RTL for Z8000 series CPU?

So why exactly would you want such a design? Do you have Z8000 binary code you must run?

As noted below other IP shops have functionally reversed engineered it. I recall one aerospace company had to run old code with timing precision and paid for the Z8000 design to be redone. A functional clone could give good guarantee that machine codes would run in same time clock for clock. The Z8000 was complicated enough but was still a fully predictable design as far as external events were concerned ie no caches.

As it happens I also reverse engineered some of the Z8000 blocks around

79 and still have paper docs for the datapath, but that wouldn't get you very far today. Also a good collection of comp arch books before the H &P bandwagon took over, would often describe the microarchitectures in some detail of most all mid 70s and earlier designs.

If you don't need cycle accuracy, why not write a ISA translator and retarget to Arm, x86, whatever. With the speed advantage you would get a few orders of improvement. Thats probably already been done too!

transputer guy

Reply to
JJ
Loading thread data ...

They did!

There were several 386 and 486 level chips designed and manufactured by IBM based on the agreement with Intel. IBM 386SLC, 486SLC and 486BL series are the examples. The chips carried Intel's and IBM's copyrights and incorporated a reasonable amount of cache. I still have some PS/2 computers with IBM 486SLC2 and SLC3 CPUs inside.

Then IBM served several x86 companies as chip foundry. They did it at least for NexGen, Cyrix and AMD.

Reply to
Bernard

ajcrm125 schrieb:

ISAs are not protected under copyright law. (The ISA documents and the CPU design are, but not the ISA)

The real pitfall are patents. As Peter noted all original pantents should be invalid by now, but

- there might be some patents that Zilog filed that were only granted many years later. These might be still alive. Check USPTO in that case. Sometimes companies threaten people with patents that were granted, but for which they stop paying patent fees years ago.

- even when building an old school uC you will likely use modern concepts some of which might be patented by zilog or others.

Companies like to invoke trademarks and trade secrets in the context. For the former just make sure that you do not use z8000 as the name of your processor but use it only in a descriptive way. ("Executes z8000 ISA") For the latter make sure, that you do not know any trade secrets from zilog. E.g. that you have never signed an NDA for an z8000 errata sheet or anything like that.

I would not worry too muc, but h I agree with Peter that it might be a good idea to try to get zilog on board. They might like what you do and provide you with advice, contacts, etc.

Kolja Sulimma

Sidenote: There is a story about a guy who built an Apple-I clone recently and got sued by apple. When Steve Wozniak realized that he informed the guy that he never signed off exclusive rights to apple. Moral: Never believe a company that threatens you. Allways demand proof.

Reply to
Kolja Sulimma
[snipped]

I often wonder how many products never got developed further because of lost documents at companies. Also how many companies can truly recover if somwhow they had to start from scratch with only their documentation in config control.

Regards Anton Erasmus

Reply to
Anton Erasmus

Believe it or not, adequate documentation and control predates the use of computers by a considerable margin. It involved such things as file cabinets with suitably dimensioned drawers to hold original drawings, prepared on paper and mylar, sometimes with India Ink, the use of Ozalid machines, proper parts list, etc.

--
"If you want to post a followup via groups.google.com, don't use
  the broken "Reply" link at the bottom of the article.  Click on
 Click to see the full signature
Reply to
Chuck F.

Oh, the documentation existed, and it was in the control of Shima, the designer. But once he left the company I don't know who, if anyone, inherited his filing cabinet. It wasn't until much later that such design materials were kept in a centralized location with a control number and access/revision control. Even then, sometimes things went in to DC never to be found again, so designers hated to give the original stuff up to DC.

Monte

Reply to
Monte Dalrymple

I collect classic arcade games and am looking to do a remake of the troublesome Pole Position boardset. I then stumbled on this page:

formatting link

and loved the idea. The PPI and PPII boardsets are Z8000 based.

-Adam ================

formatting link
================

Reply to
ajcrm125

Reply to
Peter Alfke

I looked at doing a conversion of the game, but the main problem is the CPU. Having done most of the debugging work on the T80 core, which required running a real cpu in parallel with the soft core and triggering the analyser when they went separate ways - I can assure you it is a lot of work to get it cycle accurate.

I am going to release the source of the Namco customs I have reverse engineered some time soon, some of them are on the Pole Position board which may help.

Cheers, MikeJ

formatting link

Reply to
MikeJ

Thanks Mike! That will help a ton. :-)

-Adam

Reply to
ajcrm125

Does anyone know anything about the Z80'000 that I've got a prelim datasheet/usermanual for? It seemed like a chip ahead of its time...

--
 [100~Plax]sb16i0A2172656B63616820636420726568746F6E61207473754A[dZ1!=b]salax
Reply to
Tobias Weingartner

The Z80 matched its time and became hugely popular. I don't know about the Z8000, may be it remained not that popular because it was ahead of its time (generally, products ot people ahead of their time have the destiny of simply not being understood by the majority of their time), and then may be it had to compete with other designs, like the 68k etc. Well, the 8086 became popular in spite of the 68k and the Z8000 can't possibly have been such a mess as the 8086 was (is), so the reason(s) may have been completely different, perhaps (likely, I believe) not technical at all. I hope Peter, Monte and perhaps others who have been involved could shed some light. (I am also curious about the story, but I don't know the Z8000 so I would appreciate an educated judjment).

