35S Insanity

[everettr]

So I am browsing through the Auction place. looking at calculator offerings, as one does. I look at a Swiss Micros DM42 listing, the price is okay. and then I look an anything else the seller has on offer. HP 35s for $300? Is this a listing typo?
I just did a search in the E place and $250 is the minimum asking price and NOS is over $400!! Why?!
I have an HP 35S that is barely used and that I really don't think about very much. Most of the thoughts I have had about it involved some way to kluge an input port/external storage by hot wiring the keyboard matrix. Thankfully, my 35S remains as HP intended it, so I can offer it up to the ins... discerning buyers that obviously know something that I do not.

[JAK08]

FE exam compliance is my main and really only guess.

[Jase]

(11-30-2024 12:44 AM)everettr Wrote:  So I am browsing through the Auction place. ...HP 35s for $300? Is this a listing typo?

I don't have a 35s in my collection, so 6 months ago I went to eBay to see about snagging one. Just like you, I saw the price (holy smokes!!), I then read up on its no I/O limitations, and decided I'd rather add a 41cx, a 50g, a SwissMicros... the list is long.

I'll say this though, it sure looks good. If only it had I/O and a nicer screen.
Who knows, maybe it will be like the 15LE / 15CE and get re-released with some improvements. But for now, my money is better used elsewhere.Paramore
Jase.

[Steve Simpkin]

One reason the HP 35s is in demand, at least in the U.S., is that the HP 35s (and HP 33s) are the only programmable calculators allowed by the U.S National Council of Engineering Examiners (NCEE) for PE, FE, FS and PS exams (see https://ncees.org/exams/). There were companies that sold pre-programmed HP-35 calculators to aid in taking these tests.

Other reasons for the high prices from Bob Prosperi:
1. It was discontinued, availability is limited, so prices naturally rise
2. It's the only HP/RPN machine still authorized for Professional Engineering exams
3. Greed amongst the eBay sellers aware of 1. & 2. above

[gentzel]

Good commentary on why the 35S is so expensive: Information about the HP35s Calculator Shortage

[Johnh]

Operationaly, I reckon the 35s is great. It has a good range of functions and everything you need in a non-graphic calc, a good enough Alpha mode to help with inputs and outputs, plenty of storage its easy to program and it has a worthy HP-quality keyboard.

But it has too many quirks and glitches which combine together to make it untenable except for the simplest uses because it is inevitable that at irregular and too-frequent Intervals and for several unpredictable reasons, it will lose everything and there is no way to back-up or recover short of button pressing.

[Roberto Volpi]

Not sure they actually sell at these prices, but when so many sellers quote that same high value, it is a clear indicators.

Quirks? Every calculator has its own.

[RPNerd]

(11-30-2024 03:57 AM)Johnh Wrote:  Operationaly, I reckon the 35s is great. It has a good range of functions and....

...an even greater range of bugs.

https://www.hpmuseum.org/cgi-sys/cgiwrap...i?read=735

[emefff]

Hello,

the asked prices are crazy, even 150$ would be a lot. I still have two, I paid 20€ each. Maybe someday I'll get rich,

emefff.

[Commie]

(11-30-2024 03:57 AM)Johnh Wrote:  it will lose everything and there is no way to back-up or recover short of button pressing.

The 35s is light years ahead of the hp15ce and yet it too doesn't have any i/o and how people, on this here forum, drool over the hp15ce, I just don't understand it.

Cheers
Darren

[Thomas Klemm]

(11-30-2024 01:31 PM)Commie Wrote:  The 35s is light years ahead of the hp15ce (…)

Compared to the HP-15C, the HP-35S has only mediocre support for complex numbers.

You might ask: Why would I ever use that?

Nowadays I mainly use emulators or simulators of HP calculators on my smartphone. Most of them allow me to import and export data and programs. If that is not enough, I can use WolframAlpha or a Jupyter notebook.
I can't remember the last time I used my HP-35S.

[Commie]

(11-30-2024 02:56 PM)Thomas Klemm Wrote:  Compared to the HP-15C, the HP-35S has only mediocre support for complex numbers.

Yes granted, but theres an awful lot of other positive features that the hp15ce doesn't support like EQN mode and when programming, the hp 15ce uses keycode numbers making programs difficult to read/edit whilst the hp35s uses function strings making it easier to edit a program.

Also, Ive just paid £55 for a brand new sealed ti-84+(from ebay) which has tons of features that the hp15ce doesnt have.For example, I can write basic programs on my pc and download them to my ti84+ in in record time using usb.

