PDA

View Full Version : Conversation Starter (I Hope)



olddataman
December 12th, 2003, 12:01 PM
Now I realize that I am almost ancient by comparison with most of you, but I simply can't accept the idea that the computer industry began with the development of the Intel 8008. My first hands on experience with a commercially developed digital computer was as an Air Force instructor, teaching theory and maintenance of the (then classified) AN/FST-2 computer which was the system at each radar site in the North American Ai\r Defense network. These were the sites that fed their processed data from the half dozen types of search and tracking radars into the ANFSQ-7 (etc.) Sage System computers. I had been employed prior to my AF hitch ass a technician, bilding nuclear research instrumentation from scratch, which was primarily "wired program" digital computers. My first general purpose digital computer was the LGP-30, a vacuum tube computer about the size of an office desk and costing $39,000. In my opinion it was the first true minicomputer, first deliveries in about 1956. Between then (1956) and about 1976 i worked directly or, as part of my duties) made myself familiar with just about every computer commercially available in the price range between $18,000 and $250,000 that used binary arithmetiic, was "word" oriented, had good, useful instruction sets and was suitable for the kind of work our group was engaged in. We applied computerrs raanging from basic PDP-8's to SDS (XDS) Sigma-7's. and later, used a wide range of minicomputers in a variety of applications.
I guess I would have to say that my favorite of all the machines I learned in those days the SDS 9xxx series. They consisted of 7 models. the 92, 910,920, 925, 930 9300 and 940. the SDS 92 was a 12 bit machine (word length) while all oftheotherswere 24 bit word length. Our group was chartered to develop construct, program, install and maintain real-time systems for data acqusition, contol and process data on a real-time basis from research facilities. During my five years with that group, we implimentat about eight different systems using SDS 9xxxseries, three systems using SDS Sigma series (two Sigma 7's and a Sigma 2) and about five or six more using other computers such as the PDP-8 and the CCC DDP-316 (later becoming Honeywell computer product). The SDS 910 and 920 were the most fun because they werethe most used, easiest to maintain and most reliable. I learned those computers so well that I could actually trouble-shoot them from the logic equstions so that I knew exactly what printed circuit logic card I had to find the bad component on before I actually went to the installation to see the trouble in acton by watching the console lights and looking with a scope to actually "see" the trouble occuring.
There were no floppy disks, hard discs, or 3M-type tapes in those days. No "operating systems", no peripheral Drivers, , no built-in crt displays, etc. Application software was written in-house. The only "standard" peripherals were Teletype input and output ot the console, standard IBM compatale mag. tape drives, fixed head per track disk or drum memory systems. Main memory was typically from 4096 words to 32k words of 24 bit memory. The manufacturers usually supplied a moderate to good Assembler, a Fortran sompiler and often a set of system maintenance routine Maintenanre routines. Most "standardized" trouble-shooters, translation routines and a wide varietyher software became available from the inevitible "User Groups" that were formed by customers themselves. I was Chairman of the Standing Committee on Hardware of the SDSS User Group for about five years.
There were CLONES in those dayss also. But that is another story and I'll make another posting on the subject soon.
Is noone but me interested in these old machines? I;m not very interested in the big, iron, Systems Analyst dominated and Operator Controlled Data Center tye EDP machines like those made by IBM and Unvac, mostly. Just because I was brought up in the "Hands-On" environment. But when you consider that I'm just a lowly technician and I can name at least 25 or 30 makers of computers that produced at least three to five different models of "my kind" of computers, you can see that there are at least as many of them as the totao of micro omputers now.
Ray

Erik
December 12th, 2003, 01:46 PM
Just for one reference, here is the Sage (or part of it) that's on display at the Computer History Museum (http://www.computerhistory.org):

http://www.vintage-computer.com/images/DE_VCF6/PA120027.JPG

Erik

djkrex
December 13th, 2003, 06:07 PM
When I was in the army, I had a lot of fun with these SDS computers. I went to SDS in LA to learn how to fix the 920. For the next year and half, I was stationed where they used them to track deep space probes and ICBM's. They were so reliable that I spent most of my time ( in the wee hours of the night) writing machine language programs. Some of them useful :? In school they didn't teach application programing along with maintenance.
We learned how to program in octal just enough to test basic functions and I/O. I didn't find out it had an assembler program for at least a year. By that time I knew all the octal op codes my memory.
There was quite a variety of people in that SDS class. Three guys from Norway, one from SaraLee, one from NASA. An earlier class had a Russian. Turns out that the Russians were using this thing for the same purpose we were. This was back when Louie Armstrong (or Neil ? :)) first walked on the moon. Cold War days. SDS wasn't fussy who was in there classes.
The guy from NASA said they had one of these in the crawler that the missles rode around on. This was a popular machine.
The very first real computer that I had my hands on was the NCR 500 system. I was really a giant cash register/typewriter combo with a processor attached via cable. The processor was about the size of a freezer with the cpu power of about two FLOPS. Probably closer to one FLOPS.
It had circuit cards about the size of AT mother boards. A typical card contained 4 flip flops. Memory size was 400 64 bit words. The memory diagnostic took 15 minutes to run. They could of done better by putting a bigger pulleyon the motor in one of their mechanical calculators. The Army used these things over in Viet Nam. The humidity was so high that circuits would corrode. When the thing started to fail, they would run a push broom under the processor to see if any transistor heads fell off.
The Army seemed to buy alot of this stuff, at bargan prices too :D .

