Row Grabbing Video Display Terminal Having Local Programmable Control Thereof

The present invention relates to video communication systems in which individual frames may be grabbed for video display thereof and particularly to video display terminals for use in such systems.

Video communication systems in which individual frames may be grabbed for video display are well known, such as the system disclosed in U.S. Pat. No. 3,740,465, or a system employing the Hitachi frame grabbing disc. These prior art systems such as the one disclosed in U.S. Pat. No. 3,746,780 are normally two-way request response systems requiring the user to request information by the dialing of a specific digital code which is uniquely assigned to each frame. However, such systems normally grab a group of frames for storage and then subsequently select the individual frame for display out of the group of grabbed frames as opposed to instantaneously selecting a single frame in real time. Furthermore, such prior art systems do not provide for real time updating of the grabbed video frame. Furthermore, some such prior art frame grabbing systems, such as the type disclosed in U.S. Pat. No. 3,397,283, are normally capable of only grabbing the next immediate signal in response to the provision of a starter signal or, as disclosed in U.S. Pat. No. 3,051,777, utilize a counter for frame location which must be reset to the beginning of a tape for video tape supplied information in order to locate a selected frame to be grabbed. These systems are not applicable in a real time frame grabbing environment. Similarly, other typical prior art frame grabbing systems, such as disclosed in U.S. Pat. Nos. 3,695,565; 2,955,197; 3,509,274; 3,511,929 and 3,582,651, cannot be utilized in a real time frame grabbing environment, such as one in which the video information associated with the grabbed frame is capable of being continuously updated. Accordingly, presently available prior art frame grabbing systems familiar to the Inventors are not capable of easily locating a frame to be grabbed in real time nor of being able to continuously update such a grabbed frame in real time.

Video communication systems in which the signal being transmitted is digitized are also well known. For example, U.S. Pat. No. 3,743,767 discloses a video communication system for the transmission of digital data over standard television channels wherein the digital data is transmitted in a conventional television scan line format through conventional television distribution equipment. However, such a prior art communication system merely digitizes one television scan line at a time for distribution to a video display terminal on a bit-by-bit basis in a line, 84 bits of information being provided per television scan line. Furthermore, such a prior art system is not transmission selectable by every display terminal nor is the data for a displayable video row packed into a self-contained pseudo video scan line information packet. Thus, there is no significant increase in the data transmission rate resulting from such a prior art video communication system. Similarly, U.S. Pat. Nos. 3,061,672 and 3,569,617 and German Pat. No. 2,307,414 are examples of other prior art video communication systems in which television signals are digitized without any significant resultant compression in data transmission time. Furthermore, these other prior art systems require special distribution circuitry. In addition, prior art video communication system in which a digital television signal is transmitted do not sufficiently isolate the individual rows comprising a frame so as to provide satisfactory noise immunity between these rows nor is there satisfactory data compression in the transmission time of the video information in such prior art systems nor satisfactory distortion compensation. These disadvantages of the prior art are overcome by the present invention.

A real time frame grabbing video display terminal for substantially instantaneously providing a continuous direct video display of a selectable predetermined video frame of information on a video display means, such as a cathode ray tube, from continuously receivable information which has been remotely provided over a transmission media, such as a conventional television distribution system, is provided in which the operation of the terminal is controlled by a microcomputer means, such as a central processing unit contained on an integrated circuit chip, which microcomputer has been operatively connected to the transmission media for receiving the continuously receivable information. This information comprises a plurality of pseudo video scan lines, each of which has a television video scan line format and is capable of comprising a complete self-contained packet of digital information sufficient to provide an entire displayable row of video data characters. Each of the pseudo video scan lines has an associated transmission time equivalent to the television video scan line with the packet of digital information comprising at least address information for the displayable row and data information for the displayable characters in the displayable row. The pseudo video scan line is a composite video signal. Selection means, such as a keyboard, are operatively connected to the microcomputer for variably selecting the predetermined video frame to be continuously displayed. Local storage means, such as a random access memory means, for selectably retrievably locally storing the continuously receivable video displayable data corresponding to the predetermined video frame are operatively connected to the microcomputer. Means, such as a read only memory means, are operatively connectable to the microcomputer, to the selection means and to the local storage means for selectively controlling the local storage of the continuously receivable information in the local storage means. The read only memory means comprises a condition responsive network, such as one corresponding to a permanently storable executive and loader program set of instructions for the microcomputer defining the set of conditions for controlling the loading of the selected received corresponding video displayable data and enabling control of the microcomputer processing for subsequently providing the continuous direct video display. The microcomputer processing means processes the continuously received composite pseudo video scan line for providing retrievable locally storable displayable video row data from each of the received pseudo video scan line signals pertaining to the selected predetermined video frame and a displayable video row signal to the video display means from the local storage means from each of the locally stored predetermined video frame pertaining pseudo video scan line signals for enabling the direct video display. The microcomputer processes the continuously received composite pseudo video scan lines in accordance with the predetermined set of conditions for enabling the direct video display of the selected predetermined video frame from the plurality of continuously received psuedo video scan lines. The microcomputer processing means comprises means for updating the retrievable locally stored displayable video row data in the local storage means on a displayable video row-by-row basis dependent on the real time data information content of the received pseudo video scan lines and for updating the displayable video row signal to the video display means in accordance with the updating of the locally stored displayable video row data whereby the continuous direct video display of the selected predetermined frame is updated on a displayable video row-by-row basis dependent on the real time data information content of the received pseudo video scan lines.

The terminal also comprises input/output buffer means operatively connected between at least the transmission media, the microcomputer, the video display means and the local storage means for interfacing the microcomputer with the video display means and the received information, the inputs and outputs to the microcomputer for enabling the continuous direct video display being routed through this buffer means. The buffer, the microcomputer, the read only memory means and the random access memory means each preferably comprise integrated circuit chips. The local storage means further comprises means for testing the address information portion of each of the continuously received psuedo video scan line signals for satisfaction of at least one predetermined signal reception condition, the address information corresponding to the frame associated with the received pseudo video scan line. The address information testing means provides a first predetermined output condition to the microcomputer when the reception condition is satisfied and comprises means for testing the frame information, the reception condition being correspondence between the frame information and the selected frame. The microcomputer loads the local storage means with the displayable video row data when the first predetermined output condition is provided thereto.

Each of the pseudo video scan lines received by the terminal preferably comprises a horizontal sync signal at the beginning thereof which provides a record separator between adjacent pseudo video scan lines and insures that any loss of synchronization or noise pulse will not disrupt more information than one pseudo video scan line. Furthermore, each of these pseudo video scan lines preferably comprises a three level signal having first, second and third signal levels with each of the packets of digital information varying between the second and third signal levels and the horizontal sync signal information being provided between the first and second signal levels. Significant data compression in transmission time of the signals received by the terminal is obtained due to the transmission of pseudo video scan lines as opposed to conventional television scan lines since each pseudo video scan line is a self-contained packet of information sufficient for display of an entire displayable video row containing a plurality of conventional television scan lines, such as 13, as opposed to display of one television scan line. The local microcomputer, operates in conjunction with the keyboard which provides the information request, and enables updating on a row-by-row basis as opposed to a page-by-page or frame-by-frame basis as new information is provided in real time, the selected frame being automatically updated in real time as new information is provided for a given row of the displayed selected frame.

FIG. 1 is a diagrammatic illustration of a typical pseudo video scan line format in accordance with my previous U.S. Pat. No. 3,889,054;

FIG. 2 is a diagrammatic illustration of a typical pseudo video scan line format in accordance with the present invention;

FIG. 3 is a graphical illustration of conventional vertical drive and composite sync signals illustrating the origin of the vertical sync signal in accordance with the present invention;

FIG. 4 is a block diagram, partially in schematic, of the various interface connections between the microprocessor and the keyboard in the preferred embodiment of the receiver of the present invention;

FIG. 5 is a block diagram of the preferred embodiment of the receiver of the present invention including a functional indication of the various control signals utilized therein;

FIG. 6 is a block diagram of the microprocessor portion of the preferred embodiment of the receiver of the present invention illustrated in FIG. 5;

FIG. 7 is a functional block diagram similar to FIG. 6 for use in explaining an exemplary program for the microprocessor of the preferred embodiment of the receiver of the present invention;

FIG. 8 is a block diagram of the phase locked loop portion of the arrangement illustrated in FIG. 9;

FIG. 9 is a block diagram of the timing control, memory input control and a part of the output processing portions of the preferred embodiment of the receiver of the present invention;

FIG. 10 is a block diagram of another portion of the memory input control portion of the preferred embodiment of the present invention;

FIG. 11 is a block diagram of the memory and output processing portion of the preferred embodiment of the receiver of the present invention;

FIG. 11A is a graphical illustration of the timing associated with various signals in the arrangement of FIG. 11;

FIG. 12 is a logic diagram, partially in schematic, of a portion of the timing and keyboard control portion of the preferred embodiment of the receiver of the present invention illustrated in FIG. 9;

FIG. 13 is a block diagram of another portion of the memory and output processing portion of the preferred embodiment of the receiver of the present invention;

FIG. 14 is a logic diagram, partially in schematic, of the keyboard portion of the timing and keyboard control portion of the preferred embodiment of the receiver illustrated in FIG. 9;

FIG. 15 is a logic diagram, partially in schematic, of the portion of the memory input control portion of the preferred embodiment of the receiver illustrated in FIG. 10;

FIG. 16 is a logic diagram, partially in schematic, of the portion of the memory input control portion of the preferred embodiment of the receiver illustrated in FIG. 11;

FIG. 17 is a logic diagram, partially in schematic, of the memory and output processing portion of the preferred embodiment of the receiver illustrated in FIG. 9;

FIG. 18 is a logic diagram, partially in schematic, of another portion of the memory and output processing portion of the preferred embodiment of the receiver illustrated in FIG. 9;

FIG. 19 is a block diagram of a typical preferred transmitter portion for use with the receiver of the present invention;

FIG. 20 is a logic diagram of the first in-first out memory portion of the transmitter portion illustrated in FIG. 19;

FIGS. 21 and 22 are logic diagrams, partially in schematic, of the transmitter portion illustrated in FIG. 19 except for the first in-first out memory portion illustrated in FIG. 20;

FIG. 23 is a functional block diagram of a typical preferred embodiment of a row grabbing system for use with the receiver of the present invention;

FIG. 24 is a schematic diagram of the video processor portion of the preferred embodiment of the receiver of the present invention illustrated in FIG. 9;

FIGS. 25A, 25B and 25C are graphic illustrations of the various waveforms present throughout the video processor of FIG. 24;

FIG. 26 is a logic diagram, partially in schematic, of another portion of the video processor portion of the preferred embodiment of the receiver illustrated in FIGS. 9 and 24;

FIG. 26A is a graphic illustration of the various waveforms present in the video processor portion illustrated in FIG. 26;

FIG. 27 is a block diagram of the various interconnections or interfacings of the video processor of FIGS. 24 and 26 with the balance of the circuitry in the preferred embodiment of the receiver of the present invention;

FIG. 28 is a block diagram, partially in schematic, of a typical computer/printer interface portion which may be used with the preferred receiver of the present invention illustrated in FIG. 5;

FIG. 29 is a diagrammatic illustration of a video display screen for providing 32 and/or 64 character display selection; and

FIG. 30 is a fragmentary block diagram, partially in schematic, of an additional portion of the memory and output processing portion of the receiver illustrated in FIG. 17 for providing 32 and/or 64 character selection.

