, it becomes possible to print using the defined character code.

Still, ROM and RAM are mixed in the font bank, and based on commands from host machine 1, the graphic data stored in ROM is processed (stacked, reversed, rotated) in the printer controller and transferred to RAM. It can also be treated as a new character and printed.

OX) Data recovery when a jam occurs A page that holds one page's worth of print information in a two-dimensional array.
In addition to the memory 12, by providing an auxiliary memory in the work memory area of the CPU that stores the serial code string transmitted from the host 1 as backup data in the form of a staff memory, troubles such as jams in the plotter 8 can be avoided. If a jam occurs, the position of the jam is analyzed using the status information from the plotter 8, traced back to the output print paper that became invalid, read the code string from the auxiliary memory, and again stores the print information in the page memory 12 in a two-dimensional array. and started printing, and the grotter trouble occurred on the host machine 1.
recovery without intervention.

Based on the control code transmitted from the host machine 1, the CPU 3 converts the character code into an internal character code corresponding to the character pattern storage address of the font memory 15 and stores it in one page memory 12. When character codes for one page have been stored, the data bus leaves the control of CPU 3 and transfers to DMA (Direct Memory A).
ccese) is under the control of counter 17. The page memory 12 and font memory 15 that come under the control of the DMA counter 17 become the character generation section 6, and are controlled independently of the CPU system until printing of one page is completed.

The character generation portion 6 is connected to the main scanning clock (DA) of the plotter.
TA CLOCK) and sub-scanning clock (LINESY
NC) generates an address for the page memory 12, and reads print information (character selection, unit selection, font selection, various control flags, etc.) from the page memory 12. The read print information and the font memory address where the DMA counter 17 generates the font
Memory 15 is accessed to obtain character pattern data corresponding to the character code.

The character pattern data is subjected to parallel-to-serial conversion and synthesis control by the shift register 18, and is sent as a serial image signal to the plotter in synchronization with the main scanning clock, and becomes a signal for modulating the laser beam. Then, the plotter 8 prints on a screen-by-screen basis (page).

The interface on the plotter 8 side is provided to control error information, paper size information, and start/stop of the plotter 8, and the effective time for one main scan is still limited to LINE G.
ATE signal, valid time of one page is FRAME GAT
Determined by E signal.

Since the page memory 12 is prepared for two pages, one side can be used as a character generation part and an output buffer for the plotter 8, while the other can be used as an input buffer for the host machine 1, and each can be used alternately. use.

The page memory 12 has a space width of 1/60', the basic unit of graphic information defined by character codes, and a space width of 1 line (8LPI).
, or 6LPI) width can be mapped as one control unit, and one control unit corresponds to the graphic information storage start address of the font memory 15.

Generally, character codes take discrete values, so they cannot directly correspond to graphic information storage addresses. Therefore, the work memory 5 of the CPU 3 is provided with a code conversion table for converting the character code and the corresponding 7-ont memory 150 graphic information storage start address. The character code is stored in the page memory 12 using the internal character code (figure information storage start address) using the code conversion table.

Character codes and control codes (carriage return: CR, line feed: LF, 7 ohm feed: FF) are transmitted dimensionally from the external device 1 as a serial code string (code string). P
U3 has a function of two-dimensionally arranging character codes in the page memory 12 based on control codes.

Since the laser printer 8 prints indirectly on paper via a photoconductor, there is a small difference between the time when an image signal is recorded on the photoconductor and the time when it is transferred to paper. Since it takes time to transport the paper from the printer to the paper ejection unit, the print information in the page memory 12 is no longer necessary until the output print paper on which printing is finally completed is obtained. Therefore, the print information on the page memory 12 will be destroyed in order to prepare the print information for the next page, and if a jam occurs in the plotter, even if the jammed paper is removed and printing is resumed, a series of prints will continue. Pages will be missing.

Therefore, the code string transmitted from the host machine 1 is held in the work memory 5 in a stack format (-dimensional array) so that it can be recovered to the worst jam position (paper discharge section jam) (3 pages before).

The plotter 8 outputs three types of jam signals as status signals: a paper feed jam, a transfer jam, and a paper discharge jam.9 Depending on the position of each jam, three pages of printing paper may be invalidated in the worst case. There is a possibility that it will happen.