olddataman
December 14th, 2003, 04:15 AM
Yes, those were the days. I went through the 920 maintenance coure in March/April, 1963. Our first 2 SDS computers were 910 SN 7 or 9 and 920 SN 19. (later, I was the first non company employee trained on the Sigma 7. Our first two of them were SN 2 and 3)
The "9---- Series were so reliable because it was the first comercial computer to use Silicone transistoors whicfh could run much hotter than the usual germaniums used by the competitors. With the logic levls of zero and+8volts, the SDS circuits had far better noise rejection specs too. Those computers could run reliably so hot that they could burn your hand if you wern't careful. We had one machine that had 32K core, two tape drives a head per track drum memory two cabinets of system interface hardware, the usual paper tape and IBM SelectriI/O Writer and three "swingout bins of priority interrupt cards. Those inyerrupts were "banged" constantly and got so hot that I could not touch them very long or it got most uncomfortable, But that computer went almost 21000 hours without failure and it took less than a half hour to fix it. It then went another couple years before it was retired and a new system based on the PDP-11 replaced it.
I still have the tech manual I was issued when I arrived in Santa Monica 40 years ago, It has all my notes in the margins and that kind of personal annotatioin. I used that documentation as the primary source of determining just what would be the result of executing all the unused Opcodes in the 910 and 920. Some of them proved to be very usefull and acouple proved to be disasterous if executed.
You mentioned that there was a computer at Sarah Lee bakeries. Did you know that a few years later, after SDS was acquireed by Xerox, the founder and former President, Max Palevesky, married the president, or some sort of top-dog at Sarah Lee? One more bit of history was the story that arose out of an early 920 sale to Motorola Semiconductor.The story was that the computer system was designed to monitor the yields and QC of the transistors being produced, which were sod to SDS (among others). We used to sit around during lunch and developed great science fiction plots around that one.
Well, if you want to trade more "war stories" I had a whold bunch of other SES 9-series computers to play with.
Ray

Allison
April 2nd, 2005, 07:18 PM
Well I've been futzing with computers for 35+ years. Your right the 8008 was not the first player. It was the first Micro on the block and I was building with them in '73. But by then I had time on DEC system 10s (ka cpu) and
PDP-8I and CM 2100 and 2200s. The personal computer industry is older than Altair, though that was a clear breakpoint in price of admission.

In my book the clear leader for many is having their own PDP-8 as that was
attainable (12,000$ in 1970 for a decent system, new). It was also an archetecture that was buildable with available parts. Anyone that knows the -8 also knows it was a very sparse machine but capable. Also in college
there was a nice little 5bitter that was for training. I had a whopping 128 words of memory and everything was bare minimum but you could code and test a simple program and see it in lights. Remarkably it was mostly tansistor. Before that I remember a desk sized computer for business that had drum for memory and used RTL logic for the cpu. Danged if I can remember the company. Even then I remember the engineers I'd met saying that was still too big and smaller and cheaper was on the way.


Allison

alltare
September 17th, 2006, 09:09 PM
Does anyone have any personal knowledge of Los Alamos Labs' MANIAC (Mathematical Analyzer Numerical Integrator And Computer)? It was built in the early 50s to help design the H-bomb.

Unknown_K
September 17th, 2006, 11:06 PM
I think all Intel (and Motorola to name a few) did was to start the home computer craze that ended up bringing cheap computers to the masses as means of comunication and entertainment (which is what most computers are sold for these days). By starting the craze I just mean coming out with a small, low power (electricly), and cheap processor that a computer for the home can be built around.

I was born in 1968 and had two interests when I was in highschool, computers and engineering. Since computers were getting cheaper and cheaper by the time I wanted to go to college (and I liked chemistry and getting things to work) I decided to be a chemical engineer knowing I can buy my own computers and use them as a hobby.

If I was born 15 years or more earlier I think my choice would have been much harder since the home computer revolution had not taken off yet and the engineering done on chips was very basic compared to what is done today. Both fields have changed quite a bit since then.

I have to say that the people who were working in computing in the 50's and 60's and even the early 70's were lucky to be at the right place at the right time. Today there is so much money riding on every decision that people cannot take the risks they did back in the beginning, nor do they have the big picture of what was going on as a whole because of complexity.

DoctorPepper
September 18th, 2006, 08:05 AM
Well, I was born in 1959, and electronics was my passion for the longest time. I took three years of electonics in high school, two hours a day, five days a week. After that, I joined the U.S. Navy, and after boot camp, spent the next solid year in Millington, Tennessee, going to electronics schools.

