No perfect resistors
Paw Prints: Writings of the maddog
Recently I was watching a video that Christopher “Monty” Montgomery, the founder of the Xiph.org project had produced on audio and video formats. They did a good job on the video, putting it out in both WebM format and Ogg format, with subtitles in English, French, German, Portuguese (Brazil) and Russian. They even include the SRT files so other subtitle translations could be done.
About position 7:36 on the video, where Monty was describing how "there is no such thing as a perfect transistor, or a perfect inductor or a perfect capacitor" , I burst into laughter.
In 1972 I told my college roommate *exactly* *those* *words* while he was studying digital filtering techniques at Drexel University in Philadelphia and trying to develop some of the first digital filters. Computers were just too slooooowwwwww to be doing digital filtering, so we had to develop digital filters out of discrete components and in order to keep the costs and size down, the algorithms had to be precise (as Monty pointed out in his video).
I started out studying Electrical Engineering at Drexel in the fall of 1968. I had taken three years of electronics in high school, and was very used to designing and building all sorts of analog circuits, just like Monty described in the video. My roommate was a ham radio operator, which had stimulated his interest in electrical engineering, but he had not done that much hardware work before going to Drexel. Drexel was a theoretical school, and you did not touch a soldering iron for the first three years.
After a year or so I switched to software in these (relatively) new "computer things", and my life went away from electrical engineering, although I kept a passing interest in it (and still do, which is useful for the Arduino work).
I met my roommate in the third year of Drexel (being a co-op school we went for five years, not four) and we stayed roommates until we graduated in 1973.
He continued studying digital filters while attending the University of Colorado in Boulder for his MSEE, and his Masters's thesis was on digital filtering and the mathematical algorithms to build them the fastest, least expensive way. Remember that back in those days transistors were not "billions and billions for a couple of dollars." In fact, I remember paying about one U.S. dollar for a transistor back then, and beer was .50 a pitcher at the "Railroad Bar" next to the university.
Unfortunately for my former roommate his thesis advisors at the University of Colorado found out that someone else at a different university was doing the same work (a bad thing for a thesis) and worse, the other person was getting almost the opposite results (hey, it was research right?)
Soooo....the two universities brought together the two hapless university students and had them defend their thesis against each other. Fortunately for my former roommate he was right.
Unfortunately the other student (who was wrong) had to start over.....no wonder there are so many people "ABT" (All But Thesis).
This, of course, caused a bit of a ripple in the EE world, and after my former roommate graduated he went to work for a large defense company, and then he spent the first six months of his employment flying around the world talking about his work at conferences.
After that he settled into Southern California and started to do "real work". About nine months after that his first real digital filter was working.
That was when I received a letter from him that only said:
"Now I know what you meant when you said there was no such thing as a perfect resistor."
It has been almost forty years since I thought about incident, and hearing Monty's words reminded me of a time long ago.
Carpe Diem!comments powered by Disqus
Linux Foundation's big event celebrates the 25th anniversary of Linux
Competitors get in the game with RHEL without Red Hat
Security researchers have already notified Microsoft; some fixes are available
The company is collaborating with Google and Intel to use Kubernetes as an engine for Fuel
Customers can take a free test drive of SLES for HPC on the Azure Cloud
San Francisco-based chip company announces their first fully open source chip platform.
The whole distro gets rebuilt on glibc 2.3
Ubuntu Vendor tries to solve app packaging and distribution problem across distributions.