the worst?

[John Garza (3665)]

I'm sure we all have our favorites when we vote for 'best programmable calculator'. But can we possibly agree on that's the worst?

I'll start it off by voting for the Sinclair Programmable.
These adjectives come to mind:
slow, fragile, arcane, inaccurate, uncomfortable.

I think we have, in a sense, become spoiled with HP calculators. They are literally the best of the best. Taking a walk on the "other side of the tracks" so to speak really opens your eyes. At some point you begin to feel sorry for the users of these machines and take a step back from the usual admonishments: "Eww... you're a Casio user?"


-John

[35/45/55/80/70/65/67/97/97S/91/92/21/22/27/25C/29C/19C/10/
31E/32E/33C/34C/37E/38C/41CX/28S/19B/12C/15C/33S/35S/32S/
48GX/50G/Prime/DM42]

[Maximilian Hohmann]

Hello!

(03-30-2023 01:48 AM)John Garza (3665) Wrote:  But can we possibly agree on that's the worst?

I dimly recall that we've had such votings and discussions in the past. But of course any opportunity for RPL bashing is very welcome ;-)

The all-time most useless programmable calculator bar none must be the Litronix 2290 with it's 10 program steps. A simple four banger with one memory register and nothing else (not even square root!) but three dedicated keys for program entry and execution. An absolute must-have for every calculator collector!

And as already written, for me the worst calculators are all those that come with RPL, the worst of the worst being the 28C because it has so little memory (2kBytes against the 32kBytes of the 28S) that one can not even use it's other functions like the solver or CAS properly.

Regards
Max

[Hiwi]

For me it is the newer 20S. Ugly and bad keyboard.
Ralf

[Dave Britten]

Oh boy, this will be a spicy thread...

In the HP realm, the 35S is high on my least-liked list, because it's so full of bugs. The 28C is a fine system, but it has so little RAM that you can't really do much of anything useful with it.

And I know there are some folks that aren't keen on the HP 55, but I actually really like it. I've got a 55 on my desk that's my go-to for quick calculations because it has a robust set of built-in functions and features, and a nice bright LED display. It's not terribly powerful for programming, but it's capable enough that I'll punch in a short program from time to time if I need to do something repetitive.

The Litronix 2290 might as well not even be programmable it's so utterly useless at it.

The Sinclair Programmable is also pretty awful (both to use and program), but interestingly enough, it came with a really great program library booklet. The Sinclair Cambridge Programmable is considerably better overall, but still not an enjoyable experience. But again, it too had an outstanding set of program books.

[floppy]

For me, "the worst" are all programmable calculator without an efficient safe storage of programming code. It means I am allergic to any HP16 HP15 etc.
= HP41 or HP71 are ok for me.
- others? perhaps. I dont know them
- NOGOs are.. (see above)

[John Garza (3665)]

And we should probably distinguish between simple issues like 'low capability' vs. outright design flaws/omissions.

I know there were a few 'programmable' adding machines back in the day. That would provide a precedent for the Litronix 2290. Frankly, I'm surprised HP didn't add some data registers and make the HP-10 programmable for simple tasks. That would have been so HP.

But things like using a shift key to get a decimal point?
Omitting scientific notation on an otherwise scientific calculator?

Somehow I'm reminded of a Three Stooges episode:
MOE: Think man!
CURLEY: I keep tryin' to think but nothin' happens!

-John

[35/45/55/80/70/65/67/67T/97/97S/91/92/21/22/27/25C/29C/19C/10/
31E/32E/33C/34C/37E/38C/41CX/71B/28S/19B/12C/15C/33S/35S/32S/
48GX/50G/Prime/DM42]

[pier4r]

(03-30-2023 10:50 AM)Maximilian Hohmann Wrote:  I dimly recall that we've had such votings and discussions in the past. But of course any opportunity for RPL bashing is very welcome ;-)

Hey!

[rkf]

The worst ever 'programmable calculator' I have ever seen in real live is without any doubt the Soviet Elektronica B3-21.
Besides poor numerical accuracy this wannabe-RPN-calculator shined with a circular stack, which could be rotated clock- and counterclockwise. Programming was just a PITA!

