Computer History Wiki - Recent changes [en]

EDVAC

Jnc — Tue, 15 Sep 2026 16:44:11 GMT

Fix grammar error; various clarifications

← Older revision		Revision as of 16:44, 15 September 2026
Line 5:		Line 5:
	==History==		==History==

	The EDVAC was conceived the University of Pennsylvania's Moore School in 1945, while work was still proceeding on the ENIAC~~; early~~ consideration of how the ENIAC would be re-configured for a new problem indicated that that would be somewhat difficult. The notion of programmability was known already from the ASCC, and the [[Bell Telephone Laboratories relay computing devices\|Relay Interpolator]], and that promised to speed up switching the machine to a different problem. However, the electro-mechanical devices used on those machines to read the program were an order of magnitude slower than the electronic [[arithmetic logic unit\|calculation subsystem]], as were similar devices used to hold intermediate results. Use of an electronic [[memory]] to hold both the program ''and'' intermediate results was the obvious answer.		The EDVAC was conceived at the University of Pennsylvania's Moore School in 1945, while work was still proceeding on the ENIAC. Early consideration of how the ENIAC would be re-configured for a new problem indicated that that would be somewhat difficult. The notion of programmability was known already from the ASCC, and the [[Bell Telephone Laboratories relay computing devices\|Relay Interpolator]], and that promised to speed up switching the machine to a different problem. However, the electro-mechanical devices used on those machines to read the program were an order of magnitude slower than the electronic [[arithmetic logic unit\|calculation subsystem]], as were similar devices used to hold intermediate results. Use of an electronic [[memory]] to hold both the program ''and'' intermediate results was the obvious answer.

	[[John Mauchly]] and [[J. Presper Eckert]] had already begun to consider this change in direction, when [[Herman Goldstine]], the head of the project, introduced the mathematician [[John von Neumann]] (who was familiar with the idea of programs, from his knowledge of [[Alan Turing]]'s work on computability) to them. After further discussion, they rapidly came to rough agreement on the new approach, which almost every computer from that day forward has followed: a memory which merely stores things (unlike the accumulators of the ENIAC, which also contained limited computational capability), and a [[arithmetic logic unit\|~~separate subsystem~~]] which performs all the calculations. Von Neumann wrote up the results in his famous rough draft ''First Draft of a Report on the EDVAC'' - without noting who had contributed what, thereby igniting a controversy that was never really definitively settled.		[[John Mauchly]] and [[J. Presper Eckert]] had already begun to consider this change in direction, when [[Herman Goldstine]], the head of the project, introduced the mathematician [[John von Neumann]] (who was familiar with the idea of programs, from his knowledge of [[Alan Turing]]'s work on computability) to them. After further discussion, they rapidly came to rough agreement on the new approach, which almost every computer from that day forward has followed: a memory which merely stores things (unlike the accumulators of the ENIAC, which also contained limited computational capability), and a separate subsystem the [[arithmetic logic unit\|ALU]], which performs all the calculations. Von Neumann wrote up the results in his famous rough draft ''First Draft of a Report on the EDVAC'' - without noting who had contributed what, thereby igniting a controversy that was never really definitively settled.

	The First Draft, dated June 30, 1945, was circulated fairy widely in that form by Goldstine, and was the inspiration for most of the first generation of computers; almost all computers now are descendants. The EDVAC itself was not very influential: Mauchly and Eckert left the Moore School shortly ~~thereafter~~, over a dispute about patents; von Neumann moved to the [[Institute for Advanced Study]] to build his [[IAS computer]]. Their loss slowed down the project so much that the machine was not finally operational until 1952.		The First Draft, dated June 30, 1945, was circulated fairy widely in that form by Goldstine, and was the inspiration for most of the first generation of computers; almost all computers now are descendants. The EDVAC itself was not very influential: Mauchly and Eckert left the Moore School shortly after the First Draft came out, over a dispute with the Moore School about patents; von Neumann moved to the [[Institute for Advanced Study]] to build his [[IAS computer]]. Their loss slowed down the project so much that the machine was not finally operational until 1952.

	==Details==		==Details==

Maurice Wilkes

Jnc — Tue, 15 Sep 2026 16:32:28 GMT

Add his two books

← Older revision		Revision as of 16:32, 15 September 2026
Line 1:		Line 1:
	'''Maurice V. Wilkes''' was a British computer scientist, in 1951 ~~was~~ Director of the Mathematical Laboratory of the University of Cambridge. Among his accomplishments were [[EDSAC]] (the first usable stored-[[program]] [[computer]]) and popularizing the use of [[microcode]].		'''Maurice V. Wilkes''' was a British computer scientist; in 1951, Director of the Mathematical Laboratory of the University of Cambridge. Among his accomplishments were leading the creation of [[EDSAC]] (the first usable stored-[[program]] [[computer]]), and popularizing the use of [[microcode]].

	He has written many articles and books, among them:		He has written many articles and books, among them:
Line 8:		Line 8:

	{{semi-stub}}		{{semi-stub}}

			==Further reading==

			* Maurice V. Wilkes, ''Memoirs of a Computer Pioneer'', MIT Press, Cambridge, 1985 - autobiography
			* Maurice V. Wilkes, ''Computing Perspectives'', Morgan Kaufmann, San Francisco, 1995 - essays on various topics

	==External links==		==External links==

Multics

Jnc — Tue, 15 Sep 2026 08:27:36 GMT

Fix minor unintentional error caused by loose wording

← Older revision		Revision as of 08:27, 15 September 2026
Line 10:		Line 10:
	\| architecture = [[GE-645]] initially, various [[Honeywell 6000 series\|Honeywell 6000]] models later		\| architecture = [[GE-645]] initially, various [[Honeywell 6000 series\|Honeywell 6000]] models later
	}}		}}
			'''Multics''' (''Multi''plexed ''I''nformation and ''C''omputing ''S''ervice - although the short form is ''never'' spelled in all capitals) was an early [[time-sharing]] [[operating system]] which has influenced almost every operating system created since; especially via [[UNIX]] (the name of which is a play on 'Multics'), which was created by [[Bell Laboratories\|Bell Labs]] personnel who had worked on Multics, after Bell Labs pulled out of the Multics project.
	Multics (''Multi''plexed ''I''nformation and ''C''omputing ''S''ervice - although the short form is ''never'' spelled in all capitals) was an early [[time-sharing]] [[operating system]] which has influenced almost every operating system created since; especially via [[UNIX]] (the name of which is a play on 'Multics'), which was created by [[Bell Laboratories\|Bell Labs]] personnel who had worked on Multics, after Bell Labs pulled out of the Multics project.

	Multics was intended as an ''information utility''; i.e. a service to which those who wanted information processing would connect, much as people connect to the electric grid for power.		Multics was intended as an ''information utility''; i.e. a service to which those who wanted information processing would connect, much as people connect to the electric grid for power.
Line 43:		Line 42:
	Multics ran only on special hardware, which provided hardware support for its single-level store [[architecture]].		Multics ran only on special hardware, which provided hardware support for its single-level store [[architecture]].

	It initially ran on the [[GE 645]], a modified version of the [[GE 635]]. After GE was bought by Honeywell, a number of models of the [[Honeywell 6000 series]] systems were produced to run Multics on.		It initially ran on the [[GE 645]], a modified version of the [[GE 635]]. After [[General Electric\|GE]]'s computer business was bought by [[Honeywell]], a number of models of the [[Honeywell 6000 series]] systems were produced to run Multics on.

	== Multics today ==		== Multics today ==

	Multics is dead. There are no systems running it today (the last one was ~~[http://slashdot.org/articles/00/11/13/066228.shtml~~ shut down in 2000]).		Multics is dead. There are no systems running it today (the last one was shut down in 2000).

	There is however now a working [[SIMH]]-based emulator for the Multics DPS-8/M hardware, and with most of the source code still extant, it is now possible to experience the Multics computing environment.		There is however now a working [[SIMH]]-based emulator for the Multics DPS-8/M hardware, and with most of the source code still extant, it is now possible to experience the Multics computing environment.
Line 98:		Line 97:
	* [https://softpanorama.org/History/multics.shtml Multix OS -- Pioneer of time sharing]		* [https://softpanorama.org/History/multics.shtml Multix OS -- Pioneer of time sharing]
	* [https://www.youtube.com/watch?v=jni7wk7bjxA Multics booting] - A video of a H6180 CPU front panel hooked up to a SIMH H6180 emulation booting Multics		* [https://www.youtube.com/watch?v=jni7wk7bjxA Multics booting] - A video of a H6180 CPU front panel hooked up to a SIMH H6180 emulation booting Multics
			* [http://slashdot.org/articles/00/11/13/066228.shtml The Last Multics System Decommissioned]
			** [https://tech.slashdot.org/story/17/07/09/0414245/48-year-old-multics-operating-system-resurrected 48-Year-Old Multics Operating System Resurrected]

	[[Category: Operating Systems]]		[[Category: Operating Systems]]

Notable early computing devices

Jnc — Tue, 15 Sep 2026 07:58:40 GMT

ABC most notable for being the first electronic digital computing device; add a link

← Older revision		Revision as of 07:58, 15 September 2026
Line 13:		Line 13:
	==Electronic computing devices==		==Electronic computing devices==

	At about the same time as the ASCC, in 1938, [[John Vincent Atanasoff]] took the next step, and began the creation of the first [[electronic]] digital computing device, later called the [[Atanasoff-Berry Computer]]. It was not at all programmable or general-purpose; it could only solve systems of simultaneous equations. It was notable for having probably inspired [[John Mauchly]] to use a digital electronic approach when he later set out to build the [[ENIAC]]; and ~~also~~ for being the first digital electronic computing device to use [[capacitor]]-based memory - an approach to memory now ubiquitous in the [[Dynamic RAM]] of all modern computers.		At about the same time as the ASCC, in 1938, [[John Vincent Atanasoff]] took the next step, and began the creation of the first [[electronic]] digital computing device, later called the [[Atanasoff-Berry Computer]]. It was not at all programmable or general-purpose; it could only solve systems of simultaneous equations. It was also notable for having probably inspired [[John Mauchly]] to use a digital electronic approach when he later set out to build the [[ENIAC]]; and for being the first digital electronic computing device to use [[capacitor]]-based memory - an approach to memory now ubiquitous in the [[Dynamic RAM]] of all modern computers.

	The [[Colossus]] digital electronic code-breaking devices of [[Tommy Flowers]], built in the UK during World War Two (construction of the first one started in 1943), were notable because they showed that large electronic computing devices could be made to operate reliably; because quite a few were produced; and because many of the post-War computer pioneers in the UK learned about the suitability of electronics for digital computing devices from them. They are often described as 'programmable', but this is incorrect - they had no program of any form, and they were not general-purpose; a better description is 'configurable'.		The [[Colossus]] digital electronic code-breaking devices of [[Tommy Flowers]], built in the UK during World War Two (construction of the first one started in 1943), were notable because they showed widely that large electronic computing devices could be made to operate reliably; because quite a few were produced; and because many of the post-War computer pioneers in the UK learned about the suitability of electronics for digital computing devices from them. They are often described as 'programmable', but this is incorrect - they had no program of any form, and they were not general-purpose; a better description is 'configurable'.

