• Please review our updated Terms and Rules here
  • From now on we will require that a prefix is set for any items in the sales area. We have created regions and locations for this. We also require that you select a delivery option before posting your listing. This will hopefully help us streamline the things that get listed for sales here and help local people better advertise their items, especially for local only sales. New sales rules are also coming, so stay tuned.

IBM Xenix sealed in origional box - Server room closet rescue!!

Unknown_K said:
We get a lot of people here first time posting thinking they have a solid gold ingot in their hands looking for somebody to offer them a mint for it so they don't have to pay ebay fees (who is the cheapskate now). So they come here with no clue what they have, mine the people for information (some of who were actually in the industry at the time and know what they are talking about), they flip it on ebay anyway. It gets old after a while.
Click to expand...

To be fair sometimes they have "rare" stuff but usually it is just run of the mill. Sealed IBM SW of that era is realtively common. I admit that I have not seen a sealed xenix one but it is not going to command $500 on these forums by itself. It all really comes down to what you have for sale - and if someone really wants it,
 
There's plenty of information on the web, particularly bitsavers and datasheet archives. What about the Z8000 interests you most? Doing a full-blown segmented-memory implementation with MMU is pretty ambitious. On the other hand, using the "little brother" Z8002 with non-segmented operation and 64K address space can be done very simply.

There were alternate vendors of the CPU; for example, AMD and Siemens had a joint venture called AMC that produced them. I didn't think that they were particularly rare.

There'a also the Z80000, the 32-bit version, which can execute Z8000 code.

But if I was just curious about the processor itself and wanted to work some sort of a testbed up, I'd get a Z16C01 (48 pin DIP and 44 pin PLCC) where all of the knotty problems are taken care of in a chip (as far as I can tell) that's still in production!
 
Not the CPU, there was a complete machine sold by Zilog based on it.

Your post just inspired me to go on a mission again, when you said Z8000 my brain just connected it to an entire machine (rather than just the CPU). I've emailed the owner to see if the machine is still in his garage. The more digging in google I do, the further away I feel I am from identifying the thing! I remember "Z8000" being mentioned.

Edit: bah just found it, System 8000 http://www.warthman.com/projects-Zilog-System-8000-Model-11-Datasheet.htm
http://www.secondvalleysoftware.com/hardware/z8000/System8000.html

Edit 2: owner informs me it was taken to landfill :(

but *ahem* on topic - good luck with the Xenix Sale! If it's going cheap I'm interested but can't afford too much.
 
Last edited:
There were several complete machines sold based on the CPU. The Zilog S8000 was Zilog's development system and you'd be lucky to find one. You may find that it's easiest to locate an Olivetti M20, for example. Complete and all set to go. The M24 even had an option for a Z8000 card, though they're extremely rare.

AMD manufactured, for a time, a Z8000 development system--they'd be very rare by now.

The system I used with Xenix was made by Onyx, but there were others. One of my old catalogs even contains a description of a Z8000 ISA card for the IBM PC.

So they're probably still out there--you just have to find one.
 
Well there apparently was 1 in NZ, and I got to touch it (had to lift it off an IBM RT when I picked that up back in Feb), but he just emailed me back saying it'd been land-filled - so yeah I wont be playing with one any time soon. Shame.
 
I realize that this is may be a dead thread. However, given I have first hand knowledge of this, I thought I'd reply.

First, the complete manual set for IBM Personal Computer Xenix 1.0 is 8 volumes. There are indeed 4 floppies, but they're *not* in the Command Reference Volume -- rather, they're included in the "Installation Guide" manual. All the manuals say "4 floppies" on the back.

Second, Chuck(G)'s history of Xenix development is incorrect. For the Intel versions of Xenix, Microsoft did the basic port and kernel development (and produced OEM versions) and SCO made packaged product (and did their own kernel modifications for that purpose). IBM PC Xenix went directly from MS to IBM; SCO was not involved.

