Oh hell, just bought another calculator

[tggzzz]

The Hewlett-Packard 9100A is pretty big, as well, all 40 pounds of it.

https://en.wikipedia.org/wiki/Hewlett-Packard_9100A

I saw one of those for sale at a hamfest last year. Someone physically bigger than me was determined to buy it. Half of me thinks that was for the best.

Yes, probably for the best.

Although an amazing piece of technology for the time period (late 1960's), it probably would take a massive amount of work to get (and keep) it working.

My concern was more that it would take a massive amount of work just to get it upstairs and onto my workbench. Work in J, that is.

[iMo]

Would be great to see the teardown - as it is built around transistors, no chips inside (wiki)..

[ebastler]

If only we knew which calculator tggzzz has acquired and might be planning to tear down...

[tggzzz]

If only we knew which calculator tggzzz has acquired and might be planning to tear down...

Getting it into position for use takes ~30s. Complete teardown another 30s, reassembly another 60s (a little longer to avoid damage). I tore it down while still in the auction house

Cleaning: I'm trying to find a way to that safely, but accept it would probably be unwise to try.

I've seen pictures of some devices up for auction where the surface looks like a RIFA in dangerous condition. In such cases, I cannot think of any solid or liquid (or gas!) that could be used without ruining it. Even so, such devices are almost certainly usable. You can get reproduction instruction books on fleabay for £10 (no, not from ArtekMedia!)

Auction price? I've seen estimates from <100 to >2000 "pesos" (i.e. given that range, the currency is unimportant). The 2000pesos one looks like it was produced maybe 30% into the production run. Mine was about 70% into the production run.

There are two on display in TNMoC; they inspired me to obtain mine Couldn't find room (or money for electricity!) for their Elliott 803, Witch, ICL 2906, though.

Final hint for now... Two dimensions are often quoted, but I think the larger is a cheat. Even so, the smaller quoted dimension is far larger than anything mentioned by anybody in this thread

[ebastler]

Ahh -- I think I got it now. The RIFA reference gave it away... So we are talking purely mechanical, right? And they came stock with that angled mount, which they appear to also be sitting on in the Norwich Union picture?

[tggzzz]

Ahh -- I think I got it now. The RIFA reference gave it away... So we are talking purely mechanical, right? And they came stock with that angled mount, which they appear to also be sitting on in the Norwich Union picture?

You've got it!

I wonder whether other people will fill in the details.

I wonder whether >12m/40ft long will help Hence the original hint "probably the biggest calculator, short of an old mainframe computer"

[JDubU]

https://en.wikipedia.org/wiki/Fuller_calculator
https://www.tnmoc.org/slide-rules-calculators?rq=calculator

[MK14]

The Hewlett-Packard 9100A is pretty big, as well, all 40 pounds of it.

https://en.wikipedia.org/wiki/Hewlett-Packard_9100A

I saw one of those for sale at a hamfest last year. Someone physically bigger than me was determined to buy it. Half of me thinks that was for the best.

Yes, probably for the best.

Although an amazing piece of technology for the time period (late 1960's), it probably would take a massive amount of work to get (and keep) it working.

My concern was more that it would take a massive amount of work just to get it upstairs and onto my workbench. Work in J, that is.

Yes, that makes sense (too heavy).  Also, some of the really old parts can get harder and harder to get hold of.

[tggzzz]

https://en.wikipedia.org/wiki/Fuller_calculator
https://www.tnmoc.org/slide-rules-calculators?rq=calculator

Yup. Too easy

[tggzzz]

Time for some SFW porn, split into two posts since there are 11 images.

The name, logo, and serial number, 9763/~14000, made in 1949.






Here's the calculator unfurled...




... and ready for use ...

[tggzzz]

The obligatory teardown...

There are three sliders which have move, to remain fixed, and not being abraded with use. That's non-trivial to guarantee! I've removed the top/inner brown slider, exposing how Fuller managed to achieve that: felt pads.




Some constants that ought not to be useful ...






... and some that are...



