Bringing Back the PDP8

Anne Thwacks writes " Andrew Grillet has decided that the Digial PDP8 - the first ever minicomputer, will rise from the dead. He is calling it the PDQ8. Sure others have done software emulations, and even hardware clones, but he is not just building a hardware clone, but trying to revive the whole idea of 12 bit computers!"

Comments

TRS80

[ClamClit]

...and I building a 3.5 GHz TRS80 with a GIG of RAM and 2 5 1/4" 80 GB floppys. its the cats ass

Re:TRS80

[ch-chuck]

DOn't laugh - IMSAI, which predates the TRS80, is soon to sell, for under $1000, a series two with 20Mhz Z8S180, 1 meg static memory, AND toggle switches and flashing lights! CP/M never ran so fast.

what for

[neotokyo]

while the geek factor may be high, what sort of 12-bit software is it going to run ? linux?

Re:what for

[mirko]

RT11, of course :-)

Re:what for

[Detritus]

Probably OS/8, which if I remember correctly, was a multi-user operating system. You can run a multi-user system on a PDP-8 with 32KW of core.

Re:what for

[rainwalker]

Now, I'm no expert on legacy computers, but a machine with 32KW of core sounds dangerous...suddenly my 80W Athlon looks puny!

Re:what for

[Waffle Iron]

while the geek factor may be high, what sort of 12-bit software is it going to run ? linux?

In the bottom of a box somewhere in my basement, I've still got the BASIC source code for the Star Trek game we used to run on our high school's PDP-8. For each player's turn, it printed out the map of the current galactic sector along with any Klingon ships on the line printer.

It's funny, I remember when we played that game we felt like we had godlike control over a mysterious and powerful machine. Now when I play computer games, I mostly feel like a twitching moron.

Re:what for

[dhogaza]

OS/8 was a single user system. I wrote a multi-user kernel that ran multiple copies of OS/8 customized to hook into device drivers supported by my kernel. It supported virtual paging using a hardware hack first suggested by Richard Lary (the author of OS/8). We ran four or five users on a 32KW 8/E with a couple of RK05 drives.

We never distributed it because paging performance without the hardware hack was very bad (every CDF instruction needed to be trapped and mapped in software) and the hardware hack was only developed for the 8/E and piggy-backed on one of the system boards destructively (i.e. once modified your 8/E wouldn't run without our hardware).

But it was used internally by the company I developed it for until about ten years ago.

Re:what for

[CoolVibe]

BSD might actually have ran on it. The 4.4BSD, maybe Reno or Tahoe. Heck, the 2.x BSD's might even be supported as well.

I could of course be wrong, but I really think these PDP's ran UNIX. RT11 seems unlikely, since that was a PDP11 OS. Either some V[something] or a BSD unix ran ot it, I am almost positive.

Please do correct me if I'm wrong of course. It's been a while, ya know :)

Re:what for

[red_dragon]

I remember when we played that game we felt like we had godlike control over a mysterious and powerful machine.

That must've been a nice feeling, I bet. In these days, the mysterious and powerful machine has god-like control over you!

Re:what for

[dhogaza]

Right, OS/8 gave you full control over the machine so you could write multi-user applications.

Re:what for

[dhogaza]

OK, you're corrected :)

The PDP-8 never ran anything remotely resembling Unix. The very first version of Unix ran on the PDP-1/7/9/15 18-bit family (a PDP-7 IIRC). The architecture of this family was similar in many respects to the PDP-8 and indeed preceeded the PDP-5/8 family. You can think of the PDP-8 as being scaled down from the earlier family's 18 bits to 12 bits. To make it cheaper, of course.

The original Unix-written-in-C ran on the PDP-11 (the PDP 1/7/9/15 family version was written in assembly, IIRC). The first BSD version of Unix was written for the VAX family ...

Re:what for

[JebusIsLord]

wasnt the original unix written in B?

An idea

[AppyPappy]

An idea whose time has passed. Maybe he can build an Edsel while he is at it.

Re:An idea

[b0r1s]

Maybe a '53 Corvette would be a more apt analogy in the car industry. But of course, no one would want one of those, since it's an idea whose time has passed.

A '53 vette can still drive on the roads, use normal gas from the pump, and hit speeds upwards of 100mph on the highway.

The PDP-8 simply isn't capable of competing with any current hardware; most Palms are more powerful.

Re:An idea

[Gordonjcp]

Neither is the microcontroller in your washing machine. Many PDP-8s were used for controlling complex machinery in a manner more flexible and customisable than that allowed by "traditional" relays and switches.

So, what's your point, caller?

Re:An idea

[AppyPappy]

Having been a programmer when a pdp8 was a hot item, I will say that I don't wish to return to those dark days. I've dropped enough punch cards and trimmed enough 9-track for a lifetime. And I don't want to return to calculating memory usage so I don't exceed the 48K allocated to me. I'm sure it is a good practice but I'm too old to do it now.

Give me a 300mhz with 2 gig of RAM and a wide-open prairie.

ENIAC

[kaden]

Anyone for hitting up the local vacuum repair shop and getting started on an ENIAC reconstruction project?

Re:ENIAC

[Ark42]

Funny..

I read this, then scrolled down to the bottom, and my quote of the moment said:
"Just think, with VLSI we can have 100 ENIACS on a chip!" -- Alan Perlis

Re:ENIAC

[vizualizr]

The vacuum repair shop?

no, that idea sucks.

Re:ENIAC

[fishbowl]

You're just kidding, but, I wonder if enough information about ENIAC
has survived, to make a simulator for it?

Re:ENIAC

[crawling_chaos]

The vacuum repair shop?

no, that idea sucks.

Bag it already...

Calling all Electrical/Computer Engineers

[ekrout]

Many of you probably have used Xilink's 1000, 2000, or 4000-series FPGA card during laboratories for your undergrad classes.

Well, if you'd like, you can follow this design of an FGPA implementation of the original PDP-8 computer!

If you've used Verilog (a hardware design programming language), like I have, you can even download all the code!

12 bits

[vasqzr]

I found that, where you are not primarily handling ASCII, 12 bits was a very good size.


Maybe someone would enlighten the rest of us on why a certain bit size is better than another, and why we currently use 8/16/32/64, instead of 12/24/48/96 ?

Re:12 bits

[FauxPasIII]

>> Maybe someone would enlighten the rest of us on why a certain bit size is better than another,
>> and why we currently use 8/16/32/64, instead of 12/24/48/96 ?

Because powers of 2 are easier to work with in binary.

Re:12 bits

[oliverthered]

8 bit is base 2

1 2 4 8 16 32 64 128 256 512

So I suppose 12 could be ....

1 3 6 12 24 .... hmmmmm..... now quite correct

maybe
1 2 3 6 12 24 ... 60 ... 420

anyway

12 is good because
12/6 = 2
12/4 = 3
12/3 = 4
12/2 = 6

Re:12 bits

[Viol8]

7 bits was used when the ASCII character set was initially devised (though 4 bit had been used before that) with the 8th bit being used for parity checks. At some point someone decided to hell with parity checks , lets use the 8th bit to double the number of characters available. And once 8 bit became standard in hardware then 16 bit narually followed as you could still divide a 16 bit word in 2 to get 8 bit characters. Similarly with 32 bit you can divide a 32 bit word to give two 16 bit words and hence can run 16 bit software on 32 bit hardware with a bit of help. Thats the gist of it anyway.

Re:12 bits

[sporty]

No.. not like that. 8 is a power of 2, 12 isn't :P

Re:12 bits

[sql*kitten]

Maybe someone would enlighten the rest of us on why a certain bit size is better than another, and why we currently use 8/16/32/64, instead of 12/24/48/96 ?

This article explains why base-3 systems are actually a lot better than base-2 from a theoretical perspective, but that it was much easier to design hardware in base-2, so base-2 became the de-facto standard. Nowadays we could probably fab base-3 hardware fairly easily, but it's not worth doing so with all the base-2 hardware already in existance.

As for 16/32/64 instead of 12/24/48, it's just one of those things. IBM's earlier AS/400s ran on 48-bit processors (now they are 64-bit). 96-bit floating point is an IEEE standard. And do you know why file permissions in Unix are rwxrwxrwx? It's because they borrowed that idea from another operating system designed for 9-bit bytes and a 36-bit processor.

Re:12 bits

[AlecC]

If you are using Ascii, the 8-bit bytes are a good size. Of course, you really only need 7 bits, but no programmer is going to like that. Then you add the concept of byte addressable word space, so you use the byte address to address doublebytes (short) and quadbytes(longs).

As to why 12 bits is good: simply 8 bits is a bit small for an awful lot of cases, so if you use an 8-bit fundamental word length, you are often into doubleword operations. Practically, a 12 bit word length, values +/2048 or 0-4095, seems to work out useful in a lot oc control applications. When the extra bits were expensive, the saving between 12 and 16 is relevant.

First machine I worked on was 18 bits - all 8K words of it. Enough to run a very simple Fortran compiler. From paper tape, naturally.

Re:12 bits

[lobsterGun]

This is a story that a wisened old timer told me at a convention (the guy had a whit bushy beard so you know it has to be true).

Back in the day there were two schools of thought: the 8 bit byte and the 9 bit byte. The 9 bit byte school represented the same data as the 8 bit byte school (0 to ff), and used the extra bit for parity.

8 bit bytes led to 16 bit words etc...
9 bit bytes led to 18 bit words etc...

My reaction to the old time was to ask WTH would anyone do this?

His response was to go into a discussion of 36 bit computing. "Ya see son. When you have 32 bits you can divide those bits evenly 6 ways:
32 single bits,
16 2-bit groups,
8 4-bit groups,
4 8-bit groups,
2 16-bit groups,
1 32-bit group.

With a 36 bit system you can divide those bits evenly 9 ways
36 single bits
18 2-bit groups
12 3-bit groups
9 4-bit groups
6 6-bit groups
4 9-bit groups
3 12-bit groups
2 18-bit groups
1 36 bit group

using 36 bits give you more flexibility in addressing."

He went on to tell be stories about how they explioted the advantages of 36 bit computing back when he had worked at Compuserve and how sad he was that 36 bit systems had died.

It could very well be that 3 12-bit bytes are used to make up a 36 bit word... ...or it could be that that old timer was full of crap or off his meds or something. The world may never know (Much like the number of licks it takes to get to the center of a tootsie pop).

Re:12 bits

[AJWM]

why we currently use 8/16/32/64, instead of 12/24/48/96

Because we currently use 8/16-bit character sets. Blame it on ASCII and EBCDIC (and UNICODE). (Okay, plain ASCII is 7-bit, but in communications there was a parity bit too.)