Dimiter

------------------------------------------------------ Dimiter Popoff Transgalactic Instruments

formatting link

------------------------------------------------------

Reply to
Didi

The Z80,000 was after my time. So I have nothing to say about it.

The Z8000 appeared right after the 8086, and almost simultaneously with the 68000 by Motorola, all 3 were 16-bit microprocessors. And the race was on!

The 8086 won because IBM picked its baby-brother, the 8088, for the PC, and because of Intel's massive marketing campaign ("operation Crush"), against the cleaner and technically superior 68000. Never underestimate Intel's marketing muscle. It was a brutal campaign. There is a book about it, by Mr Davidov, the Intel marketing guy.

The Z8000 became condemned to a back-water existence. It was smaller and simpler and thus potentially cheaper than the 68000, but (like the

8086) it partitioned the memory space into 64K segments, and there was no way to detect when the address counter rolled over. Ziliog used arrogant and semi-religious arguments for memory partitioning (the chip architect really believed that it was a great feature, not a handicap), but the upcoming graphics applications preferred a linear address space, and they all went to the 68000. The Z8000 was left with military and some arcade-game designs. My only encounter with Steve Jobs was when I tried (unsuccessfully) to convince him to use the Z8000 instead of the 68000 for what soon became the Macintosh. The 68000 came in a gigantic package, and just its gold plating cost almost as much as the Z8000 die. But the linear addressing won...

Of the three contenders, the obviously worst one became the winner. The

68000 did so-so, and the cleanest and leanest became the loser. Who says life is fair?. Peter Alfke, reminiscing.
Reply to
Peter Alfke

The Z8000 just didn't happen to make into anything like the PC or the Macintosh. IIRC, it was used in a fair bit of industrial, telecom, and military gear. I never used the Z80K, and don't remember much about it other than it was a full

32-bit architecture with cache, a pipeline, and a paged MMU. (I do remember seeing some datasheets). It took Intel years before they had anything even approaching the Z80K.

The Z8000 was a very nice architecture. Very PDP-11-like. Far, far, better than that dog's-breakfast that Intel puked up and named the 8086. It had a nice large set of registers (16x16bit registers that could also be used as 8x32 bit and I think

4x64bit). The instruction set was very regular (more so than even the 68K). The Z8000 daisy-chained interrupt scheme was an utter dream to work with compared to the nightmare that was the 8259 -- which was an obsolete piece of crap when it was introduced in 1980 or whenever it was. There's a very special place in hell reserved for whoever put that bit of waste into the IBM PC. The Z8000 (like the Z80) included a built-in DRAM refresh controller. The Z8000 also had some very nice peripherals in the Z8036 counter/timer/PIO chip, the Z8030 dual sync/async UART, the Z8010 MMU. The Z8530 (the version of the 8030 with the muxed bus) is still widely used today.

IIRC, the Z8000 didn't have a version with an 8-bit external bus like the 8088 or the 68008, so that limited it's application in cost-senstive products. The CPU and the peripherals were all NMOS and were pretty high power (about

200mA each, IIRC). That didn't help much.

Here's are a couple nice pages:

formatting link
formatting link

Apparently the Z8000 is still shipping in CMOS as the Z16C00 series.

--
Grant Edwards                   grante             Yow!  I just heard the
                                  at               SEVENTIES were over!! And
 Click to see the full signature
Reply to
Grant Edwards

..Or Intel's manufacturing muscle - once the IBM PC selected the x88, it really was Game Over, and helped a lot more by Microsoft's marketing Muscle, than Intel's.

~68000 is still alive, in Coldfire microcontrollers, but if the Z8000 had 64Kpages, it was too similar to the x86, and also too late to use the code base the x86 then had.

The Key in all this, is Intel was able to make, and ship x88 silicon, while the others were sampling, and/or unable to meet price targets.

There was also a time, back then, when better code size mattered due to the price of memory.

These days, the 64K issue appears again in the microcontroller sector- most 16 bit cores naturally have 64K opcode reach issues - and we see a swing into 32 bit microcontrollers : memory is far cheaper today.

An interesting 'inversion' is the sight of a mere 8K code variant ARM from Philips - LPC2101.

-jg

Reply to
Jim Granville

I was at Onyx Systems in 1980 doing the Unix V7 port for the machine that spring to meet the NCC dedline at the begining of that summer. The Z8002 had just finally became stable for a multitasking OS. We shipped a few thousand of them before Carl Berg pulled the plug on the UNIX management team wanting more MPM/CPM machines. Carl funded Onyx to sell IMI 8" disk drives, another of his venture companies. While the Z8002 easily ran PDP-11 applications that were ported to it .... larger address space applications developed for Vax, PDP10 & 20, and IBM main frames suffered horribly due to the segmentation.

