Outdated CPUs

[ronsimpson]

Oldest.

I had the first "personal computer" in the dorm. Before the word "PC" befre "windows", before "MS DOS".
I had a 2mhz 8080 with 16k of memory and tape backup. I think it had about 0.3mips. I updated to a 4mhz z80 board and added memory boards to get 48k of memory. I had two 8 inch floppy drives. Wrote the interface to get the floppys working. My first hard drive was 5Mega. It held the operating system, word processor, spell checker, spreadsheet and all my files. There was mega(s) bytes left free. I could not think of how I could fill that hard drive. $500.00
Key words: S100 bus, 8088, CPM 1.4, It had rows of switches on the front panel. I have worked on older computers but that is my oldest one.

Now I have a dual core 700mhz computer in my shirt pocket.

 

[TheNewGuy]

Because we spend so much money on being safe and documenting it there is no budget left to do work with anything modern.

Honestly I don't know why we run such old computers other than we get the hand me downs from the main offices once they have become to old and glitchy for the executives to use.

Our most recent computer at lest got us into the world of windows XP now! Granted I still had to install service pack 3 myself after we got it.

At a certain high school back in 2010, they were using PC's with Pentium II's and Windows 95. They were using Visual Basic 6 for their programming class and a REALLY old version of 3D Studio Max for their animation class. It sucks that you guys only get hand-me-downs.

Hi again,


The oldest computer i ever used was the 8080 based board that i built a long long time ago. The one after that was the Sinclair ZX80, but the one i used the most (over 10 years or so) was the TRS80 (often called the "Trash-80") and that was Z80 based too. Nice thing about the TRS80 was that it was completely understandable from all the hardware to all the software including the operating system that came with it. It was so understandable that i was able to tweek the operating system itself when it became outdated around 1988...the date of the files would not accept anything over 1987 because it only used three bits for the year, so which allowed only the years 1980 to 1987...ha ha. But even with it's single color green screen, i used it for years...even though it only had two floppy drives (180k bytes each) and no hard drive...the hard drive for that thing was about 400 dollars (USD) for about 40 megabytes...ha ha.

I since then moved on to an 8088 based computer with 16 color monitor, then to an 80386 computer which i loved, then to 80486, then to a pentium based computer, then to a dual core system, then to a quad core system, then finally to an 8 core system which i use today. But dont forget about the Android tablets, they are pretty neat too and make portability a little simpler and more economical than a laptop.

Yes you can use C to program a micro controller chip, or BASIC, or Assembler. I use Assembler mostly as it makes me feel more in touch with the hardware.

I prefer lower level languages to heigher level ones for the same reason. You are more in touch with the hardware. That is important to me due to the fact that I am a Electrical Engineering major.

The New York Times had once an investigation of "recycling centers" in poor countries.

Many metals are recycled, which in itself, is a good thing...the bad thing is the job conditions and hazards in those places.

The other bad thing is that many ICs are removed, cleaned, and sold to the gray market...to unsuspecting buyers.

If the buyers understand that they are used IC's, that is one thing. But not telling them that it is used is just wrong. I would like to see that investigation.

Not in regular use but the simplest I ever worked on was a 4-bit Hitachi HMCS2.

It was a 4 bit processor though the instructions were 10 bits wide. This was 1982 and the company I worked for tendered for a contract to make a cheap, low-sensitivity digital radiation monitor for the government to be used after a nuclear attack. It became the PDRM82 - Google it.

This chip was chosen because it could directly drive an LCD and had just enough guts to do the processing necessary to scale the input from one of several different sensitivity grades of GM tube. And it was cheap. The pulse train from the small basic GM tube went through an 8 bit prescaler before going to the chip.

From a 'software' pov it was excruciating! A very basic instruction set, just crude assembler and no tools to speak of.

RAM consisted of a notional block of 160 locations I think, addressed with X and Y coordinates.
To get a 4 bit nibble of data into the accumulator it took three instructions:
1. load x-reg with part of RAM address
2. load y-reg with remainder of address
3. load data from indicated address into accumulator.

But it was exciting too and we got the contract to make 80,000 of them!

That is pretty cool, I'm going to research that. This device was supposed to be able to work even after a nuclear attack?

Oldest.

I had the first "personal computer" in the dorm. Before the word "PC" before "windows", before "MS DOS".
I had a 2mhz 8088 with 16k of memory and tape backup. I think it had about 0.3mips. I updated to a 4mhz z80 board and added memory boards to get 48k of memory. I had two 8 inch floppy drives. Wrote the interface to get the floppys working. My first hard drive was 5Mega. It held the operating system, word processor, spell checker, spreadsheet and all my files. There was mega(s) bytes left free. I could not think of how I could fill that hard drive. $500.00
Key words: S100 bus, 8088, CPM 1.4, It had rows of switches on the front panel. I have worked on older computers but that is my oldest one.

Now I have a dual core 700mhz computer in my shirt pocket.

I have a 1 Terabyte hard drive right now, and when I bought it I thought the same thing! "HOW can I EVER fill this thing up?!". I'm already almost 600 GB used up.

