What do you do when you decide that running CP/M on a Commodore 128 with a 5.25″ drive “Isn’t CP/M enough”? If you are [FozzTexx], you reach for your trusty TRS-80 Model II, with its much more CP/M-appropriate 8″ drive.
There was one small snag with the TRS-80 though, its keyboard didn’t work. It’s a capacitive device, meaning that instead of each key activating a switch, it contains a capacitive sensor activated by a piece of aluminized Mylar film on a piece of foam. Nearly four decades of decay had left the foam in [FozzTexx]’s example sadly deflated, leaving the keys unable to perform. Not a problem, he cast around for modern alternatives and crafted replacements from a combination of foam weather strip and metalized gift wrap.
Care had to be taken to ensure that the non-metalized side of the gift wrap faced the capacitive sensor pads, and that the weather strip used had the right thickness to adequately fill the gap. But the result was a keyboard that worked, and for a lot less outlay and effort than he’d expected. We would guess that this will be a very useful technique for owners of other period machines with similar keyboards.
What is CP/M, I hear you ask? Before there was Linux, Windows, and MacOS, there was DOS, and before DOS, there was CP/M. In the 1970s this was the go-to desktop operating system, running on machines powered by Intel’s 8080 and its derivatives like the Zilog Z80 in the TRS-80. When IBM needed an OS for their new PC they initially courted CP/M creators Digital Research, but eventually they hired a small software company called Microsoft instead, and the rest is history. Digital Research continued producing CP/M and its derivatives, as well as an MS-DOS clone and the GEM GUI that may be familiar to Atari ST owners, but were eventually absorbed into Novell in the 1990s.
We’ve featured a few capacitive keyboards here at Hackaday before, including this similar repair to a Compaq from the 1980s, and this look at a classic IBM terminal keyboard.
Before CP/M there was RT-11.
OMG RT-11. Some vintage video editing stuff from 1981. The CMX system was the most widespread at the time and used RT-11 running on various flavours of PDP-11. Some were CMOS and had real problems with static. However since the memory was magnetic core, it was possible to restart the computer without losing it! I had forgotten about RT-11.
http://imgur.com/qsM9PYm
http://imgur.com/z6DLcGb
Before RT-11 there was OS/8.
That old foam rubber is the tinkerer’s bane. Nasty gunk that’s everywhere in old electronics, and is usually a gooey mess by now.
Ugh, yes. And how do I remove it from plastic without damaging the plastic?
I have a machine that came in a large hard shell case that was “padded” in the stuff. It disintegrated, spread everywhere, then all those chunks and particles melted in place.
Naphtha (lighter fluid) VMP is best. Low volatility and safe on most plastics, test first. Scrape and suck close up with a vacuum snout and get the thicker turds off first.
It’s funny how a design to eliminate metal contacts and make a keyboard last and be reliable, instead fell prey to timed life substances. Then there is the other non foam substances with poly in their names, often with rubbery or elastic uses. They tend to go messy-mono.
Some possible flammability risks with vacuuming lighter fluid..
Didn’t MS-DOS start as some sort of clone or copy of CP/M or something like that? I thought I remember reading so. If so then I would think that is a pretty important part of the answer to the question “What is CP/M, I hear you ask?”!
Ok.. so I Googled it and came back. It looks like copying was a controversial claim and probably not true. http://www.wired.com/2012/08/ms-dos-examined-for-thef/. But.. it did copy CP/M’s API.
Also, looking at a list of CP/M’s commands I see DIR, REN and TYPE which are also DOS commands that still exist today. There was also a CONFIG command which sounds familiar to me, I think DOS used to have that at one time. Also drives were handled using drive letters followed by colons. That might not sound like much but I only see 8 commands listed for CP/M! Including CONFIG that’s 50% that made it into DOS!
I’m not necessarily saying there was any wrongdoing involved in the creation of MS-DOS regarding CP/M but it looks to me like CP/M is as much a part of DOS’s history as say Unix is to Linux.
Well yeah, but most operating systems need a rename command, a command to list directories, etc. If you’re gonna give them 8-letter or less names, it makes sense to call them REN and DIR. Doesn’t mean the code or file structure was copied, ie the hard part.
Doesn’t mean it wasn’t copied either, of course.
Look at the programming API, not the user commands. The similarities are much deeper. The call to print a string uses ‘$’ as the string terminator. I can’t imagine two people independently having the same bad idea.
DOS started out as QDOS and was originally made by Seattle Computer Products in 1980. They needed something to run on their S-100 8086 board. At the time there were very few computers using the 8086/8088, it was mostly being used in traffic lights and other industrial equipment.
The IBM PC was also going to originally use a Z80, but Zilog was owned by Exxon at the time and IBM was unwilling to use an Exxon product in their computers. So they went with the 8088, expecting they could get CP/M for it. It didn’t work out and Microsoft jumped in and wrangled a deal with SCP to purchase QDOS and then licensed it to IBM.
Here’s another article about DOS vs. CP/M…
http://www.bloomberg.com/news/articles/2004-10-24/the-man-who-could-have-been-bill-gates
pirates of silicon valley all over again
maybe also mention http://hackaday.com/2015/09/26/foam-and-antistatic-bag-bring-vintage-compaq-keyboard-back-to-life/
Does this happen to every TRS-80? Looks like I might have some repair work ahead of me :(