16-Bit CPUs on the S-100 bus

General discussions related to the Altair 8800 Clone

16-Bit CPUs on the S-100 bus

Postby MajorClements » September 26th, 2022, 11:01 am

From what I’ve read, Intel 8086 cards became available on the bus in the late 1970s. If Front Panel Input had survived up to that point, would a system with 16 data switches have had to be created? Alternatively I guess you could store the lower order 8 bits of machine instruction then add the next 8 using registers. As far as I can see online, there has not been any real examples of such a setup, unless I am looking in the wrong places.
Sincerely,
Daniel
________________________________________________________________________________
“THINK!”
If you had to remember only one thing about me, just know that I’m the guy with too many ThinkPads to count.
MajorClements
 
Posts: 18
Joined: January 30th, 2014, 9:18 am
Location: McLean, Virginia

Re: 16-Bit CPUs on the S-100 bus

Postby Wayne Parham » September 26th, 2022, 7:40 pm

Look at the front-panels of 16-bit minicomputers. Take the Data General Eclipse S/130, for example:

Data_General_Eclipse_S130.jpg
Data General Eclipse S/130

The way you used its front-panel was similar to the Altair. And in fact, the Altair was patterned after the Data General systems. It was actually patterned after the Data General Nova 2, but the 16-bit Eclipse's had a very similar front-panel.

As an example, to set memory location 000100 octal to 177777, you would do this:

- Set switches to 0 000 000 001 000 000 (000100 octal)
- Toggle examine. Address lights will now show 000100 octal.
- Set the switches to 1 111 111 111 111 111 (177777 octal)
- Toggle deposit. Data lights will now show 177777 octal.

So you can see it's pretty similar to the Altair.

You could also use the front-panel to set the boot device code of a storage subsystem you wanted to boot from. So, for example, to boot the tape drive, you set the front-panel switches to 22 octal and then toggled program load.

Sadly - at least to me - by the 1980s, computer manufacturers were abandoning front-panels because of the cost. They replaced them with ROM monitors that would expect commands from the "master console." This was true for both microcomputers and minicomputers. Data General called this a "soft console" and it was used on the Nova 4, the MP/100 and MP/200 microNovas and all their 32-bit Eclipse systems. So I think probably MITS would have abandoned their front-panel too.
Wayne Parham
 
Posts: 240
Joined: March 18th, 2022, 3:01 pm

Re: 16-Bit CPUs on the S-100 bus

Postby toml_12953 » September 26th, 2022, 8:18 pm

Wayne Parham wrote:Sadly - at least to me - by the 1980s, computer manufacturers were abandoning front-panels because of the cost. They replaced them with ROM monitors that would expect commands from the "master console." This was true for both microcomputers and minicomputers. Data General called this a "soft console" and it was used on the Nova 4, the MP/100 and MP/200 microNovas and all their 32-bit Eclipse systems. So I think probably MITS would have abandoned their front-panel too.


After MITS sold the business to Pertec, the front panel did disappear and only the turnkey system was ultimately available.
toml_12953
 
Posts: 297
Joined: June 7th, 2013, 12:54 pm

Re: 16-Bit CPUs on the S-100 bus

Postby KenF » September 27th, 2022, 8:45 am

Wayne Parham wrote:Sadly - at least to me - by the 1980s, computer manufacturers were abandoning front-panels because of the cost


Not only because of cost and complexity, but unless a programmer is using assembler, masses of register lights have little use. If you use a checkpoint stop in any high level language, the values in registers is anybody's guess, and probably useless unless you wrote the compiler. When assembler began to disappear (coinciding with the increase of available ram), so did the lights. The IBM 434x series of mainframes, coming after the massive lamp panels of the systems 360 and 370, had little more than a power light and a few status bulbs. Most people coming into a data center thought it was a table. Ditto for other manufacturers.

Only Hollywood requires masses of blinking lights on a starship to give notice that this is a computer.

Of course, as hobbyists, we love them and the more the better.
KenF
 
Posts: 30
Joined: April 25th, 2022, 11:13 am

Re: 16-Bit CPUs on the S-100 bus

Postby MajorClements » October 9th, 2022, 2:44 pm

Wayne Parham wrote:Look at the front-panels of 16-bit minicomputers. Take the Data General Eclipse S/130, for example:

Data_General_Eclipse_S130.jpg

The way you used its front-panel was similar to the Altair. And in fact, the Altair was patterned after the Data General systems. It was actually patterned after the Data General Nova 2, but the 16-bit Eclipse's had a very similar front-panel.

As an example, to set memory location 000100 octal to 177777, you would do this:

- Set switches to 0 000 000 001 000 000 (000100 octal)
- Toggle examine. Address lights will now show 000100 octal.
- Set the switches to 1 111 111 111 111 111 (177777 octal)
- Toggle deposit. Data lights will now show 177777 octal.

So you can see it's pretty similar to the Altair.

You could also use the front-panel to set the boot device code of a storage subsystem you wanted to boot from. So, for example, to boot the tape drive, you set the front-panel switches to 22 octal and then toggled program load.

Sadly - at least to me - by the 1980s, computer manufacturers were abandoning front-panels because of the cost. They replaced them with ROM monitors that would expect commands from the "master console." This was true for both microcomputers and minicomputers. Data General called this a "soft console" and it was used on the Nova 4, the MP/100 and MP/200 microNovas and all their 32-bit Eclipse systems. So I think probably MITS would have abandoned their front-panel too.


Interesting, it is weird to see that for real. Now thinking back on it, I believe I recall seeing a PDP-type system in a museum somewhere in Washington, DC with 16 Bit input, but that was 12 years ago, I am not sure now.