A lot of older machines use 6-bit character sets, that's all you need if you ignore case. There's the 12-bit PDP-8, the 36-bit DEC-10 and -20, the 48-bit Burroughs machines (which could handle both 6-bit and 8-bit characters), and the 60-bit Control Data machines. The PDP-8 was superceded by the 16-bit PDP-11, the DEC-10s were dropped (superceded by 32-bit VAX), the 48-bit Burroughs machines may well still be 48-bit (Unisys) for all I know, and the 60-bit Cyber series evolved to 64-bits sometime in the 1980s.

Re:12 bits

[DaveAtFraud]

...or even the 54-bits (56?) of the CDC computers...

60 bit words actually but don't ask me why. At the time I was doing number crunching and algorithm design so, unfortunately, I didn't get down to the hardware level and find out why they used a 60 bit word.

The other interesting hardware architecture was Data General with their 36 bit words.

Re:12 bits

[MioceneMan]

12 bits makes four octal numerals.

In decimal you might type for example year 2002. In Octal that would type year 3722, in Hexadecimal 7D2.

Some of us used to quite commonly enter utterly raw machine code or data on a ten key pad. It is very easy, if you have good ten-key skills, to enter 12 bit "words" in octal.

If you have a minute, look at some 8080 family machine language in octal. You will notice the basic 8080 instruction format is centered on 3 bit fields!

Heh, slashdot as always...

Re:12 bits

[Luke]

Yep this person is right on, since 12 is evenly divisible by many numbers.

For those on the site who are musically inclined, note how often 12 and its divisors are used in music.

Re:12 bits

[pjrc]

Before ASCII in the mid-60's, computers represented characters using 6 bits. 12 bit systems were common, since 64 characters are enough to represent A-Z, 0-9, and a handful of symbols and control codes, and a 12 bit number gives enough range for a lot of types of computation.

Here's a really well written history by Tom Jennings of the early character codes and the two ASCII standards in the 60's. AT&T apparantly forced the world to use ASCII. ASCII required 7 bits, which was a huge departure from the previous 6 bit (and 5 bit, 2 set) systems.

Tom's history doesn't mention the real reason we all have 8/16/32/64 bit machines today: IBM's Extended Binary-Coded-Decimal Interchange Code (EBCDIC) encoding, which spec'd 8 bits per character (how he could leave out EBCDIC is beyond me?) IBM forced the world to 8-bits to represent characters inside the computer, and AT&T forced the world to communicate them with 7 bits. ANSI recommented in the 1967 ASCII spec (X3.4... the one that added lowercase) to use the eigth bit for parity. ASCII was more focused on communication than storage of data.

The thing I find interesting is that the old computer and telcom industry monopolies forced these fundamental architecture decisions on everyone, both discarding lots of established practice with little regard for backwards compatibility. But the modern computer and telcom monopolies have more or less adopted and adapted from existing technologies and have largely failed to push their own "standards" that discard backwards compatability, at least when it comes to how data is stored and communicated.

Re:12 bits

[Alien Being]

"12 bits makes four octal numerals."

The toggle switches on the front panel were arranged in sets of three. The color of the switches alternated between sets.

On our high school's PDP8m, you had to toggle in 5 or 6 4-digit numbers to xfer the boot loader from disk to core and set the program counter. Then you'd start the clock and the console teletype would come to life in standalone mode. From there, we would start ETOS (Educational Timesharing Operating System). ETOS was multitasking, and multiuser. IIRC, it required a special board that handled the context switches.

Perhaps the coolest thing about ETOS is that it had a facility similar to VMware. Each user had access to his own virtual PDP8.

Re:12 bits

[sql*kitten]

Very interesting, and thought provoking. While we're at it, why qwerty and not Dvorak?

QWERTY keyboards were designed for mechanical typewriters - the layout of the keys is designed to minimize the chance of two keys clashing, because on a mechanical typewriter, both keys will get stuck. In the old days, a computer keyboard was an electric typewriter (a tty), which had inherited its design from mechanical typewriters because they were already entrenched.

Despite it being inefficient, and actually designed to slow typists, we're stuck with it - probably until the keyboard itself is superseded as an input device.

Re:12 bits

[Ed Avis]

Three-state or base-3 computing systems have nothing to do with it. 12-bit machines use 12 *binary* digits. They are still base-2 hardware.

Re:12 bits

[Ed Avis]

Yeah, as if it makes a difference whether your word size is a power of two or not. A 32-bit word is no 'easier to work with' than a 42-bit or 27-bit word. Lots of things in computing are most easily handled with powers of two, but bits-per-word is not one of them - either for hardware or software reasons.

Re:12 bits

[KewlPC]

Originally the yard was based on the length from the tip of the king's nose to the end of his index finger. Nobody actually measures things that way anymore, though, because we have an official yard length which doesn't change (it's locked away at NIST, I believe). No one is going to cheat you (although if you suspect that a store is using incorrect measurements, whether on purpose or not, give your state's Department of Weights and Measures a call).

But the fact that the yard is close to the length between the tip of the nose and the end of the index finger for most people makes it easier in situations where you don't have any other way to measure, such as your yardstick getting lost.

Re:12 bits

[spitzak]

If the length is a power of two, than a "bit number" (a shift value, or an argument to a bit-test instruction, or anything else that identifies a bit by a number) will fit into a set of bits with no waste.

This DOES make a difference. More important than the space savings is the ability to know that a pattern of bits could not encode any illegal instructions that had to be tested for.

Re:12 bits

[oliverthered]

Well around 10years ago, 'when I were a lad' they used to sell cloth by the nose to index finger yard in markets. (they'd leave a bit of sag in the fabrick so as not the under measure)

Re:12 bits

[KewlPC]

Really? I used to get dragged out to a lot of fabric stores with my mother when I was a little kid, and they all used either a yardstick, measuring tape, or one of those big white measuring boards.

This is great news :)

[solostring]

I fondly remember the PDP8. My father had one installed in the garage when I was a kid, and had my first experience of programming on it when I was 8 or so.

On the subject of PDP8's, I was surprised to hear that they were used in communications in Hong Kong up until at least 1999 for a number of financial institutions. I worked with an old computer technician who earned a fortune maintaining these beasts. I wonder if they are still being used in HK after the Chinese reclaimation?.....

Re:This is great news :)

[sczimme]

I was surprised to hear that they were used in communications in Hong Kong up until at least 1999
...
I wonder if they are still being used in HK after the Chinese reclaimation?

Possession of Hong Kong went back to mainland China in 1997, so I would say yes, they were in use for at least two years after the changeover. (I don't know if they are still in use, though.)

Slashdoted already

[Tassach]

Apparantly, the PDQ8 doesn't withstand the /. effect very well. And of course, obBeowulf.

Why?

[BWJones]

but he is not just building a hardware clone, but trying to revive the whole idea of 12 bit computers!"

So, after reading the article, I am still trying to figure out....Why revive the idea of 12 bit computers? Other than nostalgia (which is why people still drive Studebakers, old Ferraris and old Porsche's I suppose), what is the point?

Re:Why?

[Kibo]

I can't be sure about Studebakers but in the case of classic Ferraris and Porsche's I'm fairly certain the idea is still to get laid.

Re:Why?

[WinterSolstice]

Having owned a classic (1973) Porsche 911t Targa, and currently owning a 1975 Corvette Stingray T-top, I can tell the idea is somewhere between getting laid, and the sheer fun of a classic performance car.

Some people, (myself obviously included) think that the pre-gas shortage cars in the US were much better looking cars. The current models seem to still be recovering from the 80's and 90's ideas of "Box on Wheels" and "Wind Tunnel Styling". I wouldn't own a new Porsche.

I think the idea with the PDP however, is not nostalgia. It was (and is) a very stable platform. Many companies either just left them, or are still using them. We just left the VAX platform not too long ago, despite having used Alphas concurrently for ages.

After all, if it works, why upgrade?

-WS

Re:Why?

[rgmoore]

which is why people still drive Studebakers, old Ferraris and old Porsche's I suppose

People care about even stranger things than that. Just yesterday, I saw a group of Edsel fans driving their cars in the Doodah Parade. When I thought that was goofy enough, the people standing next to me pulled out their pictures from a trip to a Corvair fan get together (The Great Western Fan Belt Toss & Swap Meet). ISTR that there are even Trabant fan clubs, of all things. Name a piece of obsolete technology, and there's a good chance that some people will be fanatically devoted to it.

Re:Why?

[image]

> I can't be sure about Studebakers but in the case of classic Ferraris and Porsche's I'm fairly certain the idea is still to get laid.

I'm rather sure it is easier to get laid in a vintage Studebaker than a Ferrari.

More leg room and all.

replica computers ... its logical innit?

[trash eighty]

people have been building replicas of old cars, boats and aeroplanes for years. i suppose its logical that people would start building replicas of old computers sooner or later. much computer hardware is boring these days, generic MBs and computers. ahh the good old days when we had some variety ;)

all power to him!

Re:replica computers ... its logical innit?

[plierhead]

Yeah but I can't help feeling that old cars, boats and aeroplanes are very very sexy and likely of interest to most of the inhabitants of the planet, and, yeah, to be honest might well help get you laid...whereas consider whether you could muster quite the same charisma offering to show someone over your collection of rebuilt dental hygiene equipment (for example)...

Interesting

[GeckoFood]

It would be fun to play around with something cool like that, just for the sheer ability to say "Hey, y'all watch this!" (Oops, better watch that there accent, ya rekcon?) It would especially nice to have a C compiler or something to develop apps for it, again just for the coolness factor.

With a twelve-bit computer, what is the address space, anyway? Something like 2048 words? Suprisingly, you can actually do a lot with that if you code it tightly. No, can't do weather map rendering too well or anything like that, but I bet you could pull off a stripped down version of NetHack or something...

Coolness, regardless. :-)

Re:Interesting

[johnalex]

Bad omen. Here in the South, "hey y'all, watch this!" are often the last words you hear.

Just how big *were* these things?

[ekrout]

Well, I found this old link in my bookmark manager. It details the history of EMS (sound studio) in vivid detail, including a listing of all their original equipment.

The interesting part is that they posted high resolution images of their setup, which includes PDP-8 microcomputers!

The image: http://www.ems-synthi.demon.co.uk/studiopz.gif
The PDP-8s:
Left side - Teletype for PDP8
Left bay - PDP8/L Computer ("Leo") 4K x 12 bits (=6K bytes) 1.3 s cycle (0.77MHz), 32K Hard Disk Store
Center left bay - PDP8/S Computer

Re:Just how big *were* these things?

