X-Andrew-WideReply: netnews.comp.sys.ibm.pc.rt X-Andrew-Authenticated-as: 0;andrew.cmu.edu;Network-Mail Received: via nntpserv with nntp; Mon, 6 Jun 1994 01:49:05 -0400 (EDT) Path: andrew.cmu.edu!bb3.andrew.cmu.edu!honeydew.srv.cs.cmu.edu!das-news.harvard.edu!noc.near.net!paperboy.wellfleet.com!news-feed-1.peachnet.edu!gatech!swrinde!news.uh.edu!lobster!antimatr!markw From: markw@antimatr.hou.tx.us (Mark Whetzel) Newsgroups: comp.sys.ibm.pc.rt Subject: Re: RT ESDI controller in a PC? Summary: Actually, it IS a WD1005A-ARM controller Message-ID: <281@antimatr.hou.tx.us> Date: 6 Jun 94 03:06:17 GMT References: Organization: StarTrek and Lionel trains fan, Missouri City, TX Lines: 69 In article , jma7+@andrew.cmu.edu (Jeffrey D. McMahill) writes: > I have an ESDI controller (original, not extended) from an IBM RT, > and a 70 Meg hard drive to go with it. Is there any hope at all of > getting them to work in an old IBM PC? ANYTHING is Possible. Given enough time, money and energy... read on. > I see two possible problems: > 1. Will it work at all with just the 8 bit bus? Not in a 8-bit PC bus, it was desined for the AT style 16 bit bus. (slot clearance would be the first order of incompatiblity). Western Digitial built the standard ESDI card for IBM to IBM specs. The silk screen on the board calls it a WD1005A-ARM board. A phone call to Western Digital might reveal more information about the card and what it can do. The disks it controls are fixed-sector ESDI disks, unlike the soft-sector disks usually used with PC type systems. Hard disk registers are very similar as PC/AT controller cards, and at standard I/O addresses (1F0-1F7), but it is fixed at IRQ 14 for fixed disk operation. The floppy part uses I/O addreses 3F2-3F7, IRQ 6, and DMA channel 2. > 2. I'm assuming that there probably isn't a ROM on the controller that would > have the correct software for a PC. Yes, that is correct. The RT did not rely on PC style BIOS chips, (at least none that are published in any RT manuals I have access to). Examination does not reaveal a place for a standard BIOS, nor do the manuals reveal any memory addresses that it would respond to for a BIOS location. However, there are quite a few LARGE socketed chips plastered with WD numbers on them that are NOT the disk controler nor the NEC diskette controller, they might be either PALs or ROM chips of some kind. It MIGHT run as a SECONDARY controller in an existing system that HAS an existing BIOS.. (moving its I/O addresses to 170-177), but the IRQ used is the same. > Is there some type of driver that could be loaded from floppy (or another > hard drive) on boot up? None in existance that I know of. > If not, how hard would it be to write, either from scratch or from an > existing RT driver? With the RT tech manual, and register descriptions, it should be possible to write a loadable driver from another hard drive, with this card as secondary. It is, as they say, SMOP. (Simple Matter Of Programming :-) There are no schematics published for this card, or the EESDI card. Only the PC/AT MFM card had published schematics. To my knowledge there is no published code in existance showing how to use the controller, but it is very similar to AT disk controllers in register design and should be similar. > I hooked it up just to see what would happen, and reading from it's ports, > I get something other than the FF that I generally get for nonexistant ports, > so it would seem that something is working... If it is a functioning card, it would. See my RT hardwware FAQ for jumper and other information on the card. -- AIX..... NOT just another UNIX. (tm) Mark Whetzel | My own RT system.. My own thoughts.. DOMAIN: markw@antimatr.hou.tx.us | IBM RT/135 running AIX 2.2.1 UUCP ..!menudo!lobster!antimatr!markw | comp.sys.ibm.pc.rt FAQ maintainer.