Talk:TOPS-10

Requested move

The system was far better known as DECsystem-10 than by the TOPS-10 name. The only people who used the TOPS-10 name were sysadmins (we called them system programmers back then) and some others who were actively involved with the product. Most users and other people who were familiar with it called it DECsystem-10 or by shorthand, DEC-10. --rogerd 21:18, 28 May 2006 (UTC)

Disagree. It's not that important, but when I was using the system as a student at CalTech, the OS was known as TOPS-10. — Arthur Rubin | (talk) 14:02, 4 June 2006 (UTC)

Not moved as no consensus. —Nightstallion (?) 10:34, 6 June 2006 (UTC)

It would be helpful to have "DEC-10" at least forward to "DECsystem-10", since that was a common term for the system. I'd say it would be beneficial to keep both "TOPS-10" (refering to the operating system) and "DECsystem-10" (the hardware or overall system).

Also, this dude's page has lots of links for the DEC-10:

Joe Smith's PDP-10 page User: Anonymous 16 December 2006

Disagree. Initial poster in this section was confused. PDP-10/DEC-10/DECSystem-10 all refer to *hardware*. Tops-10 was the name of an operating system that ran on PDP-10 hardware.(I worked with PDP-10 KI and KL hardware back in 1977/78.)Gcronau (talk) 12:37, 24 October 2015 (UTC)

Disagree , very strongly, per Gcronau. Tops-10 is an operating system (and is what the article is about), PDP-10 or DECsystem-10 (or just "the ten") is the machine (and this article is not about the machine, we have a separate article for that). The first sentence of the article makes this clear. Jeh (talk) 18:35, 24 October 2015 (UTC)

Nomenclature

My twopence-worth: the PDP10 CPU itself was a KA10, KI10, KL10 or KS10 (in chronological order). The hardware system collectively was a DECsystem-10 (often colloquially shortened to DEC-10 or even '-10). The DEC-supplied operating system collectively with the system software (the software environment, I suppose one could call it) was TOPS-10; however, the *actual* OS kernel itself was called Monitor, which reached version 7.04 by the end of the product development line.

Other operating systems were developed for and ran on the DEC-10; BBN's Tenex and MIT's ITS, for example, though Tenex required the addition of non-DEC paging hardware if I remember correctly, at least at first. Kay Dekker 16:45, 13 May 2007 (UTC)

I would disagree with that assessment, on 2 counts. While the KA/KI/KL/KS designations referred to the specific CPU, the "PDP-10" designation referred to more that the CPU itself. PDP-10 was the name of the entire computer, just as PDP-8 was, and PDP-11 would be. The DECSystem-10 name was something that came later, and more from the marketing department, and was the name given to an entire system that included all the I/O devices bundled together with the PDP-10 "computer" to make a complete "system".

Also, the original OS was simply called "the Monitor". Tops-10 was not something that ran on top the Monitor. Rather Tops-10 can be thought of as the product that the monitor evolved into. Or just simply a splashy new name for the monitor that the marketing department came up with starting at around version 5 of the monitor. I disagree with the concept that only the kernel was called the monitor. Mainly because, before version 5, the entire OS was called "the monitor". Are you implying that no "system software" existed before version 5? I highly doubt it. Gcronau (talk) 22:51, 24 October 2015 (UTC)

Comment on Nomenclature

Hello all, I hope I'm not stepping on too many toes with my additions! I am about a half-serious hobbyists with TOPS-10, and wanted to provide enough info to get other people curious and started.

Once you've laid out the TOPS-10 manuals in chronological order, it becomes pretty apparent from the titles: DECsystem-10 was the name stressed in the 1970's, and TOPS-10 was more stressed by the late 1980's. I believe that was to distingush TOPS-10 from TOPS-20 and the other operating systems for the PDP-10.

Please help by commenting on my style, format, and content. TOPS-10 still lives, and I hope Wiki can provide initial information for new users.

Another note on nomenclature

An "operating system" implemented on a "particular CPU" is called a "platform". A "platform" is an important distinction from the point of view of implemented software: what services does the hardware provide, and how does the OS help me request those services?

So in that light, "PDP-10" is the CPU name (implemented as types KA, KI, KL, KS), "TOPS-10" is the operating system name (distinct from other OSes implemented on that CPU), and "DECsystem-1090" or "DECsystem-2020" and whatnot are the proper "platform names". A "DECsystem-2020" is TOPS-10 implemented on a KS-10 CPU, a "DECsystem-1090" is TOPS-10 on a KL-10 CPU, and so on.

