The Once and Future Slide Rule, Great and Terrible

Because the most productive thing I’ve done today is receive a slide rule in the mail and learn how to multiply by pi in a novel way, let’s start with that as the basis for a blog post with which to rectify my sloth.

The slide rule I got is a Pickett N600-ES (“eye-saver”, meaning bright yellow) circa 1970 that I won on eBay for about $50. This model is relatively small, made of aluminum and therefore lightweight and durable, displaying a large number of “scales” or number lines with which to calculate. Those are some of the reasons this model was taken to the moon on no less than five Apollo missions, including Apollo 11 (Neil Armstrong and Buzz Aldrin both carried this very model) and Apollo 13, where, I trust, it served them well. And yes, those are some of the reasons I, in turn, bought this compact beauty.

Figure 1. Ceci n’est pas Buzz Aldrin’s actual fucking slide rule, man!

***

Figure 2. What the box for Buzz Aldrin’s slide rule would have looked like if he had lost the first one and bought a replacement a few years later out of his own pocket.

My slide-rule bender goes back only a month or so, but once I get interested in something, it’s hard to deflect me. First, I posted my top five personal retrocomputing desiderata, and then was intrigued when my friend Zenoli did the same and one of his was a decent slide rule of some kind.

Next, there was a long, unrelated thread about slide rules on the TRS-80 Model 100 list, a list which occasionally veers into other retrocomputing topics. It amused me that people were talking about whipping out their big ten-inch to impress younger engineers (I paraphrase, but only slightly), completely unironically. (Alas, I can only whip out six inches. Before today, I couldn’t even do that.)

Finally, one evening during some documentation builds at work, I started googling about, looking for slide rule manufacturers, and was astonished to discover that there are in some important senses literally none at present. I mean, not even for homeschoolers? I’m not being sarcastic; using a slide rule will improve your numeracy in ways that using a calculator or calculator app never will. It’s a mentat skill.

This is all we have: A few years ago ThinkGeek did manufacture one as a novelty, but the quality was poor and the experiment was not repeated. There is also a Japanese company called Concise that still manufactures special-purpose circular slide rules. And that’s it.

But they can hardly replace the little plastic slipstick my uncle gave me and taught me how to use when I was 10 (he worked at Boeing and had probably just gotten a calculator). For nostalgia is what this quixotic retropursuit is all about, no? (Not quite; see below.) Then, too, my wife Marty’s late father had one, and she thinks it’s here in the garage somewhere… Well, wish us luck with that.

While googling, I learned about the talismanic lunar magic of the N600-ES, and what a solid little performer it is generally, so I sniped one on eBay that was firing in a couple days. And now it’s mine. For me, that’s part of the “great” part of “great and terrible”.

What’s the “terrible” part? We built the Hoover Dam, the Empire State Building, and to a large extent, the Apollo program and its precursors with slide rules. That’s pretty badass, don’t you agree?

“Once and Future,” the “future” part

I assume I don’t need to explain the “once” part…

If you don’t want to buy an N600-ES yourself — or any other kind of slide rule — there are all kinds of “modern-day fun” (as the Retrobits Podcast says) to be had without one. For example, you can play with an N600-ES “virtual slide rule” emulator online or on your Android. There are virtual slide rules for iOS. too. In fact, there are a huge number of slide rule emulators available online and off, not just of my model.

Another bit of modern-day slide rule fun is the UltraLog, a slide rule that was scrupulously designed a couple of years ago by Zvi Doron (who is now deceased) but which was never built. The UltraLog was meant to be the slide-rule fan’s slide rule, about a foot and a half long with 40 well-chosen and ergonomically-presented scales (mine has about 20, and that’s considered a lot; the model Einstein and von Braun used had only nine). There were incipient plans for an independently-funded production run of 500; this is the kind of thing destined for success on Kickstarter, but perhaps Zvi wasn’t familiar with that business model. It might even succeed today. In any case, there seem to be high-quality digital graphics of the UltraLog available; why can’t there be an online UltraLog emulator, at least? (I’ll get right on it.)

There is, by the way, a really good course for learning how to use slide rules at the online International Slide Rule Museum, but it won’t give you much theory. If you want to learn why a slide rule works as well as how, I highly recommend Isaac Asimov‘s lucid 1965 book An Easy Introduction to the Slide Rule, available as a PDF, also at the Museum, or as a used pbook for extortionate prices from your friendly online bookseller.

The gaming/retrocomputing connection

So why am I obsessed with slide rules, and retrocomputing generally?

I’m writing a dictionary of imaginary games notionally published on the equally imaginary planet Glob. A few days ago, I was musing about the games with slide rules there, where they can calculate with things other than numbers, such as days of the week, months of the year, and planets of the solar system. (Yes, I figured out how to do this; my book is “design fiction”, so every game in it is possible to implement in theory. I hope people do, in fact, after the book is published.)