	Effectively the last step before true computers was the ENIAC, a large digital electronic computing device. As originally designed, it was only configurable, requiring considerable effort to re-configure it to perform a different computation (although it was general-purpose, unlike the Atanasoff–Berry Computer and Colossus). In 1947 an effort was started to re-configure it in a way that added a certain amount of programmability, via a program stored in the 'function-table switches' (originally intended as a ROM data source). Later, minor [[hardware]] modifications improved ENIACs efficiency when configured as a stored-program machine; it started to operate in that mode in 1948.		Effectively the last step before true computers was the ENIAC, a large digital electronic computing device. As originally designed, it was only configurable, requiring considerable effort to re-configure it to perform a different computation (although it was general-purpose, unlike the Atanasoff–Berry Computer and Colossus). In 1947 an effort was started to re-configure it in a way that added a certain amount of programmability, via a program stored in the 'function-table switches' (originally intended as a [[read-only memory\|ROM]] data source). Later, minor [[hardware]] modifications improved ENIACs efficiency when configured as a stored-program machine; it started to operate in that mode in 1948.

	==Further reading==		==Further reading==

Program

Jnc — Tue, 15 Sep 2026 07:33:59 GMT

Fix garble; minor language improvements

← Older revision		Revision as of 07:33, 15 September 2026
Line 3:		Line 3:
	* [[expression]]s to perform arithmetical and/or logical computations		* [[expression]]s to perform arithmetical and/or logical computations
	* directives to move data from one place to another		* directives to move data from one place to another
	* [[control flow]] mandates, including [[conditional branch]]es and [[subroutine]] calls		* [[control flow]] mandates, including [[jump]]s, [[conditional branch]]es, and [[subroutine]] calls

	A program almost always exists in two complementary forms:		A program almost always exists in two complementary forms:

	* as [[object code]] (often referred to as 'binary'), a collection of [[instruction]]s loaded into the [[main memory]] of the [[computer]], from where the [[Central Processing Unit\|CPU]] will [[fetch]] the individual instructions of the program, and [[execute]] them;		* as [[object code]] (often referred to as 'binary'); a collection of [[instruction]]s loaded into the [[main memory]] of the [[computer]], from where the [[Central Processing Unit\|CPU]] will [[fetch]] the individual instructions of the program, and [[execute]] them;
	* the [[source code]]; the expression, in either [[assembly language]] or a higher-level [[programming language]] (forms more suited for ~~humans to~~ humans to work with), of the steps in the program.		* the [[source code]]; the expression, in either [[assembly language]] or a higher-level [[programming language]] (forms more suited for humans to work with), of the steps in the program.

	Programs, almost all originally created in the latter form, are converted from their original human-readable source code, to the object code form needed by an actual computer, by tools such as [[assembly language\|assembler]]s and [[compiler]]s.		Programs, almost all originally created in the latter form, are converted from their original human-readable source code, to the object code form needed by an actual computer, by tools such as [[assembly language\|assembler]]s and [[compiler]]s.
Line 14:		Line 14:
	==Early programs==		==Early programs==

	In the context of early [[computing device]]s, it must be emphasized that a program was a specified series of elementary steps, which were stored in some mechanism from which they were fed into the device; not simply the ability to make ~~physical~~ changes to the device, to change what calculation it ~~performs~~ - even if those changes were instantiated in something that was added to the device for the accomplishment of that exact purpose, like the [[plug-board]]s of the [[ENIAC]]. Such 'customizability' is better referred to as 'configurability', and ''not'' 'programmability'.		In the context of early [[computing device]]s, it must be emphasized that a program was a specified series of elementary steps, which were stored in some mechanism from which they were fed into the device. They were ''not'' simply the ability to make changes to the device, to change what calculation it performed - even if those changes were instantiated in something that was added to the device for the accomplishment of that exact purpose (like the [[plug-board]]s of the [[ENIAC]]). Such 'customizability' is better referred to as 'configurability', and ''not'' 'programmability'.

	==Further reading==		==Further reading==
Line 20:		Line 20:
	* Nicholas Metropolis, J. Worlton, [https://www.osti.gov/servlets/purl/4628499 ''A Trilogy On Errors in the History of Computing''], Los Alamos National Laboratory, Los Alamos, 1972 - contains an in-depth section on the genesis of the 'stored program' idea, by someone who lived through the end of that era, and knew those who did the work		* Nicholas Metropolis, J. Worlton, [https://www.osti.gov/servlets/purl/4628499 ''A Trilogy On Errors in the History of Computing''], Los Alamos National Laboratory, Los Alamos, 1972 - contains an in-depth section on the genesis of the 'stored program' idea, by someone who lived through the end of that era, and knew those who did the work
	<!-- https://www.osti.gov/biblio/4628499 -->		<!-- https://www.osti.gov/biblio/4628499 -->

	~~{{semi-stub}}~~

	[[Category: Basics]]		[[Category: Basics]]

PDP-10

Larsbrinkhoff — Mon, 14 Sep 2026 10:13:48 GMT

Hobbyist recreations: Gary Solomon's "KI10"

← Older revision		Revision as of 10:13, 14 September 2026
Line 93:		Line 93:
	* David Bridgham: [http://pdp10.froghouse.org/ KV10 (in progress)]		* David Bridgham: [http://pdp10.froghouse.org/ KV10 (in progress)]
	* Angelo Papenhoff: [https://github.com/aap/fpdpga FPDPGA], PDP-6 and KA10 FPGA.		* Angelo Papenhoff: [https://github.com/aap/fpdpga FPDPGA], PDP-6 and KA10 FPGA.
			* Gary Solomon: [https://garysolomon199.substack.com/ A KI10-like computer]

	==Software simulators==		==Software simulators==

General Comprehensive Operating System

Jnc — Sun, 13 Sep 2026 21:26:11 GMT

Better than a stub - but not by much

New page

The '''General Comprehensive Operating System''' (usually given as the acronym, '''GCOS''') was the most common [[operating system]] for [[Honeywell 6000 series]] [[mainframe]]s and descendants, from [[Honeywell]], and then [[Groupe Bull]] - themselves descendants of earlier machines ([[GE 600 series]]) from [[General Electric]] (GE). It was still in use recently (in its most recent versions - GCOS 7 and GCOS 8) on follow-on machines from Groupe Bull.

GCOS itself was an evolution of the earlier GECOS (General Electric Comprehensive Operating Supervisor) from GE, from before Honeywell bought GE's mainframe computer business, when GE decided to leave that market (although early Honeywell documentation continued to call it GECOS). Groupe Bull, in turn, inherited GCOS when Honeywell in turn decided to exit the large-scale computer business.

GCOS did mot initially support [[virtual memory]]; the machines on which it then ran only had [[base and bounds]] [[memory management]] [[hardware]].

It was originally developed by GE for the GE 600 series in 1962–1964; GE released GECOS I in April, 1965, on the [[GE-635]].

{{semi-stub}}

==External links==

* [https://bitsavers.org/pdf/ge/GE-6xx/GECOS/ GECOS] - for the GE-6xx
** [https://bitsavers.org/pdf/ge/GE-6xx/GECOS/CPB-1002_GECOS_Jul64.pdf GE-635 Comprehensive Operating Supervisor] (July 1964)
* [https://bitsavers.org/pdf/ge/GE-6xx/GECOS-III/ GECOS-III]
* [https://bitsavers.org/pdf/honeywell/large_systems/ Large Systems] - numerous GECOS sub-folders
* [https://multics-wiki.swenson.org/gcos GCOS] - at the Multics Wiki; contains much information about GCOS
* [http://www.feb-patrimoine.com/english/gecos_to_gcos8_part_2.htm From GECOS to GCOS8: Part II - The Honeywell years]

[[Category: Operating Systems]]

Honeywell 6000 series

Jnc — Sun, 13 Sep 2026 18:53:57 GMT

Have a target, now

← Older revision		Revision as of 18:53, 13 September 2026
Line 5:		Line 5:
	They were descendants of the [[GE 600 series]] family; after GE's computer business was sold to [[Honeywell]] in 1970, the 6000 series were Honeywell's replacements. They used [[Transistor-transistor logic\|TTL]] [[Integrated circuit\|SSI integrated circuits]] and larger [[printed circuit board]]s, unlike the older (and obsolescent) discrete [[transistor]] GE machines.		They were descendants of the [[GE 600 series]] family; after GE's computer business was sold to [[Honeywell]] in 1970, the 6000 series were Honeywell's replacements. They used [[Transistor-transistor logic\|TTL]] [[Integrated circuit\|SSI integrated circuits]] and larger [[printed circuit board]]s, unlike the older (and obsolescent) discrete [[transistor]] GE machines.

	The basic [[architecture]] of the GE and Honeywell series was the same: a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] units. All models used 36-bit [[word]]s, and almost identical [[instruction set]]s; Honeywell added an [[~~Commercial Instruction Set\|~~Extended Instruction Set]] (EIS) which provided string-manipulation instructions, [[binary-coded decimal\|packed decimal]] instructions, etc.		The basic [[architecture]] of the GE and Honeywell series was the same: a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] units. All models used 36-bit [[word]]s, and almost identical [[instruction set]]s; Honeywell added an [[Extended Instruction Set]] (EIS) which provided string-manipulation instructions, [[binary-coded decimal\|packed decimal]] instructions, etc.

	Most 6000 series machines ran [[GCOS]] (General Comprehensive Operating System), an evolution of the earlier GECOS (General Electric Comprehensive Operating Supervisor) - early Honeywell documentation continued to call it GECOS.		Most 6000 series machines ran [[GCOS]] (General Comprehensive Operating System), an evolution of the earlier GECOS (General Electric Comprehensive Operating Supervisor) - early Honeywell documentation continued to call it GECOS.

Extended Instruction Set

Jnc — Sun, 13 Sep 2026 18:52:44 GMT

Can have two meanings

← Older revision		Revision as of 18:52, 13 September 2026
Line 1:		Line 1:
	'''Extended Instruction Set''' was the name given to a number of optional additions, to different models from different manufacturers, to their [[Central Processing Unit\|CPUs]]; those added [[instruction]]s to the basic set provided by the CPU of the 'base' model of any line.		'''Extended Instruction Set''' was the name given to a number of optional additions, to different models from different manufacturers, to their [[Central Processing Unit\|CPUs]]; those added [[instruction]]s to the basic set provided by the CPU of the 'base' model of any line. The term could apply to either the formal specification of the added instructions, or the piece of physical [[hardware]] which implemented them.