FIG. 2 is an explanatory diagram of the configuration of the font memory in FIG. 1.

The pixel density of the plotter 8 is 300 D P I (Dot
por Inch), and the character sizes used for general printers are 1/10' width, 1/12' width,
/1s' width, 1/2o'' width and base characters (P, S) are reeds, and the size of the dot matrix for each character size is as follows.

1/1o' character (10CPI) → 30X48 Dot 1/12' character (12c, PI) → 25 X 48 Dot "/15' character (15CPI) → 20 X 48
Dot 1/20' character (2ocpI) - + 15X4
8 dot P, S, letter -+5v
X48 dot (V = 3.4, 5, 6, 7.8) Figure 2 (a) is p, s, the sharpest letter 40
Showing an example of ``48 dots ρ character matrix country'',
Configure this size as the largest matrix of one character. Matrixes smaller than the largest matrix are left-justified, and the remaining parts are left blank to form a character matrix.

5 dots, which is the least common multiple of each character size in the width direction, is 1
Define as a unit. One unit has 5 dots (and since the dot density of the plotter 8 is 300 PPI, the actual size is "/300'-"/60'.

Therefore, all characters are controlled to be printed at integral multiples of 1/60'.

As shown in FIG. 2(F), each character size is made up of a maximum of 48 dots in the height direction, and 8 dots is defined as one block, making the whole into 6 blocks.

One unit and one block are defined as a small matrix of character matrices.

As described above, the body size of each character is composed of the maximum matrix of 40 x 48 dots regardless of the character size, but the size of the designed character differs depending on the character size.

Font memory 15 is 8 bits (I Byte
) uses a 54kbit ROM that can be accessed in units of . A small matrix of 1 unit/1 block (5 x 8 dots) is divided and stored in 5 ROMs, and when printing, 5
By accessing multiple ROMs at the same time, all data (40 dots) in a small matrix of 1 unit or 1 block can be read out at once, and the necessary bits for portrait and landscape can be selected within the small matrix. There is. In other words, when printing, in the case of the portrait (printed in the width direction of the paper) as shown in Figure 2(d), the character mark is
5 dots in the direction of the IJ sock unit (1 unit)
In the case of the landscape shown in FIG. 2(e) (printing in the longitudinal direction of paper), 8 dots (1 program, ts) are read out in the block direction, parallel-to-serial conversion is performed, and output as a serial image signal to a plotter.

As shown in FIG. 2(C), one unit and one block correspond to each *I Byte (Bit 7 to nit o) of five ROMs (chip l to chip 5), and one character consists of 48 bytes x 5 chips. = 240 bytes. However, if the first address for character storage is set every 2' (64), it is convenient for DMA, so in the ROM, 64 bytes x 5 chips - 360 bytes essentially constitutes one character. 64 kbit R as font memory 15
Since 5 OMs are used to store 1 font,
The number of characters in one font is 128 characters (including blank printing). The internal character code of 128 characters can be expressed in 7 bits, and the character code transmitted from the host (AS
CII code, JIS code) are generally not defined continuously as 7-bit or B-bit codes, but are defined intermittently, so the character code transmitted from the host machine is processed by the CPU as an internal character code, that is, the character storage start address. (7 bits) and handle it.

For ease of DMA control, the memory space per character is determined every 2' (64) addresses, but by defining it every 48 steps, it is possible to store 128 to 170 characters with five fi54kbit memory chips. be able to. However, in this case, the internal code of the character)” h
A 7-bit adder, a 9-bit adder, and a 10-bit adder are required.

FIG. 3 is an explanatory diagram of the configuration of the page memory in FIG. 1. FIGS. 3(a) and 3(b) show the configuration of the page memory. The CPU 3 converts the character code sent from the host machine (external device) 1 into an internal character code and stores it in the page memory 12. At this time, line break instructions,
Line width designation, character selection, etc. are performed based on control codes. When storing the character internal code in the page memory 12,
The storage order is different between boat rate and landscape mode, as shown in the figure. The page memory 12 is 1 unit wide (1 unit wide) so as to correspond to the font memory 15.
/60 inches) is controlled as one space.