Motorola's M68K wasn't stable yet, but did become stable about a year later while I was at Fortune Systems doing design work on that machine. After years of living in segmented architectures and crippled address spaces on small machines, the M68K was the first microprocessor that would actually handle large address spaces cleanly to compete head on with Dec Vax and PDP10 & 20 lines. Fortune took the 32:16 to Comdex in the fall of 1981 and took best of show as the classy high performance multiuser desktop UNIX box. IBM release the PC, at about 30% the performance just weeks earlier with it's new OS .... PCDOS (soon to be also sold by it's supplier as MSDOS).

While doing the M68K unix there was a LOT of pressure to find some Z80 CPM/MPM compatability and migration path for applications developed for those platforms.

By late spring 1982 the demand to CPM/MPM compatability was completely dead, as that market died in it's foot steps as IBM shipped more PC's than all the Z80 machines that existed in 1982. Going to market with the largest computer companies sales team competing against you is a very tough sell.

Now the next part of the story is the part that very few people understand ... and probably the most important part never to forget .... and that is that ....

Reply to
Totally_Lost

ok, sorry for the brief break ... in 1980 Wang was chewing up the minicomputer office market with several machines that had the premier word processing system in use in most large offices, especially the legal office market. Selectric typewriters had for a decade been the machine of choice, as it's bail lock keyboard had THE TOUCH for speed typists, as you quickly learned not to bottom keys, but release on fall thru greatly reducing finger shock by not having to bottom keys or take the recoil from the key hitting paper in your finger tips. The problem was that these typewriters were expensive to buy, and due to high maintence from the rotate tapes and complex timing they were much more complex than the standard Royal or other common typewriter. Plus, they were multi-pitch and multi-font adding a new look to business communications.

Enter word processing, daisy wheel printers, and electronic document storage and editing, and Wang had a gold mine, for about the cost of a high end selectric. IBM decided it had to compete in this new market, and in the summer of 1980 release the Display Writer, a computer based document system using the 8086 at a cost of about $8K for a single station, and about $26K in the 3 station version. The story was that IBM traded Intel bubble memory patents for the rights to the 8086 so IBM could mfg it's own processor chips. During that same spring another group in IBM was working on the PC, and decided to use the same processor design, that they already had rights too. So the Z8000, which wasn't stable yet, really was never a contender ... nor the M68K which wouldn't be stable for another year or so.

At this same time, high quality glass terminals, like the Datamedia DT80, were roughly $2K list, and low end glass terminals like the ADM-3A where just under a grand list. Similar prices for fully configured Radio Shack TRS-80 systems, especially the TRS80 Model 2 which was the high end machine.

IBM releases the PC priced at just under $2K with dual floppies, with PCDOS and inside a few months all the large Z80 programs are being ported to the PC ... especially several key word processing applications, which combined with a high end daisy wheel printer, creates an affordable word processing solution at about the same price as a good Selectric ... and a much cheaper maintence contract.

IBM killed two cash cows ... the selectric typewriter and the Displaywriter without realizing what they had done. IBM sales was giving away PC's with a huge institutional discount at just above ADM-3A prices, which combined with 3270 emulation software, also destroyed IBM's mainframe terminal business. By perchasing PC's in volume, institutional and other large buyers, were able to ratchet down IBM's multi-tiered pricing to get huge discount on IBM minicomputer and mainframe products. As a result, thousands of PC's sat every where, unopened, just to get huge prices savings on big ticket purchases. This lead to discount grey market channels for PC's, and even lowered the street price to accellerate the comdity PC word processing market.

Replacing over a decade of selectric typewriters, and several years of high end Z80 business computers, with 16 bit processors took about 18 months during 1982 and early 1983, and then the market saturated. The first huge computer tech buble ended with the tech crash in 1983. That further fuelled depressed computer prices as a huge over production inventory was liquidated at fire sale prices.

With the market very flat, depressed sales, depressed prices, it took the industry the next two years to retool, reengineer, and come out of that down turn with much better products that took another 5 years to saturate the market as demand for word processing, business computers, and home computers really became mainstream.

Reply to
Totally_Lost

...

And how many had an 8051 to interface a Selectric to a Centrnics port as a printer?

Reply to
Bill Davy

I actually did my own design from scratch, complete with all the careful testing with an oscilloscope of the reed relay signals in the IBM electronic model 85 I was working on, using an 8051 to interface it to a serial port. Included both hardware and software handshaking and buffers to handle the slow output rate of the typewriter. Worked first time, too!! Used it for years as my printer, capable of handling multi-part forms when needed.

I can't imagine how many folks did Selectric conversions -- it was because of them (and my inability at the time to find a design specifically for the electronic series that followed it) that I tried my hand at the unit I owned. I knew I should be able to get it working and, sure enough, I did. That was my very first design from scratch of any significance in electronics. I remember it well.

Jon

Reply to
Jonathan Kirwan

Interestingly, the classic drawing cabinet, to hold drawings hanging from four fingers for easy access, is credited by some sources to ... Charles Babbage.

- Brian

Reply to
Brian Drummond

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.