Slashdot Mirror
Zilog To File For Chapter 11
Frédéric writes: "The venerable company ZiLOG who was founded in 1974, and who brought us the famous Z80 CPU (used in the Timex/Sinclair ZX80/ZX81, and the Amstrad CPC/PCW computers), is filling for Chapter 11 ... I didn't find the today's news on the web, but found this article at Silicon Strategy and this one at Electronics times, which was written a few days ago to announce it."
Z80 also powered the sound chip in the Sega Genesis and a modified chip was used in the Gameboy.
Don't forget that the Z80A was the CPU for TRS-80 model 1's & 3's, too. It was the first assembly I learned and it taught me a vast respect for memory conservation (4K was all that was available at the time).
Can you do anything with an 8-bit microcontroller anymore? :)
Although putting an embedded web server on a Z80-based machine is kinda cool.
Also the Early Radio Shack Machines. TRS-80 Model 1 through 4. I once had a Model 6000 running XENIX that had both a Motorola 68000 and then a Z80 for I/O functions
As a owner of a fully decked out TRS-80 Model I, II and Timex Sinclar 1000, I'm sorry to see them go under. I bet a lot industrial and robot controller companys aren't to thrilled ethier as the CPU has been a sort of staple for them for quite sometime.
I guess the PIC / microcontroller chip market really took over, leaving little room for Zilog...
What about second sources?
ZiLOG intends to launch an exchange offer in which all holders of its notes will be offered the opportunity to exchange their notes for shares of ZiLOG common stock, plus a pro rata share of the $30 million non-recourse note. The exchange offer, which for tax and other legal reasons the company intends to complete through a prepackaged Chapter 11 filing, is not expected to have any adverse affect on its day-to-day operations or on its ability to provide a full range of products and services to its customers or pay its suppliers on normal terms.
I dont think we have much to worry about here.
Tux Games. Your complete source for native Linux games.
The Z80 is also used in the venerable TI-85 calculator, and related models.
I'd be sad to see a company that's such a piece of microprocessor history disappear.
:-(
My first computer was a 1978 NASCOM-1 kit (a board, bag of chips and seperate bare transformer!) that was based on a 1 MHz Z80 with a whopping 2K of memory - 1K for the monitor program, and 1K for the user. Back in those days we programmed from memory directly in hex - none of this fancy modern symbolic assembler stuff!
Zilog also had 16 and 32 bit microprocessors, but neither took off - the Z80 has had a long life though.
The story of the Z80 is quite interesting - the design of the Intel 8080 basically walked down the road in the head of the designers who then designed the Z80 which was close to being dual 8080's on a single die - with it's dual A and A' register sets.
Ah, the good old days....
Remember, this is not an involuntary bankruptcy (at least not completely). Zilog is filing a "pre-pack" which means that they've gotten together in advance with all of their creditors and gotten them to exchange their debt for equity. Zilog will not disappear - in fact, this really only stands to make them financially healthier.
Interestingly, the just shipped a new eZ80 webserver three days ago.
I think I'll stop here.
Intertec's Superbrain, built around 1979, had dual Z80s (one for diskette I/O and booting, the other for everything else). 16k of DRAM - expandable to 64k with a soldering iron. It ran CP/M 2.2. I last used it in 1989 for a college project.
Go ahead, tell me I am trolling, but how on earth did a company with an ancient, 8-bit architecture manage to get idiots to throw $280M at it? My cat will design you an old 8-bit architecture for 1/10,000 of that and I promise she won't file for chapter 11.
Did they REALLY expect a Z80 with a TCP/IP stack to set the world on fire enough to pay back $280M? QUARTER OF A BILLION DOLLARS!?!?!
when i first reloaded the page, i saw the Z in Zilog and thought, 'Jesus! Zapmedia went out of business already? Now that's some /. effect.'
your jesus is another mans xebu. chew on that hypocrites.
There is actually a Z80 in the new Gameboy Advance. This is for providing the Gameboy Color compability mode.
Once people started writing emulators for all of the old classic computer systems, there was no reason for anyone to go out and buy the real thing. The hackers stole the liveleyhoods of the Z80 designers. I hope they can sleep at night.
In the UK, Sinclair computers and their derivtives were huge - much more well known than the US Timex TRS-XX machines (though geeks here seem to remember them fondly:
IIRC, all these machines had some Z-80 derivative:
ZX80 1K RAM (actually named after the year it came out)
ZX81 1K or 16K RAM
Spectrum (development codename ZX82) 16K or 48K RAM [+]
Spectrum + (larger, "better" keyboard)
Spectrum 128 (a vast 128K of RAM)
Spectrum 128 +2 (built in cassette deck!)