As shown in FIG. 3(a), at the boat rate, the character direction is the main scanning direction of the raster scan of the plotter 8, and 51
2 spaces, approximately 8.5 inches (1/60' x 512
It has a width of 8.5 square meters, the row direction is the sub-scanning direction of the raster scan, and printing for 96 lines is possible, and 512 spaces x 96 lines constitutes the maximum printing area of the page memory 12. As shown in FIG. 3(b), in landscape mode, the character direction is the sub-scanning direction of the raster scan of the plotter 8, and the shift 68 spaces are approximately 12.8 inches ('/60' x 76
8 = 12.8'), the line direction is the main scanning direction of raster scan, and it is possible to print four lines at the bottom.
768 spaces x 64 lines constitutes the maximum print area of the page memory.

One row/one space is one word in the page-memory 12 access unit, and one word is composed of 31 bits. Therefore, the total size is 512 x 96 (768 x 6
4)=49152 words are the total address space of page memory 12. The bit configuration of one word is as shown in Figure 3 (C): main character select (128 characters),
Main font selection (57 onts), main character unit selection (8 units), subcharacter selection (128 characters)
, sub font select (5 fonts), sub character unit select (8 units), line feed control flags (HLD, HLU1815LPI), and underline control flags (SUL, DUL), totaling 31 bits.

As shown in FIG. 3(d), the address space of the page memory 2 has 48 words, so 16-bit address lines are required. During portrait mode, the character direction is the main scanning direction of the raster scan, so the lower 9 bits of the address line are the space address (512 spaces), and the upper 7 bits are the row address (96 lines). . In landscape mode, the row direction is the main scanning direction of the raster scan, so the lower 6 bits of the address line are the row address (
64 lines) Top 10 hitka space addresses (768
1 space). Since the raster scan direction of the plotter is the same regardless of the boat rate or landscape, the line direction and character direction of the page display memory 12 are opposite.

The page memory 1z has a bank configuration of 8 bits each, and one word has 4 banks. Since the access unit of the CPU 3 is 8 bits, the character code to be superimposed and the base character code can be written and erased independently, but it is still necessary to erase them every time form information is printed. Information without is left as is on the page.
It can be left in the memory 12. The remaining fixed addresses are stored in the work memory 5 and erased when all are cleared.

FIG. 4 is an explanatory diagram of the page memory control method in FIG. 1. During boat rate and landscape mode, the procedure for storing character codes sent from the host machine 1 in the page memory 12 by the CPU 3 is the same, except that the physical address order to be stored in the page memory 12 is different.

FIG. 4 shows an example of storage in the page memory 12. All the characters shown in the figure are characters (1/12' characters) composed of 5 units (25 dots width), and examples are shown.
The same idea can be applied to other character sizes, only the unit width is different. The printing position is in l unit units (1/6
o inch), and for areas that do not need to be printed, blank printing is performed by setting the word to "0", that is, a memory clear state. Although the subtext area used for overprinting is not shown in Figure 4, words that require overprinting can be considered in the same way, and parts that do not require overprinting can be left in the memory clear state. good.

Since the page memory 12 can be controlled in units of 1/60' space, it is possible to perform end-of-line alignment functions required by word processors, print provocative space characters, and mix character sizes.

In the mth line of Figure 4, the characters 'ABCDE' are 1/6'LP.
It is controlled so that 1B" is printed 1/2 line upwards (superscript printing, PLU), and "D" is printed downwards (subscript, PLD). In the case of a script, the current position (mth
The character code and 'PLU' control flag are set in the same space position on the previous line (line m-1),
The character code and "PLD" control flag are set in the current line (mth line).

In the case of a subscript, the character code and 'PLD' control flag are set in the same space position on the line (m+1 line) after the current position (line m), and the current position (line m)
The character code and PLU'' control flag are set in .

Also, since 'B'' and D'' are underlined, U
The "L" control flag is set together with the character code, and since 'C' is accompanied by a double underline, the "DUL" control flag is set together with the character code.

The line feed interval is generally 178'LF.
, 1/6' LF, 1/4' LP and integer multiples thereof.

The m-1st line, the mth line, the m+1st line, and the m+2th line all show examples of 115' LF, and the subscripts of each character (AO, As ! A2 , Aa I A4 s Bo
s Bls B2...) is the unit number of the font memory, and the figure shows the case where all characters are composed of 5 units. If the character size is different or the character pitch is changed, the amount of units will be different.