- Rob
 
RobF said:
Second, Chuck(G)'s history of Xenix development is incorrect. For the Intel versions of Xenix, Microsoft did the basic port and kernel development (and produced OEM versions) and SCO made packaged product (and did their own kernel modifications for that purpose). IBM PC Xenix went directly from MS to IBM; SCO was not involved.
Click to expand...

Rob, I was there--Microsoft was doing Xenix for the 80286, Intel was doing the kernel work. This was before the 80286 was released to production. MS was apologetic for long delays in Xenix 80286 deployment. I was perhaps amiss in mentioning SCO, but the collaboration between the boys in Bellevue and in Santa Cruz was very tight. Our contact for 80286 Xenix was Microsoft--in fact, since our system involved using an 80186 for I/O and the 80286 for the rest, I had a listing binder on my desk for the TRS-80 Model 16 implementation on my desk, showing me how it was done for that platform.

But Intel was most definitely involved in the kernel work for 80286 Xenix.
 
Chuck(G) said:
Rob, I was there--Microsoft was doing Xenix for the 80286, Intel was doing the kernel work. This was before the 80286 was released to production. MS was apologetic for long delays in Xenix 80286 deployment. I was perhaps amiss in mentioning SCO, but the collaboration between the boys in Bellevue and in Santa Cruz was very tight. Our contact for 80286 Xenix was Microsoft--in fact, since our system involved using an 80186 for I/O and the 80286 for the rest, I had a listing binder on my desk for the TRS-80 Model 16 implementation on my desk, showing me how it was done for that platform.

But Intel was most definitely involved in the kernel work for 80286 Xenix.
Click to expand...

Sorry, Chuck, I wasn't trying to imply that Intel wasn't involved. Intel was very involved in all the Xenix work, as Xenix was often the first customer of many of the advanced features of the Intel processors.

I, too, was there -- I was in the Xenix Group at MS from 1986 through the end. While the Salmon work happened before I arrived, I know everyone on the MS side who was involved; I worked on subsequent releases of Xenix for the 286 and 386, as well as the Merged Product work with AT&T.

As for the collaboration between MS and SCO, I'm not sure "tight" was the way I would have put it. It had its moments, let's say. Larry and Doug Michels could charitably be described as "characters".

I rather suspect we know a number of people in common.

- Rob
 
Rob, by 1986, I had long moved on and had been running a VAX 11/750 with BSD on it--that was back in BSD's Unisoft days. I don't think I ran into Xenix again until about 1988 on a Wyse server.

The funny thing was that, starting in about 1981, we'd started to look for a successor to our 8085 box running our own OS and BASIC (using the MCBA business applications). We could comfortably support about 4 users on a 3.5MHz 8085 and 192K of bank-switched memory.

We went through a massive effort rewriting the compiler and run-time, changing our own ISAM to B-Tree and getting the whole mess to run on a 6MHz 80286+80186 under Xenix with about 256K of RAM. I fought the project planning every step of the way, but eventually gave in.

The net result? We could comfortably support between 3 and 4 users at twice the price.

The company went bankrupt a couple of years later. I saw a lot of this insanity during the 1980s. The "we have to have Unix because everything will run much faster and be easier to maintain" mentality.
 
Chuck(G) said:
But if I was just curious about the processor itself and wanted to work some sort of a testbed up, I'd get a Z16C01 (48 pin DIP and 44 pin PLCC) where all of the knotty problems are taken care of in a chip (as far as I can tell) that's still in production!
Click to expand...

Isn't the Z16C01 just a SCC? In other words, you will need a CPU too if you want to do any testing?
Oh never mind, I see now that it's just the Wikipedia article that is misleading.
 
Rob and Chuck -
Awesome to see you guys comparing notes. Ironically, I now work at MS, and happen to be trying to hunt down a copy of Xenix that will run on the 8088/8086 - seems Xenix is the only Unix ever made for that processor. Anyway, if either of you are in the area still, I'd love a chance to grab a beer/coffee/beverage of choice/ and chat!
 
You must log in or register to reply here.
Back
Top