Referring now to the drawings in detail and initially to FIGS. 1, 2, 3 and 23 thereof, the preferred real time frame grabbing video display terminal, generally referred to by the reference numeral 28, of the present invention, as will be described in greater detail hereinafter, preferably substantially instantaneously provides a continuous direct video display of a selectable predetermined video frame of information on a video display device 2013 (FIG. 5) from continuously receivable information remotely provided over a transmission media, such as coaxial CATV cable 22 (FIG. 5) which information preferably comprises a plurality of pseudo video scan lines 12 or 12a, illustrated in FIGS. 1 and 2. These pseudo video scan lines 12 and 12a, as will be described in greater detail hereinafter, are preferably of the type described in my previously issued U.S. Pat. No. 3,889,054 and my copending U.S. patent application entitled "Improved Row Grabbing System", filed Sept. 10, 1975 and bearing U.S. Ser. No. 611,843. These pseudo video scan lines 12 or 12a are normally preferably transmitted to the preferred real time frame grabbing video display terminal 28 on a row-by-row basis, with the pseudo video scan line 12 or 12a preferably being identical in format to a conventional video scan line; that is, it is consistent with FCC and EIA standards for a video scan line signal format. However, the pseudo video scan line 12a (FIG. 2) actually contains a row of information such as approximately between 11 and 13 actual television video scan lines of information, with the transmission time of the pseudo video scan line 12a preferably being equal to the transmission time of a conventional TV video scan line, which is approximately 63 microseconds. This is also true, as described in U.S. Pat. No. 3,889,054, for the pseudo video scan line 12 illustrated in FIG. 1. The various portions of the pseudo video scan line 12a, for purposes of completeness, will again be described in greater detail hereinafter with reference to FIGS. 2 and 3 and it should be noted at this time that the format for the preferred pseudo video scan line 12a which may preferably be utilized with the real time frame grabbing video display terminal 28 of the present invention is similar, with certain exceptions to be described hereafter, to the previously described format for the pseudo video scan line 12, illustrated in FIG. 1, and described in my previously issued U.S. Pat. No. 3,889,054, with identical regions therein being given the same reference designations. As was shown and described in my previous U.S. Pat. No. 3,889,054 and my copending U.S. patent application entitled "Improved Row Grabbing System", bearing U.S. Ser. No. 611,843, filed Sept. 10, 1975, in a row grabbing system 10 (FIG. 23) preferably employing the preferred real time frame grabbing video display terminal 28 of the present invention, the information is updated on a row-by-row basis by transmission of a pseudo video scan line containing new information so that the frame being grabbed will effectively have this row containing new information updated when this row of information is updated in memory. Thus, in this row grabbing system 10, continuously transmitted information or messages may be instantaneously "grabbed" in real time so as to repetitively provide a video display of a selected video frame of such information which may be updated on a row-by-row basis in real time.

The video information which may be received and processed by the preferred real time frame grabbing video display terminal 28 of the present invention may be of any conventional type, such as news information, money rate information, stock market information, local advertising, television program listings, weather information, consumer information, etc., which may be conventionally supplied from conventional external information sources for such types of information, such as sources 2002 and 2004 shown by way of example in FIG. 23. These conventional external information sources 2002 and 2004 may preferably conventionally supply this information in a digital format such as from a ticker for news information or stock information, by way of example, through a conventional communication line 2006 or 2008 (FIG. 23) or a conventional local video terminal, preferably, to a conventional mini-computer 2000, such as a model no. PDP-8E manufactured by Digital Equipment Corporation. Such a typical mini-computer 2000 may preferably have an associated conventional mass memory 2010 for conventional storage of data with computer 2000 storing this information in mass memory 2010, reformatting it, such as by adding other information, and continuously providing this information as a 12 bit parallel output 2011 to a transmitter 20 which subsequently provides the pseudo video scan line 12a which is ultimately received by hy the preferred real time frame grabbing video display terminal 28 of the present invention by transmission through a conventional TV distribution network. The aforementioned 12 bit parallel output of computer 2000 may preferably present 2 characters for one word and, if desired, as described in my previous copending U.S. patent application entitled "Improved Row Grabbing System" referred to above, a 14 bit parallel output from the computer 2000 could be utilized to provide two 7 bit characters. As also mentioned therein, in ultimately providing the pseudo video scan line 12a which is subsequently received by the preferred real time frame grabbing video display terminal 28 of the present invention, the mass memory 2010 may preferably be updated by the computer 2000 in conventional fashion at the optimum transfer time for data which is, conventionally, not necessarily in the order of the reception of the external information such as from sources 2002 and 2004, this data being preferably continuously suppliable in real time to the computer 2000. Thus, the information which may preferably be continuously received by the preferred real time frame grabbing video display terminal 28 of the present invention may be supplied thereto in conventional fashion from information in computer 2000 which is supplied to the transmitter 20 which, in turn, supplies this information to the CATV cable system 22 through a conventional RF modulator 24, with composite video being supplied to modulator 24 from transmitter 20 (FIG. 23). In such a system for video displayable information which is continuously receivable by the preferred real time frame grabbing video display terminal 28 of the present invention, a typical preferred transmitter for providing such information being shown by way of example in FIGS. 19 through 23, one such modulator 24 may preferably be provided for each television channel on which information is to be transmitted, only one such channel being illustrated by way of example in FIG. 19. In addition, in such a system, the mass memory 2010 which may preferably be read in conventional fashion by computer 2000 to provide the requisite information via transmitter 20 to the CATV cable system 22 and ultimately to the preferred real time frame grabbing video display terminal 28 of the present invention, has sufficient storage capacity to store the entire page capacity of this system.

As used hereinafter throughout the specification and claims, the term "page" means one video frame of information, the term "group" means a predetermined number of pages, the term "row" is a displayable video row and means a portion of a page containing a plurality of conventional television video scan lines, the term "pseudo video scan line" means a signal which is identical in form to that of a conventional video scan line but which actually contains a row of information, such as approximately between 11 and 13 actual television video scan lines of information with the transmission time of the psuedo video scan line being equal to the transmission time of a conventional TV video scan line and with the pseudo video scan line being an entire packet of information necessary for video display of that row. The term conventional or television video scan line is used in the conventional manner.

As described in my previously issued U.S. Pat. No. 3,889,054, and my copending U.S. patent application entitled "Improved Row Grabbing System", referred to above, the mass memory 2010 can be any conventional mass memory storage device sufficient to store the requisite page capacity of the system, such as an RK-08 memory device manufactured by Digital Equipment Corporation. The output of the computer 2000 may preferably be conventionally transmitted from computer 2000 to the transmitter 20 via conventional data break of the computer 2000. All pages of information which are preferably continuously being transmitted from the computer 2000 through transmitter 20 on a pseudo-video-scan-line-by-pseudo-video-scan-line basis, that is respectively on a row-by-row basis, through the appropriate RF modulator 24 for the video channel being utilized and therefrom through the CATV cable system 22, are preferably continuously received by the preferred real time frame grabbing video display terminal such as 28 and 28a of the present invention for providing a direct video display of selected frames of information on the associated conventional video display devices, 2013 and 2015, such as commercially available video monitors, two such monitors and two such terminals being shown in FIG. 23 merely by way of example. It should be noted that the number of video display devices 2013 and 2015 and terminals 28, 28a preferably have no requisite correlation with the number of external information sources 2002 and 2004 and more such sources could be utilized than video display devices or terminals or vice versa if desired. In normal contemplated use, in the row grabbing system 10, the number of video display devices 2013 and 2015 and associated terminals 28 and 28a, respectively, will normally exceed the number of external information sources 2002 and 2004, however, this need not be the case. The aforementioned computer 2000 may conventionally recirculate the data provided thereto in continuous fashion and, as previously mentioned, eventually update the mass memory 2010 at the optimum transfer time for the data, which time is not necessarily in the order of reception of the external information from sources 2002 and 2004. The information from external sources 2002 and 2004, which may preferably be provided substantially continuously to the computer 2000 (as long as it is being generated from the external sources 2002 and 2004) may be provided to the mass memory 2010 and instantaneously to the transmitter 20 which may operate in a manner to be described in greater detail hereinafter to provide the pseudo video scan line 12a transmission of the information to the preferred terminal 28 or 28a of the present invention. Each video display device 2013 and 2015, as previously mentioned, preferably has an associated display control unit 25 and 26, terminal 28 and 28a, respectively, being comprised respectively therein, which terminals preferably function to enable the real time frame grabbing or selection of a single page of continuously receivable, continuously transmitted information for the instantaneous repetitive continuous video display, or frame grabbing thereof, this information being updatable on a row-by-row basis in real time. Preferably, each of the display control units 25 and 26 by way of example, one such display control unit preferably being associated with each video display terminal device, are identical in structure and operation. If desired, however, any display control unit 25-26, and thus any terminal 28-28a, may be modified in a manner, such as described in U.S. Pat. No. 3,889,054 or my copending aforementioned U.S. patent application entitled "Improved Row Grabbing System", so as to prevent the reception of certain categories of information while enabling the reception of other categories of information. For purposes of clarity, only one such typical preferred display control unit 25, and thus typical video display terminal 28, will be described by way of example, the structure and operation previously mentioned, being identical with that of display control unit 26 and terminal 28a. Identical reference numerals, followed by the letter "a" will be utilized in FIG. 23 for elements of display control unit 26 and terminal 28a which are identical in structure and operation with those of display control unit 25 and terminal 28. In the overall system block diagram of FIG. 23, showing a typical preferred system for providing the continuously receivable psuedo video scan lines to terminal 28 by way of example, the display control unit 25 preferably contains a conventional RF demodulator 27, one such RF demodulator 27 being provided for each channel and the preferred receiver or terminal 28, to be described in greater detail hereinafter, which video display terminal 28 preferably includes a local microcomputer for controlling the operation thereof and which receives the composite video demodulated by demodulator 27 and preferably determines, by way of example, whether the user is correct, the user has permission to receive the pseudo video scan line information being transmitted at that time, whether the signal is error free, whether the page address of the pseudo video scan line is correct, and whether a direct address condition exists, and, preferably, assuming the pseudo video scan line signal passes all these tests, then the preferred video display terminal 28 processes this signal and provides a video signal corresponding to a displayable row of information on the associated video display device 2013.

As will be described in greater detail hereinafter, the preferred video display terminal 28 of the present invention, may preferably be utilized in the row grabbing system 10 illustrated by way of example in FIG. 23. Transmitter 20 illustrated by way of example in row grabbing system 10 may preferably provide the pseudo video scan line such as the type 12a illustrated in FIG. 2. This pseudo video scan line 12a, as was previously described, is identical in format to a conventional video scan line; that is, it is consistent with FCC and EIA standards for video scan line signal format; however, this pseudo video scan line 12a actually contains a row of information, such as approximately between 11 and 13 actual television video scan lines of information with the transmission time of the pseudo video scan line 12a being equal to the transmission time of a conventional TV video video scan line, which is approximately 63 microseconds. With respect to the pseudo video scan line 12a, the horizontal sync and vertical sync portions are preferably identical to a conventional video signal as is the format for the horizontal sync and the vertical sync as well as the horizontal sync amplitude. The time and amplitude envelope of the video region of the pseudo video scan line 12a, which region is defined as areas H, J, B, C, D, E and K in FIG. 2, is identical with the format for a conventional video scan line as is the three dimensional frequency envelope. Thus, all of the above mentioned standard conditions for a conventional video scan line signal are met by the pseudo video scan line 12a of the type continuously providable by the transmitter 20 of the row grabbing system 10 and continuously receivable by the preferred video display terminal 28 of the present invention. Accordingly, any equipment that can handle conventional video can handle the pseudo video scan line 12a of the present invention which can thus be transmitted and received through a conventional television distribution system with conventional television equipment.

