2 ms·
Yup! We had two of them. One of the groups engineers was doing early Expert System stuff on it. They were a very odd duck to try to get used to just sitting col
by whartung 3y ago
Yup! We had two of them. One of the groups engineers was doing early Expert System stuff on it. They were a very odd duck to try to get used to just sitting cold at the console, and not having much inkling about Lisp didn't help either. Otherwise I'd be able to give a more cogent recollection of the machine and its environment.
The only thing of note at the time was that it was still using 10base5 "thick" Ethernet. Boy, was that stuff nasty. Huge, ungainly cables, and large boxes with truly dangerous spikes in them. As I recall you had to measure spots on the cable to add new taps, you couldn't just stick them anywhere. We were all 10base2 at that point, so it was not friendly.
Also, with all the recent hub bub about NVIDIA and it's "AI Chips", TI was a pioneer here as well. They managed to squeeze the bulk of the TI Explorer Lisp Machine onto a single chip. 550K transistors I think, 256 pins. Huge. A real marvel. Some were suggesting putting these into cruise missiles. I was a bit dismayed that they want to blow up a chip like that.
Obviously, this was very early, but also it was AI Autumn at that point.
Shame the machines just sat there in the end, we were more interested in the Suns that we had.
- imglorp 3y agoOur ECMP dept at CWRU had a couple also. One of the students used its graphics to simulate and draw roaches and their walking neurons, experimenting with what happens when you turn off a leg.