Yeah… I still remember the big square heat-sink with the round regulator can sitting on it, and all the electrolytic cap’s taking up too much space on S-100 boards. Still… *shrug*

The biggest issue with S-100 was the form factor, closely followed by the heat released by the then-extant linear regulators (there were no small-form-factor switching regulators at the time and remember, S-100 sent AC over the bus, any regulation down to DC was required to be done on the board). I actually used a Zenith Z-100 for a time in the early 80’s that was based around the S-100 bus but in an all-in-one case similar to the TRS-80 Model III, the dadburned thing was enormous compared to the IBM PC. Because it was based on an S-100 you could toss a Z-80 CPU card into the beast and run CP/M as well as PC-DOS. Not that we did so. Frankly, mostly we just ran Kermit on the damned things to talk to our mainframes.

Regarding Apple de-coupling hardware and software, one of the things they sell is that their stuff just works. There’s no fiddling with device drivers and crap like that when you’re trying to install it on a piece of hardware, it just works. There’s folks who’ve done “Hackintoshes”, added drivers to the installer disk and distribution to install it on non-Apple hardware, and you can make it work but it’s as frustrating as using Windows because nothing works *perfectly*, the way it works on Apple’s own hardware. For example, if you close the laptop lid, it may or may not go to sleep, whereas an Apple laptop *always* goes to sleep when you close the lid, and *always* wakes up when you open the lid. Installed on any arbitrary hardware, it gets as arbitrary as with Linux and Windows — on both, I’ve had trouble with things coming back up after I’ve done a sleep-wakeup cycle, on both probably 30% of the time some random driver has some random bug and decides not to wake up some random piece of hardware correctly and you end up having to reboot anyhow. Besides, selling hardware and OS together is just so *profitable* for Apple. I mean, they’re selling a $400 laptop for $995 just by slapping the Apple name and installing OS/X on it. That kind of gross margin just can’t be had by selling software alone…

OTOH, there have been a number of times when I would have killed to have that ability, having lost days worth of calculations because of a failure at a bad time.

As for “troubled and a tad paranoid”. you would pretty much have to be to have gone to the trouble of creating the Tandem. Would any sane person assume all of the possible scenarios that a Tandem could survive? He had the computer equivalent of wearing a belt and braces with a flight suit.

If you have so much ice you could just dump it in a bowl and have a fan blow over it. I did that in California for Santa Ana winds, but it wasn’t a big flat.

Just finished watching the local news. The weather report was crazy! They said that the reported temp (43-46C) were in the shade. They took a temp monitor into the sun and left it a few minutes. It was reading 52.3C! The now say that this is the first time ever we have had more than 2 days in a row over 40C, and the last time we had 2 days was 1875 (from memory).

There is a story about Tandem very few people know. 🙂 About their very rapid rise, and their very sudden fall. 😉 It’s… very interesting. I could give a hint maybe… There was a SciFi movie called ‘Colossus: The Forbin Project’ If you watch that, and you know some of the principles behind the tandem philosophy and design… you might get the idea. The Tech adviser for that movie was the genius that designed the original Tandem OS (which is a name that is never, ever to be uttered!) called ‘Guardian’. Curiously, this was the name of one of the supercomputers in the movie. 😉 It was soon discovered that the fellow was a little… troubled and a tad paranoid. 🙂

It was kinda freaky in the 80’s doing a demo on a Nonstop II and later the Cyclone. 🙂 A common demo was to have a system doing some big DB transactions and streaming data (usually a very long list of some sort) on a multi-CPU system. Each CPU and memory card had a row of LED’s at the edge like a bar graph to show the current capacity (usage) of that card. Whoever was running the demo would have a client or prospect type some kind of transaction command into a terminal which would stream data from a large DB, then the guy would open one of those black glass cabinets Tandem used, and grab the first CPU card running at max (all LED’s lit) and pull it out! The display would slow for a second and then pick up and you would see the LED’s on the other cards pick up o the slack. Then he might pull another CPU or a RAM card just to show off! LOL Everything in a Nonstop had at least two data path’s and multiple redundancy. It was beautiful from an Engineers perspective. 🙂

One day… I may just write that book! 😉 LOL

S-100 was just too logical when faced with the creative chaos of the IBM PC. I have never understood why the bad ideas all become the standards, and why none of the real pioneers never seem to profit from their inventions and concepts. “Build a better mousetrap” and either the concept will be stolen by a major corporation, or the rights will be bought for a pittance and the design will never go to market.

You could nearly heat a house with a 21-inch crt which is a good idea at 34° C. We, OTOH, didn’t hit 35° the whole of last summer, which was weird.

I well remember a few things that “seemed like a good idea at the time”, but you can’t change them, so you move on.

