Talk:IBM Personal Computer AT

Advanced Technology?
Because the AT used various technologies that were rare at the time in personal computers, the name AT originally stood for Advanced Technology. Is this accurate? True, the AT came with a hard drive which was sort-of advanced. And Mueller's book says "AT" stands for "Advanced Technologies". I think the bit-mapped display of the contemporary Macintosh was a bit more advanced than the AT video display, but Macintoshes were rare at that time, too. The AT came with a real-time clock and CMOS to set up the system, unlike the DIP switches that the origianl PC had. I suppose that's an advance. --Wtshymanski 00:38, 8 Jun 2005 (UTC)


 * It was advanced from IBM's point of view, regardless of what Apple was doing then. The XT (eXtended Technology), which predated the AT, had a hard drive.


 * It was considered 'Advanced Technology' in that it used the Intel 80286 processor. The XT (eXtended Technology) was simply a PC with slightly more room for expansion and a hard drive. Howard81 16:09, 26 June 2007 (GMT)

I was PC Specialist at IBM United Kingdom Technical Support at the time of the announcement of the PC AT (It was NEVER called the PC/AT.) and I can confirm that AT stands for Advanced Technology. It was Advanced Technology because is had an Intel 80286 and several other advanced features over the original PC and PC XT. For example: a 16-bit Data Bus, 24-bit Address Bus, and two daisy chained interrupt controllers. —Preceding unsigned comment added by 81.147.151.117 (talk) 17:40, 1 April 2011 (UTC)


 * The IBM Archives tend to say "Personal Computer/AT" e.g.: . Yet the manuals say "Personal Computer AT" repeatedly e.g.: (slow link). ( Hohum  @ ) 01:19, 22 July 2011 (UTC)

May I point out that is not "The IBM Archives" that is one document in the IBM Archives and it is incorrect. Thee person who wrote it back then, simply got it wrong. I look after an IBM Archive in the UK and I can assure readers that there is NOT ONE occurrence of PC/AT in ANY IBM manual. It is ALWAYS Personal Computer AT or PC AT. Terry, UK. — Preceding unsigned comment added by 86.134.45.70 (talk) 09:09, 22 August 2011 (UTC)
 * I didn't say that document was the entire archive. I browsed several, some of them used that nomenclature - IBM is a reliable source for IBM nomenclature. I'm pointing out that it isn't as clear cut as you are trying to make it out to be. I was an IBM qualified engineer - but personal experience isn't used as a source here. It may not be the official designation, but if it's used by IBM, it gets mentioned as being used in the article. ( Hohum  @ ) 01:01, 23 August 2011 (UTC)

Forgive me but one document does not a tendency make. It IS as clear cut as I am trying to make it. IBM NEVER used the term PC/AT. The quoted document is at variance to ALL of the official manuals and documents. Hell, I wrote one of them! Terry — Preceding unsigned comment added by 86.134.72.177 (talk) 17:52, 11 October 2011 (UTC)
 * Sigh. IBM's own support site: http://www-01.ibm.com/support/docview.wss?uid=pos1R1001804 sometimes uses PC/AT, so it's clear IBM does sometimes use that style. ( Hohum  @ ) 20:54, 11 October 2011 (UTC)

So that is two erroneous documents. (Neither of them manuals.) Keeping trying, you may find three - IBM is big company. — Preceding unsigned comment added by 86.134.24.34 (talk) 13:45, 26 October 2011 (UTC)
 * IBM is a reliable source on IBM nomenclature. You aren't. " IBM NEVER used the term PC/AT." They clearly have. Additionally the issue is whether PC/AT is sometimes used, not whether it was the official name. ( Hohum  @ ) 18:45, 26 October 2011 (UTC)

I will have a note added to those articles in the archive — Preceding unsigned comment added by 86.141.226.189 (talk) 11:56, 12 March 2013 (UTC)

Keyboard
The article should note that the AT keyboard is not compatable with the IBM PC or PC/XT or clones thereof, and that many keyboards were produced with an XT/AT or PC/AT switch to work with either until the PC and PC/XT systems began to be phased out of use in the early 1990's. (I saw many lots of PC and PC/XT systems sold at auctions for a few dollars a pallet in 1990~1992.)


 * It should be noted that the AT keyboard is incompatable with the PC and PC/XT for these two reasons. 1. Different scan codes. 2. The A20 gate control circuitry is in the keyboard. The A20 gate is used for switching to Protected Mode. An IBM AT (or any clone without a 'fast A20' BIOS setting or setting to boot without a keyboard or to not test the keyboard during POST) will not function at all without a keyboard connected. Later clones moved the A20 control onto the motherboard and had a 'Fast A20' setting in the CMOS setup to use it, then could be booted without a keyboard.


 * The A20 gate control circuitry is NOT in the keyboard. In all models of the IBM AT, it has always been in the 8042 keyboard controller chip, which resides on the motherboard. The purpose of the A20 gate is to emulate the PC's 8088 when the 80286 CPU is in real mode (that is, emulate the PC's wrapping of addresses at the 1 MB mark). The IBM 5170 can't boot without a keyboard because it presents "301-Keyboard Error (RESUME = F1 KEY)" on screen and because there is no keyboard, the F1 key can't be pressed to resume to boot sequence. 'Fast A20' is functionality found on certain clone ATs that disables/enables the A20 line faster than what the keyboard controller chip can do it. 'Fast A20' can cause problems which is why most BIOS allow users to turn the functionality off. Sphere808 (talk) 04:27, 26 December 2010 (UTC)

Power Supply
I don't recall ever seeing a pushbutton swtich at the end of a power cable. I thought the PC and XT (and possibly AT) had huge orange toggle switches, while modern ATX supplies have rocker swtiches.