	(This concept obviously only made sense when CPUs were physically large constructs, made up of many discrete components, such as [[transistor]]s or SSI [[integrated circuit\|chips]]. Only the very earliest [[microprocessor]]s had such options, often [[floating point]] arithmetic units.)		(This concept obviously only made sense when CPUs were physically large constructs, made up of many discrete components, such as [[transistor]]s or SSI [[integrated circuit\|chips]]. Now that CPUs are extremely complex monolithic constructs, the concept no longer makes any sense. Only the very earliest [[microprocessor]]s had such options, often [[floating point]] arithmetic units.)

	{{semi-stub}}		{{semi-stub}}

EIS

Jnc — Sun, 13 Sep 2026 18:45:11 GMT

Better target

← Older revision		Revision as of 18:45, 13 September 2026
Line 1:		Line 1:
	#Redirect [[~~PDP-11~~ Extended Instruction Set]]		#Redirect [[Extended Instruction Set]]

Extended Instruction Set

Jnc — Sun, 13 Sep 2026 18:44:30 GMT

Turn into a semi-stub

← Older revision		Revision as of 18:44, 13 September 2026
Line 1:		Line 1:
	~~#Redirect~~ [[Commercial Instruction Set]]		'''Extended Instruction Set''' was the name given to a number of optional additions, to different models from different manufacturers, to their [[Central Processing Unit\|CPUs]]; those added [[instruction]]s to the basic set provided by the CPU of the 'base' model of any line.

			(This concept obviously only made sense when CPUs were physically large constructs, made up of many discrete components, such as [[transistor]]s or SSI [[integrated circuit\|chips]]. Only the very earliest [[microprocessor]]s had such options, often [[floating point]] arithmetic units.)

			{{semi-stub}}

			==See also==

			* [[PDP-11 Extended Instruction Set]]
			* [[Commercial Instruction Set]]
			* [[PDP-11 Commercial Instruction Set]]
			* [[Honeywell 6000 series]]

			[[Category: CPU Basics]]

Honeywell 6000 series

Jnc — Sun, 13 Sep 2026 18:23:30 GMT

Clarify

← Older revision		Revision as of 18:23, 13 September 2026
Line 5:		Line 5:
	They were descendants of the [[GE 600 series]] family; after GE's computer business was sold to [[Honeywell]] in 1970, the 6000 series were Honeywell's replacements. They used [[Transistor-transistor logic\|TTL]] [[Integrated circuit\|SSI integrated circuits]] and larger [[printed circuit board]]s, unlike the older (and obsolescent) discrete [[transistor]] GE machines.		They were descendants of the [[GE 600 series]] family; after GE's computer business was sold to [[Honeywell]] in 1970, the 6000 series were Honeywell's replacements. They used [[Transistor-transistor logic\|TTL]] [[Integrated circuit\|SSI integrated circuits]] and larger [[printed circuit board]]s, unlike the older (and obsolescent) discrete [[transistor]] GE machines.

	The basic [[architecture]] of the GE and Honeywell series was the same: a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] units. All used 36-bit [[word]]s, and almost identical [[instruction set]]s; Honeywell added an [[Commercial Instruction Set\|Extended Instruction Set]] (EIS) which provided string-manipulation instructions, [[binary-coded decimal\|packed decimal]] instructions, etc.		The basic [[architecture]] of the GE and Honeywell series was the same: a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] units. All models used 36-bit [[word]]s, and almost identical [[instruction set]]s; Honeywell added an [[Commercial Instruction Set\|Extended Instruction Set]] (EIS) which provided string-manipulation instructions, [[binary-coded decimal\|packed decimal]] instructions, etc.

	Most 6000 series machines ran [[GCOS]] (General Comprehensive Operating System), an evolution of the earlier GECOS (General Electric Comprehensive Operating Supervisor) - early Honeywell documentation continued to call it GECOS.		Most 6000 series machines ran [[GCOS]] (General Comprehensive Operating System), an evolution of the earlier GECOS (General Electric Comprehensive Operating Supervisor) - early Honeywell documentation continued to call it GECOS.

GE-635

Jnc — Sat, 12 Sep 2026 15:07:41 GMT

Tweak DTSS mention

← Older revision		Revision as of 15:07, 12 September 2026
Line 1:		Line 1:
	The '''GE-635''' was a [[transistor]] [[mainframe]] computer; the largest of the [[GE 600 series‎]] family. It is notable mostly for being the design base of the [[GE-645]], a descendant created for the [[Multics]] project (after [[International Business Machines\|IBM]] refused to add necessary features such as [[virtual memory]] to the [[IBM System/360\|System 360]]). The [[Dartmouth Time Sharing System]], where [[BASIC]] was developed, also ran on the GE-635.		The '''GE-635''' was a [[transistor]] [[mainframe]] computer; the largest of the [[GE 600 series‎]] family. It is notable mostly for being the design base of the [[GE-645]], a descendant created for the [[Multics]] project (after [[International Business Machines\|IBM]] refused to add necessary features such as [[virtual memory]] to the [[IBM System/360\|System 360]]). The [[Dartmouth Time Sharing System]], where [[BASIC]] was developed, also eventually ran on the GE-635.

	The basic overall [[architecture]] of the GE-635 was inherited by the later Multics systems: it was a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] modules. [[Input/output\|I/O]] activities were handled by Input/Output Controller modules, similarly sharing access to the memory modules.		The basic overall [[architecture]] of the GE-635 was inherited by the later Multics systems: it was a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] modules. [[Input/output\|I/O]] activities were handled by Input/Output Controller modules, similarly sharing access to the memory modules.

	The GE-635 was intended for [[multi-tasking\|multi-programming]]; a CPU operated in either Master mode, in which there were no limitations on its operation, or Slave mode, in which [[base and bounds]] type [[memory management]] applied.		The GE-635 was intended for [[multi-tasking\|multi-programming]]; a CPU operated in either 'Master' mode, in which there were no limitations on its operation, or 'Slave' mode, in which [[base and bounds]] type [[memory management]] applied.

	The GE-635 was a 36-bit machine, as was common for scientific computers in that period. The CPU included a single double-width [[register\|accumulator]], and eight [[index register]]s (18 bits wide; the size of [[address]]es in the GE-635). The [[Program Counter]] was also 18 bits.		The GE-635 was a 36-bit machine, as was common for scientific computers in that period. The CPU included a single double-width [[register\|accumulator]], and eight [[index register]]s (18 bits wide; the size of [[address]]es in the GE-635). The [[Program Counter]] was also 18 bits.

GCOS

Jnc — Sat, 12 Sep 2026 09:35:30 GMT

Redir - short form

New page

#Redirect [[General Comprehensive Operating System]]

Michigan Algorithm Decoder

Jnc — Fri, 11 Sep 2026 14:55:18 GMT

Add minimal details on the evolution

← Older revision		Revision as of 14:55, 11 September 2026
Line 1:		Line 1:
	'''Michigan Algorithm Decoder''' (usually given as its acronym, '''MAD''') was a [[compiler\|compiled]] [[programing language]]~~, originally for the~~ [[~~IBM 7090~~]]~~. It~~ was in spirit (but not [[syntax]]) ~~similar to [[ALGOL]]~~.		'''Michigan Algorithm Decoder''' (usually given as its acronym, '''MAD''') was a [[compiler\|compiled]] [[programing language]]. It was based on [[ALGOL]] 58, to which it was similar in spirit (but not [[syntax]]).

	It was ~~later~~ moved to the [[IBM 7094]], and ~~was much used on~~ [[~~Compatible Time-Sharing System\|CTSS~~]].		The compiler was originally written in 1959 for the [[IBM 704]]; it was subsequentially moved to the [[IBM 709]], [[IBM 7090]] and [[IBM 7094]] by the University of Michigan Computing Center staff. Other groups elsewhere moved it to the [[IBM 7040]], [[Philco]] 210-211, and [[Sperry Rand]] 1107.

	{{stub}}		It was much used on [[Compatible Time-Sharing System\|CTSS]].

			{{semi-stub}}

	==External links==		==External links==

	* [http://www.bitsavers.org/pdf/univOfMichigan/mad/L2-UOI-MAD1-2-RX_MADum_62.pdf A User's Reference Manual for The Michigan Algorithm Decoder (MAD) for the IBM 7090]		* [http://www.bitsavers.org/pdf/univOfMichigan/mad/L2-UOI-MAD1-2-RX_MADum_62.pdf A User's Reference Manual for The Michigan Algorithm Decoder (MAD) for the IBM 7090] - 1962 version
			* [https://people.csail.mit.edu/saltzer/Multics/CTSS-Documents/MAD-manual.pdf The Michigan Algorithm Decoder (The MAD Manual)] - 1966 version

	[[Category: Languages]]		[[Category: Languages]]

IBM 7094

Jnc — Fri, 11 Sep 2026 14:27:59 GMT

Final CTSS host

← Older revision		Revision as of 14:27, 11 September 2026
Line 19:		Line 19:
	\| price = roughly US$3.5M		\| price = roughly US$3.5M
	}}		}}

	The '''IBM 7094''' was IBM's last commercial scientific [[mainframe]] (built at a time when computers for scientific and business computing used separate [[instruction set]]s).		The '''IBM 7094''' was IBM's last commercial scientific [[mainframe]] (built at a time when computers for scientific and business computing used separate [[instruction set]]s).

Line 29:		Line 28:

	Total production amounted to 130 7094 I's, and 125 7094 II's.		Total production amounted to 130 7094 I's, and 125 7094 II's.

			The 7094 was the final IBM model upon which [[Compatible Time-Sharing System\|CTSS]] ran.

			{{semi-stub}}

	==Further reading==		==Further reading==
Line 39:		Line 42:
	* [http://www.quadibloc.com/comp/cp0309.htm From the IBM 704 to the IBM 7094]		* [http://www.quadibloc.com/comp/cp0309.htm From the IBM 704 to the IBM 7094]
	* [http://www.frobenius.com/7090.htm IBM 7090/94 Architecture]		* [http://www.frobenius.com/7090.htm IBM 7090/94 Architecture]
			* [https://people.csail.mit.edu/saltzer/Multics/CTSS-Documents/RPQs/RPQs.html IBM 7094 Hardware Modifications for CTSS (RPQs)]
	~~{{semi~~-~~stub}}~~

	[[Category: IBM Mainframes]]		[[Category: IBM Mainframes]]

SimH Classic

Larsbrinkhoff — Thu, 10 Sep 2026 07:03:30 GMT

Newest relsease is V3.12-5. There is no GitHub repository for Classic.

← Older revision		Revision as of 07:03, 10 September 2026
(One intermediate revision by the same user not shown)
Line 5:		Line 5:
	SIMH was primarily written by Bob Supnik of [[Zork]] porting fame, and is now widely contributed to by others as well.		SIMH was primarily written by Bob Supnik of [[Zork]] porting fame, and is now widely contributed to by others as well.