[dhogaza]

The PDP-8/S was the serial version of the PDP-8, and the model I learned programming on in high school (we couldn't afford one by ourselves, it was owned by a consortium and travelled between about a half-dozen schools).

The PDP-8/S took 36 microseconds to add two words. When I first got to play with a real, parallel PDP-8 (3 microsecond add time) I was in heaven. Man, that puppy seemed FAST!

12 bit is best for the US patriot

[MosesJones]

12 bit is much better for patriotic Americans.

Think on it, power of two is a far to simplistic and dare I say it European system for the patriotic American. In Europe they use metres, kilometers, grams and kilgrams. All this regulation of structure around a number like 10 is typical of Europeans. Americans use sensible systems like 14 pounds (abbreviated sensibly to lbs as pounds clearly contains the letter l) to a Stone and 16 Ounces (again with a sensible abbreviation of oz) to a pound. Who needs these ridiculous regimented European systems that dicate that everything must follow a sensible pattern?

Patriotic Americans arise. 12 bits to a byte, 7 bytes to a word, 13 words to a sentence and 1764 bits to a chain.

Re:12 bit is best for the US patriot

[marsbarboy]

no, 12 bit is best for the BRITISH patriot. Bring back the good old imperial measurements...It's only the EU making us use the french system.
Why do you think the americans use imperial?
Because we used it first.

Re:12 bit is best for the US patriot

[david duncan scott]

You know, I have never, in 43 years, heard an American use "stone". That's a Brit thing.

And of course we inherited the whole system. As I recall, the "lb." abbreviation has something to do with the French "livre", and also led to the the "pound sterling" symbol, that fancy-schmancy "L" that featured so prominently on the Commodore keyboards of yore.

As for word width, well, there's nothing especially holy about multiples of 8. CDC used to make machines with a 60-bit word, because they mostly dealt with numbers, not text manipulation, and big fat words like that allow for big fat numbers, although storing an ASCII file in 60 bit words would be clumsy as hell (As a side note, I used to work with the CP1600, which was a real 16-bit machine. There was no way to address a byte, although there was an 8-bit shift so that you could pack ASCII into words to save space and slow down annoyingly fast programs.)

Re:12 bit is best for the US patriot

[sql*kitten]

And of course we inherited the whole system. As I recall, the "lb." abbreviation has something to do with the French "livre", and also led to the the "pound sterling" symbol

Which is interesting because the word Sterling comes from starling, which meant "small star" in mediaeval English - it was the symbol on the coin for the unit of currency. So the currency symbol should probably really be a *.

Lb is from "livre" (French for pound) and dollar comes from "taler", an old German currency.

Re:12 bit is best for the US patriot

[AlecC]

Absolutely. 12 pennies to the shilling, old bean. One of the early British computers (?Leo?) used BCD arithmetic, but had a hardware switch so that the ls digit could be used in base 12 instead of base 10 for making the pennies into shillings.

Re:12 bit is best for the US patriot

[JimPooley]

The LEO (Lyons Electronic Office) was the first ever computer designed to be used by business, and was developed by Lyons Bakery.
There's a site about it here.

Re:12 bit is best for the US patriot

[Waab]

I must respectfully disagree.

While a measurement system based on powers of 2 seems fairly straightforward and intuitive to your average techno-nerd, it is still incomprehensible to the average Patriotic American. In fact, I would venture to say that 1024 bytes in a kilobyte is every bit as opaque as our beloved, patriotic, English units.

Re:12 bit is best for the US patriot

[Anonymous Coward]

Actually, I must defend the American/British/Biblical system for a second. Granted, the metric system makes a lot of sense for modern science. But for simple math, using numbers like 12, 16, 60, etc. makes a lot of sense because they have a ton of whole number factors. For instance, if I want to divide 12 into 2, 3, or 4 different parts it's a lot easier than with 10. With modern computing devices, most people forget the ingenuity of this system when criticizing it.

Re:12 bit is best for the US patriot

[AJWM]

although storing an ASCII file in 60 bit words would be clumsy as hell

No kidding. In fact mostly they didn't. Even into the early 1980s, CDC Cyber machines stored (and maniuplated) text in one of two formats: 6-bit character, or 6/12-bit character, the latter being 6-bit plus an escape character for lower case. When they brought out the next series of machines the hardware could run in either 60-bit or 64-bit mode, and they introduced an 8-in-12 character set which was essentially 8-bit ASCII stored in 12 bits, wasting 4, until the OS was converted to 64-bit.

Doing text processing/formatting was a royal pain on that system, so I ended up writing a clone of Waterloo Script (vaguely like 'roff) in Pascal (named "Formal") that internally treated everything as 7-bit ASCII and translated on input and output. Proved pretty popular, a number of other CDC sites bought it.

(Side note -- the I/O on those 60-bit CDC machines was handled by 12-bit "PPU"s (peripheral processor units) to offload the CPU.)

Re:12 bit is best for the US patriot

[Derwen]

Americans use sensible systems like 14 pounds (abbreviated sensibly to lbs as pounds clearly contains the letter l)

the "Lb" is short for Libra (Latin for pound), the same reason that the pound sign for British currency is a fancy L (crossed with one or two lines).
- Derwen

Re:12 bit is best for the US patriot

[Derwen]

no, 12 bit is best for the BRITISH patriot. Bring back the good old imperial measurements...It's only the EU making us use the french system.

Napoleon introduced the metric system to Europe to replace the French imperial system. Nevertheless you will still find fruit sold by the pound in parts of rural France. :-)
- Derwen

Re:12 bit is best for the US patriot

[CormacJ]

You know, I have never, in 43 years, heard an American use "stone". That's a Brit thing.

I once used that with my wifes family, and had to explain to them what 1 stone equals 14lbs. Later that same day I told we'd be back again in a fortnight. That really confused them. They'd never heard someone describing 2 weeks as a fortnight.

That day I came to realise the truth of the quote that described the UK and the US as two countries seperated by a common language.

Re:12 bit is best for the US patriot

[david duncan scott]

Now "fortnight" I know, but only because I read a bunch of Enid Blyton novels as a child (long story, but they were around the house and I'd read just about anything.) Of course, "fortnight" kind of makes sense, in the same way the "bedlam" does, whereas "stone" seems to be as arbitrary as any other unit, and therefore opaque to reason.

Hey, sounds like you could answer a question for me: I sometimes watch those "World's Worst Drivers" shows (you know, when I'm feeling too smart and need to drain off a few IQ points.) Aside from driving on the wrong side all the time, which would qualify all of the UK for the show if you ask me, one other thing puzzles me. I see occasional jagged lane dividers, sort of saw-toothed or like lightning bolts. I'm guessing they don't mark the electric vehicle lane, but what do they mean?

What Gives?

[SuperDuG]

Everyone is waiting around for 64-bit chips to hit the home computer market at resonable prices, is this some kind of protest "bits, we don't need no stinkin bits".

On a related note, I'm going to be designing my own 32 bit system. It's going to be pretty cool having an asthetically pleasing case, and run most all of the common software out there, but make the operating system run on top of bsd. Then I'll make really high-end systems, and education type systems, and laptops.

Now I'm 95% of the way done with this whole project so I've hired an advertising firm to come up with some commercials. I figure I'll show joe average sixpack switching from the normal x86 windows machine, to my machine, I'll call them 'Switch-Ads'.

My proprietary systems will never run on anything else, and you will be forced^H^H^H^H^H^H encouraged to only buy via our website.

I'll call them MOC's ... and the company will be named Orange.

What for?

[archeopterix]

8 bits should be enough for everyone.

Good genes!

[ekrout]

From: http://www.grillet.co.uk/biog.html

"My mother was a Fortran programmer using computers that looked like this [picture of an ancient IBM 608-series supercomputer]"

Text based games

[Technician]

I remember playing the father of the Zork series of games on a PDP 11/35 which was the newer 16 bit machine. The game was called Adventure. We got the game on a 5 Meg 14 inch hard drive (RK05) and wasted a bunch of paper making moves on a keyboard/printer terminal. The scrollback feature of the hardcopy was great for finding your way out of a maze again. Adventure has since been ported to CPM and DOS. The game is still a great game and will challenge the thought process. Take a pencil and paper to keep from getting lost. There is no map. Do a google search to find this true classic game. You should be able to run it in a DOS window on Windows 95 before DOS and Windows 95 expire at the end of this year. I'm still trying to figure out who the shadowy figure is who tries to get my attention.

Re:Text based games

[Arandir]

It hasn't died. The C translation by Jim Gillogly is available in the FreeBSD source tree. The translation from Fortran to C was pretty basic. I've never seen so many goto statements since the days of BASIC.

a PDP8 was my home machine in 1976

[MarkWatson]

But, it was just an Intersil Intercept Jr. single board version of a PDP - same instruction set though.

It was fairly easy to program for - I wrote a simple cross-assembler on a Dec-10 that would print out my assembler source with machine code (in octal). For short programs, it was fairly quick to enter the programs in octal. Since the Intercept Jr. was all CMOS, the programs would stay in memory as long as I wanted without runnng down the batteries.

Really, it was very cool, and fun.

-Mark

What a memory...

[NormAtHome]

We had one of these in high school (25 years ago), looking back I thought it was pretty cool but man that single DEC tape drive was a royal pain. We also had an old teletype with a paper tape punch, haven't seen one of those since. In my junior year we got a grant from the state and got a PDP-11 with dual 8 inch single sided floppies, now that was living!

Why cling to the past?

[reitoei1971]

I just dont see the point of resurrecting 12-bit hardware when we have 64-bit now. Why the nostaglia? PDP is dead! the world will never see another 12-bit. Is this guy a museum curator?

Address Space

[Detritus]

The PDP-8 had a 4K (4096x12) address space.

Somewhere I have an old DEC PDP-8 handbook. They released a native FORTRAN compiler for the PDP-8. It just shows what you can do with clever coding and lots of overlays.

Didn't these things have selectable word sizes?

[GreyPoopon]

I remember seeing one of these puppies when I was in high school. It was no longer in service, but still had most of its internal components. In fact, I still have one of the "flip chips" that I took from inside it. Anyway, I believe I recall a dial on the front of the main panel that allowed you to choose the word size. I can't remember whether 12 bits was the minimum or the maximum, though. Can anybody verify that this existed, or am I just remembering some strange dream?

Re:Didn't these things have selectable word sizes?

[HotNeedleOfInquiry]

"or am I just remembering some strange dream? "

It's a strange dream. The only 8's that had a knob on the front were PDP 8/e, 8/f, and 8/m and they all shared basically the same front panel design. The knob selected the register that would display on the front panel. It had no effect at all on the operational mode of the machine.