 * Most later AT cases, especially tower styles, had the power switch (toggle, rocker, or locking pushbutton) mounted on the case front and used four wires carrying line current between the switch and the power supply. Some proprietary OEM manufacturers, like Packard Bell, used a long plastic or metal rod that pushed a switch inside the power supply case. At least one company used a toggle lever with a metal wire rod that moved an internal toggle switch mounted on the power supply.


 * I can confirm that the PC, XT and AT all had the large red toggle switches on the side of the case. Howard81 16:07, 26 June 2007 (GMT)

DMA channels
16 DMA chnnels? I don't understand. I have always thought DMA channels were from 0 to 7. —The preceding unsigned comment was added by 82.155.155.251 (talk • contribs) 19:40, June 30, 2006 (UTC)
 * It seems you are correct. The incorrect number in the article was copied from a web site with bad information. Thanks for the heads up.  --Blainster 00:41, 6 July 2006 (UTC)

Who deleted this?
Why is this article deleted? This is certainly notable enough for Wikipedia. —The preceding unsigned comment was added by Nick Warren (talk • contribs) 06:27, September 19, 2006 (UTC)
 * The article has not been touched in nearly three weeks. Did someone steal your account to make this edit?--Blainster 22:06, 19 September 2006 (UTC)

No, it looks deleted to me. I'm serious. I could take a screenshot and show you if you don't believe me. I'll go look at it through a proxy and see if it looks deleted... Nick Warren 04:58, 20 September 2006 (UTC)

Well, my initial theory seems incorect. It also looks deleted to the proxy, so it's not just me. Does it really nto look deleted to you? Weird. Nick Warren 05:01, 20 September 2006 (UTC)


 * The page looks fine to me. I recently saw an image page that was redlinked, but had the image there plain to see. I assume that was a system bug.  Is that what you see here?  --Blainster 21:46, 20 September 2006 (UTC)

Something was messed up. I re-saved the page and it comes up now. Mirror Vax 22:00, 20 September 2006 (UTC)


 * Maybe the database servers are really slow, but I'm still seeing September 1, 2006 as the last edit. --Blainster 23:18, 20 September 2006 (UTC)
 * That's right. Re-saves never appear in the history (try it). Mirror Vax 00:04, 21 September 2006 (UTC)

OK, it looks fine now. What happened?Nick Warren 11:53, 22 September 2006 (UTC)

80286 processor underutilized
Should explain how DOS did not take advantage of most of the 80286 features, so most people with PC AT computers were using the 80286 chip as a fast 8086. As late as 1991, it was fairly common for even people with 80386 computers to be running DOS with the chip in 8086 emulation mode (this only changed with the wide adoption of Windows 3.1 beginning in 1992). AnonMoos (talk) 09:11, 5 December 2019 (UTC)


 * The 286 processor was designed with a protected mode architecture that required restarting the processor to switch back, which made it unattractive for software developers to use the PM architecture. One piece of software that took advantage of this architecture, but never had to switch back to real mode was IBM Xenix 1.0, another was the EMM driver for DOS.

Deskview, and Topview took advantage of memory addressing of extended memory, by using it as expanded memory. ( extended memory was the 16 bit addressing, and expanded memory was using a window above 640K to swap memory into addresses that could be addressed by real mode ).

Mot people ran the AT with DOS, because DOS was most popular, and XENIX was $500. There were several other OSs such as THEOS. The 386 chip added a Virtual 8086 mode, but running DOS on a 386 under windows was interesting: If you had a swap file set up, the DOS window would be a Virtual 8086 mode process, but if paging/swapping was turned off, the DOS prompt would be run in Real Mode. I have not done extensive testing on this, because my interest lies in Xenix.

I used both Windows 2.0.1, and Windows 386 2.1 on a 386, but never used the VIRT8086 to check the mode the DOS prompt was running in. When the 386 came around, because XENIX for the 386 was $500, I switched to Microport UNIX, and ran dual drives, and never checked out wheather Windows 3.1 ran a DOS prompt in real mode, if paging was turned off, but I assume that it did. Most common was to either run Windows applications in windows, and not use the DOS prompt, or not start windows and run the DOS application. I would say it was uncommon to run DOS prompt in Virutal 8086 mode, and even less common to run a DOS application in Virutal 8086 mode for that reason. I supported many users with 386s.

The Maximum memory on an IBM AT was 4MB using stock IBM Cards, ( 512K on the mother board, and 7 512K cards ), and using after market adapters you could only get ONE manufacture of card to sit at an address above 4MB, and it would only add 1MB for a maximum of 5.25MB. Despite the ability of the processor to address up to 16MB - 63k, 142.254.26.9 (talk) 18:29, 26 June 2020 (UTC)

Requested move 8 February 2023

 * The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion. 

The result of the move request was: moved. (closed by non-admin page mover) Vpab15 (talk) 22:00, 16 February 2023 (UTC)

IBM Personal Computer/AT → IBM Personal Computer AT – I searched for IBM PC AT, IBM PC/AT, IBM Personal Computer AT, and IBM Personal Computer/AT in Google, but with 2 characters at the start (+") and the 2nd character as the last one. This page should be moved per WP:COMMONNAME since it is more common to write without a slash. 176.88.83.247 (talk) 18:52, 8 February 2023 (UTC) The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.