The m+3rd line and the m+4th line indicate the case of 1/4' L P', and "KL" indicates the case of "F'GHI" in the m+2th line.
115' L F 11/2 row (1/4
'LF) will be shifted. 1/4' L F is 1/4'
6' L This can be realized by shifting 1/2 line up or down with respect to P, so it can be executed using the same means as the superscript and subscript controls "P'LU" and "PLD".

"MO" on the m+3th line is a superscript for "KL", and "NP" is a subscript for "KL".

The m+sth row, the m+6th row, the m+7th row, and the m+8th row are 1
/s' LP case is shown. There are two basic line feed modes: 115' L P and 1/g' L F. These two modes are distinguished by whether or not the LF control flag is set.
P is not set, 1/s 'L P is L for the entire row
This is done by setting the P control flag. Therefore, the line feed interval can be changed on a line-by-line basis.

In FIG. 4, explanation is given by taking only the base character information as an example, but in reality, character information (font selection, character selection, unit selection) is also written in a similar manner so as to be overlapped.

FIG. 5 is an explanatory diagram of the character generation portion in FIG. 1.

FIG. 5 shows a block diagram of the character generation section. Raster
The synchronizing signal when printing one page with a mechanical printer is the output signal T7i ("D") from the plotter side.
ATACLOCK" and 'LINE 5Y-NCN, respectively, provide main scanning serial dot data synchronization and sub-scanning line synchronization. In addition to these two signals, 'LINE
There are two signals: GATE''FRAME and 9YNC. The former gives the effective time (number of dots) for one main scan, and the latter gives the effective time (number of lines) for one page (frame).

For the character generation part, the page memory 12 and font memory 15 are counted up based on the above synchronization signal,
By giving the count signal as a memory address signal, D
MA (Direct Memory Access)
It is configured to generate characters through motion. That is,
All character generation parts are controlled by the above four control signals,
The finally obtained image signal is transmitted to the plotter as serial dot data in synchronization with 'DATA CLOCK'.

In the timing control circuit 19 in the figure, 'LINE RE8
ET'' ``FRAME RESET'' signal is generated, the former is equivalent to LINE 5YNC'' and has the function of clearing the dot counter (initial state) for each main scanning line, and the latter is the signal before FRAME 5YNC''. It is generated at the edge and has the function of clearing the line counter (initial state).The timing control circuit 19 generates ``'DATA CLOCK'' at the boat rate and at the landscape time.
and LINE 8YNC" and 'LINE GATE'and'
``FRAME GATE'' and LINE RESBT''
It has a function of switching the 'FR RESET' signal (control from register 20).

The DMA counter 17 generates addresses for the page φ memory 12 and the font memory 15, and there are two types: a counter 12 in the space direction (character direction) and a counter 22 in the row direction. At boat rate, the space direction counter is DATA CLOCK'' and LINE GAT
It is controlled by the signals of "E" and LINE R, ESET" and becomes the main scanning direction counter for raster scan, and the row direction counter is "LINE 5YNC" and FB.

It is controlled by the ``AME 5YNC'' and ``FRAME RESET'' signals, and serves as a sub-scanning counter for raster scanning.

In landscape mode, the signal lines are interchanged, and the counter in the space direction becomes the sub-scan counter for raster scan, and the counter in the row direction becomes the main scan counter for star scan.

From the counter 21 in the space direction, page-memory 1
2 space address and font memory 15 CHI
A P select signal is generated, and a row address of the page memory 12, a BIT select signal and a block select signal of the font memory 15 are generated from the row direction counter 22.

As for the space address and the row address, the upper and lower address bit positions are switched by the page memory address selector 11 at the time of boat rate and landscape, as shown in the address configuration diagram of the page memory 12 in FIG.

CHIP select signal and BIT select signal are BIT/
The CHIP selector selects 5 dots for each BIT or 8 dots for each CHIP out of all the data (40 bits) in the small matrix of 1 unit and 1 block shown in FIG. Five dots per BIT are used when selecting a boat rate, and eight dots per CHIP are used when selecting a landscape.