Spectrum 128 +3 (build in 3" *not 3.25"* 2 sided (by ejecting it and turning it over) floppy disk)
There were a couple of others.
Then also things like the
MGT Sam Coupe - which was compatible
I, my family or my firends owned every single one of these fine Z80 powered machines at one time or another. Hell, I learned to program in Sinclair basic. If Zilog have gone under (Chapter 11 doesn't mean its necessarily over) this is a sad day.
[+] actually this was a marketing lie. It had 32K RAM and 16K ROM with a unified address space. I think the 16K version had the same ROM, so it would be fairer to call that a 32K, if you want to include the total...
Lord Pixel - The cat who walks through walls
A little bigger on the inside than out
Yes, even Commodore used a Z80. Specifically in the C-128, which could boot into a CP/M mode running on the Z80, or into regular C-128 mode on the 8502, or into C-64 emulation mode (also on the 8502). I don't think there was any way to run both CPUs at the same time.
I don't know how many people actually used this feature (probably not many, given how well the C-128 did in the marketplace), but it was kind of neat at the time.
Better stock up on this collector's item:
Z80 assembly language subroutines
by Lance A. Leventhal
Availability: Seller usually ships in 1-2 business days
ASIN: 0931988918
Sigh. Tears well up in my eyes. My first paying job out of school was writing Z80 assembly code for the Sega GameGear and Nintendo GameBoy (a crippled, cheapo Z80 on the Nintendo).
And I suspect there were a bunch of arcade games that ran a Z80 besides PacMan.
Rest in peace, my little 8 bit friend, RIP.
"Once we successfully address the issue of our senior notes, ZiLOG will be well positioned to compete during this difficult period and to take full advantage of the eventual economic recovery," Thorburn said.
They filed chapter 11 for strategic reasons, not because they'd gone totally bust.
Pop quiz, hot shot. Tell me what this Z80 code does...
LD BC,0FFFFH
LOOP: DEC BC
JP NZ,LOOP
...the Sharp Wizard OZ-770PC. It's amazing! 3MB of flash ram, qwerty keyboard, proportional fonts, you can code in raw Z80 assembler for it, plus they have versions of C and BASIC for it. Tons of user written programs on the net. (MyWizard.com and many other sites) Best $100 I ever spent. I much prefer this design to a Palm-style tablet. Here is a good picture.
;)
And no, I am not affiliated with Sharp or Amazon
"Mind, as manifested by the capacity to make choices, is to some extent present in every electron." -Freeman Dyson
I rememebr programming a Z80 Microtrainer in college; it had a keypad for hex code, but I managed to smuggle in a friend's copy of the ZAD cross-compiler to the lab. We had these old 286's with serial-to-headphone jacks that connected to the microtrainer. You typed in your assembly code on the 286, ran it through ZAD, and uploaded it to the microtrainer. You could even hear the data being transfered via the speaker on the microtrainer.
t or/
I remember having my first real experience with handling Interrupt Requests in a lab with the Z80. Too bad the company is having trouble.
I tried to find a pic of the old microtrainer (made by CAMI Research), but alas, they no longer support it.
I did manage to find a link to another University that used them for ECE projects. (Thanks Google!)
http://comet.ctr.columbia.edu/msl/2000class/eleva
Go ahead, tell me I am trolling, but how on earth did a company with an ancient, 8-bit architecture manage to get idiots to throw $280M at it?
By creating an architecture that is still used today in everything from calculators to embedded industrial control applications. Not every application needs ghz class CPUs, 512MB of RAM, and pipelined parallel processing.
It's guys like you that keep guys like me employed. Every problem presented to you is solved with a PC in some form factor or another. Computerized home thermostat? Mini PC on a board. Web server for monitoring temperature and pressure? PC in a rack mount case with an A/D board. Telephone voice mail system? PC in a funny looking case. Then when your product is hopelessly over the price target and behind schedule, guys like me get calls, choose an appropriate architecture, be it a PIC, Z80, AVR, or something else, and get the project back in line with reality.
Go ahead and mod this down as flamebait or troll if you want. I've got 50 Karma points as I post this so I'll live with 47 if need be.
The most popular z80 machine, I'm sure, was the game boy.