There are three variants

• type 1 is for insurance brokerage calculations. Hence many being seen in the previous photos of the Norwich Union marble hall.
• type 2 adds extra scales for sines and logs which add some precision
• the "Fuller-Bakewell" calculator incorporates suggestions made by Mr W. N. Bakewell I.C.E. that make it more useful for engineers and surveyors: cosines squares, and sines multiplied by cosines

Mine is clearly a type 1, but it also has the log scale, in the form of a linear scale split across the brass finger (two more sig figs) and the rotating cylinder (two less sig figs)

This is obviously a type 1.

As for accuracy, as with all logarithmic scales, there is better resolution and accuracy for numbers nearer 1 than 10. Given Zipf's Law, arguably that isn't as irritating as might be thought. For most of the scale, it is accurate to 5 sig fig, but at the high end it is more realistically 4 sig fig.

Here's pi², 9.86960

[IanB]

Some constants that ought not to be useful ...

Might be amusing to ask who knows what a "Cwt." is? 

Possibly unfamiliar to the younger generation or anyone from the North American continent.

[tggzzz]

I suppose I ought to justify some of my obtuse hints...

Fairly rare; there were far more HP35s produced.
~14000 Fuller calculators were produced in total

Older than an HP35.
Mine is 1949

But is bigger than an HP35... Probably the biggest calculator, short of an old mainframe computer, that has ever existed or will ever exist
The scale is 500inches long, i.e. 41ft 8inches or 12.7m. Some people, e.g. the inventor, double those figures on the basis that a 10" slide rule has the centre slider, and so can be up to 20" long when that is extended. I think that is marketing speak from 1878.

...and was in production for significantly longer.
Invented 1878, the HP35 instantly made it obsolete in 1973, so a ~94 year production run. HP35 eat your heart out

But I do have a couple of component used in the world's oldest operating computer.
Dekatrons, as used for storage and computing in the WITCH: https://www.tnmoc.org/witch Go see it!

I've seen pictures of some devices up for auction where the surface looks like a RIFA in dangerous condition
The scale is paper underneath something, presumably resin secreted by a female insect.

For example, but I've seen worse!


Auction price? I've seen estimates from <100 to >2000 "pesos
https://www.sothebys.com/en/buy/auction/2019/history-of-science-and-technology-including-fossils-minerals-and-meteorites/a-fullers-spiral-cylindrical-slide-rule-calculator

[tggzzz]

Some constants that ought not to be useful ...

Might be amusing to ask who knows what a "Cwt." is? 

Possibly unfamiliar to the younger generation or anyone from the North American continent.

Well, there's a hint in the "C"

Of course, being Americans, their "short" Cwt is smaller than our "Imperial" Cwt. That extends up to the "short ton" and "long ton".

[IanB]

Of course, being Americans, their "short" Cwt is smaller than our "Imperial" Cwt. That extends up to the "short ton" and "long ton".

Curiously, American custom tends not to make use of aggregate units. So for example you would see feet used exclusively for distance (no yards). Or in engineering perhaps inches, but not feet. It was a culture shock when I saw road signs saying "Exit 1000 feet" where in Britain I am used to signs like "Layby 400 yds".

[IanB]

On the subject of this thread, I have not heard of or seen a Fuller Calculator before today. So this thing is rather fascinating.

[tggzzz]

On the subject of this thread, I have not heard of or seen a Fuller Calculator before today. So this thing is rather fascinating.

Long long ago in a galaxy far far away I heard an old joke along the line of...

At school and as an undergrad, I used a 10" slide rule.
Doing my PhD I used a 20" slide rule.
As an industrial researcher, I used a 25ft slide rule.
When a professor, I used a 6" slide rule.

So I've long know of their existence, but only recently bothered to investigate them.

[SiliconWizard]

That is vintage computing!

[ebastler]

Yes, I forgot to say: Thank you for the little quiz, tggzzz, and for reminding me of the Fuller calculators. I had come across them years ago, but had forgotten all about them. And, in the context of this quiz, was biased too much towards electro-mechanical or electronic devices, prompted by the more obvious contemporary calculators in the picture you had shared.

It is remarkable that slide rules were still in use in parallel with those digital calculators for so long -- even in environments like the one pictured, where space and budget were obviously available to provide the digital tools of choice. But for trigonometric functions and logarithms, slide rules ruled!

I am surprised that an insurance company like Norwich Union needed all that trigonometry though. Does anyone have specifics about the use cases?