Returning once again to the pseudo video scan line 12a illustrated in FIG. 2, as is also true for the pseudo video scan line 12 illustrated in FIG. 1 which was previously described in my U.S. Pat. No. 3,889,054, the signal received by the preferred video display terminal 28 of the present invention and transmitted by transmitter 20 is in reality a digital signal which looks like a conventional video scan line to the video display terminal 28. Pseudo video scan line 12a, as will be described in greater detail hereinafter, however, may preferably employ a start bit to provide timing and phase adjustment for the phase locked loop of the video display terminal 28. In such an instance, region F which was previously contained in the pseudo video scan line 12 transmitted in the system of my U.S. Pat. No. 3,889,054, and which contained the clock synchronizing burst or pulse train at the bit rate (the frequency preferably being equal to one-half the bit rate) and comprised a pulse train of ones and zeros for two character spaces or 14 bits, is not present and the sync burst information which was previously contained therein is not required for timing and phase adjustment in the row grabbing terminal 28 of the present invention. Instead, region H, which preferably contains color burst information and region J which preferably contains one start bit are preferably inserted between regions A and B, with regions B, C, D and E being electronically shifted down in position to be adjacent region G, only being separated therefrom by a region K, which region K merely represents the standard TV spacing for providing the front porch of the signal, the back porch of the signal being defined between region A and the leading edge of the start bit in region J. The color burst signal in region H preferably is the standard FCC eight cycle signal at 3.58 megahertz. Apart from the repositioning and deletion of certain regions of the pseudo video scan line 12 of FIG. 1, the contents of regions A, B, C, D, E and G in pseudo video scan line 12a of FIG. 2 is preferably identical with that previously described with reference to FIG. 1 in my U.S. Pat. No. 3,889,054 with respect to the transmission and reception of a displayable row of data. Suffice it to say for purposes of clarity, that region A represents the horizontal sync signal which indicates the beginning of the pseudo video scan line from the beginning of the horizontal sweep for a conventional television scan line; and region B represents the pseudo video scan line 12a address which contains all the following information bit locations, a one preferably indicating the presence of a pulse and a zero preferably indicating the absence of a pulse, all of the following information bits preferably being present when data is transmitted: group, which is the section or chapter including a predetermined number, such as 1,000, of pages and comprises the most significant bits of the page address, page which represents one frame in a group, and row which occupies one character space which is preferably 7 bits and defines a portion of the page preferably containing approximately 11 to 13 scan lines which comprise one displayable character. The region B also preferably contains direct address information, which is the first transmitted bit preferably and is a zero unless a direct address condition exists which is a control condition for a selected terminal informing the terminal to supercede the requested page. This region B also preferably contains permission information which is a one bit position which is preferably a one only when the user is being given authority to receive one or more selected groups of information. It should be noted that preferably there may also be an emergency override condition which would provide control information to all terminals to override all requests including the permission request and preferably would occur on a page and group information bit location of zero, this condition preferably being utilized to display emergency information such as a civil defense warning. Region C is preferably a special character information region of 7 bits which is preferably utilized for optional functions to be performed by the individual video display terminal 28. Region D preferably contains 32 characters of displayable information in digital form. Region E preferably contains error check information, as will be described in greater detail hereinafter. Region G is preferably the same as region A and represents the horizontal sync signal. As was previously mentioned, the vertical sync is preferably provided by generating a special sequence of horizontal sync pulses during the normal television blanking period, which is after approximately 236 horizontal sync pulses, which in the invention described in my aforementioned U.S. patent application entitled "Improved Row Grabbing System", as in U.S. Pat. No. 3,889,054 is after approximately 15 pages have been transmitted. Therefore, 15 pages are transmitted before each vertical sync. The sync signal looks like a conventional composite sync signal with a vertical sync interval comprising approximately nine normal horizontal sync pulse times as illustrated in FIG. 3 which is an illustration of conventional composite sync and vertical drive signals.

Referring now to FIGS. 4, 5, 6 and 7, the preferred microprocessor and keyboard control system of the preferred video display terminal 28 of the present invention which is described in my aforementioned U.S. patent application entitled "Improved Row Grabbing System" and which preferably replaces the keyboard circuitry of the system described in my previous U.S. Pat. No. 3,889,054, will be described. As shown and preferred in FIGS. 6 and 7, the preferred microprocessor 6000 of the present invention preferably includes a conventional microcomputer or CPU 601, a conventional read only memory or ROM 603, a conventional random access memory or RAM 605, and a conventional input/output buffer or I/O 607. Preferably, the microcomputer 601, read only memory 603, random access memory 605 and input/output buffer 606 are all integrated circuit chips of the type manufactured by Rockwell International, such as what is commonly available from Rockwell International as their PPS-4 system wherein the microcomputer chip 601 is a Rockwell International 10660 CPU, read only memory 603 and random access memory 605 are Rockwell International 10432 memory chips and input/output buffer chip 607 is a Rockwell International 10696. The programming language for the permanently storable loader, executive and applications program which is preferably stored in read only memory 603 is preferably written in PPS-4 Assembler language provided from Rockwell International and a typical such conventional preferred control program for operating the microprocessor 6000 of the present invention in accordance with the desired row grabbing function for the preferred video display terminal 28 of the present invention is set forth below, with this version of the program utilizing the arrangement of FIG. 7 comprising one CPU chip 601, one 256-by-4 bit random access memory chip 605, one 1024-by-8 read only memory chip 603 and two input/output buffer chips 607: ##SPC1## ##SPC2## ##SPC3##

The various inputs and outputs to the microcomputer or CPU 601 are all preferably conventionally routed through the input/output buffer 607 for interfacing the system with the microcomputer 601 and have been omitted from FIG. 6 for clarity; however, these signals as well as their function in the system for providing displayable rows of video information on video display terminal 2013 is illustrated in FIGS. 4, 5, and 7 and shall be described in greater detail hereinafter. During operation of the system, microcomputer 601 gets the instructions required from the stored program in read only memory 603 step-by-step, such as "take data from buffer 607 and put it in random acess memory 605" or "take data out of random access memory 605 and provide it to buffer 607". Microcomputer 601 can also perform typical computer functional operations on the data. As shown and preferred in FIG. 5 which is a block diagram of the preferred terminal keyboard and microprocessor input/output for video display row-by-row of pseudo video scan lines in the row grabbing terminal 28 of the present invention, the keyboard 5000 provides selection information to the microprocessor 6000, such as the group and page address, as described in my previous U.S. Pat. No. 3,889,054 and my aforementioned U.S. patent application entitled "Improved Row Grabbing System". The balance of the circuitry of the preferred terminal 28 controlled by the mircoprocessor 6000 is generally represented by the block 8000 labeled "terminal" in FIG. 5. If desired, the input/output buffer 607 may also include a conventional special character decoder in addition to the balance of the interface functions performed thereby so as to detect the presence of a special character. The various latchings preferably occuring within the terminal 8000 shall be described in greater detail hereinafter with reference to FIG. 4. As shown and preferred in FIG. 14, the keyboard entry line for the microprocessor 6000 from keyboard 5000 preferably comprises four lines 6001, 6003, 6005 and 6006 which are set when a key is pressed and latched and are reset to the idle state only when a clear pulse is sent via line 6007 from the microprocessor 6000. The keys which can be set are numeric keys as well as special function keys such as group, page, up, back and call. In addition, as described in the copending U.S. patent applications of Robert H. Nagel, one of the joint inventors herein, and owned by the same assignee as the present application and entitled "Information Retrieval System Having Selectable Purpose Variable Function Terminal", filed Sept. 10, 1975, and bearing U.S. Ser. No. 611,937, and the aforementioned U.S. patent application entitled "Improved Row Grabbing System", additional keys such as program may be provided when the system is utilized to retrieve programs or sets of instructions as opposed to or in addition to data. As shown and preferred in FIG. 5, if a group key is depressed and preceded by a number, terminal 8000 will be informed of the new group via the group address and page address line GAD/PAD and LEN lines, with the page number preferably reset to zero. If the group is not preceded by a number, preferably the only action taken would be to release directed messages by setting and clearing the KAC line and to erase the screen via a pulse on the ERA line, all functions which change the group/page accomplishing this such as call, back, up and group. The inputs from the terminal 8000 to the microprocessor 6000 are preferably the row enable line REN, the row address read line RAR, the row address write line REW, the special character line SCH, and the permission line PER. The output to the keyboard is preferably the clear or keyboard latch reset line 6007 which is pulsed when a key has been completely processed so that the key will then be reset to the idle state. The outputs to the terminal 8000 from the microprocessor 6000 preferably comprise the erase line which erases the video display 2013, the keyboard active line KAC which is set while a group/page number is being sent to the terminal 8000 so that the terminal 8000 will take no action on it until the number is completely received, the group address/page address line GAD/PAD which collectively refers to four lines utilized in conjunction with LEN lines to preferably multiplex the group page address to the system hardware four bits at a time and which informs the terminal 8000 of a new page to grab, the latch enable line LEN and the row length line RWL. Suffice it to say that the microprocessor 6000 performs data handling and execution of the premanently stored program in the read only memory 603 with the random access memory 605 preferably functioning to store variable and intermediate data results for use by the CPU 601 in order to accomplish the row grabbing function. The data for display on video display 2013 is preferably stored in a separate video memory random access memory which is loaded by the system hardware.

Referring now to FIG. 4, the output of microprocessor 6000 preferably includes four general purpose address bits or lines 715 and five lines of enable pulses 717. Lines 717 specify which one of the plurality of latches 701, 703, 705, 707 and 709, collectively referred to by the reference numeral 699, with latches 701, 703 and 705 being the page address latches, and latches 707 and 709 being the group address latches, the address bits 715 are designated for. By utilizing the four data lines 715 and five enable lines 717, up to 20 different bits of address, which is equivalent to 1,048,576 addresses, can be established. The page address output bits collectively referred to by reference numeral 504 and the group address output bits collectively referred to by reference numeral 502 are employed in the memory input control portion of the preferred terminal 28 illustrated in FIG. 15 preferably in the same manner as described in my previous U.S. Pat. No. 3,889,054 and the aforementioned U.S. patent application entitled "Improved Row Grabbing System" with respect to the outputs of the keyboard counter 500 of the system described in U.S. Pat. No. 3,889,054 and will not be described in greater detail hereinafter since such description is specifically incorporated by reference herein.

Referring to the aforementioned exemplary program written in PPS-4 Assembler language, as was previously mentioned, a typical microprocessor 6000 arrangement which is controlled by the aforementioned program is illustrated in FIG. 7 with the various pin number designations utilized by Rockwell International for the read only memory, random access memory, CPU or microcomputer, and input/output buffer chips 605, 603, 601 and 607a and 607b, respectively being utilized in FIG. 7 for clarity. Furthermore, and as shown by way of example in FIG. 5, and described in the aforementioned U.S. patent application entitled "Improved Row Grabbing System" and particularly in the commonly owned copending U.S. patent application entitled "Interface For Enabling Continuous High Speed Row Grabbing Video Display With Real Time Hard Copy Print Out Thereof", filed Apr. 23, 1976, bearing U.S. Ser. No. 679,907, if desired, and as preferred in the latter application, the terminal 28 may additionally have a computer/printer or cathode ray tube display interface for use with a CRT character generator so as to provide a "hard copy" printout of the information being displayed on video display 2013 as well as the video display thereof. However, this is optional to the preferred video display terminal 28 of the present invention and need not be included unless such "hard copy" printout is desired. If such an instance, a printer interface 8004 and conventional printer 8006 would be required and would preferably be controlled by the microprocessor 6000 in a manner to be described in greater detail hereinafter. It should be noted that if such printing function is not desired then the functions of the terminal 28 and microprocessor 6000 relating thereto may be omitted; however, for purposes of completeness the programmed terminal will be described, as illustrated in the aforementioned program, assuming such printer and printing function is included.

Referring once again to FIGS. 5 and 7 and the aforementioned program, in order to facilitate the understanding of the functioning of the illustrative conventional program in the system of the present invention, the following detailed description of the inputs from keyboard 5000, the inputs from the printer interface 8004, the inputs from the terminal 8000, the outputs to the keyboard 5000, the outputs to the printer interface 8004, the outputs to the terminal 8000, the input/output buffer assignments, the random access memory or RAM allocations, the register allocations, and the read only memory or ROM map is provided by way of example below.