It was then I understood the reason why all this time that I’ve been sketching imaginary games, I’ve also been busy sketching imaginary retrocomputers: machines that still boot into Lisp or Forth, “writer’s notebook” computers with the “cinematic form factor” similar to the Model 100 but with higher resolution, and so on. The people of Glob are my people, as the Elves are Tolkien’s, and they share my aesthetics, as the Elves share Tolkien’s. And I like elegant things.

I am no fan of the board games Advanced Squad Leader or Star Fleet Battles; I can barely comprehend (com-prehend, “togethergrasp”) in my mind the 30-page rulebook to Elder Sign, let alone its big sib Arkham Horror — or, I flatter myself, I don’t care to. The most fun I had at game night recently was playing Nonesuch, a simple trick-taking game for the Decktet that nevertheless offers plenty of emergent complexity. I enjoy the complex, rather than the complicated. I enjoy things I can “grasp together” in my mind and turn from side to side to see, yet which still surprise me — whether they be card games, computer games, or computers.

Returning to computers, I had to install one of the smaller versions of Microsoft Visual Studio at work the other day. Do you know how big the install was? Over seven gigabytes. Do you know how big all of the firmware native to my 1987 Tandy 102 laptop is, including all the applications, and the operating system, such as it is? 32 kilobytes. Does Visual Studio do more than my Tandy 102? Sure, a lot more. Does it do five orders of magnitude more? Does it have 221,000 times more functionality? I’ve been using the previous version for most of a year, and I assure you it does not. (Don’t accuse me of Microsoft-bashing; Microsoft also wrote the Model 100 / Tandy 102 firmware I appreciate back in 1982. Famously, it’s the last code Bill Gates ever touched.) I like elegant computers too, and often that means retrocomputers.

So for games, give me chess variants, games with slide rules, and ten-dimensional card game systems that are nevertheless less complicated than Dominion. For computers, give me the Model 100 and its kin, not Visual Studio.

This is largely the origin of my retrofascination.

Model Mmm

If you have noticed that this post is longer than usual, it may be because I have been writing it on an item from my retrocomputing wishlist: a Model M keyboard, specifically a brand-new Unicomp Ultra Classic USB keyboard. After years of using a netbook as my main computer, I had forgotten until my Tandy 102 how sweet it can be to write on a real computer keyboard, and they don’t come much realer or sweeter than the Model M.

Conclusion

So that’s one item (plus a slide rule bonus) off my wishlist, and four to go. I may give up on the Jupiter Ace, though, because the FIGnition kit, modeled in some ways on the Ace, is so much more available, and I stand to learn much by building one.

Finally, for more information about slide rules than you probably care to know — but it’s curated! — let me direct you to my slide rule bookmarks on Pinboard.

Image credits

1. http://sliderulemuseum.com/Pickett.htm
2. http://www.oughtred.org/books/AllAboutSlideRules_OughtredSocietyPublication_rev121001.pdf

10 PRINT on the Model T

Last year, MIT Press released a book that examines computing through a miniature art form: the BASIC-language one-liner, or one-line program, and the subculture surrounding it. The title of the book is 10 PRINT CHR$(205.5+RND(1)); : GOTO 10 (or 10 PRINT for short). You can buy a paper copy or download a free ebook from 10print.org.

The unusual title of the book is the one-liner itself, a famous one for the Commodore 64 computer that draws an hypnotic, flowing maze pattern on the screen. This maze is the starting point for all the authors’ speculations and essays, and they range widely.

For context, here is a video of the 10 PRINT pattern in action.

As an example of what is to be found in the book, there is a history of mazes and labyrinths; a survey of repeating mathematical shapes in modern art such as dance; and a couple of chapters devoted to porting the one-liner to other platforms. I found that the Applesoft BASIC version given (for the Apple II+ and later) works just the same on my Model T (TRS-80 Model 100 / Tandy 102), because both BASICs are typical Microsoft implementations of the era. However, because both the Apple and the T lack the special graphic characters that make up the maze on the C64, this version uses slash and backslash characters instead (/ and \), and the result is a little unsatisfying.

10 PRINT CHR$(47+(INT(RND(1)*2)*45)); : GOTO 10

Here is a screenshot of this Applesoft code running on the Model T (actually the Virtual T 1.5 emulator running on my Ubuntu netbook). Note that this pattern, and all the others, is moving and constantly growing. (Click to enlarge.)

I decided to see if I could improve on the canonical Applesoft version with some of the Model T’s own graphic characters, and used 4 of them instead of 2 (characters 251 through 254). My one-liner is shorter than the Applesoft version, and the result is more graphically striking, but it doesn’t generate a maze so much as an irregular network of caves. Not quite what I was going for, but if the Model T used screensavers, this would be a pretty good one, I think. Note that this Model T pattern and some of the others below can be duplicated fairly closely on the Commodore 64 as well (see the “Variations in Basic” chapter).