The page memory 12 is accessed in units of words, and the character select (main/sub), unit select (main/sub), and font select (main/sub) signals included in the word are first input to the data selector 13. A portion is given to the font memory 15 to access the font memory 15 and the resulting data is latched in a register, and then a sub-part is given to the font memory 15 to access the font memory 15. latches the data obtained. The page memory data of the main part and the sub part are switched at time intervals that take into account the access time of the font memory 15. The obtained dot data of the main part and sub part are combined, and if there is an underline request, they are combined with the underline data, the combined data is given to the shift register, and the shift register is loaded by the shift register load signal. 18 and output to the plotter 8 as a serial image signal with DATA CLOCK''.

The shift register load signal is generated by the DMA counter 17 and is applied to the shift register 18 at a 5-dot cycle during the boat rate and at an 8-dot cycle during the landscape mode.

FIG. 6 is an explanatory diagram of the DMA counter portion in FIG. 5.

The space direction address counter 21 includes a dot/line counter 210 (dot counter at boat rate,
There are a line counter (in landscape mode) and a space counter 211, and a dot/line counter 21.
0 is a quinary counter, and the space counter 211 is a binary counter that can count up to 768 units. The dot/line counter 210 generates a carry-out signal (+1) every five clocks, which counts up the space counter 211 and serves as a shift register load signal during portrait mode.

The row direction address counter 22 has two sets of main and sub-line/dot counters 22o (line counter in portrait mode and dot counter in landscape mode) and block counter 221, respectively.
There are 29.

Main/secondary counter of line/dot karashita 22o12
22 normally functions as an octal counter (as a special case, it becomes a hexadecimal counter only when reading the 0th block at the time of a 1/8' LF request during landscaping), and it outputs a carry-out signal (+2 ) and (+3), respectively, count up the main/sub counter of the block counter and serve as a shift register load signal during landscape.

(If the FLU or PLD control flag is set in selector +-4, and the sub-counter carry-out signal (4p3) is not set, the main counter carry-out signal (-#=2) is (becomes a shift register load signal).

The count output values of both the main line/dot counter 220 and the main block counter 221 are sent to the decoder 22.
4 and generates a row direction control timing signal.

The timing signals necessary for row direction control are a row end signal (a function to reinitialize the line/dot main counter 220 and main block counter 221 and count up the row counter 229) and a sub-counter set signal (line/dot sub-counter 222 and sub-block counter 223), and a signal that inhibits data to the shift register (font memory output when scanning the upper half row when a PLU request is made, and when scanning the lower half row when a PLD request is made). There are three types of signals: a function that inhibits data.

The three types of line direction control timing signals differ in line feed width. Line feed width is 1/g ' L F and 1/6 ' L
There is a basic mode of F, but the number of lines or dots in the case of 1/g'LP is 38 lines or 38 dots (
38/300'medium'/7.9') 1 line and 1/6
'The number of lines or dots in the case of LP is 50 lines or 50 dots (50/300'=1/s') and 1
form a line.

As explained in Figure 2, the matrix of 7 on) - memory 1501 characters consists of 48 dots in the line height direction, so in the case of 115' LP there will be a shortage of 2 dots, but in the case of portrait , the 6th block, which is an unused area in Figure 2, will be read, but since dot data is not stored there is no problem in accessing the font memory 15, and in the case of landscape, the line will be forcibly read. Since the /dot counter 220 is reinitialized, no carry-out signal is generated and no inconvenience occurs. However, if you want to print continuous figures in the row direction (for example, vertical ruled lines), 1/6' line feed should consist of 48 dots or lines, and 48/300' = 1/6.25.
' It is also possible to implement it approximately.

(1) In the case of 1/s' LP, the row end signal is sent to the selector (+1
) 228, line/dot main counter 2
20 is reinitialized, and the row counter 229 is counted up. In portrait mode, both the line counter and diblock counter are initialized to '0', or clear state. In landscape mode, the dot color/block counter is initialized to 0", or clear state. The counter is 38 dots, so 8
The dot counter is preset to a state of 2'' so that each dot rotates four times and a carry count signal is output when 96 dots remain.

The sub counter set signal generates an address shifted by half a line, and the line/dot sub counter 22 is shifted from the main counter by 38/2-19 lines or dots.
2. It has a function of setting the sub-block counter 223.

The data prohibition signal prohibits data output during the scanning period of half a line, and is a signal that indicates the scanning period of line or dot numbers 0 to 18''.When a PLU request is made, the data output is prohibited during the scanning period of half a line. 18'', when PLD is requested, the data is forced to the white level state for line or dot numbers ``19 to 38''.