Anyway, z-80 isn't going to die, the company is just going for bankruptcy, not dissolving. And if it did dissolve I'm sure that people would continue to make chips.
autopr0n is like, down and stuff.
Zilog not only made processors but also a rich array of peripheral chips including SCSI chips used in earlier Sun and Macintosh workstations. Unfortunately, Zilog got too big for its britches and forgot who brung them to the dance: small independent software developers. In recent years, unless they thought you were going to place an order for one million chips, their attitude became "go away, son, you bother me."
Can't say that I'll cry any tears for Zilog.
Chapter 11 - Chapter 11 of the Bankruptcy Code is frequently referred to as "reorganization." Although an individual may file under Chapter 11, generally it is used to reorganize a business. Individuals with large federal or state tax obligations may use Chapter 11 because an extended period of time may be obtained for the repayment of the taxes. Chapter 11 generally allows the debtor to continue its business operations as it proceeds to the desired goal of a confirmed Plan of Reorganization, which must meet certain statutory criteria. A major rationale for business reorganizations is that the value of a business as an ongoing concern is greater than it would be if its assets were liquidated and sold. Generally, it is more economically efficient in the long run to reorganize than to liquidate, because doing so preserves jobs and assets. Cooperation among the various interests, however, is crucial to a successful reorganization.
Chapter 7 is liquidation where basically everything is gone. Chapter 11 doesn't necessarily mean death for a company. Hell, look at Chrysler.
Well, I am not as old as some of you guys, but I remember the very day I got my TI-85. The first thing I learned (in about a week) was TI's Basic language. I quickly became bored with that because it did not offer the programmer a whole lot of options. I then discovered that the TI-85 could be programed in ASM... at that point I did not know ASM at all (well besides an inline ASM line I used in Turbo Pascal to turn off the damned blinking cursor) so I bough a book on Z80 ASM.
:) )
After a couple of months I came out with my first ASM game, yea it wasn't all that great but it paved the road for the following games. In those next three years I released 8-9 pretty darn good games for the 85, after that I got a shiney new TI-86 and never touched the 85 again (but again it had a Z80, and a TON more ram to work with). I programed a few games for the 86, but I had slowed down a lot from when I had first started.... I guess I began to get a little bored.
I had a lot of fun with the Z80 cpu, its ASM language was pretty easy to get the hang of, and it wasn't a slouch.... my games ran fast. I no longer am a part of the TI community, I have moved onto bigger and better things (let me tell ya, knowing ASM in college was awesome.... my asm classes and computer architecture were much easier
I know there optimizing compilers exist... and they are darn good at what they do, but still there is nothing like optimizing key functions in your code by hand.... Thanks Zilog, you have given me a life skill.
I'm from Nampa, ID, just down the road from the Nampa Zilog-manufacturing plant. I can't say that i didn't see huge losses coming from this company, although the bankruptcy surprised me.
Zilog has had problems finding a niche for quite some time. In recent years (months?), they have been highly influenced by the market trends, which have affected their product directions. I mean, their main product as of recent is a z80 webserver kit.
I still think there's plenty of room in the market for a microcontrollers company, but this company needs some serious restructuring. Along those same lines, they need to keep their logos for more than a month at a time. Every time I drive by the plant they have a new logo and coloring scheme, the most recent of which is a horrid yellow-on-purple. You haven't seen tacky until you've seen a beautiful, white, futuristic-looking technology building with a giant yellow 'Z' plastered on the front, covering all the windows.
Should have seen this bankrupty coming from that alone!
Man, lots of fond reminiscing here about the Z80 (and clones). Quite rightly, too... the Z80 was a fun little beast.
... rr14). The Z8001 and Z8003 were "segmented", but they used a reasonable segmentation model to achieve 8M memory...
Zilog, however, made lots of other stuff. Some were moderately successful (Z8530 SCC), some not so (Z8000 MPU).
The Z8000 actually was fairly popular in military applications until COTS took over. I seem to recall many avionics systems used it. When it came out, it was comparable to the 68K.
It had 16 16bit registers (r0-r15), each of which could be addressed as 2 8-bit registers (rhN, rlN). R15 was the stack pointer. Nice orthagonal instruction set, with logical block moves (similar to the Z80 LDIR instruction), as compared to the intel REP instructions...
The registers could be doubled up into 32-bit registers (rr0, rr2,
The low 16 bits were the offset in the segment, and the high 7 bits were the segment number. So, you essentially had 23 bit addressing. Of course, the way you generated segmented addresses was a tad odd... I believe bits 30:24 were the segement number in a 32-bit address.