 

[tcmtech]

I have a 1 Terabyte hard drive right now, and when I bought it I thought the same thing! "HOW can I EVER fill this thing up?!". I'm already almost 600 GB used up.

I can recall the first hard drive of mine that failed. All I wanted was a 20 GB being I never filled up the old 15 GB that went bad.

Now 6 - 7 years later I am up around 10 TB of combined hard drive space, 4 something used plus a full backup, and thinking I may have add more in another year or two. Heck my music collection alone is nearing 150 GB.

 

[throbscottle]

Of course what really takes up the space is having a huge prawn collection...

Sorry, talk of hard disk space always makes me think of prawns for some reason!

 

[MrAl]

At a certain high school back in 2010, they were using PC's with Pentium II's and Windows 95. They were using Visual Basic 6 for their programming class and a REALLY old version of 3D Studio Max for their animation class. It sucks that you guys only get hand-me-downs.



I prefer lower level languages to heigher level ones for the same reason. You are more in touch with the hardware. That is important to me due to the fact that I am a Electrical Engineering major.



If the buyers understand that they are used IC's, that is one thing. But not telling them that it is used is just wrong. I would like to see that investigation.



That is pretty cool, I'm going to research that. This device was supposed to be able to work even after a nuclear attack?



I have a 1 Terabyte hard drive right now, and when I bought it I thought the same thing! "HOW can I EVER fill this thing up?!". I'm already almost 600 GB used up.

Hi again,

Oh EE major? Great to hear.

I was so used to using assembler anyway that the migration from other systems to uC was a breeze for me. Back when we had to make the CPU out of discrete chips we also had to assign codes for each type of instruction, and each instruction had to be painstakingly programmed into the ROM chip using a set of 8 switches for the data word. I later migrated to the Intel based CPU's, then to uC's. Out of all of them the uC's are *by far* the easiest to program in assembler.
Once you understand assembler for the uC's, you can then design your own higher level language too. The trick is to simply compile your new language code into assembler code, then program that into the uC. This makes it possible to design just about any language you care to come up with
The language i decided on was an improvement over assembler yet kept the same 'feel' of being in touch with the hardware. I then later added a debugging interface which allowed debugging in the language itself rather than in assembler which would then have been somewhat abstracted from the language itself so wouldnt be as nice. It doesnt take too much work to accomplish this either because some of the uC's have so few instructions in their instruction set.

 

[elfcurry]

That is pretty cool, I'm going to research that. This device was supposed to be able to work even after a nuclear attack?

Yes, I can't remember the details but that was one of the criteria for choosing that chip. I got the name wrong it was HMCS40 which you could get with a LCD drive variant. I'd like to see the instruction set to remember and wonder at the simplicity and difficulty of use.

The idea was that a network of Civil Defence workers would organise people in nuclear shelters and keep order and use radiation monitors such as PDRM82 to check when it was safe to emerge into the outside world. These were pretty insensitive instruments to measure moderate to high radiation levels and had an optional external sensor with cabling that plugged in to be used inside a shelter so you could check the outside without exposing yourself.

 

[TheNewGuy]

Of course what really takes up the space is having a huge prawn collection...

Sorry, talk of hard disk space always makes me think of prawns for some reason!

Prawn?

https://en.wikipedia.org/wiki/Prawn

 

[TheNewGuy]

Hi again,

Oh EE major? Great to hear.

I was so used to using assembler anyway that the migration from other systems to uC was a breeze for me. Back when we had to make the CPU out of discrete chips we also had to assign codes for each type of instruction, and each instruction had to be painstakingly programmed into the ROM chip using a set of 8 switches for the data word. I later migrated to the Intel based CPU's, then to uC's. Out of all of them the uC's are *by far* the easiest to program in assembler.
Once you understand assembler for the uC's, you can then design your own higher level language too. The trick is to simply compile your new language code into assembler code, then program that into the uC. This makes it possible to design just about any language you care to come up with
The language i decided on was an improvement over assembler yet kept the same 'feel' of being in touch with the hardware. I then later added a debugging interface which allowed debugging in the language itself rather than in assembler which would then have been somewhat abstracted from the language itself so wouldnt be as nice. It doesnt take too much work to accomplish this either because some of the uC's have so few instructions in their instruction set.

How long would it take you using those 8 switches? Man that must have taken a lot of patience.

That is a pretty cool look at how assembly runs underneath the languages we use like C and C++. It's funny, today in class we happen to be talking about how the CPU runs our code that we write in C. He talked about exactly that, how things translate into the assembly language underneath.

Yes, I can't remember the details but that was one of the criteria for choosing that chip. I got the name wrong it was HMCS40 which you could get with a LCD drive variant. I'd like to see the instruction set to remember and wonder at the simplicity and difficulty of use.

The idea was that a network of Civil Defence workers would organise people in nuclear shelters and keep order and use radiation monitors such as PDRM82 to check when it was safe to emerge into the outside world. These were pretty insensitive instruments to measure moderate to high radiation levels and had an optional external sensor with cabling that plugged in to be used inside a shelter so you could check the outside without exposing yourself.