[tggzzz]

Yes, I forgot to say: Thank you for the little quiz, tggzzz, and for reminding me of the Fuller calculators. I had come across them years ago, but had forgotten all about them. And, in the context of this quiz, was biased too much towards electro-mechanical or electronic devices, prompted by the more obvious contemporary calculators in the picture you had shared.

It is remarkable that slide rules were still in use in parallel with those digital calculators for so long -- even in environments like the one pictured, where space and budget were obviously available to provide the digital tools of choice. But for trigonometric functions and logarithms, slide rules ruled!

I am surprised that an insurance company like Norwich Union needed all that trigonometry though. Does anyone have specifics about the use cases?

You're welcome It has been fun.

I enjoy jolting people's out of their unnecessarily narrow preconceptions - and enjoy being jolted myself. In that vein, you gave me a new Awful Warning example of skeuomorphic interface design

I presume the insurance company used the type 1 with all the handy conversion factors, not the type 2 that was more oriented to surveying.

[artag]

I hadn't heard of the Fuller calculator either .. though I did ask about the ' tall poles dotted around the room.'.
What I do have is an Otis King cylindrical slide rule which also has a wrapped scale, though only 66 inches. Is it fundamentally the same thing ?

To be fair to HP, the 35 might have had a shorter production run but it was obsoleted by their own designers producing a successor, many times over. It was the changing manufacturing capability that caused it to be improved rather than a completely new technology replacing it.

[tggzzz]

I hadn't heard of the Fuller calculator either .. though I did ask about the ' tall poles dotted around the room.'.

You did indeed. And I chose not to comment. But I did hint that people should look at what I wasn't saying

What I do have is an Otis King cylindrical slide rule which also has a wrapped scale, though only 66 inches. Is it fundamentally the same thing ?

It appears to be. I did think of buying one, but bigger is better in this case

Strangely, I have an Otis King instruction leaflet on thin paper. Legible except for one page which has some "smears" or "pressure" marks.

To be fair to HP, the 35 might have had a shorter production run but it was obsoleted by their own designers producing a successor, many times over. It was the changing manufacturing capability that caused it to be improved rather than a completely new technology replacing it.

Counterpoint is that Fuller's concept was so remarkable that it remained in production for almost a century. Not many scientific/commercial instruments manage that.

[IanB]

Counterpoint is that Fuller's concept was so remarkable that it remained in production for almost a century. Not many scientific/commercial instruments manage that.

I was in the last year at my school that was taught how to use a slide rule (and where our parents were asked to buy one as school equipment).

I do recall there was a large scale demonstration slide rule for teaching with. It was about 8 ft long and a foot wide, so that the whole class could see it. I think that is a little bigger than your example here.

[tggzzz]

Counterpoint is that Fuller's concept was so remarkable that it remained in production for almost a century. Not many scientific/commercial instruments manage that.

I was in the last year at my school that was taught how to use a slide rule (and where our parents were asked to buy one as school equipment).

I do recall there was a large scale demonstration slide rule for teaching with. It was about 8 ft long and a foot wide, so that the whole class could see it. I think that is a little bigger than your example here.

I too learned to use four fig tables, and I still have my school slide rule.

The scale on the Fuller calculator is 500inches, i.e. i.e. 41ft 8inches or 12.7m. So the Fuller is about 5* bigger

There is the 1878 book "Spiral Slide Rule: Equivalent to a Straight Slide Rule 83 Feet 4 Inches Long, Or, a Circular Rule 13 Feet 3 Inches in Diameter" by George Fuller. I think the "83ft" is technically correct and misleading. That's ~150 years ago: advertising puff isn't a new phenomenon

[ebastler]

The scale on the Fuller calculator is 500inches, i.e. i.e. 41ft 8inches or 12.7m. So the Fuller is about 5* bigger

Following that argument, I propose to measure the size of other computers by adding the length of all the wiring in an analog computer, or measuring the length of the torsion delay line wire in a Diehl or Monroe desktop calculator, or of all the PCB traces in a desktop PC. 

I think we should stick with outside dimensions, which are impressive enough for the Fuller calculator. Not sure whether the classroom demonstration slide rule counts here, since its size was not driven by the technical requirements of implementing the computing technology.