	The current version is V3.12-2, released on ~~April 25th~~, ~~2022. More up to date developments may be found on their GitHub [[source code]] repository~~.		The current version is V3.12-5, released on July 16th, 2024.

	== Emulated Systems ==		== Emulated Systems ==
Line 48:		Line 48:
	=== DEC PDP-10 ===		=== DEC PDP-10 ===

	The [[PDP-10]] model [[KS10]] processor is emulated and will run all but the latest versions of the [[TOPS-10]] and [[TOPS-20]] operating systems.		The [[PDP-10]] model [[KS10]] processor is emulated and will run ITS and all but the latest versions of the [[TOPS-10]] and [[TOPS-20]] operating systems.

	Richard Cornwell contributed a new emulator that supports [[KA10]], [[KI10]], [[KL10]] model A and B, and [[KS10]]. It can run most versions of TOPS-10 and TOPS-20, as well as [[Incompatible Timesharing System\|ITS]], [[WAITS]], and (soon) [[TENEX]].		Richard Cornwell contributed a new emulator that supports [[KA10]], [[KI10]], [[KL10]] model A and B, and [[KS10]]. It can run most versions of TOPS-10 and TOPS-20, as well as [[Incompatible Timesharing System\|ITS]], [[WAITS]], and (soon) [[TENEX]].

SIMH/Draft

Jnc — Wed, 09 Sep 2026 01:45:00 GMT

Jnc deleted page SIMH/Draft Useless: No use now that the Draft page is gone

SIMH

Vaxorcist — Tue, 08 Sep 2026 15:46:51 GMT

SIMH Flavors: Update current SimH Classic version

← Older revision		Revision as of 15:46, 8 September 2026
Line 15:		Line 15:
	It is relatively hard for a beginner to decide which way to go. I will not give any specific advice, but some short descriptions:		It is relatively hard for a beginner to decide which way to go. I will not give any specific advice, but some short descriptions:

	'''SimH Classic''' is the oldest one, whereof the others were forked later on. It is Bob Supnik work mostly, except for some HP simulators. Development apparently stopped (or at least slowed down) in December 2022. The current version is SIMH V3.12-3.		'''SimH Classic''' is the oldest one, whereof the others were forked later on. It is Bob Supnik work mostly, except for some HP simulators. Development apparently stopped (or at least slowed down) in December 2022. The current version is SIMH V3.12-5.

	'''"New" SIMH''' was the first bigger fork and added a lot more simulated computer systems.		'''"New" SIMH''' was the first bigger fork and added a lot more simulated computer systems.

SIMH/Draft

Vaxorcist — Tue, 08 Sep 2026 15:44:16 GMT

Vaxorcist moved page SIMH/Draft to SIMH Distinction between SimH variants

New page

'''SIMH''' consists of simulators for a lot of different machines, and as such, is a very important emulator for anyone interested in computer history; in particular, that of [[Digital Equipment Corporation|DEC]] systems, as the PDPs and even some [[VAX]] models are well emulated by it.

It can even emulate network interfaces for some of the simulated computers, e.g. the [[PDP-10]], [[PDP-11]] and VAX, so the emulated systems can network directly onto the [[Internet]], if the emulated operating system supports it, like BSD or [[VMS]].

SIMH was primarily written by Bob Supnik of [[Zork]] porting fame, and is now widely contributed to by others as well.

== SIMH Flavors ==

For some time passed there are three flavors of SIMH now:

* [[SimH Classic]] ([http://simh.trailing-edge.com/ http://simh.trailing-edge.com/])
* [["New" SIMH]] ([https://github.com/simh/simh https://github.com/simh/simh])
* [[OpenSIMH]] ([https://opensimh.org/ https://opensimh.org/])

It is relatively hard for a beginner to decide which way to go. I will not give any specific advice, but some short descriptions:

'''SimH Classic''' is the oldest one, whereof the others were forked later on. It is Bob Supnik work mostly, except for some HP simulators. Development apparently stopped (or at least slowed down) in December 2022. The current version is SIMH V3.12-3.

'''"New" SIMH''' was the first bigger fork and added a lot more simulated computer systems.

'''"OpenSIMH"''' forked from "New" SIMH because of insurmountable different opinions of both maintainers and contributors.

== Usage ==

SIMH can either take in configuration parameters interactively, or you can store them in a file. If the file is the name of the emulator+.ini it will load them automatically (eg vax.exe loads vax.ini automatically). Once running, the CONTROL+E key will break the emulation and bring you back to the SIMH console where you can alter the running state.

Networking is covered in the [[Networking with SIMH]] guide.

[[Category: Emulators]]

== SIMH Media Handling ==

SIMH handles computer storage media in the form of data files called "images" for:

* [[Punched card]]s
* [[Paper tape]]s
* [[DECtape]]s
* [[Magnetic tape]]s
* [[Disk]]s

All of these image files are said to be interchangeable between the three SIMH flavors with one exception: Image files used with "New" SIMH may be unreadable with the other two flavors because of 512 byte of meta data appended to the standard SIMH images.

Sorry, Windows folks, the next two commands are for [[Linux]] only! -
If someone knows how to do the same in [[Windows]], please add it here.

You can visualize the appended metadata under Linux with (example image file trek7.rk7):

$ tail -c 512 trek7.rk7 | hexdump -C
00000000 73 69 6d 68 56 41 58 20 31 31 2f 37 35 30 00 00 |simhVAX 11/750..|
00000010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
00000040 00 00 00 00 52 4b 30 37 00 00 00 00 00 00 00 00 |....RK07........|
00000050 00 00 00 00 00 00 02 00 00 00 d2 1e 00 00 00 02 |................|
00000060 4d 6f 6e 20 41 70 72 20 20 33 20 30 39 3a 31 39 |Mon Apr 3 09:19|
00000070 3a 32 34 20 32 30 32 33 0a 00 00 00 00 02 00 00 |:24 2023........|
00000080 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
000001f0 00 00 00 00 00 00 00 00 00 00 00 00 fa 1e 03 18 |................|
00000200

In the metadata you can see:
* the simulated computer model: <code>simhVAX 11/750</code>
* the disk model: <code>RK07</code>
* the time stamp: <code>Mon Apr 3 09:19:24 2023</code>

The problem can be cured under Linux using (example image file trek7.rk7):
$ truncate -s -512 trek7.rk7
This truncates 512 bytes (the metadata) at the end of the image file.

Keep in mind that the <code>truncate</code> command truncates whatever file you specify, so look first if metadata is present.

'''Another possible pitfall:''' <br>If you leave out the <code>-</code> (minus) before the <code>512</code>, you will end up with a corrupted image of 512 bytes only!

'''Caution:''' <br>
'''Truncating does not work with SIMH disk images in [[VHD]] format!!!'''

[[Category: Emulators]]

Talk:SIMH

Jnc — Tue, 08 Sep 2026 12:12:10 GMT

Renaming / Revising SIMH pages: Redirects and moves are a complex subject

← Older revision		Revision as of 12:12, 8 September 2026
Line 68:		Line 68:

	:::: I started with moving SIMH to SimH Classic (which is used on their own web page), but the automatically generated redirect is in the way now for moving SIMH/Draft to SIMH. So this time I do really need your help, don't I? Next steps will be to revise existing/add the missing page(s). [[User:Vaxorcist\|Vaxorcist]] ([[User talk:Vaxorcist\|talk]]) 07:41, 8 September 2026 (UTC)		:::: I started with moving SIMH to SimH Classic (which is used on their own web page), but the automatically generated redirect is in the way now for moving SIMH/Draft to SIMH. So this time I do really need your help, don't I? Next steps will be to revise existing/add the missing page(s). [[User:Vaxorcist\|Vaxorcist]] ([[User talk:Vaxorcist\|talk]]) 07:41, 8 September 2026 (UTC)

			::::: Redirects and moves are a complex subject; when I go to do a move, I am given the choice of not "Leav[ing] a redirect behind", but I don't know if only admins get that option (they might). Similarly, I am given the choice to not "Move [the] associated talk page" (of a normal article page), but again, I don't know if ordinary users get that (especially, in the latest version of MediaWiki).
			::::: Also, I thought that ordinary users are allowed to move an ordinary article page on top of a redirect; you can see in [[Special:RecentChanges]] that when I put this [[Talk:SIMH]] page back where it was to start with (it seemed to me the best place for it, at a central location in the group; there's a redirect from [[Talk:SimH Classic]] to it), MediaWiki automagically deleted the redirect at the destination. Again, maybe the latest version of MediaWiki doesn't allow that for ordinary users.
			::::: Anyway, you should be all set now! [[User:Jnc\|Jnc]] ([[User talk:Jnc\|talk]]) 12:11, 8 September 2026 (UTC)

SIMH

Jnc — Tue, 08 Sep 2026 11:53:51 GMT

Jnc deleted page SIMH Useless: Get rid of the REDIRECT in the way of other moves

Talk:SimH Classic

Jnc — Tue, 08 Sep 2026 11:52:12 GMT

Jnc moved page Talk:SimH Classic to Talk:SIMH over redirect Probably the best place to place this discussion

New page

== SIMH flavors ==
I think it is time to distinguish between the three flavors of SIMH, although they have a lot in common, but the differences grow steadily.

My proposal is a generic SIMH page which might have the following section after the introduction:

<pre>== SIMH Flavors ==

For some time passed there are three flavors of SIMH now:

* [[Classic" SIMH]] ([http://simh.trailing-edge.com/ http://simh.trailing-edge.com/])
* [["New" SIMH]] ([https://github.com/simh/simh https://github.com/simh/simh])
* [["OpenSIMH"]] ([https://opensimh.org/ https://opensimh.org/])
It is relatively hard for a beginner to decide which way to go. I will not give any specific advice, but some short descriptions:

'''"Classic" SIMH''' is the oldest one, whereof the others were forked later on. It is Bob Supnik work mostly, except for some HP simulators. Development apparently stopped (or at least slowed down) in December 2022. The current version is SIMH V3.12-3.

'''"New" SIMH''' was the first bigger fork and added a lot more simulated computer systems.

'''"OpenSIMH"''' forked from "New" SIMH because of insurmountable different opinions of both maintainers and contributors.</pre>

The generic SIMH page should concentrate on the commonalities; the differences should be explained in the three specific pages linked in the gereric page.

