Path: utzoo!utgpu!news-server.csri.toronto.edu!helios.physics.utoronto.ca!ists!yunexus!xrtll!silver From: silver@xrtll.uucp (Hi Ho Silver) Newsgroups: comp.sys.ibm.pc.hardware Subject: Re: DOS and BIOS clocks Message-ID: <1991Jan8.042151.4476@xrtll.uucp> Date: 8 Jan 91 04:21:51 GMT References: <1991Jan5.181959.10764@xrtll.uucp> Reply-To: silver@xrtll.UUCP (Hi Ho Silver) Organization: Yeah, right. Lines: 18 In comp.sys.ibm.pc.hardware, iisakkil@xaloc.tky.hut.fi (Mika R Iisakkila) typed: $ Actually, it isn't quite this stupid - there's usually a one $bit 'date changed' -flag, which is set at midnight. If it's set, the $bios calls add one day to the date. So you'll have to have the system $up without rebooting for two successive midnights to lose a day. BIOS calls? They're used to access the hardware-maintained clock, which has no problem with date changes. The software-maintained clock (at least in most DOS versions) may well set a flag to say the date has changed, but it doesn't actually change it even if you run your machine from 11:58 p.m. until 12:02 a.m. Specifically, I know MS-DOS 3.20 is broken; I've run that from evening until early morning dozens of times and it doesn't change the clock. I'm pretty sure PC-DOS 3.30 is also broken. -- __ __ _ | ...!nexus.yorku.edu!xrtll!silver | always (__ | | | | |_ |_) >----------------------------------< searching __) | |_ \/ |__ | \ | if you don't like my posts, type | for _____________________/ find / -print|xargs cat|compress | SNTF