Talk:Semi-Automatic Ground Environment

Back Up Interceptor Control System
In the upgrades section it mentions BUIC and BUIC II. But from what I can tell BUIC (I) was a manual backup system not a solid state system an mentioned. See History of Strategic and Ballistic Missile Defense: Volume II: 1956-1972 p 154. — Preceding unsigned comment added by Hal9000w (talk • contribs) 10:57, 14 February 2013 (UTC)

Deployment
There are two Deployment sections! - these should be combined. Ben Finn (talk) 09:21, 16 April 2013 (UTC)

Interceptors
I worked as an intercept director (IND) and Weapons Director (WD) at the Sioux City Airbase (shared with the civilian airport) SAGE squadron and as IND, WD, Senior Weapons Director (SWD), and Senior Director (SD) at the Great Falls, Montana (Malmstrom AFB) SAGE Squadron. Sioux City in 1967-68 and Great Falls in 1969-70. The F-102, F-89, F-101 aircraft we controlled did NOT have the data link controls cited in the article. The F-106s did but it hardly ever worked, usually couldn't even be initiated on the ground before takeoff and when it was it crashed during the airborne mission.

The article refers to the large projection light boards visible to everyone in the direction center. We DID NOT have such boards. The only time I saw such a light board was in a remote AC&W site in Alaska where there was a single surveilance console which provided the input to airman behind the large lighted board who marked the board with erasable markers. 69.145.131.45 (talk) 17:26, 29 September 2014 (UTC)

This article includes this paragraph: "Aerospace Defense Command aircraft (e.g., F-94 Starfire, F-89 Scorpion, F-101B Voodoo, F-4 Phantom) were controlled by SAGE GCI. The F-104 Starfighter was "too small to be equipped with [SAGE] data link equipment" and used voice-commanded GCI, but the F-102A & F-106 Delta Dart were equipped for the automated data link (ADL). Interceptors reaching the target were able to transmit "raid assessment information back to the direction center" to allow additional dispatches.[44] Familiarization flights (fam flights) for SAGE weapons directors to fly on board two-seat interceptors allowed them to observe GCI operations. Surface-to-air missile installations for CIM-10 Bomarc interceptors displayed on SAGE consoles, e.g., "FOX" and "BED" for the New York sector (the latter at Suffolk County Air Force Base with 56 launchers)." This is a mess. F-94s didn't have the SAGE datalink installed - too early. F-4s were an ad hoc addition to the ADCOM/ADTAC family and also didn't have the datalink installed - too late. F-86L Sabres did - but arem't even mentioned here. It would be interesting to see where the F-104 quote comes from - doubt it was really size that matters here, but rather that the A-model's very limited (non-existent might be more accurate) fire control system wouldn't really benefit from the SAGE system or vice versa. This quote isn't from the only cited reference in this paragraph - for that matter, most of what is claimed in this paragraph isn't from this reference. The last couple of sentences look like someone just tossed them in - with no citations that can be verified, sounds like original research and not really all that germain. The last sentence in particular about the specifics of the New York Sector's BOMARCs is really odd - sounds like some veteran just dumping information - why the heck do we need to know that Suffolk County AFB had 56 BOMARC launchers? — Preceding unsigned comment added by 68.51.66.32 (talk) 05:12, 25 April 2013 (UTC)

Computing specs
What sort of resolution and refresh rate did the displays have? Or are some of them analog? They look...round. How many computations per second could the CPUs (did they even have CPUs?) make? I'm wondering how the system compares to modern computers. It seems a stretch for the early digital technology of the time to digitally process and display the analog radar returns, but then again SAGE was groundbreaking and incredibly expensive. Burroughs AN/FST-2 Coordinate Data Transmitting Set says that system digitized radar signals with quarter-mile resolution. -- Beland (talk) 19:26, 7 May 2013 (UTC)

I (Douglas R Bohrer) am a SAGE veteran. I was a Computer Systems Analyst Officer stationed at the 26th Air Division HQ at Luke AFB, AZ,from August, 1972, to September, 1974. I never saw the 2 story big screen at either Luke or at Fort Lee, Virginia, where I received 10 days of training. The screens I saw were 2 CRT's in a single console. The radar picture was displayed on a Situation Information Display (SID) which contained the raw radar returns. I think it was about 12-14 inches in size. To the right of it was a smaller Digital Information Display (DID) which was controlled by a light gun, a precursor to the computer mouse. The light gun worked by detecting the flash when the electron beam painted the screen. You pointed the gun at a radar return on the SID, pulled the trigger, and digital information about the radar return was displayed. This would include the last reported altitude for example. You were able to associate information you entered on a keyboard with a radar return. Then the computer would try to move the track information to what it thought was where the radar return for the tract showed up next. Remember the radar was mechanically rotating. Actually there were two tracking algorithms. SR-71 Blackbirds and YF-12's were so fast (mach 3 cruising speed) that they needed a special algorithm just for them. If I remember right, the cycle time for the SID, that is the refresh rate, was about 20 seconds. During exercises, the computer would be generating the simulated enemy radar returns as well as jamming and chaff, reflective metal confetti. During exercises the refresh rate for SIDs got as bad as 2 or 3 minutes, which would be announced over the public address system so everyone knew how long it would take for your action to show up on your screens. The instruction execution time was 6 milliseconds. However, a full 32 bit multiply took 18 milliseconds. It was a 32 bit machine. It was 1's complement, so negative zero was all the bits set, while zero was none of the bits set. Since I programmed utilities that evaluated air defense exercises, I don't know too much detail about how the active air picture was maintained, but the machine did process analog radar returns and used a tracking algorithm to keep track information associated with where it thought the actual airplane went. I want to say the memory was 300K, but I'm not sure I remember that right. I do remember each bit was a small ferrite core, with wires crossing at each little core. The memory was visible behind glass, and looked the a wire window screen. There were 4 tape drives and a drum for mass storage. Disks hadn't been invented in the 1950's. I don't remember how much storage was available on the drum. The tapes were 200 bpi. They had 3 bits of information, a sync bit and then 3 more bits of information written across the magnetic tape. The sync bit told the tape drive that there was information there. There was no parity bit. I hope this answers some of your questions.167.127.218.62 (talk) 22:09, 12 March 2014 (UTC)

