specially created instructions arriving to it.
A good solution is attained when these instructions are
155 -only of two types:
- position instructions /not stored explicitly in the display’s refresh memory/ required unconditionally mainly when updating tasks are carried out over some part of the drawing being shown;
- graphical instructions /stored in the display’s refresh memory/ required to define the different symbols,
numerals, etc. which compose the drawings. They include also the color and other modifiers related to each type of graphical element.
These instructions can be easily obeyed in a similar form as an alphanumerical display does.
Now referring to characteristics numbers 2 and 3 of drawings required in process control applications /See Sect. 2.5/» some compression of data is attained when in
formation related to horizontal and vertical straight lines of limited length is represented digitally coded /See Sect. 2.6.1/. Due to particularities of the solu
tion proposed here in order to get a given data compres
sion /a single phenomenon of repetition/, the instruc
tions related could be easily processed /or broken up/
either by software in the computer, or by hardware in the display unit or in the Controller.
The alternative of doing it in each display unit is neglected because:
- it will complicate the hardware design,'
- it would be convenient and advantageous that the behavior of the display be the same for all types of graphical instructions,
- it will increment the cost per display unit,
- it will obstruct the data transmission when some
graphical instructions must be broken up and others do
156 -not require it.
If it is done by hardware in the Controller, similar in
conveniences are present than those present when they are broken up in the display units, namely:
- the hardware design of the Controller will be more complex,
- the size of the buffer required in the Controller will be large in order to handle favorably the diversity of instructions to be broken up before being sent to dis
plays units*
- it will increment the overall cost of the whole system.
However, this solution present one significant advantage:
- it saves appreciably the computer time in relieving it of this routine work, there being thus longer available time for more serious processing.
If it is done by software in the computer, the
inconveniences enumerated above are not present, but the computer will loose time for performing these routine works. However, in this case, it is possible to get a high constant transmission rate and straightforward breaking up of the compressed instructions.
Now, instructions must be stored in compressed form in the computer memory, either if the breaking up of
compressed instructions is done in the computer by soft
ware or in the Controller by hardware or firmware. In such a way, computer memory capacity can be saved, it being advantageous that all the calculations, decisions, updating, etc., be done over the coded computer word of a single compressed instruction and not independently cell by cell. Moreover, compressed instructions must be
157
-broken up only when they are transmitted to the display units; in any other situation they must be kept in
oompressed form. It is proposed also to create direct access for I/O operations between the computer memory and the refresh memory of the display units, in order to
facilitate possible transmissions without obstructing the normal computer’s processing.
On the other hand, some complex symbols such as symbols for pumps, valves, tanks, etc., or large squares,
rectangles, etc. which require two or more cells for their graphical representations /using several single symbols with their own meaning/, can be created with graphical subroutines. The reference position of these types of subroutines could be referred or not to a
particular cell belonging to the complex symbol. Graphical subroutines are stored in the computer’s memory, being handled by the specially created routine called up by the respective graphical instruction. Color code and other modifiers, are generally specified in each independent graphical instruction composing the graphical subroutines.
These instructions can also in their turn be any type of compressed graphical instructions, meaning that a
multiple level subroutine’s handling facility will be convenient for the computer used in the system.
Moreover, if we consider that graphical subroutines be in some form relocatable, i.e. that they can be called up by every picture file created when they are needed by them, and because these graphical subroutines could be
independently called when required by means of only one graphical instruction belonging to the graphical language created, then a considerable data compression could also be attained, thus further saving computer memory loca
tions. For this reason and because of the frequency with which graphical subroutines are commonly required, they
158
-could be considered as a high level data compression means of the whole system.
Finally, it is advantageous that graphical subroutines be broken up in single graphical instructions /maybe in
cluding another type of compression of data/, only in those cases when they must be transmitted to display units.
Hitherto, handling of program subroutines has been per
formed by software in common computer systems, having widely proved its effectivity in many other applications.
For this reason among others, it could be advisable here also as the best alternative, to perform in the computer by software the breaking up of all graphical subroutines conceived for the system. If this criterion is extended, breaking up of those instruction-words which present data
compression based on the repetition of equal symbols /which is carried out more easily, requiring less time and in a more straightforward form/, would have an eco
nomical justification if it is done by software also in the computer.
2. The graphical instructions which produce the drawings on