Re:Didn't these things have selectable word sizes?

[dhogaza]

Speaking of front panels ... the PDP-8's front panel was made of glass and the design was silk-screened on it by hand. The cabinet was a basic rack with plywood siding ... though they originally offered a transparent plastic case, too - cool case mod for the mid-1960s, eh? :)

I still have the glass front panel from the PDP-8 that I worked on back in high school ... I snatched it when it was scrapped. We owned a five-cabinet beauty, 12KW core, four DECtapes, a high-speed paper tape reader, an original Centronics dot-matrix 120 CPS printer, an RF08 head-per-track system disk (full cabinet of its own) and the first Tektronix storage-tube monitor with joystick that was four feet long and weighed a lot ...

Re:Didn't these things have selectable word sizes?

[Gordonjcp]

Many years ago, before I got kicked out of Uni for never turning up because I hung about in the computer rooms all day, there was a comms rack with a big slide switch on the front. A paper label had been stuck above it, covering the original markings, labelled "P-R-N-D-L"...

Only 12 bits?

[r_j_prahad]

How can they possibly expect us to use the efficient new Radix-50 character encoding to store text? RAD-50 requires 16 bits to compact characters.

DeCastro was right, this 12 bit nonsense will never go anywhere.

Resurecting old hardware designs

[randomErr]

Some mentioned earlier on this thread jokingly about making a super fast computer based on old architecture.

Has anyone actually done that? Has anyone actually taken say, a Tandy Color Computer 3's hardware and boosted it up to something approaching our current standards? I'm not talking emulation on a x86 platform. I mean fully working with a processor with a native OS.

Those architecture are so simple, with kernels so small you could print the hex binary out on a couple of pages. Imaging how fast an accounting package would be on a 1 gHz, or even a 200 mHz.

I know this maybe off topic, but if someone could resurrect a 12-bit system to a more modern standard, why not other system. DOS is still viable in certain circumstances, why not these platforms.

Think about an 8-bit controller with a serial connection, flash memory, and a RCA video out jack that is based on a C64. There is a TON of documentation for programming on something like that. Linux guruâ(TM)s could use C/C++ and Windows users could use Commodore BASIC.

Oh well thatâ(TM)s just my ramblings.

Re:Resurecting old hardware designs

[Richard_at_work]

I used to work for a UK supermarket in the Cash Office, and the PC that ran all the Checkouts and the SSM system (stock management) was a lowley 486-66 pc with 48mb ram, and it crawled. It took 4 hours to do a standard end-of-week data compile. It ran some wierd IBM Dos OS, and used green screens etc.

Then one day they replaced it with a PIII700 system, 128mb ram the works. Still ran this IBM Dos OS, but the end of week took seconds. Literally, we could no longer get away with a 4 hour break!

The same supermarket is still using this Dos OS on p4s now, so i wonder what the speed is like now.

Re:Resurecting old hardware designs

[pommiekiwifruit]

Yeah but what's wrong with using ye olde apostrophe from ASCII? Will he be using the full-width japanese ROMAJI characters for normal text next? g...(TM)C--(TM) Z"!

Much as I love utf8, I can't see in the source code where slashdot specifies that encoding.

Not Bad for its day

[Anonymous Coward]

The PDP-8 wasn't bad for it's day. They even had a Time Sharing System that ran on it that gave each user a whopping 4 K of space do do whatever they wanted to do with it. It supported up to 16 simultaneous users. I know, I'm old enough to have gone to a college with one of those beasts as a shiny new grant from DEC. We had 2 high schools and one private school hoked up to us by the old 110 Baud ASR-33 teletypes. It was a hoot trying to make anything run on it. Assembler was about your only choice as Basic didn't have any file I/O until about 1973 - 1974.

Why bring it back? Why not? It may not ever be used for much, however who says all the cool computers have to be uber-machines? This next comment isn't meant to start a flame war, but I'd like to see some of today's bloatware folks try and make a program of any substance work on one of those puppies. I've seen some code from folks used to huge addressable and virtual memories and YIKES !

Because, as Tom Lehrer, formerly of MIT noted. . .

[kfg]

"Base 8 is just like base 10 really. . .if you're missing 2 fingers."

KFG

Six copies of Windows

[Anonymous Coward]

What else would be a 12-bit OS?

XYZZY

[MosesJones]

And of course

"with your bare hands?"
yes
you stand amazed as the dragon lies dead at your feet.

Bugger graphics you can't beat a maze of twisty passages, all different.... or was it a twisty maze of different passages.

Re:XYZZY

[serutan]

It was, "a maze of twisty passages, all alike."

Character Codes

[Detritus]

6-bit character codes were popular before the introduction of ASCII (7-bit) and EBCDIC (8-bit). The PDP-10 had a 36-bit word, which could hold 6 6-bit characters. It also supported other character codes, such as ASCII.

If you look at old assemblers and compilers, the limit on the length of a symbol/variable name is often the number of characters that could be squeezed into a single machine word.

Re:Character Codes

[dhogaza]

And the CDC 3000 series had 24 bits (4 6-bit characters) and the 6000 series had 60 (10 6-bit characters). CDC peripherals were sixbit, not 7-bit Ascii + parity.

Of course the PDP-8 came with a TTY 33 or KSR 35??? whatever the hell the heavier-duty TWX teletype machine was called. These were 8-bit.

TWX machines used 7-bit encoding with a parity bit that later was the basis for the standard Ascii/Ansi/ISO character sets we're all familiar with. I don't know if the standard was approved before or after Western Union deployed TWX machines (replacing the earlier 5-bit encoded Telex machines I believe? I'm not quite old enough to know for sure, TWX was in by the time I was in high school).

Anyway these 7-bit + parity teletype machines originally made for use on the TWX network were pretty ideal for a low-cost computer made by what at the time was a very small computer company. Rather than develop their own terminals ala IBM or CDC, Digital just bought TTY 33s as the console.

Despite the use of 7-bit Ascii using teletypes machines as consoles on the PDP-8 (and the 18-bit and 36 bit families as well), it was common to just use sixbit for many things. For instance OS/8 used six bit characters for file names - upper case only, six character file names and a two-character extension. But editors and many other applications allowed upper and lower case characters and packed 3 8 bit characters into 2 12-bit words ...

PDP8? How about a PDP11???

[Spatch3]

You know, I work with PDP11's day in and day out. DEC's last PDP processor was, I believe, a PDP 11/93. It has a whopping 2 MB of RAM and 8 serial ports on the processor board! This was a double sized board only taking up 2 slots on the DEC backplane that took the function of 5 boards that took up 5 slots each. Of course the disk controller, clock card and other boards are separate. These boards are circa 1990. They are in a custom digital dictation system that can handle 64 simultaneous audio ports where people are either dictating or transcribing. The OS is the roll off your tounge: RSX11M+. These systems, which we are replacing slowly but surely, have been absolute work horses lasting for at least 10 years.

I jsut had to reboot one this morning... :)

Chris

Because, as Tom Lehrer, formerly of Harvard, noted

[Interrobang]

...well, you could just omit using your thumbs...but that wouldn't be as masochistic (tango-y goodness!)...

Tom L. has always been a Haavaadite, not from MIT, as in the "Harvard Fight Song," and the lines, "These are the only ones of which the news has come to Haavaad/And there may be many others, but they haven't been discovaad." (Rhotic-lossy dialects bother me, since I speak one of the few English dialects that's fully rhotic.) I imagine it matters to some people (probably they go to Harvard).

Also, as far as I know, he's still there, although long since an Emeritus.

PIC is a bit more practical

[Bruce Perens]

If you want to program old microcomputer architectures, learn the PIC microcontroller, which is based on some Control Data Co. Peripheral and Input Controller (I may have the "I" wrong in that acronym). It's available in 12, 14, and 16-bit flavors. It doesn't have much of a stack. And it has the virtue that, as a $1 microprocessor, it's still practical for many projects, while a PDP-8 is really an intellectual exercise at this point.

Bruce

Re:PIC is a bit more practical

[mesocyclone]

I still do a lot of programming for embedded applications in an older 8-bit computer architecture (Motorola 6800). They are still selling millions of the embedded versions of this (MC68HC05...).

It ain't purty, but it's amazing what you can do with a 8 KBytes of EEPROM and 384 bytes of RAM!

This programming, of course, is in assembly language, because C has too much memory overhead.

Oh, btw, the typical machine executes under 1,000,000 unstructions per second.

Re:PIC is a bit more practical

[Bassman59]

gTsiros:

PIC stands for Programmable Interrupt Controller.

When talking about the 8254, yes (but that part's not really not a "CPU"), but in this case, we're talking about something like Microchip's PIC family of gizzies, which are usually called "Programmable Integrated Controllers" or "Peripheral Integerated Controllers." ahhh, acronyms...

Re:PIC is a bit more practical

[Gordonjcp]

Programmable Interface Controller, actually... It is indeed a fairly minimal CPU, but is nonetheless extremely powerful.

Reminds me of...

[Tsali]

a rough quote from Monty Python...

"I built the castle in the swamp. They said it would fall over but I did it anyways. Sure enough, it did. I build a second castle and that one fell into the swamp as well. But the third castle stayed."

Looks like he needs another iteration.

Re:Reminds me of...

[Jhan]

Less rough, though still offtopic quote:

When I first came here, this was all swamp. Everyone said I was daft to build a castle on a swamp, but I built it all the same, just to show them. It sank into the swamp. So I built a second one. And that one sank into the swamp. So I built a third. That burned down, fell over, and then sank into the swamp. But the fourth one stayed up. And that's what you're going to get, Son, the strongest castle in all of England!

(No longer to-the-word-)But I just want to sing: ... And no singing!

Re:Because, as Tom Lehrer, formerly of MIT noted.

[oliverthered]

Wow, 10 fingers, how many thumbs do you have?

I blew up a PDP8 - and survived

[Ella the Cat]

I'm old enough to have done an electronics project building a joystick interface for a PDP8 as an undergraduate. I spent ages soldering TTL chips and after a few weeks plugged the card in, to a strong smell of fish and burning insulation. It wasn't my fault, the slot in the edge connector was too wide, and every single connector on the backplane had shorted to every other. It was 6 months to get the machine repaired, so someone figured out they could take out the power transformer, scrape off the burnt mess, figure out how many primary and secondary turns were needed on the transformer, then wind them on using a reel of wire and a lathe. They go the machine going, someone else filled the board slot with epoxy and cut a new slot. My project was saved! A few weeks later i reached round the back of a PDP8 to unplug a power connector, grabbed the live pin, but was saved because my arm was earthed to the PDP8 case. I love that machine, I still have the instruction set on a sheet of paper.

