Talk:Calculator input methods

Definitions
I am considering a re-write of the first section of the article, for the following reasons.

I am not sure if the article intends to cover all calculators with the two definitions (it says there are two main types), but the definition of immediate execution - that the operations are executed “when the next operator is pressed” – may exclude some calculators with parentheses buttons or precedence mode, such as. This is not a formula calculator, but it is not an immediate-execution calculator either: with 2*(3+4)*5, the first * isn’t executed when the next operator, +, is entered, but when the second * is entered. With this example, any number of operations could be entered between the parentheses before the first * is executed, and so perhaps intermediate execution is a better term for this type of calculator. Also, if the article’s definition of a formula calculator is taken literally, it excludes calculations typed into multi-line edit boxes, which don’t use Enter to do the calculation.

If the term intermediate execution isn’t used, perhaps an immediate-execution calculator can be defined as one that executes operators during calculation entry and immediately their operands are known. Also, if calculators are differentiated according to when operations are executed and values shown, perhaps calculators that show the final value only after the whole calculation has been entered, can be known as final-execution calculators.

So I would like to clarify the definitions. Also, there is more to calculator input than the interpretation of keystrokes, and I would like to deal with several other ways in which calculators differ in their methods of input.

Fcalculators (talk) 20:23, 1 March 2010 (UTC)

Reverse Polish notation (RPN) (postfix notation)
How do you get down to five key strokes ? Is there a shorter way than Bram4 (talk) 21:16, 11 October 2009 (UTC)
 * 3 enter 2 x 1 + (six key strokes)?
 * I think it is saying that in infix notation it only requires 5 presses. &mdash; Carl (CBM · talk) 22:48, 11 October 2009 (UTC)

Why is the RPN example not done with only 6 keystrokes? * 3 enter 2 x 1 + is how it would be entered in real life. — Preceding unsigned comment added by Hopf-Rinow (talk • contribs) 02:28, 18 April 2012 (UTC)

Evolution
I think the article is should reflect historical evolution of calculator input methods. I identified the following phases:


 * 1968 RPN : The first "calculator" to use RPN was the HP9100A, which was introduced in 1968, although this machine is now regarded by many as the first desktop computer.
 * 1970 Immediate execution with out operator precedence : In April 1970, the Pocketronic appeared on the Japanese market; it was a four-function, entirely electronic calculator that retailed for about $400.
 * 1974 Immediate execution with operator precedence : The SR-50 A TI SR-50 (1974) [...] of the first calculators to implement "algebraic logic" (i.e. operator precedence) where 1 + 2 x 3 gives you 7, not 9
 * 1980 Infix (BASIC) : [...]1980 with the introduction of the Sharp PC-1211, the first pocket computer programmable in a high-level language.[...]
 * 1992 Infix (D.A.L., V.P.A.M.) : Introduced in 1992 and an industry-first, SHARP's D.A.L. allows symbols and numbers of an equation to be entered as they are written..
 * Infix (WriteView, Natural textbook display) : Could not find out when this was first introduced.

One problem for an encyclopaedic article is that not all phases have an official name.


 * An historical approach is fine by me, but it should be noted that the phrase "an industry-first" is misleading because both graphic calculators (e.g. fx-7000G introduced in 1985) and programmable calculators (e.g. fx-4500P) could do things like 1+2sin 30=. It might be closer to the truth to say the first non-graphic, non-programmable which allows symbols and numbers of an equation to be entered as they are written.


 * Also it should be noted that even today you can buy calculators for several different input methods, that is to say new methods did not completely replace older methods. Wikicgc (talk) 20:19, 1 July 2009 (UTC)


 * I agree with you on both accounts. When I first read about D.A.L. my first though was "What about my Sharp PC-1403 - that would do 1 + 2 * sin (30) when used in basic mode..." --Krischik T 09:44, 2 July 2009 (UTC)

Algebraic Entry isn't Immediate Execution
The article makes a good distinction between Immediate Execution, and Formula Entry.