[Steve Simpkin]

Regarding the HP-28C and it’s small 2K of RAM. Bill Wickes was the R&D project manager for the HP-28 and HP-48 series calculators. Around 1986 when Bill proposed designing a new advanced graphing scientific calculator with algebra capabilities, there was no budget available to develop an entirely new calculator like Bill wanted to do. Bill's choice was to either use the existing hardware of the HP-18C with its limited 2K of RAM (along with more ROM) or gamble and wait sometime in the future to see if management would approve a development budget for the hardware he really wanted to develop (basically the HP-48SX). No one at HP was certain how much of a market there was for an advanced scientific graphing calculator with algebra capabilities. HP allowed the team to test the marketplace using the already developed HP-18C hardware. When the HP-28C turned out to be a huge hit, they approved the budget needed for the HP-28S and then for the HP-48SX. I had each of them shortly after they came out and am still using my HP-48SX.

[2old2randr]

Here is a positive review of the B3-21 so not everyone thinks it is bad

[rkf]

(03-31-2023 11:40 AM)2old2randr Wrote:  Here is a positive review of the B3-21 so not everyone thinks it is bad

Yes, but even Viktor wrote:

Quote: GOTO instructions are followed by a two-digit location code that is stored as a single program step. ... the location cannot be keyed in using the number keys, you actually have to look up the key with the keycode corresponding with the intended target of the GOTO. This is made even more difficult by the fact that keycodes are not derived from the keys' physical position on the keyboard. (Oh, and did I mention that a conditional jump instruction like x>0 actually skips the address if the condition is true, and performs the jump if it's false?)

[John Garza (3665)]

... Dare I say it...
the 41C

With all it's advances, it was a pitiful machine for everyday use.
All those "hidden" functions not on the keyboard.
First you had to remember it had that function.
Then you had to remember how to spell it with all that XEQ ALPHA business.
Then, of course, you had to type in the dang thing.
And if you forgot while away from your manuals, there was always CATALOG...
...one line at a time, wasting even more of your time.
This was acceptable at the time due to "user enthrallment" over the cool advances.
Like seeing an early windows PC waste time with cutesy graphics of papers flying around when you copy a file. Back then it was "oh cool, look!". But after a while it became "Cut the BS and get on with it."

And all that cheap plastic - a thin sheet pretending to be an actual battery compartment like in the classics. The structural posts that eventually fail, and the zebra contact only connectors. Not to mention the contact only failures of the Spice series. Didn't the latest batch of HP engineers read about WHY soldering is a good thing?

And then the HP-41C was the first HP I came across that would "lock up". VERY disconcerting when this first happened, and a PITA when it still does to this day. And the solution is to remove the batts overnight to let the whole system gradually discharge. And in the morning you are met with the friendly greeting: MEMORY LOST. The die hard early adopter sallies forth. Everyone else says "Where did I put my 67..."

In all fairness, I wouldn't rate the 41C as one of the worst. But I think it deserves special mention as prior to the 41C, HP had a reputation for exceeding expectations. This was the first top of the line machine that didn't do that. In fact, it failed to meet some expectations. So the contrast was startling and a bit worrying. And I understand the business need to keep the price not too far beyond TI, and within the reach of students - and even more so with the E-Series (Spice). In the days of the 65, students were not part of the equation. Professional machines for professionals. Not even a Games Pac.

-J

[35/45/55/80/70/65/67/67T/97/97S/91/92/21/22/27/25C/29C/19C/10/
31E/32E/33C/34C/37E/38C/41CX/71B/28S/19BII/12C/15C/33S/35S/32SII/
48GX/50G/Prime/DM42]

[Maximilian Hohmann]

Hello!

(03-31-2023 12:27 PM)John Garza (3665) Wrote:  This was the first top of the line machine that didn't do that. In fact, it failed to meet some expectations.

I could not afford an HP-41C when I would have needed it most (during university that is). When I started collecting calculatos years later, I bought my first 41 with great expectations, because that used to be "best calculator" during the first half of the 1980ies. Playing around with it I soon found out about all the issues you mention in your posting, the worst for me being this "XEQ ALPHA function_name thing". The most awful way ever implemented on any machine to invoke a function!