Just use the history simulator

[ch-chuck]

run your very own pdp8, pdp11 or even an Altair with disk basic or cp/m - here. I've recently completed some serious z80 assy projects using simh on my Linux notebook. Works great w/o having to mess w/ flaky hw.

I loved the PDP-8 but I'm not convinced...

[dpbsmith]

Until I actually started programming a PDP-8 (in assembly language, of course), I would never have believed that you could program comfortably in such a seemingly restricted instruction set. And, conversely, when I moved to a PDP-11, I thought I was going to revel in the freedom and power of all those instructions, all those registers, those addressing modes, those index registers... and the ability to access 65536 bytes directly.

If C is "high-level assembly language," then the PDP-11 is "a computer that directly implements C."

To my surprise, though, I didn't really find that a lot was gained. Programming a PDP-11 didn't really FEEL much easier or more powerful than programming a PDP-8. And it was amazing how much every program expanded in size. It's been said that the PDP-8 instruction set was the most core-efficient ever devised, and I'd believe that.

On the other hand, when I tried programming a 6502, which on the face of it doesn't SEEM that much more restricted than a PDP-8, I just about went bananas.

Having said all that, I'm still not sure I see the point. The sweet design for a computer has to depend on the economics of the hardware around it. Who cares? Even IF the "core-efficiency" thing were true, and even IF you could use standard RAM with a 12-bit processor and not waste any bits, and even IF it turned out that the PDP-8 design were, say, 30% faster and used 30% less RAM for a given program than x86... how could it matter?

If the Alpha, which really WAS a superior design, wasn't superior enough to overcome Intel marketing, customer inertia, and only the normal amount of mismanagement, how can a PDP-8 be anything more than a curiosity?

Re:I loved the PDP-8 but I'm not convinced...

[dhogaza]

If C is "high-level assembly language," then the PDP-11 is "a computer that directly implements C."

Actually it's fair to say that C was developed as a "high level assembly language" for the PDP-11, in other words you've got it slightly backwards. The postfix "++" and prefix "--" operators correspond to the PDP-11's autoindexing mode and when applied to a dereferenced pointer map directly to "(Rn)++" (once the pointer's been moved to a register.

I doubt C would have these constructs if the PDP-11 didn't provide the corresponding register mode.

As far as the PDP-8 being perhaps the most core-efficient design ever, speaking as someone who once developed system software for the PDP-8 and afterwards compilers for the PDP-11, yes, I'd say you're right.

As long as you could fit program and data into 4096 12 bits words, that is. If your program could fit into 4096 12 bit words accessing data in the remaining 28KW was relatively easy due to the semantics of the CDF instruction. But once your code itself outgrew the first 4096 words things got bad in a hurry, because cross-bank subroutine calls using the CIF instruction were fairly expensive.

Gordon Bell designed both the PDP-8 and the PDP-11, and they were designed with different goals in mind. The PDP-8 was designed to be programmed in assembly code - the page and memory bank addressing structure made the development of efficient compilers impossible (it's not an accident that no system programming language like C was never implemented for the PDP-8 architecture).

The PDP-11, on the other hand, was the first minicomputer designed with the compiler writer in mind. The instruction set was very easy to generate code for, much easier than for many mainframe machines that in those days still often had a single accumulator and some auxillary special-purpose registers. The PDP-11's clean, general-purpose register design and (relatively) orthogonal instruction set made compiler writers like myself almost faint in anticipitory pleasure when the design was first announced.

While Gordon Bell designed the PDP-8 and PDP-11, the original engineering plans for the PDP-8 are signed by DeCastro, who did the implementation. He submitted a rival design for DEC's 16 bit minicomputer that was no where near as clean or compiler-writer-friendly as Bell's PDP-11 design.

When the PDP-11 design was chosen, DeCastro left and started Data General, and his 16-bit design became the oft-loathed Nova.

CDC's 12-bit PIC design was much inferior to the PDP-8's, IMO ... the PDP-8 still serves as a great example of minimalist design in an era where each bit of the accumulator was implemented by a double-width card (each BIT, thirteen of these cards in all, 12 for the accumulator bits and one for the overflow LINK bit).

Re:I loved the PDP-8 but I'm not convinced...

[dpbsmith]

'Actually it's fair to say that C was developed as a "high level assembly language" for the PDP-11, in other words you've got it slightly backwards.'

Quite right. I DID know that. Really I was making a joke. I should have put in a smiley.

Although I do think one of the things I like best about C was recognizing that "increment" and "decrement" are fundamental operations of their own that are NOT well represented by the assignment statement a = a + 1, and deserve a special notation.

Re:I loved the PDP-8 but I'm not convinced...

[Ungrounded Lightning]

Gordon Bell designed both the PDP-8 and the PDP-11, and they were designed with different goals in mind.

Actually Gordon Bell designed the PDP-5 (same instruction set as the PDP-8), then DeCastro did the PDP-8 hardware implementation.

The PDP-8 was designed to be programmed in assembly code - the page and memory bank addressing structure made the development of efficient compilers impossible (it's not an accident that no system programming language like C was never implemented for the PDP-8 architecture).

The actual target of the PDP-8 was to be tiny and cheap (yet powerful enough to be useful). Thus the RISC-anticipating instruction set. ... DeCastro ... submitted a rival design for DEC's 16 bit minicomputer that was no where near as clean or compiler-writer-friendly as Bell's PDP-11 design.

When the PDP-11 design was chosen, DeCastro left and started Data General, and his 16-bit design became the oft-loathed Nova.

What happened was IBM was moving from 6-bit to 8-bit bytes. DEC realized that powers-of-two word sizes was the way of the future - making the next logical minicomputer a 16-bitter. So they put their two star architects to proposal writing.

Gordon Bell said that the thing to do was to write a couple languages to simulate the hardware and software, co-develop the instruction set and a compiler to tune up the instruction set, co-design the processor, memory, backplane, and peripheral card skeletons to optimize the hardware design, and generally do things efficiently and beautifully - after which you'd have an inexpensive, powerful machine that would take over the market for the next generation of minis and easily evolve to handle larger tasks.

DeCastro said you could quickly hack up a 16-bit followon to the PDP-8, hit the market a year earlier, sell a bunch, enable a lot of embedded apps, and make a lot of money.

Of course they were both right.

DEC went with Bell's proposal. So DeCastro went out and founded Data General. And the Nova got to market sooner, enabled a lot of embedded applications, sold a lot of units, and made a lot of money. Then the PDP-11 came out and was also a big hit. And the PDP-11 was easier to program and easier to upgrade, so after a couple generations it took over.

The original Nova had its points. Like a very efficient hardware design. (Cute use of a single 4-bit ALU slice and a single 4-bit register file chip, 4 times as fast as the RAM's cycle, to build a 16-bit processor running at the RAM rate. Reminicent of the stuff Cray did while still at Control Data.) But the Nova was a "16-bit PDP-8" thorugh and through - with the control lines of that ALU chip showing up in the "operate" instruction's microcode bit pattern. It just didn't scale well, and compilers for it were a pain. Meanwhile the PDP-11 scaled very nicely (as a microcoded machine after the first PDP-11/20) and was a breeze for writing both compilers and multitasking OSes. So a DOS, RSX-11, and then Unix were written for it, the followon VAX fixed the few design flaws that eventually put a crimp in its ability to scale, and the Nova became a dinosaur.

More power to 'im

[Angst Badger]

Sounds cool, and reminds me of a project I've been tinkering with for about a year now -- a multiprocessor system based on the 6502 (actually, the 65c02), the same chip as was in the Apple II, Atari, the C64, and the original NES, among others. Problem is, my electronics knowledge at this point is not good enough to get beyond a good theoretical knowledge of what would be involved, though my 6502 assembly language skills are still sharp enough to write the firmware and software.

I only mention this because I hope someone who does have the requisite electronics skills will email me so we can join forces.

As for the earlier post to the effect of "what is it good for?", I can only say that it's fun to do, and old computers are good for the same things they were good for when they were new. One may as well ask what a 1965 Mustang is good for.

DecMate word processor

[jhines]

The PDP-8 had a long life as the Decmate word processor, the 12 bit word did very well, as it could handle text, with bold, underline, etc directly.

The 3rd party service man at a VAX site I worked at in the early 90's had a PDP-8 (W/ RK05's!) still under contract in a big milling machine at a local heavy manufacturor.

Re:DecMate word processor

[oh]

I know of a site that still ahs several PDP-11 (actualy PDP-11 clones) in production. This is in 2002!!!

They control a specific piece of hardware, and since they still use that hardware, the PDP's are still in use. Trouble was, when they had a disk failure, they had to find some one who could rebuild a RSX-11 system.

KW == "kilowords"

[red_dragon]

Back then, the size of core memory was generally measured in machine words, thus in the case of a 12-bit machine like the PDP/8 with 32 k-words, the core would be: 32 x 12 bits == 384 k-bits, or 48 k-bytes.

Re:KW == "kilowords"

[lfourrier]

If I remember correctly, at the time, a byte was a proceessor word long. Only ecently the meaning changed for eight bits. so 32 kw would have been 32 kB.

Re:KW == "kilowords"

[KewlPC]

Actually, IBM had nothing to do with the "a word is 2 bytes" thing. That was Intel, because the 8086, 8088, 186, and 286 all had a word that was 2 bytes long. But few people actually still go by the notion that a word is 2 bytes long, since even Intel processors now use a word size of 4 bytes (they have since the 386).

PDP-8 was not the first minicomputer

[starling]

That was the PDP-1, released in 1960, 5 years earlier.

The PDP-8's distinction was to be the first mass-produced minicomputer.

You speak heresy, Grasshopper

[kfg]

C compiler indeed.The PDP-8 was natively a FORTRAN machine. Apps can be developed perfectly well in FORTRAN. . . and the coolness factor is higher.

King Arthur: Noble FORTRAN compiler, although you are a dead language. . .

FORTRAN compiler: I'm not dead yet sire.

King Arthur: Although you are a mortally wounded language. . .

FORTRAN compiler: Actually sire I'm feeling a bit of all right.

Again, C compiler indeed. Gag my PDP-8 with a spoon. ( Actually, that would be 'anatomically' possible)

Here's an interesting little page on the history of the PDP-8 OS's and languages:

http://www.cs.uiowa.edu/~jones/pdp8/history.html

And here's an interesting computer history page with several FORTRAN links ( as well as UNIX and C links):

http://www.fortran-2000.com/ArnaudRecipes/CompMu se um.html

C compiler. . . phbbbbbt!