Opinion - REMOVE THE PAGE
It's my opinion that this page should be at least confidential and kept to USA mil personell only. Why?

It is TMI "too much information" military information. It names too many locations and system name keywords current and past. And the "encyclopedic reader" really doesn't need every particle of military information. And being from the U.S.A. I really don't feel safer where so much military info is accessed by (future) combatants.

I worked in SAGE from 1976 to 1980 as a Weapons Control Tech and as a Weapons Director Tech and any information about these systems is out dated and is not a security issue. These computers used second generation vacuum tube computers with a normal frame time of 14 seconds. In combat situations that time could easily increase to 30 seconds. With fighters moving at speeds in excess of Mach 2 that made this system difficult in being able to know exactly where your interceptor and target were located. (Rickydee2) — Preceding unsigned comment added by Rickydee2 (talk • contribs) 20:45, 14 January 2014 (UTC)

For example. The Electronics pages say plenty about Analog Computers and exactly how they work. There should be no need to know how and in which named armaments the US or British military may be using that technology. Rather, there are penty of non-military commercial systems that could be named and explored.


 * Dude, I think you've been wearing the tin foil hat too long. SAGE is an antiquated system - it was obsolete almost the day it went operational back around 1960 and hasn't been in service in what 30 years. You might as well say that an article on the Sherman tank or the P-47 gives away vital military information. — Preceding unsigned comment added by 192.158.48.9 (talk) 10:56, 26 July 2013 (UTC)

redab
Looking at the history of this page, I see that some of the links to disambiguation pages that I just fixed were previously fixed: I don't see a reason to link to the disambiguation pages, but correct me if I'm wrong. -PC-XT+ 04:00, 7 July 2013 (UTC)
 * my dab fix, July 7, 2013
 * JustAGal's dab fix, March 5, 2013

What a mess!
What a horrible mess this article is. Do editors ever read their work back and ask themself "Does any of this make sense to an average reader?". I can only assume not. Virtually every sentence is a mass of references, links (even ones that go nowhere), abreviations, parentheses and quotations. And why does every piece of information that comes from a reference have to be quoted? I'm a technical person with a science and computing background but reading this article has simply given me a headache. 3/10. 87.114.7.76 (talk) 20:07, 23 February 2014 (UTC)

SAGE System Programming and SDC
The SAGE system and the Sage System Training Program (SSTP) were maintained by System Development Corporation in Santa Monica, California, and there was a simplex SAGE computer there in the 1950s, at 2400 - 2500 Colorado. The buildings have been demolished and replaced by multi-use commercial buildings. I was employed at SDC for over two decades and worked briefly on the SSTP there, from 1958 to 1961. I do not know when the SAGE computer was taken away nor where it went. There is a separate Wikipedia entry on System Development Corporation, which had been spun off from the RAND Corporation in 1957 and which was the first company devoted specifically to software development.

SAGE cost
"IBM's production contract developed 56 SAGE computers for $½ billion (~$18 million per computer pair in each FSQ-7)[30]—cf. the $2 billion WWII Manhattan Project."

Aside from the nonstandard "$½ billion" ($500 million), this is just IBM's contract. [The same source](http://ed-thelen.org/Sage-Talk.html) says the total cost was between "somewhere between four and twelve billion dollars, in dollars of around 1960," which just to put into perspective is around $40-120 billion in 2015 depending on what kind of deflator you use. Which is to say, a fairly high price tag. _That_ is the number that should be compared to the Manhattan Project (which was $2 billion overall, not a single contractor, and that works out to ~$30 billion 2015 depending on the deflator used). So SAGE was more expensive, probably much more so, which is not at all surprising (or damning, or whatever), given that it lasted for a lot longer than the Manhattan Project. — Preceding unsigned comment added by 155.246.200.14 (talk) 21:56, 6 December 2016 (UTC)

Lead photo
The direction center in the lead photo is the one at McGuire AFB, readily identifiable from Google Maps by the non-perpendicular intersection next to it. The Stewart AFB building does not have this feature. The source identifying it as such is mistaken. 24.7.14.87 (talk) 07:18, 15 August 2018 (UTC)