It is of course understood that the above program and program description is merely provided by way of example and other programs and program arrangements could be utilized to accomplish the row grabbing video display function of the present invention without departing from the spirit and scope thereof. It should be noted that with respect to FIG. 14 which illustrates the keyboard 5000, the keyboard key switches 823 and associated lines 823a through 823d with their respective associated flip-flop latches 824, 825, 826 and 827, respectively, which are cleared by the signal present on path 6007, are preferably identical in operation with that previously described in the aforementioned U.S. Pat. No. 3,889,054 which description is specifically incorporated by reference herein.

Now referring to FIGS. 8, 24, 25A through C, 26, 26A and 27, the preferred video processor portion of the preferred terminal 28 of the present invention, which video processor is the subject of a commonly owned copending U.S. patent application entitled "Digital Video Signal Processor With Distortion Correction", filed Apr. 23, 1976, bearing U.S. Ser. No. 679,909, shall be described hereinafter. Referring initially to FIG. 24, the composite video input signal via path 402 (FIG. 9) is provided to the video processor 4000 via path 100 which is the video processor input therefor. This signal is conventionally amplified, such as by a conventional sync separator circuit 102 for providing the vertical sync and horizontal sync outputs therefrom in conventional fashion, and a video processing circuit 103 to be described in greater detail hereinafter. Video processing circuit 103 preferably extracts the correct logic data from the input data from the composite video input signal. This input data may contain significant distoritions resulting from a conventional vestigal sideband modulation scheme utilized for transmission as well as from phase delay distortion in a cable transmission system and the bandwidth limitations inherent in the FCC channel allocations. These distortions generally occur in any television transmission and are not normally compensated for due to the low level fidelity requirements of conventional television transmission and display. The nature of the aforementioned distortion is illustrated in FIGS. 25A through 25C. FIG. 25A represents the original undistorted transmission; FIG. 25B represents the received signal which is provided to a distortion compensation circuit 111, to be described in greater detail hereinafter, which is preferably included in the video processing circuit 103, this signal containing distortions as previously mentioned; and FIG. 25C illustrates the corrected or distortion compensated signal output from distortion compensation circuit 111. The nature of the distortion as illustrated in FIGS. 25A through 25C is such that zeros and ones do not always return to the same level, the level of each zero and one being dependent on the proceeding few bits of data as illustrated in FIG. 25B. It should be noted at this point that although the amplitudes of the signals in FIGS. 25A and 25C are different, if desired these amplitudes could be the same, the primary difference between the original undistorted transmission and the corrected signal output of distortion compensator 111 being the sloping sides of the pulses present in the compensated or corrected signal of FIG. 25C.

The received signal illustrated in FIG. 25B which contains the aforementioned distortions is preferably provided to the video processing circuit 103 via path 105 which provides this signal to a conventional synchronous clamp circuit 104. As shown and preferred in FIG. 24, this received signal provided via path 105 to synchronous clamp 104 is shorted through a reference voltage at a point A at the beginning of each horizontal scan line. This is preferably accomplished by field effect transistor 106 which acts as a fast switch. A gate pulse provided via path 107 (FIG. 26) which is termed the "keyed clamp gate" signal, turns field effect transistor 106 on through transistor driver 108 for a short part of the back porch of the signal, this interval being defined as after the color burst which is in region H and before the start bit which is the pulse in region J (FIG. 2). As a result, the DC level of the data line, represented by path 109, is the same at the start of each horizontal scan line. Without the presence of this synchronous clamp 104, the DC level would change from line to line depending on the relative number of ones and zeros defining the data content of each line. Synchronous clamp 104 insures that the preferred video display terminal 28 will respond properly to the first data line that follows a group of empty or no data lines. The output of synchronous clamp 104 via data line 109 is preferably provided to distortion compensation circuit 111 through a conventional isolation amplifier 110 whose input is the data line 109, with isolation amplifier 110 preferably having a high input impedance which prevents any average change of DC level during a horizontal scan line period. As illustrated in FIG. 25B, when distortion is present in the received signal, there is a significant peak-to-peak change although the value of the location of the peaks may differ. Distortion compensation circuit 111 preferably responds to the one-to-zero or zero-to-one transitions in the signal. When the transition is a zero-to-one transition, as shown and preferred in FIG. 24, the right side or output side of capacitor 112 rises positively tracking the transition up to a predetermined voltage, such as preferably the normal threshold voltage of the silicon diodes 113 and 114 utilized in the distortion compensation circuit 111, this voltage preferably being by way of example 0.7 volts. At that time, this diode 113, which as previously mentioned is preferably a silicon diode, conducts clamping the voltage at this predetermined value of 0.7 volts even though the input data may continue to rise in voltage value. This predetermined voltage is maintained until a one-to-zero transition occurs. As soon as the input voltage reverses direction, as caused by a one-to-zero transition, silicon diode 113 cuts off and the voltage at the right side or output side of capacitor 112 falls tracking the input change until, preferably, it reaches the negative equivalent of this predetermined preferred voltage which, in the example given, is -0.7 volts. At that time silicon diode 114 which preferably has the same threshold level as silicon diode 113, conducts clamping the voltage at this predetermined level, which by way of example is -0.7 volts, until a zero-to-one transition occurs at which point silicon diode 114 cuts off and the aforementioned cycle repeats. In this manner a signal is produced which has all ones and all zeros of the same respective levels as illustrated in FIG. 25C. For a string of unchanged ones or unchanged zeros, that is where no transition occurs, the clamped voltage output due to silicon diodes 113 and 114 which is provided to the gate or output field effect transistor 116 of distortion compensation circuit 111 would normally tend to decay as a result of capacitor 112 discharging. In order to compensate for this, an RC network 115 is connected in parallel across the capacitor 112--diode 113-114 network to provide a reverse tilt or compensating charge voltage to this decaying voltage to the gate of field effect transistor 116 in order to compensate for the decay. This field effect transistor 116 buffers the clamped output signal and supplies it to a conventional output amplifier 117 whose output is in turn coupled to a conventional logic converter 118 through a conventional biasing network 119 to provide a true digital data signal at the output of logic inverter 118 which is equivalent to the original transmitted signal as illustrated in FIG. 25C, amplifier 117 amplifying the amplitude of the signal illustrated in FIG. 25C to the amplitude of the signal level illustrated in FIG. 25A. Biasing network 119 preferably insures that the switching of logic inverter 118 occurs about midway on the clamped signal transition, the output of logic inverter 118 being digital data.

Referring now to FIG. 26, another portion of the video processing circuit 4000 is shown. Preferably, the digital data output of logic inverter 118 still contains the color burst which was present in region H of the signal and would normally remain in the transmitted signal if a color TV receiver was being utilized for the display terminal. However, when a monochromatic digital TV terminal is utilized, the color burst signal must preferably be omitted or gated out. It should be noted that, if desired, if only monochromatic TV terminals are to be utilized, then the color burst which would allow for color TV display would preferably be omitted all together from the received signal although the preferred terminal 28 would still function if such color burst were present, as described hereinafter. The color burst removal circuit illustrated in FIG. 26, as shown and preferred, includes a conventional separator D-type-divide-by-2 flip-flop 120 which preferably receives a 5.1 megahertz clock at the clock input and is preferably cleared by the horizontal sync provided from sync separator circuit 102 via path 102a (FIG. 24). Flip-flop 120 is preferably connected in a toggle mode so that its output, which is provided via path 121, is a series of 2.55 megahertz pulses, in the example given, that start at the completion of the horizontal sync pulse. A conventional decade counter 122 is preferably connected so as to receive these pulses and count these pulses so as to generate an output at the tenth pulse which output is provided via path 123, inverted and then provided through another conventional flip-flop 124 which is initially set by the horizontal sync pulse provided via path 102a and reset by the output signal provided via line 123. The output of flip-flop 124 is preferably a pulse which starts with the beginning of the horizontal sync pulse and ends at the completion of the count of the tenth clock pulse as indicated by the presence of a signal via path 123. The width of this pulse output of flip-flop 124 is preferably such so as to continue beyond the color burst present in region H but to and prior to the initiation of the start bit in region J. This pulse clears a conventional flip-flop 125 whose clear input is connected to the output of flip-flop 124 and whose clock input is preferably clocked by the data input provided from logic inverter 118 via path 118a (FIG. 24). As a result, the output of flip-flop 125 which is provided via path 126 is preferably low during the horizontal sync and color burst periods and is clocked high by the start bit of region J contained in the logic inverter 118 output provided via path 118a. The signal which is present on path 126 enables a conventional NAND gate 127 which has two inputs, with the other input being connected to the output of logic inverter 118 via path 118a. As a result, the output of NAND gate 127 is gated data which includes the start bit and subsequent data, such as present in regions B through K, but excludes the color burst. A second conventional two input NAND gate preferably receives the horizontal sync signal provided via path 102a as one input and the inverted gate signal from flip-flop 125 as its other input. As a result, NAND gate 128 preferably generates a pulse which starts at the trailing edge of the horizontal sync signal and terminates with the start bit. This output signal from gate 128 is preferably utilized as a delayed horizontal sync signal for use by other circuits in the preferred terminal 28 as will be described in greater detail hereinafter. It should be noted that the delayed horizontal sync output of gate 128 which is preferably terminated with the start bit essentially makes the system insensitive to any jitter or noise that might be present in the original horizontal sync signal provided via path 102a. As shown and preferred, the aforementioned keyed clamp gate signal provided via path 107 through video processing circuit 103 is preferably provided as the output of a third conventional two input NAND gate 129 whose inputs are the non-inverted output of flip-flop 124, which is the extended sync signal (FIG. 26A), and an output from counter 122 which is preferably high after counting eight pulses as opposed to the tenth pulse count output provided via path 123 to flip-flop 124. As a result, the output of gate 129 is a pulse that starts eight clock pulses after the end of the horizontal sync and ends after the tenth clock pulse; in other words, the gate output pulse via path 107 occurs between the eighth and tenth counts of counter 122. This pulse defines the aforementioned unused region on the back porch between the end of the color burst in region H and the start of the start bit in region J, this area being designated by reference numeral 11 in FIG. 2.

It should be noted that the aforementioned conventional sync separator circuit 102 is preferably identical with that previously described in the aforementioned U.S. Pat. No. 3,889,054 which description is specifically incorporated by reference herein.