KFG

Re:You speak heresy, Grasshopper

[dhogaza]

Well, damned few people used DEC's FORTRAN compiler for the PDP-8, and it was hardly "natively a FORTRAN machine". The FORTRAN compiler was a pseudo-code compiler, and the reasons for this are easy to see if you study the PDP-8 instruction set for a few minutes.

The higher-level languages most frequently used on the PDP-8 were FOCAL and BASIC.

Call it the QED8

[TheAwfulTruth]

Let it RIP.

taler

[wiredog]

and "taler" comes from the town of Joachimsthal, where the silver was mined.

ENIAC On A Chip

[Stephen VanDahm]

These dudes beat you to it.

Steve

Re:ENIAC On A Chip

[Amazing Quantum Man]

Yeah, but how do you replace the microscopic vacuum tubes (valves, for you Brits) when they blow out? That was one of the things about ENIAC, IIRC... the tubes kept blowing!

W95 and DOS will not expire at the end of the year

[kfg]

Only official support for them will. DOS will never 'expire.' It's done. Of course you can't buy MS-DOS anymore, but there's always DR-DOS and FreeD0S, both still supported.

W95/98, on the other hand, will actually expire some years in the future. I discovered this on a reinstall that went bad. Windows simply refused to install. Having a Gateway at the time I called tech support and the issue was tracked down to a buggy BIOS (gotta watch for those updates) that had reset my system clock to a future time.

"Ah, there's your problem. Windows has a 30 year time bomb built in so it thinks it's expired."

Ummmmmm, good to know. I guess that's how long we've got to port all our favorite W95/98 games to Linux ( or maybe Plan 9).

KFG

Re:Maybe...

[Jhan]

Well, with 6 bits you can have the capital letters and some punctuation.

With 12 bits, you can have all the European alphabets, and Russian and (some) Japanese.

With 24 bits, you can have Unicode. (No, you can't have Unicode in 16 bits. It's grown to 21 bits already (allthough only 2^20 code positions are actually in the code).)

To conclude: 12 bits, 4096 characters, is a very good size for a rudimentary global character set. 24 bits lets you use all of Unicode (21 bits) and is a much better fit than 32 bits.

It's like English Measurements

[hey!]

I don't think there's any fundamental DISadvantage to using 12,24,36 and 48 bit words and addresses. The question is not whether the memory addresses will be in terms of some power of two (of course they will), but whether the sizes of words used to hold values and addresses should likewise be an even power of two. Even numbers, surely, but not necessarily a power of two bits wide.

At the time when things pretty much fell into the currently accepted pattern, word sizes that were even powers of two happened to be convenient:

2^8: enough to hold the complete latin alphabet
2^16: Enough bits to handle the entire address space of a 1980s microcomputer
2^32: Enough to handle almost any day to day integer calculation

It's like the inch-foot-yard-furlong-mile of English measurements. These are well suited to the kind of day-to-day measurements that people make, as inconvenient as they are for calculation. Eight, sixteen and thirty two bit words and addresses were big enough in the 1980s, but not so big as to be wasteful.

It's interesting to speculate how things might have been different had the industry settled on twelve and twenty four bit word sizes. It may have been more convenient for people with non-latin alphabets, although not as commodious as a sixteen bit per charcter system. And a lot of effort was wasted in the nineties with the limitations of sixteen bits address spaces (the segmented memory architecture, near and far pointers etc). The need for a larger flat, memory space might have been staved off for several years until 16MB chunks were too small.

I don't know if that's a good thing or bad thing. We might be struggling with a 24-48 bit conversion today instead of happily using our P4s and Athlons and waiting for the high end users to hash out the 32-64 bit conversion.

Re:It's like English Measurements

[hey!]

But was it 8 bits plus parity?

Re:It's like English Measurements

[pjrc]

At the time when things pretty much fell into the currently accepted pattern, word sizes that were even powers of two happened to be convenient:

IBM's EBCDIC (8 bits) and AT&T/ANSI's ASCII (7 bits + parity) has a lot more to do with it than what "happened to be convenient". Before EBCDIC and ASCII, 6 bit encodings were "what was convenient". Both IBM and AT&T were huge monopolies in the 60's, so whatever they implmented was what the rest of the world had to live with. That is why computers today use 8 bit bytes. It's mostly IBM's doing, with some help from AT&T. Luckily, AT&T was generous to take the standard-based route and a fairly good encoding resulted. Heaven forbid we'd all be stuck with EBCDIC otherwise.

Of course, I'm repeating myself (a more detailed explaination awaits if you follow that link).

Odd way to go about it

[SpinyNorman]

The Harris HD-6120 "PDP-8 on a chip" is still available, and is what this guy is using:

http://www.sparetimegizmos.com/Hardware/SBC6120- 2. htm

The only point I could see in using FPGAs would be if you were trying to rectreate an early MSI PDP-8 (these things existed before the single-chip microprocessor).

On the other hand. . .

[kfg]

why disdain the past? Who knows, maybe your grandma knew something about sucking eggs that you don't.

If nothing else the guy is obviously having fun. Believe it or not when you get to be Eleventy years old your own hobby just might turn out to be keeping 16 bit Intel stuff alive.

I mean like, why ride skateboards when we have bicycles now? Why, because you *want* to.

KFG

Re:W95 and DOS will not expire at the end of the y

[JimPooley]

But why the hell would anyone still be running Windows 95 in 2025? Or running games which run on Windows 95 in 2025?

Much better things will be available by then.

Well, not to put too fine a point on it, but . . .

[kfg]

the upgraded CoCo was called. . . a Mac. You can still buy 6800 series chips, although you have to hunt around a bit. They're still very usful for a lot of things.

On the other hand the venerable Z80 not only has never gone out of production but is being updated just as you suggest:

http://216.239.39.100/search?q=cache:vPeP4Ne7p1A C: www.ebnonline.com/story/chipwire/OEG20000927S0018+ z80+chips&hl=en&ie=UTF-8

There are an awful lot of uses for small, fast, cool running, general purpose and cheap as penny candy chips.

"Charles Luther" in "Runaway" understood this full well when he used 8088's to power his nefarious robotic killing machines.

KFG

Re:W95 and DOS will not expire at the end of the y

[chas.capwell]

But why the hell would anyone still be running Windows 95 in 2025? Or running games which run on Windows 95 in 2025?

Much better things will be available by then.

You say this in a thread talking about a game that was developed around 30 years ago that has gone offtopic from an article about an architecture that is at least as old. Are you on crack or something? :)

Historical inaccuracies

[John_Sauter]

The PDP-8 was not the first minicomputer. As its name implies, it had predecessors, including the PDP-7, PDP-5, PDP-4 and PDP-1. I nominate the PDP-1 as the first minicomputer.

Also, the IBM System/360 was not the first computer with variable-length instructions. The IBM 1401 also had variable-length instructions, and I am not sure it was the first.

John Sauter (J_Sauter@Empire.Net)

Re:Historical inaccuracies

[dhogaza]

Well, the earlier 18-bit models weren't marketed as mini-computers. Tney were marketed as Personal Computers, though :) For lab scientists ...

The PDP-1 was pretty huge and was the first general purpose computer sold in the US for under $100,000 (back when that figure was real money!). The 12-bit PDP-5 was the first sold under $40K and the PDP-8 the first under $20K.

The PDP-6 was older than the PDP-8 and was by no means a minicomputer, being a 36-bit predecessor to the PDP-10!

Anyway ... the term minicomputer came into common usage because of the PDP-8. It was small enough to sit on a desktop and didn't need a rack in its basic configuration. The earlier models you mention, i.e. the 18-bit family PDP-1/4/7/9/15 and the PDP-8's immediate predecessor the PDP-5 were all required standard 19-inch racks.

The table top configuration was the distinction the PDP-8 brought to the market, though nearly all were sold in racks because you couldn't really expand the desktop version.

Why we use base 2 instead of base 3

[Anonymous Coward]

As nice as a base-3 system my be in theory, there are very good reasons for sticking to a base-2 system in hardware. As we are moving to smaller and smaller fabrication processes, it is necessary to lower the supply voltage Vdd. For example, now that we are approaching the 0.1um and 900nm levels (at least in research labs), Vdd is getting down around 1 Volt. However, the Vt (the threshold voltage needed to turn "on" a MOS transistor) stays the same, because it is determined by physical properties of silicon (mostly). That means we're losing headroom. To implement ternary logic, we would need 3 different voltage levels. We're simply running out of room to do things like that. You need to leave a noise margin around your "1" and "0" values for reliable operation. (For example, if Vdd=1V, you might consider 0.0-0.4 = "0", 0.6-1.0 = "1". Then a logic gate that "sees" 0.0-0.5 interprets it as "0", etc. If you had a "0" that was really 0.4V, you would hope that "noise" wouldn't bump it up above 0.5V, or else it would look like a "1".)

The threshold voltage for transistors is somewhere under 0.2-0.3V usually (depening on the technology & lots of other parameters). So, you absolutely need a 0.6V supply. (0-0.3 = "0", 0.3-0.6 = "1".) Unfortunately, even with Vdd=1V, you'll get voltage drops happening throughout the chip ("IR drops" - as in I=current times R=resistance) so that the 1V may only look like 0.8V to some parts of your circuit.

From the above discussion, it should be obvious that there really isn't room to shoehorn in a third voltage level. Also, a nice feature of CMOS design is that when a gate is sitting in a "0" or a "1" state, it is drawing no (well, negligible) power. Power is only dissipated while a value is switching from a 0/1 or vice versa. Off hand, I can't think of a way to do that with a third logic-value. Consider drawing even a tiny amount of current while a gate is sitting at logic "2" (or whatever you want to call the 3rd value). 1mA (milliAmp) times 1 million transistors on a chip = 1000 Amps. That chip's going to get a little hot!

Ok, so you've probably got at least two questions, which I will try to answer in advance. If you've got other questions - I'll just let someone else tackle those.
Q1) Why don't we just use a higher Vdd (supply voltage)?
A1) If you're using smaller transistor widths, you simply can't. When you use a really thin gate (i.e. 0.1um) on a transistor, the breakdown voltage of the gate is reduced. If you use a higher voltage, the transistor melts. (You could use larger transistors, but that kind of defeats the whole purpose! We make transistors smaller because we can fit more on a chip, and they operate faster and use less power.)