What do my fellow writers think? [[User:Vaxorcist|Vaxorcist]] ([[User talk:Vaxorcist|talk]]) 18:12, 2 April 2023 (CEST)

: Sounds good to me. [[User:Jnc|Jnc]] ([[User talk:Jnc|talk]]) 19:14, 2 April 2023 (CEST)

== New content ==

Don't do a 'cut and paste' to move the content from [[SIMH/Draft]] to here; let me know, and I can merge the two pages. [[User:Jnc|Jnc]] ([[User talk:Jnc|talk]]) 18:02, 3 April 2023 (CEST)

: O.k., I will [[User:Vaxorcist|Vaxorcist]] ([[User talk:Vaxorcist|talk]]) 18:23, 3 April 2023 (CEST)

== Renaming / Revising SIMH pages ==

I think it is time (or rather, long overdue) to rename the following pages:
* SIMH to "Classic" SIMH
* SIMH/Draft to SIMH
and revise both pages.
What do you think? [[User:Vaxorcist|Vaxorcist]] ([[User talk:Vaxorcist|talk]])

: Is the term "'Classic' SIMH" really in use to refer to the old one? Does that version get enough use to make it worth having a separate article for it (as opposed to just covering it in the same article as the current one)? Mind, I don't have a strong preference either way; I am just trying to work out what makes the most sense.
: I took a look at [[SIMH/Draft]], and the article here, and there is a lot of content in this article that isn't in the Draft one, so we'd need to do something a little more complicated than just do some moves. But let's work out the previous point first. (If we wind up editing the good ideas/content from Draft in to this one, or something, don't worry; I can merge the two histories into one once the editing process is complete.) [[User:Jnc|Jnc]] ([[User talk:Jnc|talk]]) 21:25, 2 September 2026 (UTC)

:: There are three variants of SIMH - whether that is desirable or not. And in my opinion, they differ sufficiently from one another to keep them separate.
::The term "Classic SIMH" has become established to refer to the original SIMH written by Bob Supnik. There's still development on this variant, although somewhat sluggish.

::The "New SIMH" is my own wording - the most important difference to the Classic SIMH is the introduction of asynchronous I/O, an improved clock and several other useful additions. Sometimes you can see that this version is called the "V4 SIMH", but I think that's not a good idea as some time sooner or later it might become V5. I was even tempted to call it "Closed SIMH", but I do not want to provoke anyone, even if I do not like what he's doing.

::"OpenSIMH" is an official name given by the founders of the newest SIMH version.

::All in all I think it needs clearly distinguishable names to avoid more confuision as there already is. Imagine a beginner stumbling into this chaos of three similar, but still very different in detail, versions!

::If I had one wish in "bit heaven", I would pray for the reunification of "New SIMH" and "OpenSIMH."

::There are probably good reasons to keep the "Classic SIMH" and a modernized SIMH apart, but noone needs a third one. What a waste of precious time, ideas and resources! Life is too short for such (some rude words here).

::And you're right, all the SIMH related pages need to be more or less revised or created:

::* SIMH/Draft, which shall become the "common" SIMH page
::* SIMH, which shall become the "Classic SIMH" (it is already, but needs an update)
::* New SIMH, which is probably the hardest to write about (at least for me)
::* OpenSIMH, which carries the spirit and recommandation of Bob Supnik

:: [[User:Vaxorcist|Vaxorcist]] ([[User talk:Vaxorcist|talk]])

::: After thinking about it for a while, I basically agree with you; the way to handle this is to move the current [[SIMH]] page (and history) to [[Classic SIMH]] (since, as you indicate, that name has been "established to refer to the original SIMH"); [[SIMH]] needs to be a new page, basically a disambiguation page, but including a ''short'' history of SIMH. After looking at [[SIMH/Draft]], I guess that will do for this, but it has more detailed content than I would have put there (e.g. the info on file formats); but I guess if those are the same for all the versions, it makes sense to put them there.
::: [[OpenSIMH]] you will have to create; it might start as a stub with just links to the online area/work. For 'New SIMH', I would recommend asking the New SIMH community what they think its formal name should be. Then, you will have to write it! [[User:Jnc|Jnc]] ([[User talk:Jnc|talk]]) 18:07, 6 September 2026 (UTC)

:::: I started with moving SIMH to SimH Classic (which is used on their own web page), but the automatically generated redirect is in the way now for moving SIMH/Draft to SIMH. So this time I do really need your help, don't I? Next steps will be to revise existing/add the missing page(s). [[User:Vaxorcist|Vaxorcist]] ([[User talk:Vaxorcist|talk]]) 07:41, 8 September 2026 (UTC)

Talk:SIMH

Jnc — Tue, 08 Sep 2026 11:52:11 GMT

Jnc deleted redirect Talk:SIMH by overwriting Deleted to make way for move from "Talk:SimH Classic"

Talk:SimH Classic

Vaxorcist — Tue, 08 Sep 2026 07:41:06 GMT

Renaming / Revising SIMH pages

← Older revision		Revision as of 07:41, 8 September 2026
Line 66:		Line 66:
	::: After thinking about it for a while, I basically agree with you; the way to handle this is to move the current [[SIMH]] page (and history) to [[Classic SIMH]] (since, as you indicate, that name has been "established to refer to the original SIMH"); [[SIMH]] needs to be a new page, basically a disambiguation page, but including a ''short'' history of SIMH. After looking at [[SIMH/Draft]], I guess that will do for this, but it has more detailed content than I would have put there (e.g. the info on file formats); but I guess if those are the same for all the versions, it makes sense to put them there.		::: After thinking about it for a while, I basically agree with you; the way to handle this is to move the current [[SIMH]] page (and history) to [[Classic SIMH]] (since, as you indicate, that name has been "established to refer to the original SIMH"); [[SIMH]] needs to be a new page, basically a disambiguation page, but including a ''short'' history of SIMH. After looking at [[SIMH/Draft]], I guess that will do for this, but it has more detailed content than I would have put there (e.g. the info on file formats); but I guess if those are the same for all the versions, it makes sense to put them there.
	::: [[OpenSIMH]] you will have to create; it might start as a stub with just links to the online area/work. For 'New SIMH', I would recommend asking the New SIMH community what they think its formal name should be. Then, you will have to write it! [[User:Jnc\|Jnc]] ([[User talk:Jnc\|talk]]) 18:07, 6 September 2026 (UTC)		::: [[OpenSIMH]] you will have to create; it might start as a stub with just links to the online area/work. For 'New SIMH', I would recommend asking the New SIMH community what they think its formal name should be. Then, you will have to write it! [[User:Jnc\|Jnc]] ([[User talk:Jnc\|talk]]) 18:07, 6 September 2026 (UTC)

			:::: I started with moving SIMH to SimH Classic (which is used on their own web page), but the automatically generated redirect is in the way now for moving SIMH/Draft to SIMH. So this time I do really need your help, don't I? Next steps will be to revise existing/add the missing page(s). [[User:Vaxorcist\|Vaxorcist]] ([[User talk:Vaxorcist\|talk]]) 07:41, 8 September 2026 (UTC)

SIMH/Draft

Vaxorcist — Tue, 08 Sep 2026 07:34:52 GMT

SIMH Flavors: another "SimH Classic" in the text

← Older revision		Revision as of 07:34, 8 September 2026
(One intermediate revision by the same user not shown)
Line 9:		Line 9:
	For some time passed there are three flavors of SIMH now:		For some time passed there are three flavors of SIMH now:

	* [["Classic~~" SIMH~~]] ([http://simh.trailing-edge.com/ http://simh.trailing-edge.com/])		* [[SimH Classic]] ([http://simh.trailing-edge.com/ http://simh.trailing-edge.com/])
	* [["New" SIMH]] ([https://github.com/simh/simh https://github.com/simh/simh])		* [["New" SIMH]] ([https://github.com/simh/simh https://github.com/simh/simh])
	* [[OpenSIMH]] ([https://opensimh.org/ https://opensimh.org/])		* [[OpenSIMH]] ([https://opensimh.org/ https://opensimh.org/])
Line 15:		Line 15:
	It is relatively hard for a beginner to decide which way to go. I will not give any specific advice, but some short descriptions:		It is relatively hard for a beginner to decide which way to go. I will not give any specific advice, but some short descriptions:

	'''"Classic~~" SIMH~~''' is the oldest one, whereof the others were forked later on. It is Bob Supnik work mostly, except for some HP simulators. Development apparently stopped (or at least slowed down) in December 2022. The current version is SIMH V3.12-3.		'''SimH Classic''' is the oldest one, whereof the others were forked later on. It is Bob Supnik work mostly, except for some HP simulators. Development apparently stopped (or at least slowed down) in December 2022. The current version is SIMH V3.12-3.

	'''"New" SIMH''' was the first bigger fork and added a lot more simulated computer systems.		'''"New" SIMH''' was the first bigger fork and added a lot more simulated computer systems.

Talk:SIMH

Vaxorcist — Tue, 08 Sep 2026 07:28:06 GMT

Vaxorcist moved page Talk:SIMH to Talk:SimH Classic Distinction between SimH variants

New page

SIMH

Vaxorcist — Tue, 08 Sep 2026 07:28:06 GMT

Vaxorcist moved page SIMH to SimH Classic Distinction between SimH variants

New page

'''SIMH''' consists of simulators for about 20 different machines, and as such, is a very important emulator for anyone interested in computer history; in particular, that of [[Digital Equipment Corporation|DEC]] systems, as the PDPs and even the [[VAX]] are well emulated by it.

It can even emulate network interfaces for the [[PDP-10]], [[PDP-11]] and VAX, so the emulated systems can network directly onto the [[Internet]], if the emulated operating system supports it, like BSD or [[VMS]].

SIMH was primarily written by Bob Supnik of [[Zork]] porting fame, and is now widely contributed to by others as well.

The current version is V3.12-2, released on April 25th, 2022. More up to date developments may be found on their GitHub [[source code]] repository.

== Emulated Systems ==

=== Altair 8800 ===
This is the Altair 8800 emulator which can run both AltairDOS and CP/M 2.2

=== Altair Z80 ===

SIMH emulates the [[Altair]] architecture with your choice of either a [[Z80]] or [[Intel 8080]] processor. It can run [[CP/M]].

=== Honeywell H316/H516 ===

Emulates the work horse behind the [[ARPANET]] IMPs. In 2013 the emulator was modified to bring up the original IMP code and connect them together (tunnelled over IP). Demo instructions are available in the latest source. The IMP side of the 1822 is there, but there is not much of a host side yet. I don't see anything in simh that would be able to plug in anyway. An Associated Computer Consultants (ACC) 1822 card for the pdp-11 would be a start.

=== HP2100 ===

I really don't know much about this machine, but I included a way to run the basic1 program I found online. It kind of reminds me of the ROMBASIC back 'in the day'... It's basic with line numbers so I know it's kind of old.. I really don't know much otherwise about this machine.

=== Interdata 8/32 ===

The Interdata 8/32 micro was the first non [[PDP-11]] to run UNIX. SIMH can run both v6 & v7 research editions for the Interdata.

=== Nova ===

SIMH can run [[RDOS]] 7.5 for the [[Nova]].

SIMH also includes [[Data General Eclipse]] support, but there is no available OS.

=== DEC PDP-1 ===

The original [[PDP-1]] is emulated although most of the software is expected to be toggled from the console, or loaded from paper tape. SIMH does not emulate the display needed for [[spacewar]].