Referring now to FIG. 8, the preferred phase locked loop 4100 (FIG. 9) of the preferred synchronization timing portion of the preferred terminal 28 of the present invention is shown. This phase locked loop 4100, which is the subject of a commonly owned copending U.S. patent application entitled "Phase Locked Loop For Providing Continuous Clock Phase Correction", filed Apr. 23, 1976 bearing U.S. Ser. No. 679,701, preferably being utilized in place of the phase locked loop arrangement described in the aforementioned U.S. Pat. No. 3,889,054. Phase locked loop 4100 preferably includes a conventional voltage controlled crystal oscillator 130 whose nominal frequency, which is preferably by way of example 5.1136 megahertz, is set by a conventional crystal 131. This frequency can preferably be varied over a small range by adjusting the voltage present at input 132. The output of voltage controlled oscillator 130 is preferably fed back to a conventional D-type flip-flop 133 at the D input as well as being provided via path 401. The gated data output from gate 127 (FIG. 26), which is preferably provided via path 127a, is preferably provided to the clock input of flip-flop 133 which flip-flop preferably acts as a phase detector. Preferably, whenever a zero-to-one data transition occurs while the clock is high, flip-flop 133 is in a set state. If it occurs when the clock is low, flip-flop 133 is preferably then in the reset state. Thus, a change in the output of flip-flop 133 occurs only when the phasing of the clock changes with respect to the phasing of the data. It should be noted that preferably the voltage controlled oscillator 130 is the same as the voltage controlled oscillator forming part of the phase locked loop described in the aforementioned U.S. Pat. No. 3,889,054. As shown and preferred, a pair of conventional two input NAND gates 134a and 134b connect the output of the phase detector 133 to the oscillator control circuit only during the valid data period which is preferably defined as the time of the horizontal scan line including the start bit when data can be present. One input to gates 134a and 134b, which input is connected in parallel thereto, is the data gate output from flip-flop 25 provided via path 125a (FIG. 26). The other input to gate 134a provided via path 135 is the inverted output of flip-flop 133 while the other input provided via path 136 to gate 134b is the non-inverted output of flip-flop 133. If path 135 is high, it denotes that the clock leads the data in phase whereas if path 136 is high it denotes that the clock lags the data in phase. When path 136 is high, a diode 137, which is preferably connected to the output of gate 134b through an inverter, charges a capacitor 138 connected to the output thereof in a positive going direction. As long as the phase lag condition remains, the voltage continues to rise. In the leading phase condition, that is with path 135 high, this preferably causes capacitor 138 to discharge through a diode 139 connected to the output of gate 134a. In the normal closed loop condition, the phase varies between a very small leading and very small lagging angle, such as by way of example, plus or minus 10 degrees, as necessary to maintain a constant voltage on capacitor 138. This voltage is preferably amplified by a conventional FET operational amplifier 140, such as an Intersil 8007C, to provide the control voltage for the voltage controlled oscillator 130 via path 132. It should be noted that for the phase locked loop 4100, if a full line of data is present, phase locked loop 4100 will utilize every data transition to continuously correct the clock phase whereas in the case of a series of empty or non-data lines being transmitted, phase locked loop 4100 will make a single correction each line utilizing the start bit which correction will be adequate to insure that phase lock exists at the beginning of the first non-empty or data line. Thus, phase locked loop 4100 can maintain phase lock to a single start bit rather than to the entire clock burst as well as utilizing every data transition to continously correct clock phase. Thus, phase locked loop 4100 is an improvement over the phase locked loop arrangement described in the aforementioned U.S. Pat. No. 3,889,054 which only utilizes the clock burst for phase lock rather than utilizing every data transition. In order to prevent large over corrections during periods when lines are not transmitted such as when not even the start bit is transmitted, for example during the vertical blanking interval, during which periods it is important that the last error state at the output of gates 134a and 134b does not continue to charge or discharge capacitor 138, such as if the last error state was in a leading condition where over correction would continue to drive through into a lagging condition, gates 134a and 134b are turned off except when an active line, which is defined as a line including a start bit, such as in region J, is received via data gate path 125a. This data gate path 125a preferably goes high coincident with the start bit and goes low at the beginning of the following horizontal sync pulse so as to indicate the presence of an active line and to thus prevent the occurrence of correction during periods when lines are not transmitted.

Referring now to FIG. 12, which is a logic diagram, partially in schematic, of a portion of the timing and keyboard control portion of the preferred terminal 28 of the present invention illustrated in FIG. 9, the circuit shown therein is essentially similar to that described with reference to FIG. 8 the aforementioned U.S. Pat. No. 3,889,054 with the exception of the improved phase locked loop portion 4100 previously described with reference to FIG. 8 and with the exception of an improved error check circuit 432 in place of the error check circuit described in the aforementioned U.S. Pat. No. 3,889,054. As shown and preferred in FIG. 12, the input to error check circuit 432 is preferably resynchronized data out of flip-flop 760 which, as described in the aforementioned U.S. Pat. No. 3,889,054 is an output which preferably follows the input data line except that it will be synchronized with the clock B signal by the clock of flip-flop 760 provided via path 435 which is the same data as utilized by the balance of the preferred video display terminal 28 of the present invention. The resynchronized data preferably goes to the A input of a conventional single bit adder 437 which produces the sum of two input bits present at inputs A and B thereof with the B input initially being zero. This sum is preferably provided via path 439 to the input of a conventional seven bit shift register 441. The output of shift register 441 is preferably provided to the B input of adder 437 through a conventional inverter 447. As a result of the seven bit delay provided by register 441, adder 437 at any given time adds an input bit of a given character, since the system preferably utilizes seven bits per character, with the same bit of the previous character. At the beginning of the line, shift register 441 is preferably cleared so that the B input of adder 437 is again logic zero for the duration of the first character. When the first bit of the second character is received at the A input of adder 437, the first bit of the first character is then present at the B input of adder 437 and the output of adder 437 provided via path 439 is the sum of these two bits. At the beginning of the third character, the B input of adder 437 represents the sum of bit one from the first two characters. In this manner for the remainder of the line, that is the pseudo video scan line, the individual bits of the characters are added and accumulated. This is preferably a serial process which is serially repeated for each of the seven bits of the character; for example, for bit one of character 10, the B input of adder 437 is equivalent to the sum of bits one of characters one through nine and for bit two of character 10 the B input to adder 437 is equivalent to the sum of bits two of characters one through nine, etc. Adder 437 provides a carry output where required by binary addition; namely if one and one are added, the sum is zero in that position plus a carry of one to the next position. This carry output is present on path 451 and is preferably applied to the D input of a conventional flip-flop 443 via a conventional two input NAND gate 445. Flip-flop 443 is preferably clocked by the system clock provided via path 449 so that the non-inverted output of flip-flop 443 represents the carry output of adder 437 delayed by one bit. Thus, the carry output present via path 451, such as, by way of example, the output that results from the addition of bit one, is present at the C input of adder 437 when the bit two addition is taking place. The bit two addition then is a full addition resulting from the data bits at A and B of adder 437 as well as the carry state at C of adder 437. Preferably, a carry output of bit seven is ignored which is accomplished by gate 445 which turns off the input to flip-flop 443 during bit seven in response to a bit seven pulse from the decoder 412 whose operation is described in the aforementioned U.S. Pat. No. 3,889,054, which pulse is provided via path 453. It should be noted that the output of adder 437 via path 439 preferably represents in serial form the accumulated sum of individual character bits. Preferably, during the 38th character, which is preferably the error check character, the accumulated sums will all be ones after the error check character is added in if no error is present. The inverted output of adder 437 is provided via path 455 to another conventional flip-flop 457. Path 455 preferably must be in the zero state during the seven bits of the 38th character for the received pseudo video scan line to be considered a true or valid line. Flip-flop 457 tests for this condition by being kept in a set state due to a negative preset signal at all times except during character 38; in other words, except when the 38th character pulse is provided. Path 455 is connected to the K input of J-K flip-flop 457 so that if path 455 is high during any bit of the 38th character, flip-flop 457 will be reset. Thus, a negative pulse at the noninverted or Q output of flip-flop 457 during the 38th character will indicate an error. Accordingly, the improved error check circuit 432 of the present invention adds on a word-by-word basis as opposed to a bit-by-bit basis as described in the aforementioned U.S. Pat. No. 3,889,054 and accomplishes this summing or addition in serial fashion rather than parallel fashion.

Now referring to FIG. 17 which illustrates a portion of the preferred memory output processor portion of the preferred terminal 28 of the present invention, this portion is preferably identical with that described in the aforementioned U.S. Pat. No. 3,889,054 with the exception of the erase circuit and the special character latching functions utilized with 32-or-64 character selection, if such selection is desired, with these exceptions to be described in greater detail hereinafter with reference to FIG. 17. Identical reference numerals are utilized for identical functioning components in FIG. 17 with those previously described in the aforementioned U.S. Pat. No. 3,889,054 with reference to FIG. 12 of that patent which description was specifically incorporated by reference herein and will not be repeated. It should be noted that reference numeral 901 represent a special character latch, with the fifth character preferably being the special character, and, accordingly, latch 901 is strobed via path 903 during the fifth character. Any time a new valid pseudo video scan line is received, the special character, which is as previously mentioned preferably the fifth character, of that pseudo video scan line is entered into latch 901. The output of latch 901 is preferably provided to microprocessor 6000, as represented by the symbol "SCH" to provide an input thereto. Preferably, microprocessor 6000 does not act on all special characters such as for example not acting with respect to 32-or-64 character selection, although, if desired, the microprocessor could also be programmed to accomplish this.

Now considering the preferred erase circuit portion 905 of the memory and output processor portion illustrated in FIG. 17, the erase circuit 905 provides a means for making the screen of the video display device 2013 go blank. This is desirable when the preferred terminal 28 of the present invention is utilized in row grabbling system 10, as at initial turn on the main memory would come on with a random bit pattern which in turn would display a random assortment of characters which would be meaningless and, perhaps, confusing to the viewer. Erase is accomplished by loading all character locations in memory which comprises stages 1030 through 1042 preferably with an octal 40 value which is the value which corresponds to a "space" in ASCII code. The erase cycle can be initiated also by the microprocessor 6000 in response to the appropriate special character assigned for the erase function or whenever a keyboard cycle takes place, if desired. The erase cycle is preferably initiated by a conventional flip-flop 907 which gets set initially when power is on as a result of an RC charging network 909 through inverters 911 connected to the preset input of flip-flop 907. Flip-flop 907 can also preferably get set by microprocessor 6000 via the clock input provided via path 913 from microprocessor 6000. Once set, flip-flop 907 preferably enables a conventional two input NAND gate 925 which also receives the vertical sync pulse via path 923. Therefore, the first vertical sync pulse that occurs after flip-flop 907 is set appears as a negative pulse at the clock input of a second flip-flop 915 to which the output of gate 925 is connected. Flip-flop 915 is clocked to its set state preferably at the trailing edge of the vertical sync pulse. When set, the output of flip-flop 915 preferably switches a conventional multiplexer 917 such as a Texas Instruments SN 74157N, which accomplishes the switching necessary to load the memory octal 40 values; specifically, switch 917a and 917b illustratively shown separate in FIG. 17 actually comprise the multiplexer 917 with switch 917a of multiplexer 917 loading a logic one into the correct bit position of character latch register 468 whose operation is described in detail in the aforementioned U.S. Pat. No. 3,889,054, to obtain octal 40 therefrom. All of the other bits are preferably set to zero via path 919 which is connected to the inverted output of flip-flop 915 with the signal present thereon turning off the input to shift register 466 previously described in detail in the aforementioned U.S. Pat. No. 3,889,054, via a conventional two input NAND gate 921 whose inputs are the output of shift register 457, which is also previously described in the aforementioned U.S. Pat. No. 3,889,054, and the inverted output of flip-flop 915. As a result, all logic zeros are present at the outputs of shifts register 466 which logic zeros are provided to the inputs of character latch register 468 except for the previously mentioned single bit. Switch 917b of multiplexer 917 switches multiplexer 1020, which is described in detail in U.S. Pat. No. 3,889,054, so that the read row addresses are applied to the memory stages 1030 through 1042 even though the memory stages 1030 through 1042 are in the write mode during the erase cycle. This insures that all memory locations are addressed during erase. Termination or the erase cycle is preferably accomplished by the second vertical sync pulse that occurs after flip-flop 907 has been set. The second vertical sync pulse clocks flip-flop 915 back to its original state and also clears flip-flop 907 via another conventional two input NAND gate 927 whose output is provided to the clear input of flip-flop 907 through an inverter 929. Thus, both flip-flop 907 and 915 are returned to their original state. The period of the erase cycle is thus preferably one vertical field period, this time being adequate to insure that all memory stage locations 1030 through 1042 have been addressed. As was previously mentioned the balance of the circuitry illustrated in FIG. 17 is adequately described in detail in the aforementioned U.S. Pat. No. 3,889,054 which was specifically incorporated by reference herein.