Q2) Can't we lower the threshold voltage?
A2) Yes, to some extent. (It's not always easy.) But we don't want to. Even when a transistor is "off", there is still a very small amount of leakage current flowing through it. If you reduce the Vth, you also increase the amount of leakage current. In older technologies, this hasn't been much of a problem, because the leakage current was so small in comparison to the dynamic power consumption. But as we are putting more and more transistors on a chip, the leakage power consumption in modern chips can easily add up to 30%-40% of the total power consumption. There's also another reason. If you did that, you would be lowering your noise margin. And you don't really have much control over the noise (which is why it's called that). If you reduce noise margins too much, you'll find it almost impossible to create a circuit that actually functions reliably.

Well, I hope that satisfies some of you (and doesn't get the rest of you too upset). VLSI circuit fabrication is a really neat field. Some of the tricks that are being used these days to fabricate that chip sitting in your computer and get it running at 2GHz (or aren't they up to 3GHz now?) are quite amazing - they're doing their best to cheat physics! Using a ternary counting system to build computers may have a lot of nice theoretical properties, but I can't see it displacing binary any time soon, except possibly in some really specialized applications. (There are always exceptions.)

That's my $0.03 worth. (Hey, I typed a lot. I think that's worth at least $0.01 extra. Maybe $0.025?.) Any errors in the above are mine, but I won't admit it.

Re:Why we use base 2 instead of base 3

[pjrc]

Off hand, I can't think of a way to do that with a third logic-value. Consider drawing even a tiny amount of current while a gate is sitting at logic "2" (or whatever you want to call the 3rd value).

It could be accomplished (fully static CMOS, no steady state current to maintain a 3rd logic level) with a second power supply, and circuitry designed to connect the output to either Vss, Vdd or Vmm (m for middle, for lack of any other name.. hmm) Brian Hayes's flawed assumption is that circuit complexity increases linearily with the number of logic levels. He writes "An obvious strategy is to minimize the product of these two quantities", refering to the radix and number of symbols to represent a number... but he just pulled that out of a hat. The required circuit complexity is not linear function of the radix, and a realistic model would quickly prove that binary is the most efficient. A fully static ternary output requires a minimum of four transistors, whereas binary requires only two.

That chip's going to get a little hot!

With a static CMOS circuit designed this way, power consumption would be approx 0.5 * C * f * V^2 (as it is in normal binary circuit). C will probably increase somewhat, as nearly twice as many transitors would be needed per circuit, yet fewer trits are needed that bits for the representing the same numerical range, so the increase in C probably wouldn't be by a factor of two. Presumably f (the clock frequency) would stay the same (well... I'll get to that...), and V stays the same (50% of transitions in binary are full supply voltage, in ternary 33% are full voltage and 33% are half voltage). Power comsumption would probably be similar.

Saddly, f probably won't stay the same. C gets larger on each signal, and when driving to half voltages, the transistors that would connect to the Vmm supply get only half the effective gate voltage applied. So doubling the load and cutting the drive significantly is really going to hurt the circuit's speed.

Dynamic logic tricks (pre-charged busses) and bicmos circuits add another interesting dimension that's too complex to worry about, though it'd be important for any microprocessor.

But power consumption isn't likely to be a problem.

Getting back to the old PDP-8, as I recall it was a binary machine. The motivation behind 12 bits was that 6 bits was ideal to represent both upper and lower case characters and plenty of symbols, and 12 bits (two chars) was plenty for useful math. I don't recall the popularity of 6/12 bit systems having anything to do with base-3 signaling.

Re:Why we use base 2 instead of base 3

[lindsayt]

I'm a sysadmin and not an electrical engineer, so I may be far afield here. But why does everybody talk about ternary as being 0 voltage, full voltage, and some random number in between? Why do both non-zero values have to be on the same side of zero? Wouldn't it make more sense to have zero for one state, and then the same voltage for each of the other two states, just with opposite polarity? That way, there would be no reduction in signal-to-noise level, and trits could be determined by separate presence-or-absence of voltage on either side of zero (+- of course). There's probably some obvious reason why this wouldn't work, but the whole point of digital circuitry is to not have to measure the exact voltage of the current to work (as analog equipment does). This would then be a simple two-step process - absence or presence of voltage, and polarity. No reliance on measure amount of voltage.

Ideas? I've always been intrigued with digital (non-analog) techniques at determining ternary states...

Re:Why we use base 2 instead of base 3

[pjrc]

Nowhere in the paper could I find that Brian Hayes claimd "that circuit complexity increases linearily with the number of logic levels."

In the paragraph just above figure 2:

Everything hinges on the assumption that rw is a proper measure of hardware complexity, or in other words that the incremental cost of increasing the radix is the same as the incremental cost of increasing the number of digits.

Scroll back up to figure 1, and the third paragraph up is where the assumption is first made:

By one plausible measure, it is the most efficient of all integer bases; it offers the most economical way of representing numbers.

How do you measure the cost of a numeric representation?

[snip, absurd cases of base 1 and base 1e6]

Evidently we need to optimize some joint measure of a number's width (how many digits it has) and its depth (how many different symbols can occupy each digit position). An obvious strategy is to minimize the product of these two quantities. In other words, if r is the radix and w is the width in digits, we want to minimize rw while holding rw constant.

So what he's saying is that the "cost" of a representing a number is the cost per digit multiplied by the number of digits required. But he makes the assumption that the cost of each digit is a linear relationship with the radix, which is simply not true in almost any system (certainly not in electonic circuitry nor in telephone menu systems).

Speaking mathematically, r is the radix, and w is the number of digits (or symbols, words, or whatever you call them) required using that radix. The cost is F(r) * w, where F(r) is some model for the cost to implement that radix.

The words "An obvious strategy" are plain wrong. It's not obvious at all. It's simple-minded and ignorant. It's devoid of any anaylsis or thought about any real system. Even from a purely theoretical standpoint, it's academically dishonest to gloss over this critically important point rather than write "r * w" instead of "w * F(r)" and state the assumption of a linearly increasing cost per digit as the radix changes.

Well, maybe that's a bit strong. Who am I to judge what's academically proper. But the paper clearly begins by saying:

People count by tens and machines count by twos ... I want to offer three cheers for base 3, the ternary system ... They are the Goldilocks choice among numbering systems: When base 2 is too small and base 10 is too big, base 3 is just right.

The general arguement that base-3 is actually superior for computer arithematic is also quite evident in the "Trit by Trit by Trit" section (just below figure 1). I'll avoid quoting much of it, but at the conclusion he writes:

Why did base 3 fail to catch on? One easy guess is that reliable three-state devices just didn't exist or were too hard to develop. And once binary technology became established, the tremendous investment in methods for fabricating binary chips would have overwhelmed any small theoretical advantage of other bases.

Now the rhetorical question "Why did base 3 fail to catch on?" is answered by postulating (not even any real knowledge) that way-back-then it was too tricky to design and base-2 gained so much momentum and became so well established that base-3 never caught on. Notice how he concludes with the words "overwhelmed any small theoretical advantage of other bases", reaffirming once again the standpoint the base-3 has some advantage, if small, over base-2, theoretically speaking. He's clearly talking about implementation of circuitry.

The ugly truth is that rolled up in the theoretical advantage of base-3 for circuitry is that assumption that the "cost" is "r times w" (r for radix, w for number of digits). Any engineer can tell you that cost has units of dollars, and r and w are both unitless quantities. To compute the cost of using a particular number system, you need to use a function (above I called it "F(r)") that transforms the abstract number "r" into the cost of implementing that radix. The unitless number of possible digits needs to be turned unto a quantity in units of dollars (or some other currency) before it can be multiplied by "w" to obtain the cost of implementing that radix.

Re:Daler

[david duncan scott]

In fact, reaching deep into my trove of useless trivia, I seem to recall that we in the US snitched the term from the Spanish, along with an arcane bit of slang: the Spanish dollar, circulated in the old West, was often broken into 8 pieces (it may even have been scored to facilitate this, but I may be making that up.) One quarter of a dollar was, therefore, two of these bits, leading to such Americanisms as "two-bit whore" and the ever-popular secret knock pattern, "Shave and a Haircut, Two Bits".

Even older..

[pigeon]

And I thought I was old fashioned, owning a PDP-11..

Re:Because, as Tom Lehrer, formerly of Harvard, no

[kfg]

Tom has always been a *student* of Ha'vard, Tom's spelling, ( as has been most of my family, I'm the "black sheep" and went to Bard), but he spent some years teaching at MIT. He also spent some time teaching at UC Berkeley, where he not only taught math but. . .musical theater.

I chose to highlight the MIT link because of their infamous connection to the PDP line of computers.

And yes, he's still at Ha'vard.

KFG

Believe it or not. . .

[kfg]

an object may have a proper noun all to itself and still retain membership in a larger general class.

Just as a "Seacock" and a "Butterfly," despite having proper nouns of their own are, nonetheless, still just valves.

Draw the Venn diagram, you'll figure out the set/subset relationship eventually.

In short, I have just as many thumbs as I have "pinkies."

KFG

I suppose for the same reason I still play. . .

[kfg]

Chess, Checkers and Go, even though "better" games have been available for thousands of years.

And by the way, Infocom rulz d00d!

KFG

PDP-8 microprocessors

[panurge]

This is so long ago I can't even remember the manufacturer, but there were at least 2 CMOS implementations of a PDP-8 processor. Has anybody observed that the instructions were based around core memory, so the accumulator was cleared when it did a store? If you could get the core again, a home computer with a PDP-8 processor and core memory would really benefit from a modbox with windows and lights.

For some industrial control jobs, something like a PDP-8 or PDP-11 is in many ways ideal because you can see everything that goes on. It is actually possible for one person to understand the hardware, the microcode, and every single bit of the software. For me, that is the great pleasure of small embedded designs. I really think it would be good to have a teaching tool for CS that actually meant that the student could do a project and have a complete overview of the entire thing in this way. I'm far from knocking progress, but there are comments on this thread that are a bit about the kind of alienation we have now between hardware and software - most people have no real idea at all what the hardware does, and use terms like "cache" without even stopping to think about what is going on. So yes, let's have someone build an understandable modern PDP-8. It's less weird than the RCA1802 and easier to get your head around than the 8080.

Joachimstal.

[Grendel Drago]

Hey, did you watch "Connections" too? That's kind of an obscure factoid, and I'm wondering where else it shows up.

--grendel drago

Re:Joachimstal.

[wiredog]

Got it from "Making of the Atomic Bomb" by Rhodes. It was mentioned in passing when he was talking about Oppenheimer's visit to Joachimsthal. Which is also a source for pitchblende, btw.

Meter.

[Grendel Drago]

Actually...

