Last edit Oct 18 2011 (c) Herb Johnson, link fixes July 26 2015.
In June 2009, Bill Beech contacted me to trade some Multibus manuals. We have since talked about a number of Intel Multibus projects he has: the linked Web for links to all those activities, and about Bill and his other work on the Web.
Bill is a computer engineer from the 1980's who worked on Intel systems including the Intel 310 and 320 series of Multibus systems on which ran Xenix. He's got a 310 system now but also is building up an 80/10 (intel 8080) system. Bill also produced a Web page about his 80/10 and related systems, and the tools he's developed. This Web page is about our discussion of his System 310 work. My responses are in italics. - Herb Johnson
June 2009: I am trying to find the iSBC 88/45, iSBC 188/48 and iSBC 286/10 hardware reference manuals. I have a complete Intel 310 and hope the iSDM manuals might list the commands. The 310 works but does not boot off the 140 MB HD. I also have about 50 Intel multibus one boards on hand as backups....
The Intel 310 is the single chassis with a 286/12 processor, a couple of 020CX 2 MB memory boards, a 552 ethernet card, a 221 hard/floppy controller. The 221 has an SBX 217 and 218 on it. The system has a floppy, tape drive and 140 MB HD. The HD does not boot, but the system comes up and goes into the SDM. That is why I was curious if those SDM manuals - especially the 286 one gives the SDM monitor commands. We never used the SDM in the Army since the Intels always booted into Xenix. They were very reliable.
All our 310's and 320's ran Xenix. We used OpenNet to connect to them over ethernet for file and printer sharing. This one was an Army evaluation machine that somehow never ended up on a property book anywhere. So it was passed around to various people in the shop where I worked until it ended up with me. I have an 386/38 processor card so I suspect I will bring it up with IDE drives vice the MFM, and load a version of Linux on it. -
I have used both the 310 and the 320 for many years while the Army had them fielded. They were a very good machine. I have two more 547 boards for the 320 in my stock. The 214 is a winchester controller for ST-506 drives. I just bought the hardware manuals for the SBC 208, SBC 012CX and the SBC 546/547/548. The hardware manuals are a must for getting jumper settings and developing new code for the boards.
Herb says: I have three sets of Intel 320 system boards, only one set works completely.
I'd like to get the SBC 547 serial board supported in my Xenix. I often offer a page for page trade on manual copies. Let me know if you'd like to do that with the 547 manual, then I have something to work with. Next I'd need to find how to write Xenix drivers. It's a little hard to find docs for Xenix at the driver level. But, I have not tried hard.
I had hoped to run iRMX, but (as my site says), the shop I got the hard drive from claimed they did not have a 386 version. Another shop said later that they did have one, but they did not offer one to me. - Herb
MFM to IDE? I did not know a converter was readily available. But it would be possible to drive an IDE with an 8255, it's been done around the Web. Linux on a '386 is a little tough, you might need to use the 2.4 kernel. If I had more docs on my cards, I'd consider doing something of that sort.
I encourage you to put this stuff on your Web site. Keep the Multibus alive, pass on the knowledge, etc. - Herb Johnson
Bill says: Be glad to trade you the manual on the 547 board. I have it in the queue to scan into a pdf. I do this at 600 dpi and generate manuals and schematics that are much sharper than the 300 dpi stuff. You can really see the difference in the schematics.
There was a xenix manual called "Device Drivers for Xenix" which covered it quite nicely. I had a complete SCO XENiX for the PC here 20 years ago which I wish I never got rid of. I had all the manuals on the Intel boards in our systems as well. Hindsight....
I have already dumped the ROMS and used the SDM to disassemble the code on the actual 310. About 1/2 way through the complete ROM image. It contains several programs - the initial system tests, the SDM and the actual boot code for whatever is on the HD.
Forget [my request for Intel hex format] SDM pages. I have it roughed out now, anyway. A bit more cleaning here in the room and I will bring it back in and hook it up.
Actually, if you look at the iSBX standard for the blue expansion connectors on the cpu boards (16-bit, not 8-bit), you will find all the data, address and control you need to talk to an IDE drive directly. And the cpu board includes DMA for those connectors. I believe an IDE drive can be directly connected to that connector and operated.
I would like to get an earlier version of Linux running on the 386 processor. For the 286, have to probably deal with Xenix off the net, if it can be found, or port Minix to it. Either way, it will be fun to beat on it. - Bill Beech
Dec 2009: On multibus, you are right, there aren't many sites [on the Web]. I believe yours is the only one that shows up after you purge those currently doing remanufacturing.
I have the Intel 320 sitting here in the radio room hooked to one of the ports on a terminal server. It will not boot the Xenix off the 140 MB drive but does come up in the SDB debugger. I have been disassembling the ROM so I could get some more insight to its operation.
I need to talk to you about seeing if I can get some help from [a Mulitbus support company]. The Intel 310 is an 80286-based system. Both it and the 320 would run iRMX but I would much prefer Xenix. I have the hard drive but believe I have a controller problem in the 310. Never really ran it down. iRMX would be interesting to have to play with as well, but the Xenix would network into the rest of the TCPIP machines if I can find the TCPIP stack for it.
We never used iRMX so I can't dispute what they told you. All our Intels were Xenix and networked with OpenNet. If you would like to borrow my 386 processor, I am sure we could arrange it. It has the fully populated addon memory board.