(2) In the case of 1/6 WL F, the row end signal is a selector +1 for every 50 dots or lines.
The line/dot counter 220 and block counter 221 are reinitialized, and the line counter 229 is counted up. The line/dot counter and block counter are initialized to 0'', that is, to a clear state, regardless of whether it is portrait or landscape.

The sub-counter set signal initializes the line/dot sub-counter 222 and the block sub-counter 223 to a value shifted by half a line, and has the same function as the line end signal except that it is shifted by 50/2225 dots or lines. It is.

The data prohibition signal is a signal indicating the scanning period of line or dot number "θ~24", and is a signal that indicates the scanning period of line or dot number "θ~24". It functions to forcibly keep the data at the white level for a scanning period of ~49''.

As described above, the line/dot sub-counter 222 and the block sub-counter 223 are the line/dot main counter y:P
220, the block main counter 221 is counted up with a value that is shifted by the number of rows, and the count output of both is set to PLO/PL with selector 11 = 2 and -43.
It is switched depending on the presence or absence of the D signal.

The count output value of the line/dot counter obtained from selector +2 becomes the font memory 15 (DBIT select signal (valid only in portrait mode)) and is sent to selector +3.
The count output value of the block counter obtained from the block counter becomes the block select signal of the font memory 15.

Also, carry-out signals +2, -11 = 3 of line/dot main/sub counter are selected by selector +4, and PLU/PLD
+3 force when requested, +2 force when not requested; selected. Furthermore, the shift register load signal is selected by selector +-5, and carry-out signal +1 is selected in portrait mode, and either carry-out signal 41-2 or 4+3 is selected in landscape mode.

The block select signal output from selector +3 is
When the adder 231 manually requests 1/8" LP, +1 is added, and the block number is forcibly advanced by one block. This means that in the character dot matrix shown in FIG. The block number ``θ'' is forcibly skipped and the block number ``1'' is started.In other words, in a character height of 48 dots, the upper 8 dots are removed to form 40 dots, and the lower 2 dots are also removed. The character matrix is printed as 38 dots by forcibly scraping. Therefore, the characters designed across the entire 48 dots in the height direction of the characters will be partially chipped, so the characters to be printed with 1/g' LF. is limited to fonts that have been designed in advance to the extent that they are not missing.

FIG. 7 is an explanatory diagram of the configuration of the font bank in FIG. 1. FIG. 7(a) shows a control block diagram of the font memory 10.

Five font banks are prepared, and each font bank 15 is 64 kbit (8 bit
/<Larel output) ROM or RAM (5 pieces) provides 17 onts, and 128 characters can be stored per 17 ont bank. Each font bank includes '/10' characters, 1/12" characters, 1/1s" characters, 1/20' characters,
It is possible to set either P or S character.
When configured with RAM, it can be set in advance before printing dot matrix data from the host machine, and the font itself can be changed each time.

As the address information given to the 7-ont memory 15, the main/sub character select signal, main/sub character unit select signal, main/sub font select signal and block select signal read from the page memo 912 are used as the main/sub character select signal. The data selector 13 first applies the main select signal to the 7-ont memory 15 when the data is determined after accessing the page memory, and the font store memory 15 is accessed. After the access to the font memory 15 is completed and the data is determined, a sub-select signal is given to the font memory 15, and the font memory 15
access. That is, in order to print one character, the font memory 15 is accessed twice, with the first access obtaining character dot data as the base, and the subsequent access obtaining character dot data to be superimposed. Even if there is no need for characters to be overlapped, blank data will be overlapped. Even if the same font bank is overlaid, different 7
Ontobank is also fine.

As shown in FIG. 7(b), the font memory 15 is structured so that five memory elements can be accessed at the same time.
0 dot data is obtained. The address line has a unit select signal, a block select signal, and a character select signal for each font memory, and a font select signal for each font bank.

The obtained 40 dot data is BI at boat rate.
The T select circuit selects 5 dots in the unit direction of the small matrix, and in landscape mode, the CHIP select circuit selects 8 dots in the block direction of the small matrix.
It becomes dot data. That is, during portraiture, the data of 40 dots in the small matrix is outputted eight times each with five dots, and the land data is outputted five times with eight dots each.