Referring now to FIG. 16, the preferred permission memory circuitry utilized in the preferred row grabbing terminal 28 of the present invention shall be described in greater detail. Except where otherwise specifically noted, the portions of the circuit in FIG. 16 are preferably identical with the permission memory circuitry described in the aforementioned U.S. Pat. No. 3,889,054, with reference to FIG. 11 of that patent and identical reference numerals are utilized therefor. Thus, this description which is incorporated by reference herein will not be repeated and the following discussion of the preferred permission memory circuitry will primarily be directed to the differences over the permission memory circuitry described in U.S. Pat. No. 3,889,054. In the previous permission memory circuit of U.S. Pat. No. 3,889,054, counters 974 and 976 were initially cleared to zero rather than being preloaded to a predetermined number such as 254. Thus, this previous system could possibly provide a couple of extraneous clock pulses before the first permission bit was received so that the counter was at a value of one when the first permission arrived and, accordingly, group 0 would not be available. In the preferred permission memory circuit shown in FIG. 16, counters 974 and 976, which are four bit counters, constitute a 256 bit counter, although if desired a 128 bit counter could be utilized instead. This counter, which is comprised of counters 974 and 976, is preferably utilized for addressing the permission memory 462 during the permission write cycle. Counters 974 and 976 are preferably initially loaded to a predetermined value, such as preferably 254, which is accomplished by utilizing the load input 931 of counter 976 in conjunction with the preset inputs 933a and 933b of counters 974 and 976, respectively. As a result of the preload via path 931, at the beginning of a permission write cycle, the first two clock pulses advance the counter 974-976 to 0. The second clock pulse occurs just prior to the availability of the first permission bit. This timing sequence makes it possible for the first permission bit to represent group 0. Thus it is possible to insure permissions for groups 0 through 127 on one pseudo video scan line. As shown and preferred in FIG. 16, 128 groups are provided via the connection of permission memory 462; however, as will be described in greater detail hereinafter, if desired, additional counters may be provided in conjunction with permission memory 462 so as to provide up to 1000 groups.

The preferred permission memory circuit illustrated in FIG. 16 also enables self-termination of permission write to return the system to the normal mode if a permission line occurs at the end of a field. This is accomplished in the following manner. An inverter and NAND gate constitute a decoder 935 that generates a pulse preferably at a count, such as 192, which is greater than the number of groups, which were previously mentioned as being 128 by way of example in the arrangement illustrated in FIG. 16, and less than the number of bits in a data line to permission memory, which number of bits preferably constitutes 238 by way of example. The number 192 is preferably chosen for ease of decoding although it could be any number between 128 and 238, the criteria for these limits being defined as being greater than the number of groups and less than the number of bits in a data line to permission memory. This pulse is provided through an inverter 937 to one input of a conventional two input NOR gate 939 whose other input is the keyboard active line 941 and whose output is utilized to reset the permission flip-flop 960. Thus flip-flop 960 is guaranteed to get reset even if the completion of permission write occurs during vertical blanking. In such a case, flip-flop 960 would not otherwise get reset because the decoded character 41 pulse present on path 413 is not normally generated during vertical blanking. During a keyboard cycle, the group address is preferably not a valid signal. Therefore, the output of the permission memory 462 would not be valid. Accordingly, in order to prevent an erroneous authorization, flip-flop 960 is held in a reset state during the keyboard cycle. This is accomplished by the keyboard active line or KAC 941 which is preferably generated by microprocessor 6000, which as was previously described, is applied to the other input to NOR gate 939. Preferably, the preferred permission memory of FIG. 16 provides permission initialization when the power is turned on, the permission memory circuit automatically selecting group 0, page 0 at such time. When the first permission line is received by receiver 28, the permission memory circuit of FIG. 16 will then preferably revert to normal operation. This initial mode, group 0, page 0 on turn on, is preferably established by flip-flop 953 which is initially preset by line 955 (FIG. 17) when power is turned on. The output of flip-flop 953 is preferably applied to a conventional NOR gate 957 via line 959 and sets the permission O.K. line 556 through an inverter 961. This asserts permission. Line or path 959, which is also preferably connected to gate 963 of the page address circuit (FIG. 15), and is termined pre-permission, also sets line 508 low. This simulates an address of page zero, group zero.

Referring now to FIG. 18, the preferred permission memory update circuitry for updating the permission memory 462 (FIG. 16) is shown and will be described. As will be described in greater detail hereinafter, the purpose of the preferred permission memory update circuitry is to avoid any flicker which might otherwise occur in the video display on display device 2013 during update of the permission memory as a result of such update. In the preferred circuitry of FIG. 18, the previously considered clock B signal is replaced by the prime clock provided via path 401 from the voltage controlled oscillator 130 (FIG. 3). This prime clock input provided via path 401 to the permission memory update circuitry is inverted by an inverter 965 and applied to a conventional single-shot 967. Preferably, the purpose of single-shot or one-shot 967 is to generate a symmetrical square wave which is required by the conventional frequency doubler circuitry comprising inverters 1074 and 1076 and exclusive OR gate 1078. As described in the aforementioned U.S. Pat. No. 3,889,054, inverters 1074 and 1076 provide a predetermined delay, such as 100 nanoseconds in the prime clock signal provided via path 401, this delay time preferably representing a fraction of the clock period. It is this delayed prime clock signal which is preferably supplied to one input of the two input exclusive OR gate 1078 whose other input is directly connected to the prime clock input provided at the output of single-shot 967. Exclusive OR gate 1078 preferably provides an output only during the period of time that the delayed prime clock signal overlaps the undelayed prime clock signal, which occurs twice per clock period and, as a result, two output pulses are available from gate 1078 for each input pulse. As shown and preferred in FIG. 18, the output of exclusive OR gate 1078 is fed to a conventional switch 1079 which selects between the 32 character position and the 64 character position (assuming such selection is desired) depending on the desired character display, with the exclusive OR gate 1078 output being connected to the 64 character position and the non-doubled or direct prime clock signal being connected directly to the 32 character position of switch 1079. The output of switch 1079 is provided to the clock input of a conventional divide-by-eight counter 1080 previously described in the aforementioned U.S. Pat. No. 3,889,054, such as the type manufactured by Texas Instruments under the designation SN74161N, which is a four bit binary counter connected as a divide-by-eight counter, although, if desired, a conventional divide-by-eight counter could be utilized. Thus, the frequency doubled clock signal is preferably utilized as the clock for counter 1080 only during 64 character operation while, during 32 character operation, that is 32 characters per video row versus 64 characters per video row, counter 1080 is clocked directly by the prime clock signal. The prime clock signal provided via path 401 is preferably unaffected by the permission write mode (FIG. 12). The normal horizontal sync signal provided via path 969 is preferably utilized for the horizontal timing of all of the display circuits including the memory read circuit. The horizontal sync on path 969 is also unaffected by the permission write mode as it is the normal sync during the permission write mode.

The clear signal for counter 1080 is preferably developed as follows. Counter 971 in conjunction with flip-flop 973 establishes the start of a display video row. During horizontal sync, counter 971 is preloaded to a count which depends on whether the preferred terminal 28 is operating as a 32 or 64 characters per row terminal, the preload condition for counter 971 is preferably selected as one value, such as preferably 15, and when the preferred terminal 28 is operating in the 64 characters per row mode the preload for counter 971 is preferably selected as another value, such as preferably 11. Whichever value is slected, that value is selected so as to obtain the correct starting position on the display screen. A different preload condition is preferably required for the 32 and 64 characters pers per row modes because preferably a different clock frequency is required for these two modes.

At the conclusion of the horizontal sync pulse, counter 971 preferably starts counting at the prime clock rate which prime clock is provided to the clock input of counter 971 from single-shot 967. At the completion of the count, the output of the two input NAND gate 975, which has one input connected to the non-inverted output of flip-flop 973 and the other input connected in parallel to the clock input to flip-flop 973 prior to inversion thereof, goes low and stops the counting using the P inhibit input of counter 971. Preferably a predetermined count value, such as preferably a value of 31, terminates the count cycle. The negative level present at the output of gate 975 is also provided in parallel through an inverter 977 and applied to the clear input of counter 1080 which enables counter 1080 by removing the clear state and counter 1080 starts its count sequence. Thus, by utilizing the prime clock for row one and the normal horizontal sync 969 together with the preferred circuitry for enabling counter 1080, the display circuits are preferably independent of the permission memory update and, thus, any flicker which might otherwise occur in the display during such upate as a result thereof is minimized and preferably avoided. The balance of the circuitry associated with FIG. 18 is preferably identical with that described in U.S. Pat. No. 3,889,054 with reference to FIG. 13 thereof.

Referring again now to FIG. 16, the preferred main memory update circuitry shall be described in detail hereinafter. Preferably, under normal conditions when the main memory which comprises stages 1030 through 1042 (FIG. 17), is in the write mode, the outputs of these stages 1030 through 1042 are not valid. Thus, as shown and preferred in FIG. 16, update control circuitry 979 is provided to prevent the main memory updating or writing if a valid pseudo video scan line is received while a row is being displayed. If a valid pseudo video scan line is received while a row is being diplayed, other than during the occurence of dead space on the screen, this line will preferably be stored in shift register 457 (FIG. 17) until completion of that displayed row and the line then would be written into memory 1030 through 1042 during the dead space following the row. When a valid pseudo video scan line is received, path 981, which is connected to the output of NAND gate 986, goes low and sets a flip-flop 983 which remains set preferably until the display sweep completes the 12th line counted of a row which is equivalent to line 11 which is preferably a dead line of a row with lines 1 through 9 of the row preferably being considered active lines for a row. The information corresponding to line 11 is provided to a three input NAND gate 985 which decodes the count of 11 provided from the output of line counter 1056 (FIG. 18) and provides a negative pulse to the clock input of flip-flop 983. Flip-flop 983 is then preferably reset at the end of that pulse; that is, at the completion of the 12th line counted which is equivalent to line 11. In this regard, it should be noted that the zero line is the first line counted and, therefore, the 11th line or a count of 11 is the 12th line counted. The time during which flip-flop 983 is set preferably establishes the time during which the input data must be stored in the one line shift register 457 (FIG. 17) which preferably stores this information as long as flip-flop 983 is set plus one more line to enable for shifting out and writing into memory 1030 through 1042. Update control circuit 979 also preferably includes a two input NAND gate 987 which has one input connected in parallel to the Q or non-inverted output of flip-flop 983 and the other input connected to the output of NAND gate 985. The output of gate 987 is preferably provided through an inverter 989 to one input of another gate 991 whose other input is the data gate signal provided via path 993 from flip-flop 696 (FIG. 12). The output of gate 991 is the gate hold signal which is provided via path 997 to flip-flop 1002 (FIG. 17) and is the control line which stopes the clocking of shift register 457 (FIG. 17) during the period that flip-flop 983 is set excluding the 12th line counted; in other words, the output of gate 991 ivia path 997 goes low when a valid pseudo video scan line is received and goes high at the beginning of the 12th line counted, which is equivalent to line 11. The output of gate 991 preferably cannot go high until the inverted data gate line 993 goes low. This continues to hold data in the serial shift register 457 (FIG. 17) preferably until the fourth character. Thus, the clocking of shift register 457 is preferably enabled at the beginning of the fourth character of the transmitted pseudo video scan line after the 12th line counted (line 11) of the displayed row is started. As was previously mentioned, the output of gate 991 is preferably applied to the preset input of flip-flop 1002 via path 997 (FIG. 17). The Q output of flip-flop 983 is also preferably connected in parallel to one input of another two input NAND gate 999 whose other input is connected through an inverter to the output of gate 985. Gate 999 preferably produces a negative level during the 12th line counted (line 11) which terminates a hold period. This 12th line counted is the time during which data is preferably written into the main memory 1030 through 1042 (FIG. 17). The low level at the output of gate 999 is preferably applied to the main memory control (read) line 446 via a two input NOR gate 1001, whose other input is the inverted erase signal, through an inverter 1003. Gate 1001 preferably causes the memory control read line 446 to also go low during an erase cycle as a result of the inverted erase input applied thereto. As was described in the aforementioned U.S. Pat. No. 3,889,054, the output of gate 994 which is provided through an inverter 996 preferably generates the memory pulse R/W via path 995 which is provided to the memory 1030 through 1042 through inverter 1047 and gate 1046 via path 1044 (FIG. 17). Thus, the operation of the update control circuit 979 preferably prevents any possible flicker in the display during the main memory write mode. The balance of the circuit of FIG. 16 is preferably identical with that described with reference to FIG. 11 of the aforementioned U.S. Pat. No. 3,889,054 with the exception of the differences previously noted above.