I should be able to run an iRMX 86 or 286 on my Intel 310. Just need the image file to dd onto a drive. I really need to get out there and find some wini's before they become too pricey! - Bill
Herb said: so I'd need more RAM and of course another hard drive for iRMX. I have some MFM drives, chances are a few of them may be large enough. I'll have to see about reviving the drives and testing them on really old PC's I have for such purposes. Do you know offhand, what drives (brand model) or drive configurations (head cylinder) are accepted by iRMX and the controller I have (you have)?
Bill says: Well, it is indeed true that bitsavers.org archive is adding new multibus stuff. They have placed the tar.Z files for Xenix 286 and iRMX 86 on the site. I have pulled it all down, along with the new manuals they have loaded.
I [was] confused by the format of the tar.Z files. [Bill and I discussed this.] I have [sionce] looked at the catweasel stuff and the disk images and now understand them. They are simply full track reads from index hole to index hole. I suspect using a track write would restore the image to the 360 kb 5 1/4 inch disks. Now I need to find a source of 5 1/4 inch DD disks.... Looks like we will need a bunch and I don't have a stash of them.
Herb said: MY understanding of the recent ISIS image files on bitsavers in the Intel directory, is that they are NOT "full track reads". They are copies of SECTOR DATA ONLY, in "logical" sector order. They are in the order an ISIS-II program running on an ISIS system would "read" the diskette as a bunch of files plus boot track plus directory sectors. However in the world ISIS "links" sectors to create one program, file, etc - that's what there!
Bill said: This is true. If I can find the right ROM image, I could boot up those disks on hardware or the altair 8080 emulator.
The .dmk stuff contains the sectors in the order they appear on the disk with the address marks set however they were. If the disk format included sector skewing, then the address marks are out of order. I suspect these types of formats were designed to preserve this information. It would be lost with a img or iso image.
I have mounted a iSBX-218 and iSBX-350 on the [Zendex] ZX-80/05 [CPU card I acquired]. I have two iSBC-064 memory boards. I suspect software wise, I have created the Intel multibus equivalent of the N8VEM. I believe I should be able to get the same software to run on it.
I also found multibus I prototype boards for under $30 each. If I can find an extender board for multibus I plus iLBX (P2) I will be in like flint.
I have been collecting the intel literature and pulling the individual documents to build a web page on all the stuff I can find. This includes all the old data sheets and such. I found the diagram of the iSBC-215D and iSBC-213A which are versions of the winchester controller in my 310 in the MDS IV schematics.
I want to bring up the emulation of an intel 310 on the SIMH group. So far, all their machines are single processor. My intel 310 has an 80286, 80287, 2 ea. 80188, 80130, 8742 and an 8089 all working in concert to make it operate. That would be a very complex emulation to build. But it sure would be fun! And would require I get the actual 310 to operate as well. One addition I would like to make is to replace the winchester with and IDE.
HErb says:Looks like a good start for your Web pages on your 8080 Multibus system. Hope you add something about your Intel 310 system. YOu know I have a working 320 and two non-working 320's?
As for ISA floppy controllers....you mean all those old 8-bit floppy controller cards won't run single density? Or be modified to support Single density? It's so long ago I've forgotten....HErb
Bill said:Yes, I remember a discussion about the the 320 and your getting an imaged disk from one of the vendors. I am not to the point of doing much with the 310. It comes up, passes the basic tests, fails to boot, and enters the SDM. SDM works fine. I have captured and disassembled about half the 286-10 ROM with SDM. I cannot see any disk activity on the floppy or HD during the boot attempt, indicating there may be a problem on the iSBC 215G disk controller. I will have to put the scope and/or logic analyzer on it and see what I can find. With the Xenix images on Al's site, I should be able to bring it back up with load media. I really need to find a few more WD 140 MB hard drives. Be nice to find a 311 chassis, as well.
What kind of shape are your other 320's in? Remember I have two 8-port serial boards for the 320. - Bill
Jan 2011:[Months went by, and in Jan 2011 I asked Bill about his progress. MOstly he was working on the 80/10 stuff and on its related 8080 software. But on the 310 he said:]
I have powered up the 310 occasionally, to check the connectivity through the terminal server. I want to get the 80/10 and emulator completed, then attack the 88/45 and 188/48 boards. I'll need to create emulators for the 8088, 8086, 80188 and 80186 to get that stuff working as I like. I found the 8086 emulator for the Altair 8800, but I prefer to roll my own. - Bill
Sept 2011: I am working through disassembly of the 88-45 and 188-48 ROM images. Intel did extensive processor testing in the ROM code before they set up the I/O started processing. I believe the 88-45 waited for an external module to be loaded to actually configure it. We used it for X-25 connectivity to the milnet.
Oct 2011: I am looking at the disassemblies of the ROMs on some of the Intel boards. I am right now trying to build a decompiler to generate plm86 from the assembler source. I have placed on my site a set of files that contain .EXE versions of plm86, the librarian, linker and loader. Apparently a cross-compiler set from Intel for the PC. Using that to figure out what each plm instruction generates. This will never be 100%, but might get close enough to compile into a correct executable. I would be able to modify the code on the 188/48 if it were in a "higher" level language. The assembler could be decompiled into any higher level language, but I will stay with plm.
If I can get this to do something useful, I will create another version to decompile the ISIS-II 8080 assemble code. This will allow me to run it on other than an exact MDS hardware or emulator. It is tied to the 2 board disk controllers, either SD or DD prior to the iSBC-2XX controllers. - Bill
Contact information:
Copyright © 2015 Herb Johnson