The ordinary calculator that isn't a scientific calculator is an immediate execution calculator.

2+3*4 will give you 20.

Algebraic Entry is a type of Formula Entry that saves the whole expression, and executes the multiplications before the additions.

Then,

2+3*4 = 24

Formula calculators have been made for boyh Algebraic Entry and for RPN.

...infix and postfix.

So it's like this:

1. Immediate Execution

2. Formula Entry

...2a) Algebraic Entry

...2b) RPN

The article makes a mis-statement that needs to be fixed, when it says that Algebraic Entry is Immediate Execution.

Algebraic Entry calculators don't execute each operation immediately. They save the whole thing, and then execute the multiplications before the additions.

Michael Ossipoff — Preceding unsigned comment added by 66.169.246.137 (talk) 18:25, 23 July 2017 (UTC)


 * There are important distinctions between what this article is describing as "immediate execution with operator precedence" versus true infix notation. Consider, for example, the following series of (all current-production) scientific calculators from Casio, Sharp, and Texas Instruments. Casio fx-260, Sharp EL-501XT, TI-30Xa; versus Casio fx-991MS, Sharp EL-510RT, TI-30X IIS.
 * In examples using 2-operand operators, such as +-*/, etc, all of the above calculators will use the exact same sequence of keystrokes.
 * So, to evaluate 1 + 2 * 3, we would use the exact same sequence of keystrokes for every one of those models:
 * [1] [+] [2] [*] [3] [=] and obtain the correct result 7.
 * But for single-operand operators, such as sin, cos, log, e^, the former models from each brand, upon encountering such an operator, would evaluate all of the operations which had been input prior to that point within the current level of precedence, and then use whatever result ended up in the working register as the input to the operator; whereas the latter models from each brand would expect the inputs to those operators to follow the operator.
 * So, in degrees mode, to evaluate 1.5 + sin(30), we would have different keystrokes for the former models than for the latter models:
 * For the former models, it would be:
 * [1] [.] [5] [+] [3] [0] [sin] [=] and obtain the correct result 2.
 * For the latter models, it would be:
 * [1] [.] [5] [+] [sin] [3] [0] [=] and obtain the correct result 2.
 * All of these models are commonly referred to as "Algebraic", but only the latter category are actually using true infix input. 142.68.80.147 (talk) 20:46, 29 November 2022 (UTC)
 * Indeed, in the former models, if you perform the trig example, you can actually watch it immediately compute the sin(30) step at the very instant the sin key is pressed, yielding the intermediate result 0.5, before giving the overall result when you press [=].
 * The fact that immediate execution really is going on, becomes even more obvious if you compute the (contrived) example 1.5 + sin(30) * 2 + 0.5
 * In the former models, you would input
 * [1] [.] [5] [+] [3] [0] [sin] [*] [2] [+] [.] [5] [=] and obtain the correct result 3
 * You will see the intermediate result 0.5 immediately upon pressing [sin] (establishing the first elevation of operator precedence), then another intermediate result 2.5 immediately upon pressing the second [+] (ending the second level of priority elevation which had previously started when you had pressed the [*] key), and finally the correct result 3 after pressing [=].142.68.80.147 (talk) 21:01, 29 November 2022 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 2 external links on Calculator input methods. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20070207044654/http://www.uclic.ucl.ac.uk/harold/srf/allcalcs.pdf to http://www.uclic.ucl.ac.uk/harold/srf/allcalcs.pdf
 * Added archive https://archive.is/20120723100846/http://www.articlesalley.com/article.detail.php/7013/178/Education/Internet/36/Declarative_Programming_-_Strategies_for_Solving_Software_Problems to http://www.articlesalley.com/article.detail.php/7013/178/Education/Internet/36/Declarative_Programming_-_Strategies_for_Solving_Software_Problems

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 04:12, 29 July 2017 (UTC)