By setting RAM in the font bank 15, various printing functions can be performed.

■It is possible to transmit font matrix data from the host machine 1, treat it as one character, connect it two-dimensionally into a dot matrix of one character, and develop it into a graphic function.

■ Existing dot matrix data existing in other font memories can be accessed by CPU 3, combined multiple times, stored in the RAM font bank area, and newly defined as one character and printed. .

■The existing dot matrix data existing in other font memory is accessed by the CPU, rotated 90 degrees, stored in the RAM font bank area, defined as a new character, and printed. You can partially change the text direction.

■Access the existing dot matrix data existing in other font memories with the CPU, invert the black and white, store it in the B and AM font bank areas, and create a new one.
It can be defined as a character and printed.

In addition, a figure code is simultaneously defined for each new figure for which font matrix data is transmitted from the host machine, and the figure code, the start address stored in the font memory, and the figure size are registered as a table in the CPU's work memory. After that, you can freely use the graphic code for printing.

FIG. 7(C) shows the address when accessing the font memory.

FIG. 8 is an explanatory diagram of the synthesis circuit in step 1-. 8th
The figure shows a functional block diagram of underline and overlay composition. The decoder 27 has a function of detecting the block number and BIT number of the character matrix in order to insert an underline.

In the character matrix in Figure 2, 1/6' L
In case of P, the underline insertion position is BIT'''0'' and '1'' of the 5th block are single underline, B
IT"0" f+IN, n4n, N5" is the insertion position of the double underline. In the case of 1/8 #LF, the underline insertion position is the BIT of the 5th block.
″′2″.

+3″ is a single underline, BIT″2″, +3
″.

n6n and 'n7' are the insertion positions of the double underline.

During boat rate printing, all 5 bits of the shift register 36 are forced to the black level when the underline insertion position is detected by the decoder 27, and during landscape printing, the black level is set when the underline insertion position is detected. Forcibly sets the bit position to black level.

The line generation circuit is provided independently because the line insertion position differs depending on whether it is portrait or landscape, and the output data for both can be switched by a data selector. When the SUL and DUL control flags are not set in the page memory, the line generation circuit is in an inactive state and the multi-function is not performed.

The V5 dot parallel data finally obtained from the font memory is latched in the data register 4I-1, ≠2, but in the data register ≠1, the dot data of the base character is latched in the data register 412. Character dot data is latched to be superimposed. The latch timing is that after the page memory access is completed and the data is determined, the font φ memory is accessed for the first time, and at the end of the access time, the data is latched in the register ≠ 1, and the delay circuit After the second font memory access, data is latched into register +2. The latch data outputs of both data registers are combined with the underline data by a combining circuit (logical OR), and become parallel manual data of the shift register 36.

The data prohibition signal gives half the scanning time of a row, and when the FLU control flag is set, data output is prohibited in the lower half of the character matrix (corresponding to the upper half of the row scan). The register +1 is made inactive and the parallel data is forced to the white level. When the PLD control flag is set, the data register +1 is inactivated to inhibit data in the upper half of the character matrix (corresponding to the lower half of the row scanning), and the parallel data is forced to the white level.

Note that it is also possible to synthesize by detecting the insertion position of a vertical ruled line in a character matrix using a method similar to underlining. It is also possible to print suit scripts and subscripts using the overlay function.

In the embodiment shown in FIG. 7, the base character dot data is accessed in the first access to the font memory 15, and the character dot data to be overlapped is accessed in the subsequent access. , character dot data to be overlapped may be accessed first.

Furthermore, in the embodiment shown in FIG. 5, the main part of the page memory 12 is accessed first, and then the sub part, but the sub part (character codes to be overlapped) is accessed first. can be accessed. Also, in the embodiment of FIG. 8, when the main part is obtained from the font memory 15, it is latched into the data register 33, and when the sub-part is then obtained from the font memory 15, it is latched into the other data register 34. However, if you only need to latch the first data obtained and send the next data directly to the synthesis circuit 35, there will only be one data register. nothing.

FIG. 9 is an explanatory diagram of the code conversion table of the font memory.

