Talk:Whirlwind I

Nobody calls this thing "MIT Whirlwind"

Nobody calls this thing "MIT Whirlwind". It's always just plain "Whirlwind", and the context make it obvious that it's not the aeroynmaic event. I'll do a stub disambig page from "Whirlwind", indicating the aerodynamic event, and linking here.

Noel 20:08, 18 Aug 2003 (UTC)

What came of the Whirlwind?

Two things on this section:

• It mentions only Ken Olsen, and claim he "led" the TX-0 effort. I believe Was Clark was the lead on the TX-0 design, and at least should be mentioned.
• It mentions that TX-1 came after TX-0, was "too ambitious" and was scaled back to the TX-2. However, The Dream Machine claims that TX-1 was the original proposal for a "Whirlwind with transistors", was nixed, TX-0 done in its place, and that TX-2 was then used as name for the follow-up to TX-0. Anyone know the facts here?

• • The claim that Whirlwind is the predecessor of TX-0, or for that matter that TX-0 and friends are the predecessor of the PDP-1, seems quite a stretch. Those other machines are 18 bit machines. And a 2 minute inspection of the relevant manuals reveals that (a) the Whirlwind instruction set is nothing like that of TX-0, (b) the TX-0 instruction set is nothing like that of the PDP-1. Paul Koning (talk) 18:31, 6 September 2017 (UTC)

In what way was the whirlwind the first that did not replace a mechanical system?

There were no anolgues mechanically used in Tnnny breaking by Colussus.

Project Whirlwind began in 1944, the letter of intent ("Project Whirlwind", p.14), with an intended electromechanical control system. Later that was replaced by the digital computer ("Project Whirlwind" chapter 3 "The Shift to Digital"). Whirlwind (computer) DID replace a mechanical (electromechanical) system, even if that system was never completed. 69.106.238.70 (talk) 06:06, 26 June 2009 (UTC)

My perspective on reading this was that the systems which computers like ENIAC, EDSAC, and BRLSEC replaced were organizations of human computers—considered to be clerical workers—working at desks in a "computer room" with pencil, paper, and mechanical desk calculators. There were also mechanical cash registers, adding machines, analog mathematical function analyzers, aviation computers (e.g. the Norden bombsight, and mechanical analog computers for fighter aircraft flight control), etc. that were eventually replaced with programmable electronic computers, eventually ones that used the same memory for program instructions and data. So the statement in the article makes general sense to me: it is historically plausible. But I have no idea if Whirlwind was the first electronic computer not intended to replace a system or adapt a process that previously had been realized without an electronic computer. Extraordinary claims call for very strong support, so very credible citations are needed for the claim that Whirlwind was the first electronic computer used for such a novel application.

As for COLOSSUS at Bletchley Park, I read a good bit about that a day or two ago, from a serious historical book published in 1980, and I can confirm that none of the predecessors of Colossus were purely mechanical, though the original HEATH ROBINSON machine used more mechanical components than COLOSSUS, having two punched paper tapes kept synchronized by sprocket wheels. HEATH ROBINSON used electronic counters and logic, and the chief advancement of COLOSSUS seems to have been the elimination of the second tape and the sprocket wheels by computing Boolean functions of the original cyphertext electronically rather than reading them from a second tape. But COLOSSUS itself used mechanical-optical memory—the paper tape—which was read over and over while the machine ran, so COLOSSUS was not an all-electronic system. Also, another type of machine used at Bletchley Park to aid code breaking, called a "bombe" (if I spelled that right), was entirely (electro-)mechanical, as I understand. But I have no information that a COLOSSUS or a HEATH ROBINSON was at all similar in function and specific purpose to a "bombe", and my hazy understanding is that it was quite different, and that the bombes continued to be operated after the COLOSSUS machines were in full operational use, while the HEATH ROBINSONs were retired long before that (because they did not work very well, hence the development of better machines, culminating in COLOSSUS).

173.49.122.138 (talk) 11:36, 9 March 2019 (UTC)

Merger proposal

Rationale: Duplicate subjects

The new article Information Technology: At the dawn of the computer age is about Whirlwind, including many claimed firsts for Whirlwind and an enumeration of later projects that are said to have benefited from Whirlwind. There is no Wikipedia user benefit from a 2nd Whirlwind article; the Whirlwind details should be merged into the existing Whirlwind article (a 60th or 2^6th or ... anniversary is not justification for a redundant article).

btw, the many claims of "firsts" in that article should not be merged into the Whirlwind article, but instead should be placed on this talk page together with the suggestion that anyone moving those claims to the article also add explicit, non-controversial, references for each claim moved. 69.106.238.70 (talk) 05:08, 26 June 2009 (UTC)

NPOV?, citable facts in summary?

I'm wondering about the statement "...and indirectly to almost all business computers and minicomputers in the 1960s." I don't think you're required to cite things in the summary, but that's a pretty broad statement, and how it influenced this development is not mentioned or clarified later in the article (only how it affected SAGE). Wolverine00000 (talk) 23:30, 12 July 2009 (UTC)

I strongly disagree with this merge, I beleve that this specific model of computer should have it's own computer article. —Preceding unsigned comment added by 98.108.201.126 (talk) 04:55, 14 March 2010 (UTC)

N42

The infobox says that the building is now called N42. However both floorplans.mit.edu and whereis.mit.edu say that such building doesn't exist. I can only find 3 mentions of N42 building: this Wikipedia article, draft blueprints for the building from 1996, and an old news article from 1998. — Preceding unsigned comment added by 176.36.231.82 (talk) 22:13, 12 July 2016 (UTC)

Origin of hexadecimal notation using ABCDEF?