There are a large number of TOPS-10 manuals at Bitsavers.org. Is it OK to reference them in the article?

I hope I'm not making you nuts with this discussion..

Cbmira01 02:45, 14 May 2007 (UTC)

Case distinctions to observe...

I know it may seem incredibly nitpicky, but the correct capitalisations are DECsystem-10 and DECSYSTEM-20. See DECSYSTEM-20 for the little nugget of history :) Kay Dekker 23:36, 28 May 2007 (UTC)

Historical Importance

It's too bad that this important operating system doesn't have a better article. DEC's operating systems formed the basis for concepts like CP/M, DOS, and even UNIX. The commercial minicomputer OS's were much simpler than mainframe systems like IBM's OS/360, and they were generally the center of most university student timesharing cultures in the 1970s. DonPMitchell (talk) 17:46, 18 November 2007 (UTC)

My Input

I was the systems programmer at Cerritos Collage in Norwalk Calif. from the time the DEC-SYSTEM-10 was installed.

The operating system was called TOPS-10. An I am fairly sure it was called that from around 1968 or 69. I took a TOPS-10 internals course in 1968. TOPS-10 is a multi user time sharing operating system with real time application capability.

The TOPS-10 operating system was configured for the number of tasks and high priority queues by editing assembly files that were assembled and linked into a core image file. Devices were also configured as well. A symbol file could also be generated that allowed for some patching and debugging of the running system. The debugger called DDT was the first dynamic debugger on a commercial system. DDT was short for Dynamic Debugging Technique. I would not say that CPM or DOS commands were modeled after TOPS-10. There closer to RSX-11M command structure. TOPS-10 commands were of the form:

output = input

The command:

pip xxx.yyy = abx.zzz

Would copy the abx.zzz file to xxx.yyy

pip was a general copy and transfer program. pip was short for Peripheral Interchange Program.

An interesting thing about TOPS-10 was that you had a memory image. You could run a program, Stop it or have it programed to halt. The memory image would be there at command level and it could be saved using a save command. Or memory examined and displayed bye commands at a user terminal. The idea was to give each user the same abilities as if he were setting at the console debugging a program stand alone. Each user had their own virtual machine. I wont to be clear that users could only access their own memory image. Unless you had privileges enabled on your account. TOPS-10 also had a UNIX like file protection system. That allowed programmer(owner), project and other access rights to files. Basically the user(owner), group, other of UNIX.

The DEC-10 was no minicomputer as I have seen claimed else ware here. At Ramada Inns we had as many as 80 or more users logged in at a time. PLUS running the Ramada reservation system that supported more the 1000 terminals in Inns across the country. The people I worked with at Ramada came from TWA whre they had 3 IBM 360 mainframes running full out to handle the transaction load we were doing on our reservation system using less then 10% of the duel KI10 processors. We had less then 3% downtime the first year running the reservation system. And that included PM down time. I do not know of any Monitor module being a part of TOPS-10. There were quite a few files that went onto the build. The most basic or heart of the operating system being the sched module. There was a sysmon program that ran and output statistics similar to systat. There was a later scheduler in the late 70s or early 80s that allowed control of task scheduling. It allowed job types: batch, terminal, user, and project to be divided into classes and the processor resource by % to be divided among the classes. We never used it at any place I worked. I think Monitor maybe was the control application. — Preceding unsigned comment added by Steamerandy (talk • contribs) 12:37, 29 July 2012 (UTC)

TOPS-10 and monitor came before Unix. You really can't say Tops-10 had any Unix-like features. Unix had Tops-10-like features. Its well before me but the [group,user] and [project,programmer] file system structure goes back to CTSS with "Problem numbers" being the basis of the project or group idea. — Preceding unsigned comment added by Sciencia61 (talk • contribs) 01:41, 8 July 2018 (UTC)

Minor disagreement regarding similarity between TOPS-10, CP/M and DOS. I do not dispute that DOS and TOPS-10 are different, e.g. DOS uses "copy source destination". However CP/M was closer to TOPS-10 in that it used "command destination = source" format. Ironically "PIP DEST.TXT = SOURCE.TXT" is a valid CP/M command that copies SOURCE.TXT to DEST.TXT. So not only did Kildall use the command structure of TOPS-10 when he wrote CP/M, he also borrowed a few command names verbatim. Citation: https://dfarq.homeip.net/origin-cpms-pip-command/ — Preceding unsigned comment added by 216.110.214.99 (talk) 13:57, 31 March 2024 (UTC)