Calculators

pc app: gnuplot, dunno if there's a GUI though.

Ditto, though I rarely graph things by hand. Can't remember the last time I did it.

Why not Excel? It works. ;-)

just get octave or scilab, matlab ~clones that can read and plot anything y ou can dream up in more ways than you can imagine

I use Desmos and Geogebra on my PC, phone, Chromebook, etc. Web apps are nice because beause they look much the same on any device. Desmos is quite simple and has a feature that slowly changes variables over a range of values so that the effect can be seen on the graph. More 2D:

3D:

3D with anaglyph (red-blue) glasses:

Was it HP who created the world's first 'scientific' calculator, do you know and if so, how long after that did Sinclair bring out his 'scientific' model? I merely ask because you seem to be something of an authority on classic calculators.

The first handheld. The HP35. I got one of the first ones, for $400 in

Before that, I had a desktop 9100, which was a fabulous machine, $4900, which would have bought two cars in those days. It was programmable, with core memory, and a CRT display. All discrete transistors. I have two of those.

The first scientific calculator was probably the HP-9100A. I've repaired a few of these and they are awful to work on.

The first POCKET scientific calculator was the HP-35. I have several, including one of the first versions.

I don't know much about the Sinclair Scientific: The HP-35 was introduced in 1972, while the Sinclair in 1974. TI also introduced the SR-16 and SR-50 scientific calculators in 1974.

When the HP-35 was first introduced, I really wanted one. However, the initial $400 price was too much for my budget. I needed something, so in 1973, I went to Radio Shack and bought a TI SR10 electronic "slide rule" for about $150. Big mistake, but better than nothing.

Some years earlier, I decided that analog computers were the future. I threw together several Burr-Brown analog log/anti-log amp modules, four 10 turn Bourns Helipots for input, and a mirrored panel meter for output. The precision pots and turn counting dials were sufficient for 2 digit accuracy or 3 digits with a magnifying glass. Everything, including a power supply was crammed into a briefcase. Since many instructors initially banned all calculators from exams, they were somewhat undecided as to what to do with what was essentially an analog computing electronic slide rule. I had to demonstrate my contraption to a few instructors to convince them it was not a calculator in disguise. I eventually determined that a slide rule was generally faster and that my device was far too expensive. That's when I went shopping for a real scientific digital calculator.

I'm not an authority on calculators. I have some experience fixing them, mostly for collectors. Most HP repairs are mechanical, not electronic. I also collect old HP LED calculators, which I have about

More:

Drivel: At one point, I was using (and fixing) Monroe, Marchant, and Friden mechanical calculators to grind numbers. Electronic calculators were a huge improvement over mechanical: (6:04)

Wow! What an amazing pile of old junk! But seriously, look at the amount of engineering, design and workmanship that must have gone into bringing that into production. It's a work of art in its own right.

Hey, I have two that I'd like to get working!

I don't think anyone has schematics, except for the power supply.

I talked to the HP historian: she said yes, they have schematics and no, I can't see them.

I worked one summer on a research project in microwave spectroscopy. Two grad students spent the entire summer calculating rotational resonances on a Friden mechanical beast. You could do all that work now on a cheap laptop, in Python, in milliseconds.

Yep. My father's lingerie factory used a Borroughs Tabulator in the office. I couldn't find a close enough photo with Google. It was similar to the Friden in the above video, but with a typewriter roller for tabulator sheets towards the rear. Something like this but bigger: It had to be bolted to the table or it would try to walk away.

Regular maintenance was costing the company too much money, so my father asked me to look at the machine. It didn't take much to figure out the problem. The "mechanic" was using an oil that would turn to gum after a few months. That produced jamming followed by an expensive service call. I cleaned out the gum with alcohol and replaced it with non-evaporating sewing machine and clock oils. After that, there was no more "regular maintenance" needed.

He eventually replaced it with an Altos something CP/M system and later with an Altos 986 Xenix AOS (Altos Office System).

Yeah, I know. The problem goes back to 2005: It reappeared in 2012: (Google Groups sure is running slow today).

Some more:

As I vaguely recall, the video board should be on the miserably scanned HP archive print, but nothing else. I got lucky because all of the problems were in the power supply (aging electrolytics, high voltage sneak paths, shorted HV diode, and marginal 2N1701). More clues:

There's always reverse engineering and espionage: Grab the "PDF with Text" which is 12 pages (3.44MB). Looks like it's all there but will be difficult to clean up the scans and decode the jargon. There are also no component and circuit side board photos and drawings. The "PDF" is much larger (27MB) but seems slightly easier to read. Grab both. Incidentally, that scan was not there in 2012 when the topic last appeared in S.E.D.

Perhaps an emulator will be sufficient for now?

My experience with mechanical devices was centered around teletype machines. I worked for a summer for a refurbisher of these machines. The most important thing I learned is that many people have zero mechanical abilities. I still have a poster of an ASR-33 machine with a fire axe imbedded in the cover.

Let me think about the HP-9100A repair. I'm planning on spending the summer loafing and vacating and don't want anything like work getting in the way.

Wanna swap for a week in the cabin in Truckee?

Thanks, but no.

I took the JP2 images of the 12 page HP-9100B schematic collection and did some rotation and image enhancement. About 32MBytes total: The originals are at: The images were photographed, no scanned. Printed on A size paper, they are unreadable. However, in the original C size format, they seem to be quite readable and usable.

Also, I looked at the HP-9100 service manual at: As you noted, only the schematics for the power supply and display sections are included. The service manual seems to be self consistent, with no pages or drawings missing. I would guess(tm) that this was all that HP ever supplied in the service manual and that the digital logic and core memory schematics were never published for public consumption.

There's a graphing calculator among the Microsoft "power toys."

Even without schematics, you can do a lot with a meter that does diode test. That's because that beast has a LOT of diodes (there weren't any ROMs, all the firmware was diode-matrix).

I once had a Sinclair Scientific. I scrapped it when I discovered (the hard way) that it was less accurate than a slide rule for some calculations. Dividing one number by the difference between two similar numbers was a disaster and gave errors of tens of percent.

The reason it was so slow was that it was clocked far below the minimum clock speed for the TI chips and relied on its dynamic registers not leaking charge away between clock pulses. I assume this was done in order to get better battery life.

John

It's a *loooonnnnnggggg* time ago, but I thought the outrageous delays with certain types of calculation using the Sinclair Scientific were due to it taking repeated guesses to progressively home-in on the correct answer. Or have I mis-remembered that?

All calculations, especially transcendental functions are like that!

The Sinclair Scientific's ROM held *320* instructions *total*, including keyboard/display and arithmetic operations. Most people would be hard pushed to do +-*/ in that; it is remarkable that transcendental instructions were (sort of) squeezed into that!

FFI, see an emulation of the ROM at