The origin of the hexadecimal ABCDEF letters is typically attributed to IBM (though we are still searching for a reliable reference for this), but I found the following comment in an old archived thread of the hexadecimal talk page:

IBM certainly was not the first to use A-F. Such was in use from the late 1940's through the late 1950's at MIT's Wirlwind Project - a 16 bit binary computer - I joined the project in 1952.

[...]

—Preceding unsigned comment added by Kp2a (talk • contribs) 16:31, 31 July 2008 (UTC)

I've gone through a few Whirlwind I documents at Bitsavers but could not find this ABCDEF notation being mentioned there so far. If you stumble upon Whirlwind related documents using the ABCDEF notation, please add this info to the hexadecimal article or join the discussion at Talk:Hexadecimal#Origin_of_hexadecimal_notation_using_ABCDEF. Thanks.

--Matthiaspaul (talk) 18:59, 5 June 2017 (UTC)

Operation time stats according to DIGITAL COMPUTER NEWSLETTER

Operation

A number of tests have been run without error for hours at a time in 1949 (32 registers of test storage) and in 1950 (electrostatic storage).

In January-March 1951 Whirlwind has been operating usefully about 30 hours a week (with satisfactory operation about 85% of the time). From April it was 35 hours a week (about 90% trouble-free).

From March/April 1952 the number of hours was Increased from 30, through 50, to about 70 hours in June. From August average weekly operation time increased to 99 hours (85% useful time).--89.25.210.104 (talk) 22:46, 21 May 2018 (UTC)

References:

"Records" box at bottom of article

The "Records" box near the bottom has no sources and doesn't link to anything related to its topic. E.g. "Fastest computer in the world" ends up as a link to "Mainframe Computers" which isn't at all the same thing and which doesn't mention or involve Whirlwind at all. While it might be fun "original research" to make such a list, or to find sources for such a list, this thread of random unsourced claims across the bottom of multiple articles doesn't seem to be a good idea in its current form. Gnuish (talk) 22:25, 28 September 2020 (UTC)

I think that there used to be an article about the fastest computer through history, but it must not be there now. Other articles have this, for instance IBM NORC. It is probably useless to link to the mainframe article, unless it has a section about it. Bubba73 ^{You talkin' to me?} 22:54, 28 September 2020 (UTC)

Bill Wolf/Wulf

I edited this article to replace "Bill Wolf" by "Bill Wulf", but the change was reverted. Although I have no firsthand knowledge of the reference, Bill Wulf was a noted computer scientist who was interested in computer history; I've never heard of a Bill Wolf. Rather than get into an edit war, which I have no patience for, I'm simply noting here that the change should not have been reverted. (As a senior at MIT in 1952, I actually did a bit of programming on Whirlwind.) Paul Abrahams (talk) 23:42, 27 March 2024 (UTC)

You made the change from an IP address with no edit summary, so it was indistinguishable from vandalism.

I have a theory. There are two prominent Bill W[ou]lfs in computer science. I'm thinking Bill Wolf and Wolf R&D were the Whirlwind, Bill Wulf is the computer scientist you are thinking of.

Some refences for Wolf: the Computer History Museum has a collection of Whirlwind items from "Bill Wolf" (example, google search), including CHM's oral history of Gordon Bell, which says "A guy by the name of Bill Wolf had acquired the Whirlwind in the late ‘60s" pdf, and references to the "Wolf Research and Development Corporation". This MIT report talks about Wolf R&D (though not Bill). Here's a "progress report" signed by William M. Wolf in 1963.

Some references for Wulf: there's a scattered reference or two to Wulf in the MIT archives: 1 2. The National Academy of Engineering president is "Dr. Wm. A. Wulf". The Smithsonian orial history of Gordon Bell mentions Wulf and says "Bill Wulf, a fellow professor who eventually became the President of the National Academy of Engineering". None of these seem to be in the context of Whirlwind. tedder (talk) 00:07, 28 March 2024 (UTC)

And more- here's William Martin Wolf's obituary, which mentions [ his autobiography "No E"] (he's not Wolfe). The 1964 LIFE article on him makes it very clear. I think this is just two prominent people who have close names- there's no connection from Wulf to Whirlwind or Wolf R&D. tedder (talk) 00:31, 28 March 2024 (UTC)

You're absolutely right. I had never heard of Bill Wolf, and (unlike Bill Wulf) he's not listed in Wikipedia. Furthermore, they had mostly overlapping lifespans: Bill Wolf (8/29/1928-4/25/2015 and Bill Wulf (12/8/1939-3/10/2023). But despite his not being in Wikipedia, Bill Wolf was well known in the computer field during his lifetime and is indeed associated with Whirlwind. Paul Abrahams (talk) 02:25, 29 March 2024 (UTC)

It seems like William M. Wolf deserves a Wikipedia entry. Unfortunately I'm not the one to create it. Paul Abrahams (talk) 14:45, 29 March 2024 (UTC)

Historical recollection

I don't know if any of this belongs in the article, but I'm one of the very few (perhaps the only) living people who actually wrote and executed a program on Whirlwind.

I graduated from MIT in 1956.My senior thesis was on linear programming, and as part of it I wrote a program for Whirlwind. (I don't remember what that program actually did). I prepared it on paper tape that I generated on a Flexowriter. Whirlwind was visually very impressive; lots of flashing lights and clacking sounds. When a program aborted, the programmers would take a picture of the machine tom capture the contents of the registers. Paul Abrahams (talk) 02:49, 29 March 2024 (UTC)