=== DEC PDP-6 ===

There is a new [[PDP-6]] emulator by Richard Cornwell.

=== DEC PDP-8 ===

The [[PDP-8 family|PDP-8]] was the first [[minicomputer]]... [[OS/8]] runs under SIMH.

=== DEC PDP-10 ===

The [[PDP-10]] model [[KS10]] processor is emulated and will run all but the latest versions of the [[TOPS-10]] and [[TOPS-20]] operating systems.

Richard Cornwell contributed a new emulator that supports [[KA10]], [[KI10]], [[KL10]] model A and B, and [[KS10]]. It can run most versions of TOPS-10 and TOPS-20, as well as [[Incompatible Timesharing System|ITS]], [[WAITS]], and (soon) [[TENEX]].

=== DEC PDP-11 ===

The PDP-11 is emulated enough to run almost all PDP-11 operating systems including the early Unixes (v1, v4, v5, v6, v7 are known to run).

SIMH can be configured to provide simulations for the following PDP-11 models:
* PDP-11/03
* PDP-11/04
* PDP-11/05
* PDP-11/20
* PDP-11/23
* PDP-11/23+
* PDP-11/24
* PDP-11/34
* PDP-11/40
* PDP-11/44
* PDP-11/45
* PDP-11/53
* PDP-11/60
* PDP-11/70
* PDP-11/73
* PDP-11/73B
* PDP-11/83
* PDP-11/84
* PDP-11/93
* PDP-11/94

=== MicroVAX II ===

The SIMH '''microvax2''' simulator can run the [[MicroVAX II]] with as much as 16 MB of emulated [[RAM]].

Both [[VMS]], [[ULTRIX]], [[NetBSD]], [[OpenBSD]] and original [[BSD]] can run on this emulator.

With the following SIMH commands you can choose between three MicroVAX II models:

* Standard MicroVAX II: '''SET CPU MODEL=MicroVAX'''
* MicroVAX II with one Plane Graphics ([[VCB01 Video Controller|VCB01]]) = [[VAXstation II]]: '''SET CPU MODEL=VAXstation'''
* MicroVAX II with four/eight Plane Graphics ([[VCB02 Video Controller|VCB02]]) = [[VAXstation II/GPX]]: '''SET CPU MODEL=VAXstationGPX'''

=== VAX 11/780 ===

SIMH can also simulate the first model of the VAX, the [[VAX-11/780]], the name of the simulator is '''vax780'''.

This is appropriate for older operating systems like VMS release 1.x (see: [[Installing VMS V1.0 on SIMH]]) and the original [[Unix/32V]], [[3BSD]] and all 4BSD releases, e.g. [[4BSD]].

== Usage ==

SIMH can either take in configuration parameters interactively, or you can store them in a file. If the file is the name of the emulator+.ini it will load them automatically (eg vax.exe loads vax.ini automatically). The CONTROL+E key will break the emulation and bring you back to the SIMH console where you can alter the running state.

Networking is covered in the [[Networking with SIMH]] guide.

== Support ==

There is a mailing list here: https://groups.io/g/simh

There was an earlier mailing list here [http://mailman.trailing-edge.com/pipermail/simh/ http://mailman.trailing-edge.com/pipermail/simh/]. I would recommend searching the old archives, and you'll be amazed how much good information there is in there. The [https://github.com/open-simh/simh public source repository] is also a great source of information.

==See also==

* [[:Category:SIMH Tutorials|SIMH Tutorials]]
* [[SIMH Simple Sound System]]
* [[Mkdisttap.pl]] - make 'tapes'

==External links==

* [http://simh.trailing-edge.com/ Home page]
* [http://sourceforge.net/projects/simh/ SIMH on SourceForge]
* [https://github.com/open-simh/simh Open SIMH on GitHub]
* [https://github.com/simh/simh Mark Pizzolato's fork of SIMH on GitHub]

[[Category: Emulators]]

Computer Systems Research Group

Jnc — Mon, 07 Sep 2026 23:44:34 GMT

External links: +The Berkeley Network − A Retrospective

← Older revision		Revision as of 23:44, 7 September 2026
(One intermediate revision by the same user not shown)
Line 7:		Line 7:

	Kirk has released CD sets containing as much of the work of the CSRG as possible. You can download ISO's of them [http://ftp.wuppy.net.ru/pub/pub/BSD/CSRG/ here].		Kirk has released CD sets containing as much of the work of the CSRG as possible. You can download ISO's of them [http://ftp.wuppy.net.ru/pub/pub/BSD/CSRG/ here].

			{{semi-stub}}

	==External links==		==External links==
Line 12:		Line 14:
	* [https://www.oreilly.com/openbook/opensources/book/kirkmck.html Twenty Years of Berkeley Unix: From AT&T-Owned to Freely Redistributable] - An excellent overview of the history of the CSRG, as told by Marshall Kirk McKusick		* [https://www.oreilly.com/openbook/opensources/book/kirkmck.html Twenty Years of Berkeley Unix: From AT&T-Owned to Freely Redistributable] - An excellent overview of the history of the CSRG, as told by Marshall Kirk McKusick
	* [http://www.groklaw.net/article.php?story=20050505095249230 The Daemon, the GNU and the Penguin: Chapter 7 - BSD and the CSRG]		* [http://www.groklaw.net/article.php?story=20050505095249230 The Daemon, the GNU and the Penguin: Chapter 7 - BSD and the CSRG]
	~~{{semi~~-~~stub}}~~		* [https://maibriz.de/unix/ultrix/berknet/netintro.pdf An Introduction to the Berkeley Network] - something from the roots of CSRG
			* [https://www.krsaborio.net/bsd/research/acrobat/8002.pdf The Berkeley Network − A Retrospective]

	[[Category: Research Organizations]]		[[Category: Research Organizations]]

CB-UNIX

Jnc — Mon, 07 Sep 2026 23:28:05 GMT

External links: +CB-UNIX Programmer's Manual

← Older revision		Revision as of 23:28, 7 September 2026
Line 63:		Line 63:
	* [https://www.tuhs.org/Archive/Distributions/USDL/CB_Unix/SourceCode/ Source Code] - source (scan of kernel listings)		* [https://www.tuhs.org/Archive/Distributions/USDL/CB_Unix/SourceCode/ Source Code] - source (scan of kernel listings)
	* [https://www.tuhs.org/Archive/Distributions/USDL/CB_Unix/ CB-UNIX] - Edition 2.3 documentation		* [https://www.tuhs.org/Archive/Distributions/USDL/CB_Unix/ CB-UNIX] - Edition 2.3 documentation
			** [https://www.tuhs.org/Archive/Distributions/USDL/CB_Unix/raw/cbunix_intro.pdf CB-UNIX Programmer's Manual - Edition 2.3]

	{{Nav Unix}}		{{Nav Unix}}

IAS computer

Jnc — Mon, 07 Sep 2026 23:03:20 GMT

Papers: expand characterization of 'Trilogy On Errors'

← Older revision		Revision as of 23:03, 7 September 2026
Line 168:		Line 168:
	* Gerald Estrin, [https://dl.acm.org/doi/pdf/10.1145/800259.809005 ''A description of the electronic computer at the Institute for Advanced Studies''], Proceedings of the 1952 ACM national meeting (Toronto), September 1952 - very complete, but low-level, description		* Gerald Estrin, [https://dl.acm.org/doi/pdf/10.1145/800259.809005 ''A description of the electronic computer at the Institute for Advanced Studies''], Proceedings of the 1952 ACM national meeting (Toronto), September 1952 - very complete, but low-level, description
	<!-- https://www.cs.unh.edu/~pjh/courses/cs520/05fall/p95-estrin.pdf -->		<!-- https://www.cs.unh.edu/~pjh/courses/cs520/05fall/p95-estrin.pdf -->
	* Nicholas Metropolis, J. Worlton, [https://www.osti.gov/servlets/purl/4628499 ''A Trilogy On Errors in the History of Computing''], Los Alamos National Laboratory, Los Alamos, 1972 - contains a section on the mis-identification of the IAS machine as the 'MANIAC'; also historical notes about both machines		* Nicholas Metropolis, J. Worlton, [https://www.osti.gov/servlets/purl/4628499 ''A Trilogy On Errors in the History of Computing''], Los Alamos National Laboratory, Los Alamos, 1972 - contains a section on the mis-identification of the IAS machine as the 'MANIAC'; also has extensive technical and historical notes about both machines
	<!-- https://www.osti.gov/biblio/4628499 -->		<!-- https://www.osti.gov/biblio/4628499 -->

Talk:J. C. R. Licklider

Jnc — Mon, 07 Sep 2026 22:17:07 GMT

My Lick story: new section

← Older revision		Revision as of 22:17, 7 September 2026
Line 2:		Line 2:

	It's a major crock that Lick doesn't have a [[Turing Award]]. We just never got organized at MIT (his home institution) to press for one. But the way Turing's are given out is (or were, in that era - I don't know about now) semi-disorganized; they didn't do a wide search, etc. The only reason that Corby has one (in 1990, 7 years ''after'' Ken and Dennis, who worked in Corby's footsteps) was that ''I'' bestirred myself to organize an effort to do so; I remember getting Joel Moses (then head of the CS dept at MIT) to support it. I wonder if they do posthumous Turings (they don't, for Nobels); if they do, I'll have to send an email out to some people and see about organizing one for Lick. [[User:Jnc\|Jnc]] ([[User talk:Jnc\|talk]]) 14:29, 26 December 2023 (CET)		It's a major crock that Lick doesn't have a [[Turing Award]]. We just never got organized at MIT (his home institution) to press for one. But the way Turing's are given out is (or were, in that era - I don't know about now) semi-disorganized; they didn't do a wide search, etc. The only reason that Corby has one (in 1990, 7 years ''after'' Ken and Dennis, who worked in Corby's footsteps) was that ''I'' bestirred myself to organize an effort to do so; I remember getting Joel Moses (then head of the CS dept at MIT) to support it. I wonder if they do posthumous Turings (they don't, for Nobels); if they do, I'll have to send an email out to some people and see about organizing one for Lick. [[User:Jnc\|Jnc]] ([[User talk:Jnc\|talk]]) 14:29, 26 December 2023 (CET)

			== My Lick story ==

			So, I have my own Lick story - admittedly not very interesting, but here it is. When I first started 'doing stuff' at [[LCS]], I was eventually (this would have been about 1978) given an office - a half-room in the very corner of the 5th floor. One office away (almost in the other corner, across the narrow end of the building) was Lick's office. I am ashamed to admit that I had no idea who we was (so I never interacted with him at all, unlike Corby and Jerry). I liken this to being a young physicist getting an office next to Einstein (q.v. "Who Got Einstein's Office?: Eccentricity and Genius at The Institute for Advanced Study", by Ed Regis), and them not knowing who he was. Sigh. [[User:Jnc\|Jnc]] ([[User talk:Jnc\|talk]]) 22:17, 7 September 2026 (UTC)

Program

Jnc — Mon, 07 Sep 2026 21:38:48 GMT

Further reading: +Metropolis on the genesis of the 'stored program' idea

← Older revision		Revision as of 21:38, 7 September 2026
Line 15:		Line 15:

	In the context of early [[computing device]]s, it must be emphasized that a program was a specified series of elementary steps, which were stored in some mechanism from which they were fed into the device; not simply the ability to make physical changes to the device, to change what calculation it performs - even if those changes were instantiated in something that was added to the device for the accomplishment of that exact purpose, like the [[plug-board]]s of the [[ENIAC]]. Such 'customizability' is better referred to as 'configurability', and ''not'' 'programmability'.		In the context of early [[computing device]]s, it must be emphasized that a program was a specified series of elementary steps, which were stored in some mechanism from which they were fed into the device; not simply the ability to make physical changes to the device, to change what calculation it performs - even if those changes were instantiated in something that was added to the device for the accomplishment of that exact purpose, like the [[plug-board]]s of the [[ENIAC]]. Such 'customizability' is better referred to as 'configurability', and ''not'' 'programmability'.

