Pre-AOS TIs--And they say RPN is complicated!?

[Matt Agajanian]

Hi all.

In TI's sales literature (especially with the SR-52/56 literature), TI mentions about their choice of AOS instead of RPN. Even as I have my eldest sister's SR-50, lookng at its parentheses-lacking keyboard, I'm still stunned, befuddled, and perplexed that, although pre-AOS, TI's entry system was an improvement over RPN.

Do any of you here see any logic (yes, pun intended), common sense, or intuitiveness to problem solving on an SR-50/50A, 51/51A?

[d b]

The problem I have in using them (I have one) is that I'm in the habit of using RPN. RPN is consistently Postfix Notation. AOS in any and all it's variants is not. I could learn to use it comfortable but why bother?

[Gene]

Eh, the non-parentheses TI hierarchy followed MDAS (Multiplication/Division before addition and subtraction). It did this by a "sum of the products ability using a Z register to store the previous result (see the SR-51A owners manual online). It also used hierarchy that evaluated functions (trig, square root) as they were pressed.

It really wasn't THAT hard to use or understand, IMO. I used an SR-51A for years and found it very helpful.

My 2 cents.

[Thomas Radtke]

AOS offers immediate access with some headaches afterwards maybe. But seriously, RPN isn't w/o problems either. Remember the different bugs and/or implementation-specific quirks that went unnoticed for decades.

The first clean RPN implementation came with RPL.

[Nigel (UK)]

(02-18-2016 09:59 PM)Matt Agajanian Wrote:  Hi all.

In TI's sales literature (especially with the SR-52/56 literature), TI mentions about their choice of AOS instead of RPN. Even as I have my eldest sister's SR-50, lookng at its parentheses-lacking keyboard, I'm still stunned, befuddled, and perplexed that, although pre-AOS, TI's entry system was an improvement over RPN.

Do any of you here see any logic (yes, pun intended), common sense, or intuitiveness to problem solving on an SR-50/50A, 51/51A?

I didn't own an RPN/L machine for many years - HP's prices were way too high - and so I grew up with algebraic entry. To me, calculators were miracles and if I had to jump through a hoop or two to get to the correct answer, perhaps storing or even writing down intermediate results, it was still far far faster than the alternative. Algebraic /AOS works fine if you are used to it!

Also, I can see the marketing appeal of a machine that lets you type "2 + 2 =" and has "4" appear in the display, rather than "2 ENTER 2 +". If a product is aimed at home users or students then the marketing advantages of algebraic cannot be overlooked. Even HP succumbed eventually, I imagine for this reason!

Which do I like best today? I like what the Prime offers - algebraic, with the ability to scroll back, retrieve previous results and edit previous entries. My Prime is not set to RPN mode. Sorry!

Nigel (UK)

[Paul Dale]

(02-19-2016 07:20 AM)Thomas Radtke Wrote:  The first clean RPN implementation came with RPL.

I have to agree here. Classic RPN with its weird stack lift behaviour is not intuitive.

The commands that are prefix rather than postfix (STO, FIX, ...) are easier to grasp but are still contrary to the pure postfix way.

I believe it would be possible to have a classic RPN without stack lift -- keep a separate command line which is parsed on ENTER and which only goes into X.


- Pauli

[Thomas Radtke]

(02-19-2016 10:42 AM)Paul Dale Wrote:  I believe it would be possible to have a classic RPN without stack lift -- keep a separate command line which is parsed on ENTER and which only goes into X.

The 20b/30b implemented it this way! :-)

[Paul Dale]

(02-19-2016 11:11 AM)Thomas Radtke Wrote:  The 20b/30b implemented it this way! :-)

All my 20b/30b devices have been improved


- Pauli

[Matt Agajanian]

(02-19-2016 01:47 AM)Gene Wrote:  Eh, the non-parentheses TI hierarchy followed MDAS (Multiplication/Division before addition and subtraction). It did this by a "sum of the products ability using a Z register to store the previous result (see the SR-51A owners manual online). It also used hierarchy that evaluated functions (trig, square root) as they were pressed.

It really wasn't THAT hard to use or understand, IMO. I used an SR-51A for years and found it very helpful.

My 2 cents.

Thanks Gene. That helps clarify the methodology. As I will browse the 50A/51A manuals again, I'll keep that in mind. I want to see if, by re-reading the examples, I can extrapolate how other parentheses-included equations could be input.