Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!purdue!bu-cs!buengc!art From: art@buengc.BU.EDU (A. R. Thompson) Newsgroups: comp.arch Subject: Re: WOM(Write Only Memory): also the SED and the DED Message-ID: <2936@buengc.BU.EDU> Date: 22 May 89 20:19:49 GMT References: <184@berlioz.nsc.com> <1227@motmpl.UUCP> <2365@wpi.wpi.edu> Reply-To: art@buengc.bu.edu (A. R. Thompson) Followup-To: comp.arch Distribution: usa Organization: Boston Univ. Col. of Eng. Lines: 97 In article <2365@wpi.wpi.edu> jhallen@wpi.wpi.edu (Joseph H Allen) writes: >In article <184@berlioz.nsc.com> nelson@berlioz.UUCP (Ted Nelson) writes: > >>Or we could use National Semiconductor's new memory product: 1 Megabit > ^^^^^^^^^^^^^^^^^^ >> Write-Only Memory (WOM). This is extremely inexpensive, has an access >> time of only 10 ns, and will be available in a dual-port version in only >> a few months. If you wish to order any of this great part, pleast >> contact me directly -- it is such a secret project that we haven't let >> Marketing in on it yet. The WOM is a great product and has been around since at least 1971. The first I saw of it, it was as a real looking spec sheet from the manufacturer (who was the original manufacturer, anybody remember?) I see by your posting that there have been many improvements in the design since then. Chief among them a substantial reduction in power requirements, down from the 6.3A of the original. > >Speaking of new products, I'd like to introduce my new display devices: the >SED (smoke emitting diode) and DED (dark emitting diode). Actually, normal LEDs >can be resold as SEDs by simply modifying the specs with reguard to power >requirements: instead of 5 volts through a current limiting resistor, use 9 >volts (or if you'd like to try to use the diode in the 'EED' range (explosion >emiting diode) use 12 volts) without the limiting resistor. Note that a power >supply capable of 10 amps is preferable. Here in the College of Engineering we have several students who have become quite skilled at designing with SED technology. In fact, it is a graduation requirement that all students do at least one substantial experiment with an SED. As one of our graduates put to me a couple years ago, "I've had a meltdown, hell you're not an engineer until you've at least one meltdown." The program has been quite successful and, unlike nearly all other graduation requirements, we have never had a student petition for a waiver from this particular requirement. > >Unfortunatly, the SED has several marketing problems: > > * short lifespan (only one puff of smoke) We have two graduate students, Pleishman and Fons, who are attempting to produce a voltage controlled SED. Basically their approach is to attempt to induce a low temperature, voltage controlled closure of the "Smoke hole". They call their process "room temperature fusion". Their work is controversial with some groups claiming to be able to reproduce the effect and others denying it. Ok, it's war story time. In 1971 I was a post-doc in the biology department of a large university. Computers had just started to make their way into the lab and several of us picked up on them (so much so we stopped doing biology, but that's another story). I had a friend upstairs, a young hotshot assistant prof who was into complex behavior and neural modeling. As part of his startup package the University bought him a PDP/11-20 (one of the first) with batch of memory (12k), a 64k head per track disk (yes, those are "k's" not "M's"), a DECtape (you don't know what that is? Go ask your mother.), a fancy A/D converter and a CRT (no more messy Teletypes). In those days the whole shooting match cost over $40,000. (Correct that for inflation and see what you get.) I was not part of the group, but we were close enough buddies that I shared in the excitement of the wait for the arrival of the real computer. Well, the great day dawned bright and clear. There it was proudly humming though its acceptance tests while my buddy and the DEC service man patted each other on the back and laughed. They were having the time of their lives. We set it up and had the FOCAL interpreter running and we were even talking about putting Algol up. The next day it was not so bright and clear. I walked into the lab and there was the grad student frantically running back and forth between the front and back of the computer. He ran behind the machine and shouted, "Is the DECtape turning?" "No." I replied. "Oh Shit." He said. It seems he was fooling around in the backplane and somehow had managed to jumper the logic to the 100V ac. The smoke leaked out of everything. It was a mess. On the power supply board diodes had exploded. Where a diode once sat now we only had two chunks of silver wire and a burned spot on a circuit board. Whole runs evaporated. It is very weird to look at a circuit board and see only neat black lines where those little metal ribbons used to be. You get a sort of sick feeling. He also blew the timing tracks right off the disk and completely lost contact with the upper 8k of memory. This was about the biggest computer mess I have ever seen. I would like to report that they put it all back together and they all lived happily ever after, but that didn't happen. The repair bill was over $20,000 and, since it was such a gross case of user mistreatment (remember it was our boy who connected the wall to the logic) DEC (rightly) said it was not covered by the warranty. They offered to give him the cheap parts free, sell the expensive ones at cost and supply labor at cost, but the damage was so total it was still too expensive to fix given the departmental budget. Sic transit gloria.