The HP-35 Design: A Case Study in Innovation (2011) [pdf] | Hacker News Reader

The HP-35 Design: A Case Study in Innovation (2011) [pdf]

archive.computerhistory.org

79 pointskeenerd15 comments

My dad had this calculator in the 70s. He was an engineer for the local power utility and told me that before getting the HP-35 he would submit calculations at the computer center and get his results back about a week later. With the calculator he could run his computations immediately, at his desk. It was life changing, at least in the context of his work.

It was one of the more expensive small things he owned so he never left it at his desk, carrying it around in his jacket pocket (engineers wore suits to work in the 70s). He told me he also got a reputation for this, since it was such a unique thing to be carrying around.

Calculators went through a dramatic price drop over a very short period of time. As I recall, a 5-function TI calculator was around $100 in 1975 while HP calculators were maybe 5 times as much. (So about $2500 today using standard CPI measurements.)

Within about a year, full TI scientific calculators were down to about $100 and by a year after that HP models like the HP-55, which I owned for a long time, were around the same price point.

Calculators essentially went from something unavailable to unaffordable at the student level to not inexpensive but ubiquitous for engineers (even students) over less than five years.

Back in the early 1990s, I purchased a HP 48G calculator for $150, IIRC. It was at a university bookstore, so it may have been overpriced compared to the actual market value at the time.

Same thing with my dad, but at university level as he could write engineering programs to help with homework and avoid booking time on the mainframe.

Btw: Would you please ask your father what type of calculations he ran on his calc as a power systems engineer? Fortran would've been around for state-estimation and contingency analysis by that time if he was in transmission. For distribution I could see him doing fault analysis or calculating line parameters. If he was on the generation side I'm not really sure what he'd be calculating (not my area). Of course utilities were fully vertically integrated during that time frame, so he could have done a mixture.

I'll ask the next time I talk to him. I know he worked on high voltage transmission.

Awesome...same as me. I'd love to hear how they did things in more resource constrained times.

I also have a hobby interest in HP calcs (own a 15-c), so double interest here. Not sure if you're a blogger, but that is some legitimate history with the early usage of handheld computing to aid science & engineering. A blog post would be great, but I'd settle on a reply here.

I'm a past blogger, but I haven't kept up with it for a while.

Here's what my dad wrote when I asked him:

The work often involved transmission line impedance calculations. These and other power calculations required a great many translations of rectangular to polar coordinates and back again. We had an Oliveti vacuum tube machine in the office that would only do this one operation in one direction or the other. Otherwise it was back to slide rule, pencil and paper. We had an IBM 360 in the basement that would do grid system power modeling. These calculations required 24 hours to get an answer.

Mom got me an HP-35 for Christmas. It was like touching an alien spacecraft technology.

Later on around 1976, I operated a consulting and substation design business. In this activity I needed to do hand calculations of short circuit current for industrial customers again using the HP-35. The HP-35 was an important tool and a real game changer for electrical engineering work.

Thanks for the reply! It was a very interesting read for sure. The polar->rectangular conversions can be done with an $15 Casio FX-115ES from Walmart pretty easily although RPN is much nicer than scrolling around a bunch when the equations get long. He might already know that, or find it interesting. I've never heard of an Oliveti vacuum tube machine, but will look it up. The IBM 360 machine he mentioned is probably not that different to what we do today (I bet the software has only changed in minor placed). The biggest difference is it's moved to Nix or Windows and off the mainframe and we can simulate massive scale problems in seconds. Of course that means we steadily increase the scope of the problem until you get a process that takes 24 hours ;). I'm impressed with anyone who can use a slide-rule for anything more complex than multiplication. Props to your dad btw. I love my job and it's fascinating to see how past engineers managed to essentially do the same things with tools that seem archaic today. I guess this is similar to modern day software engineers on HN getting really excited bout the Xerox Alto Smalltalk machine & the Symbolics Lisp Machines. The past is neat.

> I'd love to hear how they did things in more resource constrained times.

I found this to be very good: http://www.hpmuseum.org/software/soft41.htm

And I've got this book from 1976: https://www.abebooks.com/book-search/isbn/0830668535/

It covers several different calculators too.

Thanks for the gems. It's been awhile since I've been to the HP Museum page. The Circuit Solver example is pretty cool, but that is a lot of keystrokes haha:

http://www.hpmuseum.org/software/41/41elcirc.htm

I actually have my dad’s HP-35 on my desk, and use it quite often. I patched it to a USB cable for power, just a quick hack. Would be nice to drop in a rechargeable battery pack someday.

Very well written and makes me realize what a pioneer company HP was at that time. It was the Apple of its time. Best anecdote:

> Stanford Engineering Dean Fred Terman, the person responsible for bringing Bill Hewlett and Dave Packard together to start HP was one of the first. He was overwhelmed, looking for an umbilical cord connected to a big computer doing the precision calculations.

I bought a used one in the 80's and found it an amazing device: simple yet powerful. A Turing Complete pocket computer. It should be noted that HP did NOT invent reverse polish notation, and the HP-35 allegedly borrowed a lot of ideas from the Olivetti Programma 101 computer, arguably the first desktop computer.

Still, the HP-35 was an amazing packaging of existing but relatively new ideas (at least as actual products). This kind of thing is what made Apple big: take different new but existing ideas and combine and package them well by keeping the good and tossing the bad or low-use features. Steve Jobs had a nose for how to use new ideas invented by others by carefully picking, choosing, and associating features. What you leave out is often more important than what you put in.

I've also read that HP was hesitant to advertise such devices as "computers" because that carried export restrictions per military concerns. Calling it a "calculator" got around that. The convention of defining "computer" as something Turing Complete came later I believe.

Very interesting article. My dad, an engineer, had the successor to this model, the HP-45. It was an amazing machine for it's time. Because of it, I fell in love with RPN and refused to use anything else. When I went to college I got an HP-48sx. I think it was the zenith of scientific calculators with so much functionality crammed into it.

Really enjoyed that.

Reply on news.ycombinator.com