Referring now to FIGS. 4 and 28, the computer/printer interface 8004, which is the subject of a commonly owned copending U.S. patent application entitled "Interface For Enabling Continuous High Speed Row Grabbing Video Display With Real Time Hard Copy Print Out Thereof", filed Apr. 23, 1976, bearing U.S. Ser. No. 679,907, which was generally referred to with reference to FIG. 5 with respect to the option of utilizing a printer to provide hard copy text in addition to the video display of information shall now be described in greater detail for completeness. As was previously mentioned, the above described preferred system will function as a row grabbing terminal 28 without the additional computer/printer CRT display interface 8004 and printer 8006, if such hard copy text material is not desired, without departing from the spirit and scope of the present invention. However, assuming such hard copy text is desired, the computer/printer interface 8004 shall now be described with reference to FIGS. 4 and 28. As will be described in greater detail hereinafter, the interface 8004 preferably utilizes character information when available at a high speed rate so as to enable continuous high speed video display of the information which is normally preferably provided with the preferred row grabbing terminal 28 of the present invention, as well as with the row grabbing terminal described in U.S. Pat. No. 3,889,054, while also enabling real time pick off of this information for printing. The printer 8006 is preferably a conventional matrix printer, such as an Extel Model No. AF-11, whose operations are preferably controlled by the microprocessor 6000. The following functions are preferably commanded by the microprocessor 6000: PRINTER WRITE, PRINT, LINE FEED, CARRIAGE RETURN, SPACE COMMAND, and the 32/64 CHARACTER COMMAND, as indicated by the control lines illustrated in FIGS. 4 and 5. The PRINTER WRITE COMMAND from the microprocessor, provided via path 1007 from microprocessor 6000, preferably causes the printer interface 8004 to write a row of characters into a buffer memory 1011 from the terminal main memory 1030 through 1042 as will be described in greater detail hereinafter. The PRINT COMMAND, provided via path 1013 from microprocessor 6000, causes the interface 8004 to output the row of characters to the printer 8006 at the proper baud rate for the printer 8006, such as at 110 baud, in serial EIA standard format including start and stop bits for the printer 8006. The LINE FEED COMMAND, provided via path 1015 from microprocessor 6000, preferably causes the interface 8004 to issue an ASCII line feed character to the printer 8006 in the same format as the characters; that is, in serial EIA standard format at the same rate, such as the 110 baud rate. The CARRIAGE RETURN COMMAND, provided via path 1017 from microprocessor 6000, preferably causes the interface 8004 to issue an ASCII carriage return character to the printer 8006 in the same format as the characters. The SPACE COMMAND, provided via path 1019 from the microprocessor 6000, preferably causes the interface 8004 to issue an ASCII space character to the printer 8006 in the same format as the characters. The 32/64 CHARACTER COMMAND, provided via path 1021 from microprocessor 6000 to interface 8004, preferably causes the interface 8004 to write into its memory 1011 the correct number of characters. Thus, as will be described in greater detail hereinafter, the microprocessor 6000 can establish each printed page format.

Now describing the memory write mode for the interface 8004. When the interface 8004 receives a PRINTER WRITE COMMAND via path 1007, this signal is provided to the clock input of a flip-flop 1023 which is clocked to a set state. This causes a second flip-flop 1025 to subsequently be clocked to its set state by the first horizontal sync pulse occurring after flip-flop 1023 is clocked to its set state. When flip-flop 1025 is set, its Q or inverted output preferably resets or clears flip-flop 1023 via path 1023a. The set output of flip-flop 1025 preferably operates a conventional multiplexer 1027, such as a Texas Instruments SN74157N which is illustratively represented in FIG. 28 by further illustrating its various sections 1027a, 1027b, 1027c and 1027d as switches which are located in FIG. 28 in their appropriate functional positions for purposes of clarity. Thus, the operation of multiplexer 1027 preferably puts the interface 8004 into the write mode. Multiplexer section 1027d connects the horizontal sync pulse to the clock input of a conventional bit counter 1029 which will therefore advance one count for each TV line after the start of the write cycle. The write cycle preferably lasts for eight horizontal lines. During each of the eight horizontal lines, one bit from the main memory 1030 through 1042 of each character is preferably written into the buffer memory 1011. Thus, for example, for line one, no bit is selected; for line two, bit 7 for every character in line two is selected; for line three, bit 6 for every character in line three is selected; for line four, bit 5 for every character in line four is selected; for line five, bit 4 for every character in line five is selected; for line six, bit 3 for every character in line six is selected; for line seven, bit 2 for every character in line seven is selected; and for line eight, bit 2, for every character in line eight is selected; however, it is inverted. A conventional bit select multiplexer 1031 preferably selects the appropriate main memory stage 1030 through 1042 line for each of the eight counts. During the first count, no line is selected if there are only 7 bits per character. During the second count, the memory line corresponding to the 7th bit is selected. This process continues for each line with the memory line for the second bit being selected on the seventh count. As was previously mentioned, on the eighth count the memory line for the second bit is again selected, but it is applied to multiplexer 1030 through an inverter 1033. The purpose of the above procedure is to preferably convert the standard 6 bit ASCII code from the main memory 1030 through 1042 into a 7 bit standard ASCII code which is used by the printer 8006, such conversion preferably being conventional. The output of multiplexer 1031 is preferably applied to the D input of a flip-flop 1035 which is clocked by the PRINTER LATCH CLOCK provided via path 1037 from gate 1094 (FIG. 18) and is preferably the same waveform that operates the character generator 570 (FIG. 18). Flip-flop 1035, as was previously mentioned, is a D flip-flop and, thus, the output follows the input but is delayed by an amount determined by the PRINTER LATCH CLOCK 1037. The purpose of flip-flop 1035 is to preferably provide a logic level which is stable for the full character period to the buffer memory 1011. The clocking of memory 1011 is preferably accomplished via R/W line 1039 through multiplexer sections 1027a which is shown in the write position in FIG. 28, this signal being a clock signal provided by multiplexer 1041. The select inputs a multiplexer 1041 which are 1043a, 1043b and 1043c, preferably select either a low or high level to provide a clock waveform via path 1039 with a period equal to a character period. The timing of the clock waveform 1039 is preferably such that data is clocked in memory 1011 after the data has been loaded into the flip-flop 1035. Memory 1011 is preferably a one-by-1024 bit static random access memory. Thus, it preferably has one data input line and ten address lines with three of the address lines preferably being used for the bit address, allowing 8 bits per character, and six of the address lines preferably being used for the character address, allowing for up to 64 characters, one of these character address lines not being utilized when 32 character lines are to be printed as opposed to 64 character lines. The tenth address line is preferably not utilized. The bit address for memory 1011 preferably comes from the same bit counter 1029 that operates the bit select multiplexer 1031, while the character address preferably comes from a pair of conventional character counters 1045 and 1047. During the write mode of operation, the character counters 1045 and 1047 are preferably utilized simply as latches to store the character address for the main memory 1030 through 1042 and to apply it to the buffer memory 1011. The character address is strobed into the character counters 1045 and 1047 by the PRINTER LATCH CLOCK 1037. The load input to counters 1045 and 1047 is preferably held low during the write mode by flip-flop 1025. It should be noted that the bits need only be selected at the character rate not at the bit rate since a given bit number for each line in the character contains the same information to the character generator input for that character; therefore, for example, bit 1 provides the same bit 1 information for all nine lines; similarly bit 2 provides the same bit 2 information for all nine lines, etc., thus enabling the use of a lower speed circuit thereby slowing things down so that the printer can be operated at normal speed. Thus, the preferred system takes advantage of the eight-to-one differential between the bit and character rates such as, by way of example, in the 64 character mode the bit rate being 10.2 megahertz, while in the 32 character mode the bit rate being 5.1 megahertz and the character rate being approximately 0.64 megahertz. Summarizing the above described memory write cycle, this cycle starts with the bit counter 1029 preferably set for bit 1 which corresponds to count 0, the character counters 1045 and 1047 addressing characters 1 through 32 or 1 through 64 and writing all zeros into memory 1011. At the start of the next TV line, the bit address out of counter 1029 is advanced by one and the character counters 1045 and 1047 again address characters 1 through 32 or 1 through 64 depending on whether it is in a 32 character mode or a 64 character mode and writes bit 7 into memory 1011. This preferably continues through eight counts or bits of the bit counter 1029, writing bits 6, 5, 4, 3, and 2 into memory 1011 as the line changes. At the beginning of the ninth TV line, the D output of bit counter 1029 preferably goes high to the K input of flip-flop 1025 causing flip-flop 1025 to be reset at the trailing edge of the horizontal sync pulse. The resetting of flip-flop 1025 thereby terminates the write mode with flip-flop 1023 having previously been reset after flip-flop 1025 was set.

Now describing the output mode, which with reference to FIG. 28 is a mode in which all multiplexer sections or switches are set at the R or read position, all of these sections or switches being at the W or write position in the write mode. The output mode preferably comprises the print mode, the line feed mode, the carriage return mode and the space mode. During the output mode, the bit clock is preferably generated by a conventional oscillator 1049 which preferably comprises a conventional integrated circuit oscillator such as an NE555V, which is set at a frequency of preferably 110 hertz for the 110 baud rate described above by way of example. This corresponds to a print rate of 10 characters per second using an 11 bit per character format. It should be noted, however, that the printer rate can be set at any desired value merely by changing the oscillator 1049 rate, although the presently preferred printer rate is 110 baud. An 11 bit format preferably consists of one start bit, 7 character bits, one parity bit, which is preferably not utilized, and 2 stop bits. During the output mode, multiplexer 1027 is preferably not activated and the output of oscillator 1049 is preferably applied through multiplexer section 1027d to the clock input of bit counter 1029, which counter can preferably count up to 16 for the example given, although counter 1029 preferably operates in a count 11 mode. Thus, preferably when the output of counter 1029 is equal to binary 10, a gate 1051 connected thereto acts as a decoder and applies a negative level to the clear input of counter 1029 through a two input NOR gate 1053 whose other input is connected to the J output of J-K flip-flop 1025. Preferably, counter 1029 has a synchronous clear so that it is cleared to zero on the next clock following count 11. In addition, during the output mode, counter 1029 preferably establishes the bit sequence of the serial line to the printer 8006 with the counter 1029 count value 0, corresponding to count one, establishing the printer start bit and with counter values 1 through 7, corresponding to counts two through eight, establishing the data bits, count values 9 and 10, corresponding to counts 10 and 11, establishing the printer stop bits, and with count value 8, corresponding to count 9, preferably not being utilized.

With reference to the print mode, microprocessor 6000 preferably initiates a print cycle, as was previously mentioned, by pulsing line 1013. This pulse, provided via path 1013, is the PRINT COMMAND and clock a flip-flop 1055 to its set state. The set output of flip-flop 1055 is preferably applied to the enable inputs P and T of character counter 1045 which then advances when clocked by the D output of bit counter 1029. Thus, character counter 1045 advances at count values 8 of the bit counter 1029 which is the D output of counter 1029. As a result, the memory 1011 is addressed sequentially through the bit and character values that constitute one row. Thus, the output of memory 1011 is a serial bit stream when bits 1 through 8 of character 1 first appear with bits 1 through 8 of subsequent characters following. In each case, bit 1 is a zero level and bits 2 through 8 are the bit values of the character retrieved from memory 1011. The output of memory 1011 is preferably applied to one input of a two input NOR gate 1057, the other input thereto preferably being provided from the reset output of flip-flop 1055 which is low during the print mode. Thus the output of gate 1057 is the inverted bit stream from memory 1011, this inverted bit stream preferably being applied to one input of a negative NAND gate 1059, the other input thereto being low during the print mode so that the non-inverted bit stream is present at the output of gate 1059. The output of gate 1059 is provided as one input to a two input NOR gate 1061, the other input thereto being the D output of bit counter 1029 which is preferably high during bits 9, 10 and 11. Thus, the output of gate 1061 is inverted data including the printer start bit during bits 1 through 8, but is a steady low level during bits 9, 10 and 11. The output of gate 1061 is preferably inverted by an inverter 1063 to provide the final serial output through a level changer 1065 to the printer 8006. The output of inverter 1063 preferably has a high level during bit 1, which is the printer start bit, and follows the data during bits 2 through 8, with this output being low during bits 9, 10 and 11. This corresponds to the prescribed preferred EIA format for a 0 start bit and two 1 start bits. Level changer 1065, which is also a line driver, preferably generates an output signal such as, by way of example, with -12 volts representing binary 0 and -12 volts representing binary 1, these levels being prescribed by the EIA serial line interface standard. At the end of the row, a decoder 1067 preferably provides a negative level to reset flip-flop 1055 with the decoder 1067 preferably being programmed via line or path 1021 from the microprocessor 6000 by setting the appropriate input levels to provide a negative output in the case of a 64 character format at count value 64, which is at the beginning of count 65 of the character counter 1047, and to provide a negative output in the 32 character mode at count value 32, which is at the beginning of count 33 of character counter 1047. This resetting of flip-flop 1055 preferably terminates the print mode.