			==Further reading==

			* Nicholas Metropolis, J. Worlton, [https://www.osti.gov/servlets/purl/4628499 ''A Trilogy On Errors in the History of Computing''], Los Alamos National Laboratory, Los Alamos, 1972 - contains an in-depth section on the genesis of the 'stored program' idea, by someone who lived through the end of that era, and knew those who did the work
			<!-- https://www.osti.gov/biblio/4628499 -->

	{{semi-stub}}		{{semi-stub}}

	[[Category: Basics]]		[[Category: Basics]]

Template:Nav Unix

Jnc — Mon, 07 Sep 2026 15:55:04 GMT

rm dup entry

← Older revision		Revision as of 15:55, 7 September 2026
Line 9:		Line 9:

	<b>BSD Descendants</b> [[386BSD]] • [[NetBSD]] • [[FreeBSD]] • [[OpenBSD]] • [[NeXTSTEP]] • [[Darwin]]		<b>BSD Descendants</b> [[386BSD]] • [[NetBSD]] • [[FreeBSD]] • [[OpenBSD]] • [[NeXTSTEP]] • [[Darwin]]
	\| cbody = <b>Other</b> - [[xv6]] • [[AMIX]] • [[SunOS]] • [[Solaris~~]] • [[ULTRIX~~]] • [[A/UX]] • <!-- [[VENIX]] • --> [[XENIX]] • [[AIX]] • [[Dell UNIX]]		\| cbody = <b>Other</b> - [[xv6]] • [[AMIX]] • [[SunOS]] • [[Solaris]] • [[A/UX]] • <!-- [[VENIX]] • --> [[XENIX]] • [[AIX]] • [[Dell UNIX]]

	<b>Clones</b> - [[DEMOS]] • [[Coherent]]		<b>Clones</b> - [[DEMOS]] • [[Coherent]]
	}}		}}

Installing and Testing Asynchronous DECnet over DDCMP between two MicroVAX Computers

Jnc — Mon, 07 Sep 2026 15:18:55 GMT

Fix cat name

← Older revision		Revision as of 15:18, 7 September 2026
Line 452:		Line 452:
	* [https://bitsavers.org/pdf/dec/vax/vms/4.5/AA-JR84A-TN_MicroVMS_Installation_Guide_4.5B_Mar87.pdf MicroVMS VAXstation 2000/MicroVAX 2000 Installation Guide (AA-JR84A-TN)]		* [https://bitsavers.org/pdf/dec/vax/vms/4.5/AA-JR84A-TN_MicroVMS_Installation_Guide_4.5B_Mar87.pdf MicroVMS VAXstation 2000/MicroVAX 2000 Installation Guide (AA-JR84A-TN)]

	[[Category: ~~VAXen~~]]		[[Category: VAX]]
	[[Category: SIMH Tutorials]]		[[Category: SIMH Tutorials]]
	[[Category: DECnet]]		[[Category: DECnet]]
	[[Category: VMS Practical Guides]]		[[Category: VMS Practical Guides]]

Micro-LINC 300

Jnc — Mon, 07 Sep 2026 15:14:30 GMT

External links: +micro-LINC 300 in a museum

← Older revision		Revision as of 15:14, 7 September 2026
(One intermediate revision by the same user not shown)
Line 20:		Line 20:

	The '''MSC 7''' instruction disables interrupts.		The '''MSC 7''' instruction disables interrupts.

			==External links==

			* [https://archive.computerhistory.org/resources/access/text/2024/06/102785480-05-0001-acc.pdf SPEAR micro-LINC 300 Interim Data Sheet]
			* [https://suomentietokonemuseo.fi/laitteet-esineet/#micro-linc-m1965 Finnish Computer Museum] - a micro-LINC 300 in a museum

			[[Category: Minicomputers]]
			[[Category: Personal Computers]]
			[[Category: 12-bit Computers]]

SPEaR, Inc

Jnc — Mon, 07 Sep 2026 14:57:17 GMT

+cat

← Older revision		Revision as of 14:57, 7 September 2026
Line 2:		Line 2:

	{{stub}}		{{stub}}

			[[Category: Computer Manufacturers]]

Meta key

Jnc — Mon, 07 Sep 2026 14:53:48 GMT

+cat

← Older revision		Revision as of 14:53, 7 September 2026
Line 16:		Line 16:
	==External links==		==External links==

	* [http://xahlee.info/kbd/meta_key.html History of Meta Key] by Xah Lee.		* [http://xahlee.info/kbd/meta_key.html History of Meta Key] by Xah Lee

			[[Category: Terminals]]

PDP-11 Unix overlays

Jnc — Mon, 07 Sep 2026 13:06:02 GMT

Add 'manual' overlays and "auto-overlay"s in 2.9BSD

← Older revision		Revision as of 13:06, 7 September 2026
(2 intermediate revisions by the same user not shown)
Line 1:		Line 1:
	The small [[address space]] of the [[PDP-11 architecture]] meant that larger computations, both [[operating system]]s and [[application]]s, needed to make use of [[overlay]]s.		The small [[address space]] of the [[PDP-11 architecture]] meant that larger computations, both [[operating system]]s and [[application]]s, needed to make use of [[overlay]]s.

	Early versions of [[UNIX]] did not support them, but support for them was added in [[2.9BSD]]: use-of/support-for overlays in ''[[process]]es'', in user mode; and also for overlays in the [[kernel]]. This support was then taken up in [[UNIX V7M-11\|V7M]].		Early versions of [[UNIX]] did not support them, but support for them was added in [[2.9BSD]]: use-of/support-for overlays in [[process]]es, in user mode; and also for overlays in the [[kernel]] (using mechanisms different in detail in the two cases, although they both used the [[PDP-11 Memory Management]] [[hardware]]). This support was then taken up in [[UNIX V7M-11\|V7M]].

			(2.9BSD actually supported two different kinds of overlay mechanisms for processes; 'manual' overlays, which required the process to explicitly perform an 'exec()' [[system call]] to read in each overlay [each kept in a separate [[file]]], and "auto-overlay"s, where all the overlays are kept in the same file, and the 'current' overlay is automagically switched when a [[subroutine]] call is made from a subroutine in one overlay to a subroutine in another.)

	''Running Large Text Processes on Small Unix System'' (below) says:		''Running Large Text Processes on Small Unix System'' (below) says:
Line 7:		Line 9:
	: "We describe a set of simple modifications to the Unix system, which permit larger programs to be run than has previously been possible. In particular, the 'f77' and 'a68' compilers and version 2 of the 'ex' editor, which previously would not run on the non-separate I/D machines such as the 11/23, 11/34 and 11/40, may be run, without source code modification, using this scheme. This scheme will also allow processes larger than 65K bytes of instruction space to run on all 11/ cpu's with segmentation hardware."		: "We describe a set of simple modifications to the Unix system, which permit larger programs to be run than has previously been possible. In particular, the 'f77' and 'a68' compilers and version 2 of the 'ex' editor, which previously would not run on the non-separate I/D machines such as the 11/23, 11/34 and 11/40, may be run, without source code modification, using this scheme. This scheme will also allow processes larger than 65K bytes of instruction space to run on all 11/ cpu's with segmentation hardware."

	V7M contains notes ~~that~~ more or less state that its use-of/support-for overlays in the kernel is based on the prior support for overlays in ''processes'', done at Berkeley; and also that overlays were only used in the kernel:		V7M contains notes which more or less state that its use-of/support-for overlays in the kernel is based on the prior support for overlays in ''processes'', done at Berkeley; and also that overlays were only used in the kernel:

	: "Covld is derived from Bill Joy's covld but several things (such as the format of thunks) have been changed .. The paper in ovpap.n describes the original Berkeley overlayed text scheme which was intended for use in user mode programs. I use overlays only for the kernel itself."		: "Covld is derived from Bill Joy's covld but several things (such as the format of thunks) have been changed .. The paper in ovpap.n describes the original Berkeley overlayed text scheme which was intended for use in user mode programs. I use overlays only for the kernel itself."

	This is somewhat incorrect: BSD2.9 ~~contained~~ overlays both in the kernel, and in processes; and V7M seems to have supported use of overlays in processes.		This is somewhat incorrect: first, BSD2.9 did contain overlays in both the kernel, and in processes; and second, V7M seems to have supported use of overlays in processes, also.

	==External links==		==External links==
Line 22:		Line 24:
	* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/man/man1/ld.1 ld (1)]		* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/man/man1/ld.1 ld (1)]
	* [https://www.tuhs.org/cgi-bin/utree.pl?file=V7M/src/cmd/covld/README README] - use of overlays in V7M		* [https://www.tuhs.org/cgi-bin/utree.pl?file=V7M/src/cmd/covld/README README] - use of overlays in V7M
			* [https://bitsavers.org/pdf/dec/pdp11/ultrix-11/Unix_V7M_Release_2.1_Software_Description_Sep81.pdf Unix/v7m Release 2.1 - Software Description] - notes use of overlays in V7M

	[[Category: ~~Unix~~]]		[[Category: UNIX]]

PDP-11 Memory Management

Jnc — Mon, 07 Sep 2026 10:07:35 GMT

+links

← Older revision		Revision as of 10:07, 7 September 2026
Line 1:		Line 1:
	[[PDP-11]]s which provide [[memory management]] use a standard '''PDP-11 Memory Management''' [[architecture]]. When memory management is enabled, the basic 64 Kbyte [[address space]] of the [[PDP-11 architecture]] is divided into 8 '[[segment]]s' (the terminology used in the first version of the documentation; it was later changed to 'pages', for unknown reasons, but this is ~~incorrrect -~~ the functionality provided is that of segmentation, not [[virtual memory\|paging]]).		[[PDP-11]]s which provide [[memory management]] use a standard '''PDP-11 Memory Management''' [[architecture]]. When memory management is enabled, the basic 64 Kbyte [[address space]] of the [[PDP-11 architecture]] is divided into 8 '[[segment]]s' (the terminology used in the first version of the documentation; it was later changed to 'pages', for unknown reasons - probably marketing - but this is incorrect; the functionality provided is that of segmentation, not [[virtual memory\|paging]]).