I was introduced to consumer-level computers (vice the computers we had in the aircraft I worked on) in 1982. I got hooked, and wanted to do nothing else but work with computers after that. I spent the last almost 10 years I was in the Navy working with computers as a "collateral" duty (something I did in addition to my main job, which was working on the aircraft), in every command I was in. I got out of the Navy in late 1992, and got a job as a hardware/softwar/networking tech. for a government contractor. my job there was to keep the computers running, perform upgrades to hardware and software, teach users how to operate the computers and software packages we used, and to perform some light programming for our site. I did that for four years, then moved down to Florida. Down here, all I've done is program (work-wise, anyway). This year marks my 10th year down here in Florida.

I recently had a conversation with one of my Amateur Radio friends, and I confessed that if I couldn't work on/with computers, I don't know what I'd do for a living, and that's pretty much true. I haven't worked in electronics in a very long time, and the whole industry has changed since I last did. Besides, while I was in the Navy, I worked with 1960's technology!

Mike Chambers
October 4th, 2006, 07:38 PM
olddataman, thats pretty neat. you were in the AF, huh? i looked at your profile and it says you were born in 1931.. so you're a year older than my grandfather. he was also in the AF, which is why i bring him up. i'm sure you've never met him but i'll ask anyway... do you remember a Harry Chambers? he got out of the service as a Lt. Colonel.

you probably never met him... he was a test pilot. one of the very first few to pilot the original fly-by-wire. he also did some work with IBM punch card programming at one point.

chuckcmagee
October 4th, 2006, 08:53 PM
Life is getting SO complicated, I can't believe it. This actually happens to me sometimes.... I'm about the throw away a platic soda bottle. I look at the bottle and think how it would be close to impossible for me to make a bottle like that on my own. Now, imagine trying to repair some broken part on your cell phone. The point of all this is that modern life has gotten too complicated and interrelated. Have your power go out for 14 hours and you will see what I mean.
What brought all this on? Remembering our great memory aid - bad boys r--- our young girls but violet goes willingly. Nowadays, the resistors are so tiny you can't even properly position it on the board, let along solder it in.

Mike Chambers
October 4th, 2006, 09:26 PM
Life is getting SO complicated, I can't believe it. This actually happens to me sometimes.... I'm about the throw away a platic soda bottle. I look at the bottle and think how it would be close to impossible for me to make a bottle like that on my own. Now, imagine trying to repair some broken part on your cell phone. The point of all this is that modern life has gotten too complicated and interrelated. Have your power go out for 14 hours and you will see what I mean.
What brought all this on? Remembering our great memory aid - bad boys r--- our young girls but violet goes willingly. Nowadays, the resistors are so tiny you can't even properly position it on the board, let along solder it in.

yeah. all anybody needs is an altair. who needs a monitor when you can just have a bunch of LEDs on the front of your computer :D

old technology > athlon 64

Robby
October 20th, 2007, 02:07 PM
It is fascinating to me to read about early console computing users, technicians, or anyone who has some experience to relate. History is important, and considering the impact of computing technology on the world over time it becomes a most engaging pursuit to sift through what ever information remains from the period of development and computing hardware.
My interest is at times seemingly futile since most all of the 50s and 60s hardware is long gone, and compared to later "vintage" computer interest there seems much less material to look at. That is I suppose the nature of time, the immediate past is not of concern until enough time has passed it becomes a mystery of nostalgia. A thought to bring back some piece of the past in a challenging and useful way I have a notion or two might be to simulate the workings of a Burroughs 205 console or others to learn about the machine and try to come up with an "emulator" that outputs via USB or serial to a replica console, just for the fun of it. Thing is I have not too much of an idea where to start as to the emulation process, the idea being to have the console function as it did originally but with a PC replacing all the rest of the hardware. I don't know how much interest there might be in something like this but considering the 205 and other hardware was made somewhat famous by hollywood's use of such machines as props through the 60s and 70s. Because of this there at least exists blueprints for exterior dimensions of the 205 that could aid in building a replica, it's the idea of having the lights on the console flash in some meaningful way that intrigues me, just a basic "out of the box" when new sequence would be nice, maybe some programs of the time period when actually in use could then be added?

I found this thread by searching for info on Sage Computing and really had no idea how grand these old machines really were in design and function until reading more on the web. Some of the smaller panels from Sage hardware unknown to me were used in the "Lost In Space" tv series from 1965 to 1968, and the Burroughs B205 was used on both LIS and Batman, which obviously made them much more famous than when they were classified from the public just a few years prior. Now with so much time past It seems appropriate to honor them and the people that worked with them with something, an article gathering any stories had come to mind and to try submitting to a related publication, even better a book, and even better still, a book and an emulation software compatible with modern computers. Not being an expert by any measure, not even close would make this a daunting challenge, but given time, a few years perhaps to gather information could be doable, I think anyway.

Terry Yager
October 20th, 2007, 04:09 PM
Did you try bitsavers.org? There might be something there:

http://www.bitsavers.org/pdf/burroughs/

--T