Ah ok, that makes sense. That must have been really cool to be able to work on something like that, and probably good to have on your resume. I'll try and see if I can find a page on it.

 

[throbscottle]

Prawn?

https://en.wikipedia.org/wiki/Prawn

You know, prawn videos, prawn photos, prawn magazines...

 

[TheNewGuy]

You know, prawn videos, prawn photos, prawn magazines...

They have videos, photos and magazines about crustaceans?

lol

 

[3v0]

I was wondering how a thread about something that is near impossible got so long

Oldest.

I had the first "personal computer" in the dorm. Before the word "PC" before "windows", before "MS DOS".
I had a 2mhz 8088 with 16k of memory and tape backup. I think it had about 0.3mips. I updated to a 4mhz z80 board and added memory boards to get 48k of memory. I had two 8 inch floppy drives. Wrote the interface to get the floppys working. ....

Pre IBM PC times. I loved it. I was in HS when Popular Electronics came out with the first computer kit. I first used main frames and mini computers. The first I owned was an H-11. The H-11 got me back into collage so it was a good investment. I still recall the day I soldered the backplane. My soldering iron was 40W because I did not know better. But it let me do each joint quickly and move on to the next. It was very dependable.

 

[ronsimpson]

My first looked like this with out the floppys. Then I added the 8 inch floppys and then the 5 inch hard drive.

There is my first "cpu" board with no memory. I got memory 16K at a time in boards this size. Had 3 boards. Serial and parallel was on one board. Floppy controller was a board. Hard drive was a board. I think I had 10 boards at the end.

 

[3v0]

Neat stuff thanks for sharing. I wish I had some images of our old TRS80 with a home brew expansion interface.

 

[cowboybob]

Went to work for Apple in 82' (sales). Got a ||+ with 2 5.25 drives, lower case chip and a B&G monitor. Came with Basic and assembler built-in.

Started the family out with Zork and Snakebyte (after my wife asked "Why did you bring THAT home...). Now, of course, it was her idea to get it .

Later (84') got one of the first Macs plus ImageWriter printer, Multiplan, MacPaint, Draw,Write,Project all for $1,000. Helluva deal at the time.

 

[MrAl]

How long would it take you using those 8 switches? Man that must have taken a lot of patience.

That is a pretty cool look at how assembly runs underneath the languages we use like C and C++. It's funny, today in class we happen to be talking about how the CPU runs our code that we write in C. He talked about exactly that, how things translate into the assembly language underneath.



Ah ok, that makes sense. That must have been really cool to be able to work on something like that, and probably good to have on your resume. I'll try and see if I can find a page on it.

Hi,


It took SEVERAL hours to program the ROM doing it that way, but we had no choice because programmers (boards or whatever) where not around yet.
Add to that misery, when the head engineer ordered the new ROMs, he ordered the wrong type and no one knew it until we programmed one (taking maybe four hours) and found that the bits were flipping after it was fully programmed. We had to experiment for a few more hours to find out what was going on after which we finally determined that the problem was the ROM chips needed a different programming pulse than the old ones did. All said and done, we wasted at least two days fooling around with this alone.

 

[unclejed613]

first book i ever saw about programming was an IBM mainframe oriented textbook. you looked up the instruction you wanted, a chart told you the binary for that instruction and you punched a hollerith card with the instruction and data for that instruction. i don't remember what IBM system that book was for, but it was very interesting, especially for a 10yr old....

 

[ronsimpson]

How long would it take you using those 8 switches?

By the switches A=1001 1101 1010 1111 Instruction = 1101 0011 Push the store button.
Then in your head add one to the A and flip the switches.
A=1001 1101 1011 0000 Instruction = 0010 1010
and pray you did not add wrong or get off a line on the paper copy.

H......O...............W..........................L.........................................O.................................N................G........?

 

[Toe Cutter]

Hello ETO,
I have a bunch of old CPUs I've been collecting from junked computers that I've been taking to the recycling centers. I have them on my desk, and they are giving me these googley eyes!

To name a few:
Intel Celeron
Intel Pentium III
AMD Athlon 1100

I just wish I could use them for something! Maybe a small embedded system? I'm guessing though they would require a lot of supporting circuitry to get them to do anything though.

EDIT: Thanks to everyone who responded!

I've considered the same question before. I have some high power pulse lasers. I was thinking of making a time-of-flight range finder, but I would need a high speed counter. Time domain reflectometry would be another application that could use a high speed counter. I'm sure there is ECL logic and other specialized counters. Then I thought of an old Pentium 4 with a 3 ghz clock. A Pentium cpu can count. But of course there are all the problems of operating a complex cpu of this type that others have mentioned. With a 3 ghz counter I could get 5 centimeter resolution.

I bet someone who is motivated could hack something together, but I'm not that person. How does a modern CPU bootstrap? Modern BIOS is stored on a serial Rom/Flash chip. If I remember, intel chips start executing in real mode at address F000:FFF0. Power supply might be difficult, 1.5-1.8 volts at 50 amps? I wonder if you could run it off of a 'D' cell battery. LOL! It would be interesting to see a bare minimum CPU system.