	Each segment of a virtual address space can be individually assigned to any location in physical [[main memory]]. Each segment can be set to any length between 0 bytes and 8 Kbytes, in 0100 (64.) byte increments. Segments can grow either up from their base address, or down; the latter to ~~accomodate~~ PDP-11 [[stack]]s, which typically grow down (from higher addresses to lower).		Each segment of a virtual address space can be individually assigned to any location in physical [[main memory]]. Each segment can be set to any length between 0 bytes and 8 Kbytes, in 0100 (64.) byte increments. Segments can grow either up from their base [[address]], or down; the latter to accommodate PDP-11 [[stack]]s, which typically grow down (from higher addresses to lower).

	The [[Central Processing Unit\|CPU]] can be in one of three modes; 'Kernel', 'Supervisor', and 'User'; each has its own separate mappings, to provide separate virtual address spaces for each mode.		The [[Central Processing Unit\|CPU]] can be in one of three modes; 'Kernel', 'Supervisor', and 'User'; each has its own separate mappings, to provide separate virtual address spaces for each mode.

	An additional enhancement is that [[instruction]] and data ~~fetches~~ can be set to go to separate 64 Kbyte address spaces, the so-called '''Split I+D space''' capability; this increases the memory available to the operating system, and each user, to 128 Kbytes. Since immediate ~~operands~~ and absolute addresses are stored contiguous with their instructions, fetches of them are considered to be instruction fetches, for the purposes of deciding which space to use for them.) When split I+D is not enabled in any particular mode, only a single 64 Kbyte address space is provided, using only the I-space registers for that mode.		An additional enhancement is that [[instruction]] and data [[fetch]]es can be set to go to separate 64 Kbyte address spaces, the so-called '''Split I+D space''' capability; this increases the memory available to the [[operating system]], and each user, to 128 Kbytes. Since immediate [[operand]]s and absolute addresses are stored contiguous with their instructions, fetches of them are considered to be instruction fetches, for the purposes of deciding which space to use for them.) When split I+D is not enabled in any particular mode, only a single 64 Kbyte address space is provided, using only the I-space registers for that mode.

	[[Trap]] and [[interrupt vector]]s are fetched from Kernel Data space, when memory management is enabled; this allows bad physical main memory at [[absolute address]] 0 to be bypassed, if it is failing.		[[Trap]] and [[interrupt vector]]s are fetched from Kernel Data space, when memory management is enabled; this allows bad physical main memory at [[absolute address]] 0 to be bypassed, if it is failing.

Overlay

Jnc — Mon, 07 Sep 2026 09:51:17 GMT

UNIX V7M-11

Jnc — Mon, 07 Sep 2026 02:02:02 GMT

Link PDP-11 Unix overlays

← Older revision		Revision as of 02:02, 7 September 2026
Line 12:		Line 12:
	'''UNIX V7M-11''' (the official name; it is usually referred to as '''V7M''', '''Unix/v7m''', etc) was [[Digital Equipment Corporation\|DEC]]'s first official [[UNIX]]. It was an "enhanced and modified version of" [[UNIX V7]], for [[PDP-11]] machines.		'''UNIX V7M-11''' (the official name; it is usually referred to as '''V7M''', '''Unix/v7m''', etc) was [[Digital Equipment Corporation\|DEC]]'s first official [[UNIX]]. It was an "enhanced and modified version of" [[UNIX V7]], for [[PDP-11]] machines.

	The primary changes/enhancements in V7M were [[device driver]]s for the then-full range of DEC devices (see the "Unix/v7m Release 2.1 - Software Description", below, for a full list), support for [[~~overlay~~]]s, and "significant work for hardware error recovery". "Much work was put into producing a release that would reliably bootstrap from many tape drives or disk drives".		The primary changes/enhancements in V7M were [[device driver]]s for the then-full range of DEC devices (see the "Unix/v7m Release 2.1 - Software Description", below, for a full list), support for [[PDP-11 Unix overlays\|overlays]], and "significant work for hardware error recovery". "Much work was put into producing a release that would reliably bootstrap from many tape drives or disk drives".

	<!-- ==See also==		<!-- ==See also==

UNIX

Jnc — Mon, 07 Sep 2026 02:00:33 GMT

PDP-11 Unix overlays

Jnc — Mon, 07 Sep 2026 01:33:08 GMT

A poorly documented corner of later PDP-11 Unixes

New page

The small [[address space]] of the [[PDP-11 architecture]] meant that larger computations, both [[operating system]]s and [[application]]s, needed to make use of [[overlay]]s.

Early versions of [[UNIX]] did not support them, but support for them was added in [[2.9BSD]]: use-of/support-for overlays in ''[[process]]es'', in user mode; and also for overlays in the [[kernel]]. This support was then taken up in [[UNIX V7M-11|V7M]].

''Running Large Text Processes on Small Unix System'' (below) says:

: "We describe a set of simple modifications to the Unix system, which permit larger programs to be run than has previously been possible. In particular, the 'f77' and 'a68' compilers and version 2 of the 'ex' editor, which previously would not run on the non-separate I/D machines such as the 11/23, 11/34 and 11/40, may be run, without source code modification, using this scheme. This scheme will also allow processes larger than 65K bytes of instruction space to run on all 11/ cpu's with segmentation hardware."

V7M contains notes that more or less state that its use-of/support-for overlays in the kernel is based on the prior support for overlays in ''processes'', done at Berkeley; and also that overlays were only used in the kernel:

: "Covld is derived from Bill Joy's covld but several things (such as the format of thunks) have been changed .. The paper in ovpap.n describes the original Berkeley overlayed text scheme which was intended for use in user mode programs. I use overlays only for the kernel itself."

This is somewhat incorrect: BSD2.9 contained overlays both in the kernel, and in processes; and V7M seems to have supported use of overlays in processes.

==External links==

* Charles Haley, William Joy, William F. Jolitz, [https://www.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/doc/ovpap ''Running Large Text Processes on Small Unix Systems'']
* Barbara Bekins, Bill Jolitz, [https://www.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/doc/ovtutorial ''How to use the UNIX Automatic Text Overlays: A Tutorial'']
* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/doc/2.9_kernel.ms Changes in the Kernel in 2.9BSD] - includes discussion of the use of overlays in the 2.9BSD kernel
** [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/include/sys/koverlay.h koverlay.h] - "Definitions relating to kernel text overlays."
* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/man/man5/a.out.5 a.out (5)] - this and the next entry describe the support for overlays in 2.9BSD
* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/man/man1/ld.1 ld (1)]
* [https://www.tuhs.org/cgi-bin/utree.pl?file=V7M/src/cmd/covld/README README] - use of overlays in V7M

[[Category: Unix]]

← Older revision		Revision as of 15:07, 12 September 2026
Line 1:		Line 1:
	The '''GE-635''' was a [[transistor]] [[mainframe]] computer; the largest of the [[GE 600 series‎]] family. It is notable mostly for being the design base of the [[GE-645]], a descendant created for the [[Multics]] project (after [[International Business Machines\|IBM]] refused to add necessary features such as [[virtual memory]] to the [[IBM System/360\|System 360]]). The [[Dartmouth Time Sharing System]], where [[BASIC]] was developed, also ran on the GE-635.		The '''GE-635''' was a [[transistor]] [[mainframe]] computer; the largest of the [[GE 600 series‎]] family. It is notable mostly for being the design base of the [[GE-645]], a descendant created for the [[Multics]] project (after [[International Business Machines\|IBM]] refused to add necessary features such as [[virtual memory]] to the [[IBM System/360\|System 360]]). The [[Dartmouth Time Sharing System]], where [[BASIC]] was developed, also eventually ran on the GE-635.

	The basic overall [[architecture]] of the GE-635 was inherited by the later Multics systems: it was a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] modules. [[Input/output\|I/O]] activities were handled by Input/Output Controller modules, similarly sharing access to the memory modules.		The basic overall [[architecture]] of the GE-635 was inherited by the later Multics systems: it was a [[tightly-coupled]] [[multi-processor]], with all the [[Central Processing Unit\|CPUs]] sharing access to a collection of [[multi-port memory]] modules. [[Input/output\|I/O]] activities were handled by Input/Output Controller modules, similarly sharing access to the memory modules.

	The GE-635 was intended for [[multi-tasking\|multi-programming]]; a CPU operated in either Master mode, in which there were no limitations on its operation, or Slave mode, in which [[base and bounds]] type [[memory management]] applied.		The GE-635 was intended for [[multi-tasking\|multi-programming]]; a CPU operated in either 'Master' mode, in which there were no limitations on its operation, or 'Slave' mode, in which [[base and bounds]] type [[memory management]] applied.

	The GE-635 was a 36-bit machine, as was common for scientific computers in that period. The CPU included a single double-width [[register\|accumulator]], and eight [[index register]]s (18 bits wide; the size of [[address]]es in the GE-635). The [[Program Counter]] was also 18 bits.		The GE-635 was a 36-bit machine, as was common for scientific computers in that period. The CPU included a single double-width [[register\|accumulator]], and eight [[index register]]s (18 bits wide; the size of [[address]]es in the GE-635). The [[Program Counter]] was also 18 bits.

← Older revision		Revision as of 09:51, 7 September 2026
Line 5:		Line 5:
	Overlays perform the same basic function as the later [[virtual memory]] ('jam three gallons into a one-gallon bucket'), but do so explicitly, not automatically and invisibly.		Overlays perform the same basic function as the later [[virtual memory]] ('jam three gallons into a one-gallon bucket'), but do so explicitly, not automatically and invisibly.

	==~~External links~~==		==See also==

	* [~~https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/man/man5/a.out.5 a.out (5)] - this and the next entry describe the supports for overlays in the later~~ PDP-11 ~~[[BSD]]s~~		* [[PDP-11 Unix overlays]]
	* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/man/man1/ld.1 ld (1)]
	* [https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.9BSD/usr/doc/2.9_kernel.ms Changes in the Kernel in 2.9BSD] - describes use of overlays ~~in the [[2.9BSD]] [[kernel~~]]

	[[Category: Software Concepts]]		[[Category: Software Concepts]]

← Older revision		Revision as of 02:00, 7 September 2026
Line 91:		Line 91:
	* [[Repairing early UNIX file systems]]		* [[Repairing early UNIX file systems]]
	* [[BSD Fast File System]]		* [[BSD Fast File System]]
			* [[PDP-11 Unix overlays]]
	* [[Unix dump/restore tape format]]		* [[Unix dump/restore tape format]]
	* [[The Unix Heritage Society]]		* [[The Unix Heritage Society]]