Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!sdrc!thor!scjones From: scjones@thor.sdrc.com (Larry Jones) Newsgroups: comp.sys.ibm.pc.hardware Subject: Re: How to access all of a 676MB ESDI Seagate/WREN under DOS 5 ? Summary: Why DOS can't (won't) support >1024 cylinders Message-ID: <173@thor.sdrc.com> Date: 24 Jun 91 21:00:23 GMT References: <1991Jun17.124112.9416@ufhx1.ufh.ac.za> Organization: SDRC, Cincinnati Lines: 46 In article , cg108w3@icogsci1.ucsd.edu (Steve - Happy Hacker) writes (and others have written similarly in other articles): > My AMI BIOS is fully aware that I have 1313 cylinders and never once > complained. (I didn't need a utility to make it accept >1024, I just > typed it into the config screen.) UHC Unix had no trouble with the > >1024 cylinders. Why on earth is Microsoft so bent on the notion > that DOS machines simply *must* have small drives?! The problem with more than 1024 cylinders is that DOS uses the BIOS to access the disk and the BIOS interface only allows 10 bits for a cylinder number. This is an unfortunate fact of life. Since this is an interface problem, there is nothing that either DOS or the BIOS can do to avoid it -- both must agree on any change. Since most BIOSes are in ROM, this is a nasty problem. The only reasonable way to fix this is for DOS to completely avoid the BIOS and access the disk hardware directly, but that means that DOS must now become much more complex since it will have to be intimately familiar with every existing disk controller, as well as future controllers. Most BIOSes are not very useful for multitasking or protected mode operation, so the Unix vendors have been forced to adopt this route and that's why there's no similar problem for Unix. Of course, they have the other problem that you can't just go buy the neatest new whiz-bang disk controller and use it, you have to wait for them to write a driver for it and then you have to configure it into your system and then live with various problems for three or four releases until they get all the bugs out. Then you get a new whiz-bang motherboard and discover that there's some subtle interactions and you need to go through another three or four releases until it all works again. Microsoft really doesn't want to get themselves in that position, and I think that's understandable. Most disk controllers these days have enough intelligence that sector translation is a viable alternative, although that screws up software that does real low level access like non-destructive low-level formatters. I don't know of any really good way to solve the problem -- if you do, share it with us! I'm sure Microsoft would jump at the chance to support large disks as long as they don't have to paint themselves into a corner to do it. ---- Larry Jones, SDRC, 2000 Eastman Dr., Milford, OH 45150-2789 513-576-2070 Domain: scjones@sdrc.com Path: uunet!sdrc!scjones Any game without push-ups, hits, burns or noogies is a sissy game. -- Calvin