Cheers
Darren

[Thomas Klemm]

(11-30-2024 03:19 PM)Commie Wrote:  (…) the hp15ce doesn't support like EQN mode (…)

From A Sigma Function in the 35s Solver!:

Quote:You may call it a bug. On the other hand why would somebody want to do such a strange thing?

With this behavior of the solver I suspect it's quiet impossible to calculate a few hundred digits of pi.

Compare it to what is possible with the solver of other models: N-queens benchmark and the HP-17BII.
But then you might wonder why on earth would you want to use the solver to solve the N-queens problem instead of a simple key-stroke program: Calculator Benchmark.
To me it looks like a bit of everything was crammed into this calculator but nothing was done properly.

[naddy]

(11-30-2024 04:42 PM)Thomas Klemm Wrote:  To me it looks like a bit of everything was crammed into this calculator but nothing was done properly.

I just skimmed through the manual, which is very much derived from the 32SII one, because featurewise the HP-35S is indeed a slightly extended 32SII. It has more memory, so a stack level can now hold a complex number or a three-element vector, and there are a lot of unnamed registers that can be addressed indirectly. Oh, and it has an algebraic mode.

Is it possible that they didn't have/couldn't use the source code and re-implemented everything from scratch, modeled on the 32SII?

[naddy]

(11-30-2024 02:56 PM)Thomas Klemm Wrote:  

(11-30-2024 01:31 PM)Commie Wrote:  The 35s is light years ahead of the hp15ce (…)

Compared to the HP-15C, the HP-35S has only mediocre support for complex numbers.

It's matrix support is also lacking, FWIW.

[Commie]

(11-30-2024 04:57 PM)naddy Wrote:  [It's matrix support is also lacking, FWIW.

Yep, never thought of that, just had a quick wiz through the hp32sii manual and it looks like it too doesn't have matrix support, although perhaps I'm wrong?

[Johnh]

(11-30-2024 01:31 PM)Commie Wrote:  

(11-30-2024 03:57 AM)Johnh Wrote:  it will lose everything and there is no way to back-up or recover short of button pressing.

The 35s is light years ahead of the hp15ce and yet it too doesn't have any i/o and how people, on this here forum, drool over the hp15ce, I just don't understand it.

Cheers
Darren

But what makes the HP15C-CE appealing and feasible for use now is that it does indeed have i/o support , as developed by the wonderful people here. And it works, plus you can use any of the very good phone emulators to complement it, as I do every day. It's also a beautiful piece of classic design, and I enjoy that every time I use it.

HP35s was almost ok, if it was what it should have been I'd just be using that. But sadly it wasn't. I could put up with the annoying quirks, but not the gut-punch of the lock-ups and data loss.

[naddy]

(11-30-2024 05:33 PM)Commie Wrote:  

(11-30-2024 04:57 PM)naddy Wrote:  [It's matrix support is also lacking, FWIW.

Yep, never thought of that, just had a quick wiz through the hp32sii manual and it looks like it too doesn't have matrix support, although perhaps I'm wrong?

The 32S was marketed as the successor to the 11C. It added SOLVE, ∫, limited support for complex numbers, a number of minor things, and took advantage of the possibilities added by the new dot-matrix display.

The 32SII was a strict superset of the 32S. It added equations, fractions, again a few minor things, and reworked the user interface.

The 11C/32S/32SII manuals include example programs to solve simultaneous linear equations in three unknowns.

After the 15C, support for matrices and full support for complex numbers was limited to the 42S, which was positioned as the successor to the 15C, and to the RPL line: 28C/S, 48S(X) and successors.

[Steve Simpkin]

(11-30-2024 04:55 PM)naddy Wrote:  ...
Is it possible that they didn't have/couldn't use the source code and re-implemented everything from scratch, modeled on the 32SII?

Yes, that is exactly what happened. HP did not design, code or manufacture the HP 35s calculator. HP provided a detailed set of specifications for the functionality of the HP 35s and its form factor to Kinpo Electronics. Kinpo then designed the calculator circuitry, hardware, wrote the software and manufactured the calculator for HP. The HP 33s and HP 35s went through a similar process. They are all using a variations of a Sunplus CPU which has a core based on the 6502 microprocessor.
https://en.wikipedia.org/wiki/HP_33s
https://en.wikipedia.org/wiki/HP_35s

[Thomas Klemm]

(11-30-2024 08:01 PM)Steve Simpkin Wrote:  Kinpo then designed the calculator circuitry, hardware, wrote the software and manufactured the calculator for HP.

It appears that these are not the only models.
The “Scientific Calculator” on Kinpo's Consumer Electronics looks like an HP 10bII+ to me.