Electronics > Projects, Designs, and Technical Stuff
Experimenting with TTL Cpu, 74LS chips, old vs New? Retro style switches?
rwgast_lowlevellogicdesin:
Just out of curiosity, if most of the time you just did manual entry on the front panel in order to tell the computer to load off paper tape or whatever input device you were using; wouldn't it have been easier and cheaper in labor to build indavidual diode based eproms with loading instructions for each input device and just insert that in a socket. Seems a lot faster than Dicking around with switches with every new software load, or no? Of course the computer would have had to be able to except the diode card but I'm sure modification or manufacturing in that feature wouldn't be a huge issue.
rstofer:
--- Quote from: duak on June 06, 2020, 05:18:08 pm ---Regarding the IBM PC being the first computer without switches and lighs. The Apple II, Commodore PET and TRS-80 all preceeded the PC starting in 1977.
--- End quote ---
Obviously, you are correct. All of these were 'personal' computers as opposed to 'commercial' or 'hobby' computers. By 'commercial', I'm thinking about machines like the PDP-11, not Vaxen.
My beginnings with the 'hobby' computer were with the Altair 8800. It certainly required a bit of toggling until EPROMs were used. Even then, we had to set the starting address to force the CPU to the beginning of the EPROM. Then we had sophistications like mapping and it became possible to just use the Reset switch and start from address 0000h
Those were good days! A mere mortal could understand every aspect of the hardware and software.
rstofer:
--- Quote from: rwgast_lowlevellogicdesin on June 06, 2020, 06:09:26 pm ---Just out of curiosity, if most of the time you just did manual entry on the front panel in order to tell the computer to load off paper tape or whatever input device you were using; wouldn't it have been easier and cheaper in labor to build indavidual diode based eproms with loading instructions for each input device and just insert that in a socket. Seems a lot faster than Dicking around with switches with every new software load, or no? Of course the computer would have had to be able to except the diode card but I'm sure modification or manufacturing in that feature wouldn't be a huge issue.
--- End quote ---
The IBM 1130 had a coldstart hardware arrangement whereby it would read a cold start card from the card reader, unpack the code into low RAM (from 0000h) and execute it. There were various coldstart cards and the scheme was also used to load diagnostics.
I don't know how the paper tape version worked but I suspect there was a cold start tape. The machine I used (circa '70) used the card reader approach.
We had a lot of lights (on the order of 160) along with toggle switches and a bouncing ball typewriter at the console. The typewriter might not be used very often because scientific jobs tended to be batch oriented with little to no operator intervention other than to confirm the paper was aligned on the plotter if the job required plotter output.
It was possible to use the switches and a big rotary selector switch to enter programs and do various debugging kinds of things. I didn't play with that feature, I was a 'guest' user and I didn't want to rock the boat.
This image gives a feel for the console lights but is only presented because all of the photos I can find are even less helpful:
http://www.ibm1130.net/functional/Console.html#figure25
Here's a survivor:
http://computermuseum.informatik.uni-stuttgart.de/dev/ibm1130/ibm1130.html
Only gurus did reboots on minicomputers (PDP-11, again) and they weren't about to abandon their switches. Knowing the coldstart code was a right of passage. They most certainly did not have to refer to the crib sheets taped to the cabinet.
rstofer:
I miss the toggle switches and blinking lights. That's why I have a couple of the PiDP-11 computers running BSD2.11 and a bit of a web server. There's something comforting about knowing when your program runs off the rails by watching the lights.
https://obsolescence.wixsite.com/obsolescence/pidp-11-overview
The fun bit is using the original Unix tools with the original C compiler. OK, the editor is a PITA but if you don't know vi, you don't know much about computers. Emacs always seemed like too much effort... Real K&R C, not this modern rubbish!
Note that BSD2.11 is very nearly identical to the more recent BSD4.3. This isn't some stripped down, obsolete, OS. I was never a PDP-11 user so I have a LOT to learn.
rstofer:
The PDP11-70 had boot code in ROM(s). All the user had to do was set a start address and load the boot device address in the console switches. Press Start and it was off to the races.
Figure 3-9
http://www.bitsavers.org/www.computer.museum.uq.edu.au/pdf/EK-11070-MM-002%20PDP11-70%20Maintenance%20And%20Installation%20Manual.pdf
This is similar to the cold start for the S-100 machines (like the Altair 8800) once we got more sophisticated memory boards. Examine 0xF000 (or wherever) and push Run. Magic commences now!
Remember, the 2102 RAM chip (1k x 1bit) was king. A 64k machine generated a LOT of heat and the power supply on the Altair was totally inadequate.
https://www.nteinc.com/specs/2100to2199/pdf/nte2102.pdf
350 ns access time was described as high speed!
Those were good days but things are a lot simpler today.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version