Here it says:

The meter (m) is the Si unit of length and is defined as the length of the path traveled by light in vacuum during the time interval of 1/299792458 of a second.[3] This replaces the two previous definitions of the meter: the original adopted by CGPM in 1889 based on a platinum-iridium prototype bar, and a definition adopted in 1960 based on a krypton86 radiation from an electrical discharge lamp. In each case, the change in definition achieved not only an increase in accuracy, but also progress toward the goal of using fundamental physical quantities as standards, in particular, the quantum mechanical characteristics of atomic systems.

--grendel drago

Re:Meter.

[Amazing Quantum Man]

Wasn't the meter originally defined as the 1/10^8 the distance from the North Pole to the Equator along the meridian passing through Paris?

Re:Meter.

[Grendel Drago]

*shrug* All I did was google for it. But it seems plausible that that distance was measured and the platinum-iridium bar was created to match it---it would be much, much easier to go by a canonical bar than a measurement of the planet, I suppose.

--grendel drago

Re:Meter.

[Sri Lumpa]

Yes, From http://physics.nist.gov/cuu/Units/meter.html

"The origins of the meter go back to at least the 18th century. At that time, there were two competing approaches to the definition of a standard unit of length. Some suggested defining the meter as the length of a pendulum having a half-period of one second; others suggested defining the meter as one ten-millionth of the length of the earth's meridian along a quadrant (one fourth the circumference of the earth). In 1791, soon after the French Revolution, the French Academy of Sciences chose the meridian definition over the pendulum definition because the force of gravity varies slightly over the surface of the earth, affecting the period of the pendulum."

Re:Meter.

[Derwen]

Some suggested defining the meter as the length of a pendulum having a half-period of one second; others suggested defining the meter as one ten-millionth of the length of the earth's meridian along a quadrant (one fourth the circumference of the earth). In 1791, soon after the French Revolution, the French Academy of Sciences chose the meridian definition over the pendulum definition because the force of gravity varies slightly over the surface of the earth, affecting the period of the pendulum.

"And so in June 1792, just as the Terror was beginning to crank into its dreadful gear, two eminent French astronomers set out in opposite directions from Paris. Their aim was to measure the distance of the meridian arc between Dunkirk and Barcelona: once obtained, this distance would be divided by 10 million to give a definitive and utterly impartial length for the new metre. Pierre Méchain travelled south to Barcelona, while Jean-Baptiste Delambre went north to Dunkirk. The savants were to work towards each other, reunite in the middle of France, do their sums and present the revolutionary government with the most classless unit of measurement conceivable. The whole business was to take them no more than a year."
But of course it went wrong:
"Somewhere near Barcelona, at the very start of his triangulations, Méchain made a small error of computation. Once this error had entered his system, it was impossible to eradicate it. For years the knowledge of this error haunted Méchain - the thought that he alone had managed to falsify what was intended to become 'the fundamental scientific value, the measure which would for ever more serve as the foundation for all scientific and commercial exchange'. In 1804, suffering from acute depression apparently brought on by guilt, Méchain returned to the Valencian coast to try to atone for his error. There he caught malaria and died."
Read more in The Measure of All Things: The Seven-Year Odyssey that Transformed the World by Ken Adler
- Derwen

Re:Measurements

[Chandon Seldon]

Tell me something. The atmospheric pressure at sea level is 15 psi using imperial measurements. Using metric, how would you specify that figure?

Re:Measurements

[pommiekiwifruit]

~1000 millibars or 100,000 Newtons per square metre or 100 KiloPascals. What could be easier? But it varies quite a bit around 1013 millibars. (It's 999 millibars just south of Sardinia at the moment. Alas the weather map doesn't show psi or inches of mercury - do US ones?)

Re:Instruction sets are a matter of taste

[dhogaza]

And the 68000 was simply a beefed-up rip-off of the PDP-11, extended to 32-bit addressing.

It didn't include the PC as a general purpose register because that was one of the design innovations in the PDP-11 that Gordon Bell/DEC patented ...

pdp8 microprocessor

[scharkalvin]

Intersil used to make a CMOS microprocessor chip that used the pdp8 instruction set. It was a 12 bit micro processor. They also made a chip that acted as the memory address extension controller. You still might find some of these chips for sale someplace. DEC even used the 'cmos 8' inside one of their terminals to run pdp8 based word processing software. If you want to build your own pdp8 today, I'd look for some of those old intersil chips.

Re:PDP8 instruction set

[scharkalvin]

spy this instruction, CIA. Then again the PDP 10 had a 'floating add register triple' or FART instruction.

Re:Because, as Tom Lehrer, formerly of Harvard, no

[Amazing Quantum Man]

He's been at UC Santa Cruz since the early '80s. He teaches "The history of the American Folk Musical", IIRC.

Re:PDP-8 was not the first minicomputer-Alleleuia

[starling]

I used to repair PDPs and Novas for a living before I got into programming (MUMPS, in case anyone remembers that), so I have a certain interest in this story.

Still got the scar from a Nova 4 power supply which zapped me - geek chicks dig that sort of thing :)

Re:Daler

[zulux]

And of course Doller is a basterdised version of the Thaler wich is a basterdised version of Joachimsthaler.

more info

Re:Because, as Tom Lehrer, formerly of Harvard, no

[kfg]

Yes, but he's bicoastal, just as in his UCB days. His home remains in Mass. and last I knew he still maintained "ties" with Harvard ( but then one of the problems with becoming an old fart is that "last I knew" turns out to be about 1980 which you think of as "just a while ago").

I could recall this wrong, but I think at one point he was, at least officially, teaching at Harvard, MIT and UCB all at the same time.

( And note that I make no claims to know the man, I just know several people who know/knew him)

KFG

Another point...

[TheConfusedOne]

I'd say your chances of getting laid are better if you tell 'em you're resurrecting a '53 Corvette than a PDP-8.

Re:Another point...

[cburley]

I'd say your chances of getting laid are better if you tell 'em you're resurrecting a '53 Corvette than a PDP-8.

I dunno; sing a few verses of "Old MAC Hacker" ("Old MAC hacker had an '8, EAE IO...") and you've evened the odds, I'd say....

I overspoke when. . .

[kfg]

I termed FORTRAN "native" to the PDP-8 ( and I suppose it would only be correct to term the appropriate assembler "native").

It would be more correct to merely note that FORTRAN was the first 'higher' language for which a compliler was available.

I've at least seen FOCAL. I've *never* seen BASIC on a PDP-8 ( thank God), but my experinece with the machine was strictly in a hard science academic setting where FORTRAN was King.

Don't get me wrong though. I'm no FORTRAN zealot. I'm an APL zealot who always resented the forced move from one to the other.

KFG

Knob on the front?

[serutan]

Hey I remember that, but I always thought it was the choke.

Form factor

[Ed Avis]

If this thing fits on a single FPGA, then it could be made into a handheld called the PDA8.

Re:What the Hell?

[KewlPC]

I think he may have meant $180,000. IIRC I read somewhere that the PDP-1 cost $100,000 and the size of a refrigerator, yet for the time was dirt cheap and incredibly small.

I actually have an emulator for a PDP-11 around here somewhere (don't remember if it's still on disk or on a backup CD-R in the closet). SCO used to offer an "Ancient Unix" license, where you could get it and then download old versions of Unix (actual Unix, not BSD or Linux or whatever). I used to run Unix on it (not sure what version it was anymore, probably V or VII) just for kicks.

I hate to break it to you...

[twoslice]

I used to work for Digital before Compaq took us over and the Digital name died right there. Now that HP has taken over Digital - Digital is deader than dead!

Perhaps it is time for a new logo for the topic (RIP comes to mind)

NB: Digital was the best company to work for...

Re:Measurements

[swillden]

The atmospheric pressure at sea level is 15 psi using imperial measurements.

14.7 psi is closer to the average value. It varies a bit, of course.

Using metric, how would you specify that figure?

1 bar.

Re:W95 and DOS will not expire at the end of the y

[Zaiff Urgulbunger]

But why the hell would anyone still be running Windows 95 in 2025? Or running games which run on Windows 95 in 2025?

Ask the man with the beard and the Dec-PDP8. He speaks in 12 bits. He knows everything!

A tiny OS for 8080s

[Ungrounded Lightning]

Those architecture are so simple, with kernels so small you could print the hex binary out on a couple of pages. Imaging how fast an accounting package would be on a 1 gHz, or even a 200 mHz.

If they do reconstitute the 8080 and a couple peripheral chips for it (like clock, uart, parallel-port), I recently found the source listing of an RTOS I did for it.

- ROMable
- Real-time
- Preemptive multitasking (arbitrary number of priorities)
- Reentrant code
- Application and driver tasks are "actors". (Think "objects" where each instance of a task object subclass is a thread of execution).
- Intertask communication supported by the kernel was:
- Semaphores
- message queues ("communicating semaphores")
- freespace memory allocation queues
- Written in assembler, but with a system service calling convention compatable with the CP/M compiler output's subroutine calls.

The kernel, queueing system, idle task, init routine, memory allocator, and an "empty" (just the idle task and stop flags) task & preallocated memory table totaled a few bytes under a half-K. (Yes, Virginia, all that in less than 512 bytes.) Or a few bytes over if you want to keep some extra per-task state so a debugger can figure out what a task is up to.

The size was very important, because the system it was written to run in had only 8K of ROM available, max. This had to hold a complete energy management system, including OS, device drivers, real-time task debugger, com stack, real-time clock handler, schedule handler, command interpreter, data collector, thermostat input, relay contact input, event counters (watthour meter data collectors), relay drivers, relay logic simulator (ladder-diagram interpreter), and a table of relay logic to simulate. The total software came in under 4K, leaving one of the two ROMs available for the relay logic, sensor, and other configuration definition tables. (At a couple bytes per relay contact and a bit per interconnect you can define a VERY complicated hunk of relay logic in 4K. B-) )

The trick was the insight that adding Mark Weiser's "T" (non-blocking "P") to semaphores' "P" and "V" lets you use communicating semaphores for EVERYTHING, including communication between the below-the-line interrupt routines and the above-the-line part of device drivers (which compete with other real-time tasks in the normal scheduling process). Think "message-passing minikernel", with a vengance.

Religiously using communicating semaphores, in a handful of idomatic ways, for all intertask communication (and even some task-internal functions), makes the tasks themselves tiny, and leads to a software organization that amounts to object-oriented programming. (I believe that this effect, in a direct ancestor of this operating system, IS the origin of OOP.)

I kept the rights to the supervisor core and tools, and had been thinking of open-sourcing it if I ever found it again (which I now have - at least the hardcopy version). Let me know if anybody needs it and I'll get in touch.