Hobbies page

Michael Robinson

I pursue some "typical" hobbies, such as playing the cello and studying languages. However, I also have some more unusual (or at least more documented) interests as I describe here. I also have an online build log called "Engineering Mathematical Machines". Feel free to contact me about my projects!

(The content on these pages may be freely used for educational, noncommercial purposes provided appropriate references are provided. Commericial users are kindly asked to contact me first.)

Amateur Radio

Tower climbing When I was a child, I always looked at my father's old radio gear (and his call sign KA1GCZ) in my parents basement with a certain sense of awe. Radio is a bit magical, even once you know how it works. Right before heading off to college, I got my amateur radio license (call sign KB1DDS). Once at college, almost my first act was to join W2SZ, the amateur radio club. That was fortuitous as I met my future wife (call sign N2SZ) in that organization, but it also gave me the opportunity to explore some of the more interesting possibilities of the hobby. I learned to climb radio towers to install and repair antennas! I alo met some of W2SZ's illustrious alumni, who call themselves the Mount Greylock Expeditionary Force. They focus on competing in VHF, UHF, and microwave contests. (They also get mistaken for being alien-hunters and talking to the dead, but that's another story.) They run an exciting operation three times a year: I've helped out on occasion, and it is very impressive.

Building computers from gate ICs

Lots of people like to build their own computers from stock parts. It's fun, and I've done it both for work and amusement. However, in high school, I encountered an old 74xx-series logic databook on my father's shelf. The book contained descriptions of small integrated circuits that do very simple things, like AND gates, OR gates, and registers. I thought to myself, "wow, these are just the things I'd need to build a computer entirely from scratch!" I was delighted to find out that this is what such chips were intended for and that they can still be bought. So I set about in the somewhat eccentric hobby of constructing my own computer processors, in the style of the minicomputers in the past! (Here's a different take on the same idea; build your own gates from transistors! There's even plans for a 4-bit computer...)

My first attempt (in high school) was an asynchronous processor, with an elaborate state machine controller. (Ill fated plans are available.) This experience is of course is why I am still interested in asynchronous circuits. Sadly, the machine didn't work, at least as a single unit. The different components didn't work together due to synchronization problems.

That wasn't the end, as I tried again with a synchronous design in my sophmore year of college. I constructed a machine that has a single instruction: subtract and branch if negative. Never mind the difficulty of programming such a machine, it is a complete computer! While I chose a synchronous architecture (timing diagram, complete schematic (blurry), logic diagram), and a carefully constructed controller, I didn't realize the importance of filtering the power supply on the processor board itself. Therefore, the machine behaves erratically at certain points in the execution cycle. I'm pretty sure that adding bypass capacitors at each chip would fix the problem, but I haven't tried.

My third attempt scaled back considerably from the previous attempts. I chose a Harvard architecture (in which code and data are separate) and permit computed branches to occur after every instruction. This simplifies the execution cycle as there are fewer conditional states to worry about. I was a whole lot more careful about signal quality and synchronization on this machine, and it works! Here is its instruction set. The design is summarized in a one page schematic and these Verilog sources. You can also use a custom assembler, which also includes example programs. To build one of these, you need an EEPROM programmer and data files. The latter link contains source for a small Linux program to drive the programmer.

Building on the success of my third machine, I tried a more ambitious machine, with more memory, instructions, and slightly more disciplined branches. This machine's instruction set, Verilog source, plans,and EEPROM files (use above programmer) are available. I rather like this machine, as it is a nice balance between usability and complete simplicity. I had lost it for a while, but recently found it buried in my garage at the bottom of a box!


I'm not just a fan of using 74xx-series logic to build machines of my whims. I also enjoy the machines that were built originally with these chips. For instance, I have a pdp-11/45 (shown here being operated by my son, Edwin), that I actually found on a trash heap at RPI. It had no memory, no peripherals, no power supply, and had signs of physical abuse. After several years of careful testing, detective work (including leads from RPI staff, who were able to locate the engineering diagrams and some core memory), and luck, I brought the machine back to operation. It's a joy to program, especially from the front panel, and certainly "looks the part" of a real computer. Unfortunately, after following me for a few moves, it seems to be a bit flaky.

At roughly the same time, I intercepted one of RPI's VAX 8530 being hauled for scrap. That machine, while not being as badly damaged as the pdp-11, still needed some care. It had a tape drive, racks of hard drives, and the like. It was missing the console computer that was required to load the microcode (and hence was unbootable), but a bit of effort on the part of myself and a few friends found one. I haven't seen that machine much, since I graduated from RPI and a friend took over its care.


You can read about my recipes at my Recipes blog, which largely functions as a repository for the meals that "worked." It is too often the case that we forget what made a particular meal especially memorable, so at least with a blog that information is not lost.


I would not have expected to get drawn into gardening, but when our family lived in Ithaca, we had an ample yard that was well-suited to gardens. A previous owner had a old, scrambling rose plant that bloomed enthusiastically each summer. I was intrigued, and since Cornell University has a good horticulture school, it also has a good library for plants. So I dug through old manuals of roses, explaining in detail all the hundreds of varieties of roses that were available. After much searching, I found that the rose in question was a "White Dawn," a somewhat rare variety that usually isn't hardy to the area. But since the plant was growing against a fence, it thrived.

At that point, though, I was hooked. I had seen enough interesting roses and other plants to know what I wanted to grow. I have a few at our house now, and the collection is slowly growing.

Camping, Hiking, and Geocaching

I've always enjoyed being outside, and going hiking, boating, and the like. More recently, our family has gone geocaching. I usually like to use my n900 phone's built-in GPS with AGTL to control it. However, we have had good success with a simple Garmin eTrex (no maps) as well.