And of course the fact that now you need to buy three HP-41s if you want to have one that's actually working. On the other hand almost every HP-67 still works and these calculators are five years older than most 41s.

Regards
Max

[Massimo Gnerucci]

The 67. With those stupid shifted legends below the keys.

Oh, and you two are on my foe list now.

:) :D

[Maximilian Hohmann]

(03-31-2023 01:55 PM)Massimo Gnerucci Wrote:  Oh, and you two are on my foe list now.

I've been that list for quite some time because I never made a secret of my feelings about the 41 :-) Luckily you are not Arya Stark who takes her list of foes seriously!

And yes, the shifted key legends below the keys can be distracting. I just recently received a rare(ish) "Aristo Unilog S", not programmable so not eligible in this thread, which has the same feature. Playing around with it I got my keypresses wrong most of the time...

Saluti
Max

[Massimo Gnerucci]

(03-31-2023 02:06 PM)Maximilian Hohmann Wrote:  I've been that list for quite some time because I never made a secret of my feelings about the 41 :-) Luckily you are not Arya Stark who takes her list of foes seriously!

I know, I know you're not in love with LCD displays...

However ditching the 41 for its "XEQ ALPHA function_name thing" is ludicrous: how would you cope with all its standard functions? With 4 or 5 shift keys?
And all those added by plug-in modules and peripherals?

If you repeatdly need some hidden function, you can always assign it to another key in USER mode.

To each their own, of course. :-)

[John Garza (3665)]

I've used both the 65 and 67 for years and never had a problem switching between the two. With the 67 I always 'go down the slope' of the keys to get to all the shifted functions - unlike the 65 where you go both up and down. And gee, I though that was an improved feature!

I think the problem started with the 67 re-using the case/keyboard of the 65. If you want more user definable keys - there is no room to put the script above the keys. But I suppose you could put it ON the key, on the slope as an h function. But then, technically, it is still below the key.

And how to deal with all those functions on the 41C? It's obvious. Use menus. Even HP fixed that problem on the successor 42.
Not to mention the brute force approach to solving the majority of the lockup problems - get rid of the ports!


-J

[Bill Duncan]

As other's have mentioned, I'd have to say the HP-28, either one. The clamshell design broke too easily. I've had two and neither works.

The other least favourite was the HP-27s. Nice design but, not RPN like the original woodstock hp-27. Also no way to self-adjust the clock which, at least on the one I have, gains a few minutes every week. I still pull it out and use it sometimes.

[aurelio]

(03-31-2023 01:11 PM)Maximilian Hohmann Wrote:  Hello!
I could not afford an HP-41C when I would have needed it most (during university that is). When I started collecting calculatos years later, I bought my first 41 with great expectations, because that used to be "best calculator" during the first half of the 1980ies. Playing around with it I soon found out about all the issues you mention in your posting, the worst for me being this "XEQ ALPHA function_name thing". The most awful way ever implemented on any machine to invoke a function!

And of course the fact that now you need to buy three HP-41s if you want to have one that's actually working. On the other hand almost every HP-67 still works and these calculators are five years older than most 41s.

Regards
Max

Max, I can carbon copy the whole experience and feelings you wrote here about the 41, but Massimo please do not put me also in the foe list for that, because I must say that after getting the first one, I immediately fell in love with this small computer that I would sometimes take on vacation if my sweetheart could afford it once

[John Garza (3665)]

Perfection takes time.
The first pass at something innovative is usually not a very smooth implementation.

The 35, 65, and 41C were all incredible advances that made them 'game changers'.
And each had their share of issues and design oddities.
And the successor machines were all better:
The 45, the 67, and the 42.

Regardless of the 'innovation tax' that must be paid, the 41C was considered revolutionary, sold well, and made an impact on the competition. And as I recall, both TI and Novus cancelled new products in development when they saw the new 41C.

-J

[35/45/55/80/70/65/67/67T/97/97S/91/92/21/22/27/25C/29C/19C/10/
31E/32E/33C/34C/37E/38C/41CX/71B/28S/19BII/12C/15C/33S/35S/32SII/
48GX/50G/Prime/DM42]