Honestly, I didn’t even know minicomputers existed. Looking at photos and documentations now, it looks to be in a similar in footprint to the ubiquitous server rack unit, but larger than a Altair, which we know it to be called a “microcomputer” at the time. I assume these were for medium to small businesses who needed that extra power from mainframes, but lacked either the room or cooling infrastructure for a room-sized one?

Nowadays, It seems that any thing with a “mini” in the title always seems to disappear first. USB 2.0, it was micro, mini, and type-A. SD cards: micro-, mini-, and typical-SD’s. HDMI: Micro, Mini, and Full-size. In retrospect, I remember seeing all 3 around at the same time, but was it more like the situation: a normal and small-size are created, then a newer extra small-size comes out later? That concept makes more sense then just making all three sizes at the same time.
Sincerely,
Daniel
________________________________________________________________________________
“THINK!”
If you had to remember only one thing about me, just know that I’m the guy with too many ThinkPads to count.
MajorClements
 
Posts: 18
Joined: January 30th, 2014, 9:18 am
Location: McLean, Virginia

Re: 16-Bit CPUs on the S-100 bus

Postby Wayne Parham » October 10th, 2022, 9:01 am

Minicomputers ran most businesses in the 1970s and into the 1980s. They appeared in the 1960s - even before ASCII was popular - and they really started disappearing pretty fast in the 1990s. But in the 1970s, they were everywhere. Minis were the computing standard of that era.

Most minis were designed to fit into EIA standard 19-inch racks. They also usually included peripheral subsystems that often fit into 19-inch racks too. Things like expansion chassis, tape drives and smaller disk drives fit into the same rack. That was most common back then. The only subsystems that didn't fit into those racks were line printers and (ten-platter) cartridge disks, which were about the same size as washing machines. But floppy disks, smaller (two-platter) cartridge disks and Winchester disk drive subsystems often were made to fit into 19-inch racks.

The Altair was a microprocessor implementation of what was commonly seen in minicomputers of that era, with front-panel switches and lights grouped in octal. I would assume that Ed Roberts intended the Altair to be a desktop system though, since he didn't use a cabinet that mounted in the same size 19-inch rack as the minis.
Wayne Parham
 
Posts: 240
Joined: March 18th, 2022, 3:01 pm

Re: 16-Bit CPUs on the S-100 bus

Postby MajorClements » October 10th, 2022, 9:47 am

Hah, I think that was what Pertec ended up with in the end, a no-frills, business-oriented computer: MITS 300/55 Business System. I guess it is neat that you get a hard drive, floppy, and Altair in one large rack (almost how PCs are configured today), but it's very much a far cry to Ed's vision. I appreciate his mindset, especially today: a free copy of the computer schematics in the Theory of Operations manual. No black boxes, proprietary ICs (aside from Intel), or components. Today, no company that has people paying 4 figures for a new cellphone would even do that. That said, I guess that "open-ended-ness" of the MITS was pretty much its own downfall: S-100 compatible clones came out quick, half-priced in some areas too. It's a shame.

Sorry for the tangent, what would have been the first true minicomputer from IBM? I also looked up the PDP-11, and it was exactly how I remembered. Very similar to that the Data Eclipse S/130. I really like those switch colors and designs, Red/Orange and Blue/White. I assume these were still very capable systems despite their mini-name. Also, to add to the removal of the front panel switches, I really appreciated the "Single-Step" switch on the Altair, especially from understanding the way the instructions execute. I haven't really used the Turnkey Module yet, does it have single-step support? (I cannot recall if the clone has that feature implemented)
Sincerely,
Daniel
________________________________________________________________________________
“THINK!”
If you had to remember only one thing about me, just know that I’m the guy with too many ThinkPads to count.
MajorClements
 
Posts: 18
Joined: January 30th, 2014, 9:18 am
Location: McLean, Virginia

Re: 16-Bit CPUs on the S-100 bus

Postby Wayne Parham » October 10th, 2022, 7:00 pm

IBM had the Series/1 and then the AS/400. They might have had other minis that completed with DG and DEC, but those are the ones I remember. Seems like all the IBM shops I knew used 'em.
Wayne Parham
 
Posts: 240
Joined: March 18th, 2022, 3:01 pm

Re: 16-Bit CPUs on the S-100 bus

Postby MajorClements » October 12th, 2022, 3:27 pm

Wayne Parham wrote:IBM had the Series/1 and then the AS/400. They might have had other minis that completed with DG and DEC, but those are the ones I remember. Seems like all the IBM shops I knew used 'em.


Awesome! I’m going to check them out. You mentioned IBM shops, were they authorized dealers to sell IBM systems or did IBM have physical brick and mortar stores for clients to checkout the latest and greatest?
Sincerely,
Daniel
________________________________________________________________________________
“THINK!”
If you had to remember only one thing about me, just know that I’m the guy with too many ThinkPads to count.
MajorClements
 
Posts: 18
Joined: January 30th, 2014, 9:18 am
Location: McLean, Virginia

Re: 16-Bit CPUs on the S-100 bus

Postby AltairClone » October 12th, 2022, 4:17 pm

Hi Daniel,

Saw your message on FB. I don’t really use that forum anymore. Reach out to me by email at https://altairclone.com/contact.html. Thanks!

Mike
AltairClone
Site Admin
 
Posts: 632
Joined: April 5th, 2013, 10:55 am

Next

Return to General Discussions

Who is online

Users browsing this forum: No registered users and 19 guests