The microprocessor 6000 can also preferably command a single special function character such as for example LINE FEED, CARRIAGE RETURN or SPACE. Preferably, if the microprocessor 6000 wants a repeated special function character, it must re-command the character after adequate time, such as for example, 0.1 seconds for a 110 baud rate, has elapsed for the first special function character to have been issued to the printer 8006. A multiplexer 1069, such as a Texas Instruments SN74151AN, is provided which preferably functions as a programmable character generator to provide the proper bit sequence that corresponds to the special character which has been requested. The microprocessor 6000 commands a line feed preferably by pulsing line 1015 with the LINE FEED COMMAND. This clears flip-flop 1071 which is normally in a set state with flip-flop 1071 setting a low level at bit input 2 of multiplexer 1069 via path 1071a, and also sets bit input 4 low via NOR gate 1072, all other bit inputs to multiplexer 1069 preferably being high. Multiplexer 1069 preferably sequentially switches the bit inputs 0 through 7 to the output Y under control of the inputs A, B, C from bit counter 1029. As a result, a serial output is applied from multiplexer 1069 to one input of gate 1059. At this time, the other input to gate 1059 is preferably a steady low level so that the single character bit stream is provided to the NOR gate 1061. This bit stream is preferably combined with the stop bits and is applied to the output line to printer 8006 from level changer and line driver 1065 in the same manner as previously described with respect to the print mode. At the completion of one special character, flip-flop 1071 is preferably reset by the D output of bit counter 1029 via multiplexer section or switch 1027c, NAND gate 1073 and inverter 1075, with the output of inverter 1075 preferably being connected in parallel to also provide the character clock for the microprocessor 6000. This preferably completes a line feed cycle.

When a CARRIAGE RETURN COMMAND is requested by the microprocessor 6000, it pulses line 1017 and clears flip-flop 1077 which then sets bit input numbers 1, 3 and 4 low to multiplexer 1069, multiplexer 1069 then generating a single character return ASCII code in the manner as previously described with respect to the LINE FEED code. Flip-flop 1077 is also preferably reset in the same manner as previously described with respect to the LINE FEED by the next character clock from inverter 1075.

When the microprocessor 6000 commands a space by providing the SPACE COMMAND by pulsing line 1019, flip-flop 1079 is cleared which then sets bit input number 6 low to multiplexer 1069 causing multiplexer 1069 to then generate a single space ASCII code in the same manner as previously described with respect to LINE FEED. Flip-flop 1079 is also preferably reset in the same manner as previously described with respect to LINE FEED by the next character clock provided from inverter 1075.

It should be noted that preferably the character counters 1045 and 1047 are cleared during either a line feed or carriage return by the low level from NOR gate 1072. Thus, one row cannot follow another to printer 8006 unless a LINE FEED or CARRIAGE RETURN is issued. However, normally microprocessor 6000 will generate such a CARRIAGE RETURN and LINE FEED between each row. Thus, the interface 8004 enables the microprocessor 6000 to make the following types of decisions: what row to be printed, how many rows, whether to print a 32 or 64 character row, how much marginal space for the row and whether or not extra spaces are to be placed between rows. These decisions are based on keyboard inputs from the operator provided to the microprocessor 6000, and inputs from the transmitted data in the form of special characters, directed messages or override messages provided to the microprocessor 6000. Due to all of the above, the computer/printer interface 8004 enables a system to have the following capabilities: the operator can print a complete page of displayed information or any selected row or group of rows of displayed information; it provides the capability to print an override message, such as an emergency message, without operator involvement as is also true with respect to a directed message; it permits the operator to select a special print mode wherein a row or page which he has selected will be reprinted whenever data on the selected row or page is changed with operator involvement not being required after initial selection of the mode, this operation being activated by a one time per update special character on the updated row; and special messages such as override messages can be emphasized by using extra line feeds between rows.

Lastly, referring now to FIGS. 29 and 30, with FIG. 30 being a modification to a portion of FIG. 17, a preferred circuit 9000 for providing a row-by-row determination of 64 character or 32 character display of a row in the system of the present invention shall be described. This circuit 9000 has been described in the aforementioned U.S. patent application entitled "Improved Row Grabbing System" but is being repeated herein for completeness. Circuit 9000 may be omitted without departing from the spirit and scope of the present invention if such row-by-row determination of 64 or 32 character display is not desired. As shown and preferred in FIG. 29, assuming such a 64 or 32 character row-by-row determination is desired, the display screen for the display device 2013 is preferably considered as comprising two half screen widths with the left hand half of the screen preferably displaying even rows which are rows 0 through 62, and with the right hand half of the screen preferably displaying odd rows, which are rows 1 through 63. It should be noted that during normal display as previously described, a row is defined as being contained in the full screen width; however, in the instance of a row-by-row determination of 64 or 32 character display, it is preferable to consider half of the screen as comprising a row with each row in this instance comprising half a message for the pseudo video scan line. Preferably, in determining whether to provide a 32 or 64 character display for a given row, if 32 characters are contained in the even row, that is rows 0, 2, 4, 6, etc., up to row 62, then only the even row is displayed for the full screen width irrespective of the presence of an odd row in memory. If, however, there are 64 characters contained in the even row, then both the odd and the even row are displayed. In the preferred circuit arrangement 9000, no weight is given to the presence of this bit in the even row being considered. As will further by explained in greater detail, the control of the clock rate determines whether a 32 character or 64 character row is displayed, the memory being read twice as fast for display of a 64 character row as for a normal 32 character row, although the write speed for both a 32 and a 64 character row is the same because of the same speed of transmission in the preferred system of the present invention which receives row-by-row or pseudo video scan line-by-pseudo video scan line. In the system described in the aforementioned U.S. Pat. No. 3,889,054, the 32 or 64 character determination was based on a page-by-page basis and each row had to have a 32 character or 64 character bit according to the page sent or the display screen would flash between 32 character and a 64 character display affecting the readability of the display. As shown and preferred in FIG. 30, the memory 1030 through 1042 preferably includes another memory stage 2000 for purposes of determining whether the row is to be a 64 or 32 character row display, memory 2000 only looking at the row position and not at the character position. Preferably, the even row message which is contained in row 0, 2, etc., through row 62, includes the 64/32 character bit. When the characters are loaded into memory 1030 through 1042 for the even row, from shift register 466 (FIG. 17), a decoder latch 2002 preferably decodes the 64/32 character bit which is then loaded or written into memory stage 2000. The row position in memory stage 2000 is determined by row latch 470 through multiplexer 1020. The character position of memory stage 2000 is preferably ignored. On read out, the 64/32 character bit from memory stage 2000 is then preferably read into multiplexer 2004 which then makes a switch connection to gate 1090 (FIG. 18) via path 2004a (FIG. 18) and to divide-by-eight counter 1080 via path 1079 (FIG. 18) to double the clock frequency of the prime clock to provide for the 64 character row by proper timing. For a 32 character row, multiplexer switch 2004 is left in the normal position with no output being provided via 2004a and 1079 which, as shown in FIG. 18, are in the 32 character position, so that only the prime clock non-doubled output is provided for 32 character row timing.

Now describing the adjacent odd row message gating although, if desired, the odd row gating can be omitted if the odd row is not transmitted with a 32/64 character bit. Furthermore, the 32/64 character bit in the odd row message can be discarded in other conventional ways than to be described hereinafter if desired. As shown and preferred, when the odd bit of row latch 470 through multiplexer 1020 is on, indicating the presence of an odd row, which information is provided via path 2006a to a two input NAND gate 2006, and the system is loading or writing into memory 1030 through 1042 and stage 2000, as indicated by the presence of a write signal via path 2006b to gate 2006, the memory location for the odd row in memory stage 2000 is preferably changed to an unused memory location and the output of memory stage 2000 to multiplexer switch 2004 remains the same as for the previous even row message in that line. The ignoring of the first bit in the digital output of multiplexer 1020 (FIG. 17) will always provide the even row input to memory stage 2000. This is accomplished by preferably grounding the first bit row input to memory stage 2000 from multiplexer 1030. In order to ignore the 64/32 character bit when reading the odd row, then when writing the odd row, the 32/64 character bit is preferably put in an unused location by enabling this unused location in memory stage 2000 through gate 2006 whose input is the first bit position from row latch register 470, which is always preferably a 1 for the odd numbers and a 0 for the even numbers. Therefore, gate 2006 is enabled only when the input provided via path 2006a is a 1 and the write mode, also indicated by a 1, is provided via 2006b or, in other words, only for an odd row.

Lastly, discussing erase of memory stage 2000, when the normal erase is provided to the memory 1030 through 1042, preferably a common space character is put in all positions of the memory. When the 64 character row is erased, in order to insure that this row remains a 64 character row or, similarly, for a 32 character row, in order to insure it remains a 32 character row, in erasing the memory location in memory stage 2000 this memory location is positioned to an unused location without erasing the previously loaded memory bit 2010 from memory stage 2000, the erase signal via path 2008 together with the row input determining the location in memory 2000.

It should be noted that the description of the balance of the circuitry disclosed herein relating to the row grabbing system 10 which is common to the aforementioned U.S. Pat. No. 3,889,054, and which has not been repeated herein, is specifically incorporated by reference herein and the identical reference numerals therefor are utilized herein for clarity. Furthermore, any other disclosed circuitry not specifically described in detail herein is conventional and readily understandable by one of ordinary skill in the art without further explanation and, accordingly, will not be described in further detail. In addition, it is to be understood that all logic described herein is conventional unless otherwise specified.

Although the preferred video display terminal 28 of the present invention has been described above with reference to preferably utilization of a local microprocessor 6000 for control of the operation thereof, which microprocessor is programmable in accordance with the control program stored in the read only memory means 603, it should of course be understood that equivalent conventional logic circuitry could be utilized to replace the microprocessor 6000 if desired. In addition, if desired, the microprocessor 6000 could be programmed by storage of a different control program in the read only memory means 603 to accomplish additional control functions than those previously enumerated by way of example herein, with the omission of corresponding previously described circuitry which would accomplish these functions in the absence of programmed control by the microprocessor 6000. Furthermore, if desired, some of these enumerated control functions of the microprocessor 6000 could be replaced by equivalent logic circuitry with the balance of the functions enumerated above still being performed by the programmed controllable microprocessor 6000 or any other desired balance between logic circuitry and programmable microprocessor control 6000 can be achieved without departing from the spirit and scope of the present invention.

By utilizing the row grabbing terminal of the present invention in which grabbed frames may be updated on a row-by-row basis, conventional television transmission techniques and distribution equipment can be utilized for transmission and reception of data which has been packed into pseudo video scan lines which look like a conventional TV scan line to television equipment but contain a complete packet of information suitable for display of an entire row of video information with enhanced noise immunity to any jitter or noise present on each received pseudo video scan line being processed as well as clock phase correction for the terminal on every data transition of the received pseudo video scan line whereby the received information will be essentially noise insensitive.

It is to be understood that the above described embodiments of the invention are merely illustrative of the principles thereof and that numerous modifications and embodiments of the invention may be derived within the spirit and scope thereof.