Only problem was, they never got the Z8070 FPU working. Bummer.
Fascism starts when the efficiency of the government becomes more important than the rights of the people.
Holly: I was in love once - a Sinclair ZX-81. People said, "No, Holly, she's not for you." She was cheap, she was stupid and she wouldn't load - well, not for me, anyway.
Lister: What are you trying to say, Hol?
Holly: What I'm saying, Dave, is that it's better to have loved and to have lost than to listen to an album by Olivia Newton-John.
Cat: Why's that?
Holly: Anything's better than listening to an album by Olivia Newton-John.
Maybe this could be a model for new IPO's... ;)
A feeling of having made the same mistake before: Deja Foobar
... was the dual register set. It was used to good advantage in embedded systems where the executive would use one set and the app would use the other. If you remember the chess program Sargon, it used one set for white's state and one for black's. It was a pretty neat machine. But like many of the 8-bit stuff out there, it didn't make the jump to 16 bits gracefully. The Z8000 was pretty dismal...
That is all.
not true. the Empire State Building has gone bankrupt dozens and dozens - literally - of times in its life.
the interplay of debt and equity offer the capability to create investments with non-differentiatable payout patterns. Sometimes these make sense. Bankruptcy means that the current equity holders's stake has gone to zero, so their rights dissolve and the debt holders become the new equity holders. All of the assets continue to exist, simply their ownership changes.
I ported 3COM's UNET TCP/IP stack to an Onyx Z8000 box around 1982. This may have been the first single-chip microprocessor on the Internet. It was at IP address [128.5.32.5]. Note that that's class B network #5; this was early stuff.
Yes there really is a 20MHz z80.
Maybe now the Z80 is just for embedded devices. But from about 1980-1984, I worked at a place that sold Z80-based word-processors (when a 'word-processor' was a physical box). We had all the usual features for that era, such as footnotes, spell-checking, mail-merge, and even a builtin spreadsheet.
And it was all written in hand-coded assembler.
I'm one of those veterans who built his own computer at the dawn of the Personal Computer era. The 4 MHz Z80 was the ne plus ultra of the processor world, and we thought we were hot stuff when we managed to overclock one above 5 MHz. Their Serial I/O chips were also the most advanced thing going for serial comms, and we implemented many protocols like X.25 and PARS that formerly required a complete circuit board for what the SIO did. It's a bit sad to see what was once the cutting edge company brought so low.
I agree with you about the 6502, but not about the Z-80. IMHO the best 8/16 bit CPU is the 6809, or perhaps the Hitachi 6309 (which extended the 6809 architecture nicely...too bad folks in the US didn't know about it for years). The 63C09 has been run at 5 MHz, and there's a company called INICORE that counts among its products iniCPU, a logic design burnable into FPGA or ASIC that implements the 6809 instruction set--the web page cites its performance running at 40 MHz.
...but there are other things she won't be able to do that will hinder the success of her chip in the embedded market. My cat could probably also design an 8-bit processor, but the documentation process - in fact, the entire support process - needed to bring a chip to market would be a little beyond her. Without opposable digits, she would be unable to type up the whitepapers and specs, and without the power of speech, she would not be able to dictate this information. So, unless my own cat blatantly copied an older processor, no one except her would be able to develop for it.
Of course, if my cat did manage to successfully document her chip, I'm sure she's sell it for a reaonable price - a lifetime supply of canned food, say. She and I both know that's a little steep for an 8-bit chip, but who can say no to my adorable little cat?
I'm the stranger...posting to
Look, nobody's gonna buy an 8-bit personal computer these days. The only use for Zilogs is in embedded apps, and that can't be emulated away.
I'm the stranger...posting to
Actually, it's also used (in GBA mode) as the "sound cpu".
You're confusing the GBA with the Sega Nomad. I have written some software for the Game Boy Advance. The technique of letting a second cheap CPU handle some sound chores is common on Sega Genesis and necessary on Super NES (which has very little bandwidth between the sound side and the CPU side of the system), but in Advance mode, the GBA completely cuts power to its GBZ80 processor.
Read more about the GBA hardware here.
Will I retire or break 10K?
... of my first time to the NCC, in the early '70s - when Zilog had first announced the Z8000.