10 PRINT CHR$(251+(INT(RND(1)*4))); : GOTO 10

I wondered if any M100 mailing list members  who’d worked with the graphics characters on the Model T would like to try improving my version and posting their code back to the list, and several did. Jan VANDEN BOSSCHE was the first to respond, saying, “I don’t have a Model T handy – not even an emulator – but what about this?”

10 PRINT CHR$(229.5+RND(1)); : GOTO 10

A nice pattern, especially for writing it in his head. I also like how he managed to keep the “char.5” trick from the original. Jan said, “I could keep the .5 trick because these characters are also next to each other, just like the original C=64 ones.”

Willard Goosey’s judgment was, “Neat pattern, but too closed in to be a maze.” Next, Donald Kyllo offered the following one-liner, to which Willard responded, “Now that’s a maze generator!” I concur.

10 PRINTCHR$(240+INT(RND(1)*10));:GOTO10

Finally, Ken Pettit, a VIP in the Model T community (he runs the Club100 user group and invented the popular NADSBox storage device for the Model T), had more or less the last word. He wrote, ‘Okay, well this sorta breaks the spirit of “10 PRINT”, but I have a one-liner that produces very similar graphics to the C64’s one-liner.  I make the first statement a “10 PRINT” by printing a CHR$(12) (clear screen) in an attempt to keep with tradition of “10 PRINT”.  Otherwise this uses the line command.  Take it for what it is.’

10 PRINTCHR$(12):FORC%=0TO5000:V%=INT(RND(1)+.5):X%=(C%MOD30)*8:Y%=INT((C%/30)MOD8)*8:LINE(X%,Y%)-(X%+7,Y%+7),0,BF:LINE(X%,Y%+7*V%)-(X%+7,Y%+7*(1-V%)):NEXT

I think this and Don Kyllo’s are my favorites. Don’s is more elegant, more in keeping with the original code, but Ken reproduces the original Commodore graphics better. It doesn’t scroll like the others, but fills the screen and starts at the top again with an intriguing ripple effect.

I don’t think that Ken breaks the spirit of 10 PRINT entirely. At least it’s a one-liner, unlike most of the Processing ports in the book, which take a page. On the other hand, the Processing programmers were trying to reproduce the Commodore 64 experience as exactly as possible, as Ken was. At the other extreme, one of the 10 PRINT authors has a student who ported 10 PRINT to (if I recall correctly) 8 bytes of assembly language.

Juan Castro spoke up with a Mattel Aquarius port (!). He remarked, “[The Aquarius] has the convenient ‘spans all character cell’ slashes.” That’s not a port I expected to see.

To my remark that unlike the Aquarius, the Model T doesn’t seem to have graphics characters that fill their cells, John R. Hogerhuis responded, “Alphanumeric and punctuation characters generally don’t fill the cell because of the reserved ‘descender’ line. But the line drawing and block  graphics characters fill the cell.”

Apart from the usual retrotechnical knowhow on the list, there was an unusual amount of artistic and cultural back-and-forth, as well as a fair amount of negative reaction. Paper copies of 10 PRINT have been selling very well, but it has been a little controversial. Some people think it’s pointlessly academic. I read it from front to back and enjoyed it a great deal, and I will point out that some of the authors of the book are technically adept professional developers of games and other applications for platforms as diverse as Facebook and the Atari 2600, not atechnical pomo flyweights.

Books recommended on-list for people who enjoy 10 PRINT include:

• 3:16 Bible Texts Illuminated by noted computer scientist Donald Knuth
• Le Ton Beau de Marot by Douglas Hofstadter

And that is where we left the conversation, Reader, but it’s not over. Please leave a comment with your own perspective on the 10 PRINT phenomenon. Have a port to an obscure platform, a link to a video of one, a favorite one-liner with a different purpose, news about a similar project, or hate mail? As long as it’s not pink and doesn’t come in a can, please comment away.

Fun with the Apollo Guidance Computer

There I was, pursuing my interest in retrocomputing and reading a good book on the Apollo Guidance Computer, an unavowed exemplar of the new genre of platform studies, and I idly wondered, “Did anyone ever write an AGC emulator?”

Oh, man, did they. Versions for Windows, Mac, and Linux. Fan videos and hardware workalikes. Canned moon missions. Poster-size PDFs explaining every detail in Italian. Instructions on how to make the Apollo Guidance Computer your desktop clock. (As famous astronaut Jar-Jar Binks once said, “How rude!”)

Please, take one of the highest achievements of the American space program and do what you will with it. It’s the spirit of making and open source.

(later)

Well, I did it. I installed the emulator, punched in the right commands, and made the Apollo Guidance Computer my bitc… er, my desk clock.

Because the AGC accepts commands in verb/noun combinations, an interactive fiction “GET LAMP”-style connection is vaguely gesturing at me. Script the AGC with ZIL? Port the emulator to Inform for an “abuse of the Z-Machine”? The possibilities are endless, if somewhat implausible and, as I say, vague.