Mix up some cookie dough and drop it on top of the system unit before you power up. At least you get something for the heat.

They created a JV company with Panasonic called Panafacom and made PC’s. They released a dual-CPU 8086 based system that was awesome! But, they refused to license DOS and only used CP/M-86, and then DRI/ICL/Panafacom created MP/M for it and some ICL PC’s.

My company was actually blacklisted by IBM. We had converted some major IBM sites to Facom. 🙂 Heh… I loved it! 😀

Then, for certain reasons, I enlisted and ‘went off to war’. When I returned, and my sanity returned, my company was dead and I was offered a job in Spooksville.

I’ve made several mistakes… They seemed like a good idea at the time. 😉

In the early 90’s, I was maintaining the Melb POP for an ISP that started in ’88 with Telnet and UUCP access. We started with a pair of 64k leased lines, and gradually increased to 8. When there wasn’t much traffic, I used to enjoy the speed from my HP workstation. I actually became a Mosaic and NCSA developer, and then moved on to Netscape. I had a lot more fun in those days.

I had a KSR33 hooked up to a PDP 11/03, then an 11/78, and eventually a VAX 11/750 from ’78 to ’81. In ’79 I was given a Beehive CRT terminal (I forget the model now). I also used a Northstar Horizon. I loved that system! LOL Ran a CP/M variant (North Star DOS I think) and had 2 big Shugart FDD’s. The PSU was huge and heavy I remember. Hmmm… S100 bus too. I developed a couple cards for it. 🙂 We ended up with a Pr1me 250 Mini in ’80, and I had fun learning PrimeOS. The guy who developed their OS (Bill something or other) and other software left to start Apollo. I really enjoyed playing with a network of Apollo Workstations in the late 80’s (3000 and 4000 series, and eventually, a prototype 10000 monster server!) And of course, the DomainOS and Aegis OS’s they had. Sadly, they were taken over and destroyed by HP (As HP does with all their acquisitions. I also worked on Tandem Nonstop mainframes which HP inherited via DEC).

I became a SunOS Sys Admin. When we were forced to go Solaris, we decided to go with HP and I gained HP-UX accreditation (v6 I think). None of them were as good as Apollo’s Aegis or DomainOS IMHO. Anyway, we discovered that moving to HP wasn’t such a great idea after all. Oh well… 🙂

Ahhh yes… those good ol’ days! *sigh* 😉

It’s stinking hot (34C in my room here according to mo temp monitor here). I just turned on the PC to check email and do a bit of snooping. Having a high-end PC and a big 21″ Trinitron CRT in Summer sux! They generate a lot of heat. Winter is good though. 😉

AT&T was really annoying because they had the R&D as well as the facilities to kick everyone else’s butt and they wimped out. Bloody Xerox gave away the whole GUI world, and didn’t anything with what was created at PARC.

My younger brother spent the bulk of his working life in the mini, then micro world, so I got to see some amazing things that never made it to the market. There was software and hardware created that would have rivaled almost anything available today, but the “suits” didn’t get it. They locked themselves into the concept that they sold to other corporations, and just couldn’t conceive of a retail sale. That’s why Dell HP owns DEC and Compaq. Dell HP understood customers, while the old line guys spent too much time watching Wall Street Week. [senior moment there]

The new Linux versions are fine, and eventually that’s where people will end up, because Microsoft has lost the thread. Vista was a disaster, Microsoft has finally figured that out and is moving to it’s next iteration, but the GOPression we’re in will start draining even their cash reserves.

Apple could make a killing if it delinked its software from its hardware. They certainly have the slickest X GUI interface. I don’t expect them to do it, any more than I expect Microsoft to offer Office on Linux. I would do it in a heartbeat.

What most amused me about those Suns was Sun’s funky window manager, OLWM. That was the biggest piece of trash I had ever seen compared to real window managers such as on the Mac or Amiga.. I discovered that twm was still hidden away somewhere on the system, read the documentation on how to set it up, and promptly hacked my .xsession file to start up twm rather than olwm. twm did no more than olwm and was even uglier, but it at least was small and fast. Then came the day when Sun disowned SunOS and forced-marched everybody to Solaris, and everybody had to use CDE. CDE was even worse than OLWM. By that time Windows 95 was out, and I looked at Windows 95, then looked at CDE, and I was, like, “excuse me? Are these Unix workstation geeks smoking some really nasty hashish or somethin? It’s like they ignored everything we learned from the Mac, Amiga, and Windows!” But of course it was that whole NIH thing. The Mac, Amiga, and Windows were “toy” computers, not “real” computers like Unix workstations. So there was no need to look at what “toys” were doing. Assholes. No wonder Unix lost the desktop wars… crap, they never even fired a shot, they just pretended there was no war in the first place and then one day woke up astounded to find that Windows ruled the world.