As shown in FIG. 1 and FIG. 7, in this embodiment, the font memory 10 is provided with five 7 (7) links from font e no (k O to font bank 4). It is assumed that font bank 0 consists of ROM and a certain font is set therein, and font bank 4 consists of RAM, and its contents are set with fonts with the same structure as font bank 0. ] (The character code stored in link 0 and the pattern data correspond to each other in the code conversion table (see FIG. 9(a)) stored in the work memory 5 of FIG. 1.

That is, the character pattern corresponding to the character code from the external device 1 is indicated by the internal code that provides the storage start address of the font memory.

For example, when the character code "20H" shown in FIG. , after converting it into an internal code "OOH" and storing it in the page memory 9, when reading it and accessing the font memory 10, the pattern at the start address "0OOOH" of the font bank O is
Data is read. The width of this character pattern is 6 units, and the p1 unit is 1/60', so it is controlled at a width of 1/10'.

On the other hand, since the font bank 4 is constituted by a RAM, the code conversion table is also undefined at the initial stage, and is defined each time based on the designation from the external device R1. For example, (1) If you want to create a new character number by combining multiple characters in font number 0, you need to specify the newly defined character code and the character to be combined from external device 1. After the CPU 3 reads out the pattern data and synthesizes them, it is stored in the free area of the font bank 4 and registered in the code conversion table (see FIG. 9(b)) in the work memory 5.After that, , you can print using the newly defined character code (download).

(n) The above can be processed in exactly the same way for black-and-white inversion, rotation, enlargement, and reduction of an image.

In this way, the font bank organized in RAM can be freely used and downloaded. For example, in Figure 9(b), a certain unique image is
It is input as a figure, given a character code "AOH", and stored in the page memory 6 as an internal code "OOH". When accessing the font memory 1o with the internal code "OOH", the font・Start address of bank 4 “o”.

00H" is read out.The character width is defined to be controlled by 6 units.

In this example, the maximum character width is 8 units.

It is also possible to create new types of characters using the characters stored in the ROM of font bank 0. For example, if Gothic characters are stored in ROM,
R by the program of CPU3 according to the specification of external device 1
Read the character from OM, perform image processing, create a diagonal character and store it in RAM, or repeat the same character several times little by little to create a thick character.
It is also possible to store it in AM.

In this way, in the present invention, graphic information is stored in advance not only in ROM (RA with the same capacity as ROM) but also in
By arranging M in parallel and processing the graphic information by the CPU 3, a multifunctional printer or display device can be realized. That is, (a) create a new figure by combining information in ROM, transfer it to RAM, and treat it as one character. (b) R.O.
Reverse the black and white information of M to create a new eastern form, R, AM
, and treat it as one character. (c) Rotating the information in ROM to create a new figure, transferring it to RAM and treating it as one character. (d) Enlarging or reducing information in ROM to create a new figure, transfer it to RAM, and treat it as a single character. (e) Obtain graphic information from an external device, create a new graphic, transfer it to RAM, and treat it as one character. etc. are possible.

〔effect〕

As explained above, according to the present invention, by configuring the font memory in banks, it is possible to freely set the RAM, ROM, and improve the functionality of the printer or display device that is configured with the RAM. It is possible.

[Brief explanation of the drawing]

Fig. 1 is an overall block diagram of a printer controller showing an embodiment of the present invention, and Fig. 2 is a font/controller diagram in Fig. 1.
FIG. 3 is an explanatory diagram of the configuration of the page memory in FIG. 1. FIG. 4 is an explanatory diagram of the page memory control method. FIG. Block diagram, Figure 6 is a block diagram of the DMA counter part in Figure 1, Figure 7 is an explanatory diagram of the configuration of the font bank, Figure 8 is a block diagram of the synthesis circuit in Figure 1, and Figure 9 is the font bank diagram. - It is an explanatory diagram of a code conversion table of memory.

Claims (1)

[Claims] Equipped with a page memory that temporarily stores code in page units and a font memory that stores graphic information corresponding to the above code, and accesses the above page memory and font memory using the count value of the rask scan synchronization signal. In an output control device that directly obtains graphic information, multiple ROs are provided in banks for each character size and font.
A font memory consisting of M and at least one RAM, a page memory that stores character selection information including bank specification information of the font memory, and graphic information read from the ROM of the font memory and processed. and a processor for transferring data to the RAM.