TIME DILATION ONC PC'S: "I hope the EXPERTS jump in with both feet"
The following are some snippets from the FAQ on the 'time dilation'problem in (older) PC's:
gmi.edu
Such older PC's may be contained in systems such as POS, ATM's, Printers. In cases where these systems do use the RTC, it might have this problem. More investigation is needed by EXPERTS (i.e. BIOS / RTC experts).
The compilation has been made by Jace Crouch; great work!
--------------------
NOTE: This file is a collection of messages that address the y2k-related phenomenon known as "Time Dilation" that have been posted to the newsgroup comp.software.year-2000.
Here's hoping that this phemonenon is some sort of an extremely rare computer aberration!
Peace & Preparedness,
Jace Crouc
--------------------
Date: Wed, 27 Aug 1997 12:14:26 -0400
From: Jace <jcrouch@gmi.edu>
Newsgroups: comp.software.year-2000
Subject: y2k - strange results in DOS 6.22
On Wed, 27 Aug 1997, J. L. wrote:
> I haven't done any WORK on the PC's SIMULATING Y2000 and beyond - because I don't want my date and time stamps on some files messed up. But I'm prepared to take a tape backup of my PC and then to run a proper Y2000 test to see if anything freaky happens.<
I did this with one of the computers in my office:
AMD 286-12, 8 megs RAM
SCAT motherboard, AMI BIOS
<NOTE: it is a C&T SCAT BIOS, I was mistaken when I wrote this. JC>
210M IDE hard drive
New multi-I/O card w/ 2 fast serial ports
MS-DOS 6.22 (no HD compression)
Running a combination of DOS & win 3.1 apps
Don't laugh. It's still a screamer on Word Perfect 5.1, and with a SLIP connection is really fine for running a UNIX shell account! There are plenty of other boxes on campus that are even older; professors tend to keep their newer machines at home and use their old ones at the office.
I did a backup, set the clock to 12-31-99, 11:55:00 and watched. The date rolled over to 00, and various WP and Microsoft applications showed a date of January 1, 2000. Files were saved to the HD with a year date of 00. The win 3.1 file manager read a year date of :0 on these files.
I let the system run for two weeks, and used it from time to time to make overheads, handouts, or handle e-mail. No files were uploaded or downloaded to our UNIX server, although I did keep using the SLIP connection to run Procomm as a terminal emulator. I'd leave the machine on all day, turn it off on nights and weekends. Normal stuff.
Here's what happened:
1. The system clock "ran" extremely rapidly. After two weeks, the system date was mid-December 2000. The date reported in CMOS and reported by various WP and Microsoft applications was identical. Whatever date the RTC reported, the applications displayed. Files were saved with a 00 year date, but win 3.1 file manager displayed a :0 date. These files were readable, writable, and seemed otherwise normal.
2. After about ten days, the system would not recognize that I had two serial ports. For whatever reason, every system test that I ran reported only one serial port. This was the case with Norton Utilities 7, MSD, and an old WP utility. Nothing else started shutting down, but I figured that if a serial port went down, anything could be next, maybe trash the hard drive in some strange time-warp way.
3. Once I set the system clock back to the correct 1997 date, both serial ports were recognized, and they worked fine. I have no idea why the clock"ran" so fast in y2k, nor why the system stopped recognizing the second serial port. This was enough to convince me that even on a simple PC, the y2k problem can cause hardware failures. Fortunately, I was running a word processor to prepare a few handouts and overheads for a history class, not administering the power grid for a three county area.
Peace & stocking up on sterno and propane and MRE's,
Jace
----------------
Date: 10 Oct 1997 16:05:09 GMT
From: RonKenyon <ronkenyon@aol.com>
Newsgroups: comp.software.year-2000
Subject: Re: Time Dilation thread posted
<snip>
I hope the EXPERTS jump in with both feet. What we have now is some anecdotes taken by credible observers, some sporadic, some reproducible -- all of which would not be surprising if it turns out to be a memory over-write in varied configurations whose instructions or data streams were munged at varied
states.
It's possible we're looking at more than one root problem.
It's possible some of this has been going on forever, just never has been closely observed before, and dismissed amid the din of "%#@& PC" problems .
It's possible some inadvertent misrecording or uncontrolled testing has crept into the mix, or that we've been testing spares that landed in "spares" inventory because they never worked to begin with.
It's unlikely c.s.y2k will sort this out by reverse-engineering the BIOS logic and board design of varied makes, models, configurations and patch levels. We're at the point where the EXPERTS should be walking forward through their designs, looking for windows of vulnerability, starting from the best-documented, most consistent failure-prone units we can find. Who are the EXPERTS? Depends on the build and BIOS.--
RonKenyon
------------------
Date: Mon, 20 Oct 1997 14:13:08 GMT
From: Mike echlin <echlinm@aecl.ca>
Newsgroups: comp.software.year-2000
Subject: Re: Time Dilation thread posted
<snip>
Test machines for the new test.
286-failed.
386 dx-33 passed
386-40 failed
486-33 passed
486-33 failed
486-66 passed
486-66 failed
pentium 2 266 passed
pentium 75 failed.
(all those that failed this test, with the exception of the p 75 also
failed the two week test, the p 75 was never tested on the two week
test.)
<snip>
Mike. |