One of the things I did was drop by the Zilog booth - twice (the second time when the head of the project was there), to comment on the instruction set. Went something like this:
Me: ~"This is a really great instruction set. But there's one thing missing. When an instruction is aborted by an external memory controller and the interrupt taken, the state isn't preserved well enough to restart the instruction after the memory fault is fixed. You could do true virtual memory if you fixed that.~"
Him: ~"We're not planning to do that. We already looked at it, and it would expand the microcode by about 50%~"
Me: ~"Oh, good. Then (given Moore's law improvements in silocon fabrication) it could be done in 6 to 9 months.~"
Him: ~"Nobody would ever want to do virtual memory on a microprocessor.~"
So they didn't do it. And a few years later the Motorola 68000 family (which DID have restartable instructions on memory faults) became the canonical processor for the "cheap unix box" explosion.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Bzzt. The 68000 could not completly re-execute an instruction that was aborted due to a bus error.
That's why I said the "68000 family" rather than the 68000.
Like Zilog, Motorola's first "x000" chip couldn't restart instructions that failed due to memory faults. Unlike Zilog, Motorola did a followon which COULD.
Perhaps it was more cluefulness on Motorola's part than Zilog's. Perhaps it was pressure from Sun, or their two-CPU box which PROVED that people REALLY wanted to do virtual memory with microprocessors - badly enough to put in a second expensive (in those days) CPU chip just to make it possible.
So the 68010 was the first microprocessor chip that could do true virtual memory. Combined with Unix its family became the foundation of one era of the microprocessor explosion.
This lasted until it was displaced by the horribly asymmetrical, but extremely cheap (at the time), Intel 86 family, on the (almost inadvertently) open and expandable IBM PC/clone defacto-standard platform.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
My 1985 IBM XT has a disk controller (presumably ESDI?) with a Z80 to run it, in the same way that 68000-series chips show up on RAID cards nowadays. (our mailserver at work has a disk controller more powerful than an Atari TT :) )
"don't fall into the fallacy of believing that Perl can solve social problems. Maybe Perl 6 can, but that's a ways off"
Although your rant has merit, you picked on the wrong rantee. I work for a fabless semiconductor company architecting extremely low-cost 8051-based smartcards. And it is 8051s that own the mass 8-bit market; Z80 & HD64180 are passe. This has been the case for years, hence my original point about complete dunderheads being required to cough up that amount of money on a game plan based on a dead architecture.
There was,
LDIR Load Increat Repeat
LDI Load Increment
LDD Load Decrement
LDDR Load Decrement Repeat
but no LDIF
I cut my teeth write ZX Spectrum games in Z80 assem, and i'm very sorry to see Zilog go.
Z80s and varients are still in use in places, embedded controllers etc, and i hope someone keeps
making them
6809 RULES!
Ahem. I had a 6809 machine once, and it was a pleasure to program in assembly. 6809 also had OS/9, a multi-user multi-tasking OS which was better than any other OS for a system of the period.
And you can't dislike a processor with the opcodes BRA and SEX!
(BRanch Always, and Sign EXtend, as if you didn't know.)
"Information wants to be paid"
I work for a fabless semiconductor company architecting extremely low-cost 8051-based smartcards. And it is 8051s that own the mass 8-bit market; Z80 & HD64180 are passe.
Bad news: 8051s are becoming passe as better architectures like the ARM and the Atmel AVR RISC microcontrollers come on the scene.
But Z80s are still used in many devices. They aren't usually the best choice for a new design, but if the company has an existing product or existing code base, then they may well be justified. Look at how many of them TI used in their calculators (though I don't know for certain that Zilog built them).
An 8-bit CPU can address only 64K AT ONE TIME. The venerable Commodore 128 managed to have 512K available, despite having the 8-bit 6510 as a CPU. It takes some tricks, but you could, for example, make the 1st page of RAM (256 bytes=2048 bits) specify which 64K block of RAM is to be available, giving you a total of 2^2048 *64K of RAM (give or take 2^2048 * 256 bytes :)
It's not going to be fast, and it's not going to be pretty, but you could access huge amounts of information with only an 8-bit microcontroller.
Turing developed some theoretical underpinnings of computer science; the bit I was referring to was the idea that a Turing machine can simulate any other computing device. This can be interpreted (not entirely correctly) that any computer can simulate any other computer (performance considerations notwithstanding). Von Neumann made some of this practical (or at least, was given credit for it).
P.S. To my comment of just over a minute ago:
I didn't note the bit about bankswitching until I'd hit post and back...DOH...sorry for pointing out the obvious.