test-drb October 1998

test-drb@ccmp.vtda.org

175 participants
1141 discussions

64bit data 128Bit address...Re: Building a Z-80 (64bit!!!)

by allisonp＠world.std.com

< > It is good design in my opinion to have your address bus twice as wide With a 128 bit wide data bus for a predominatly byte oriented machine all that means is every memory cycle gets you 16bytes at memory cycle speeds. It's a bus bandwidth saving trick as it effectively can give what ever your word time is to put up another 16bytes. This goes back to the core days when memory was slower than the CPU. The penelty is a whopper when all your doing it altering or writing one byte as you have to resort to RMW (read, modify, write) cycle at memory speeds. The alternate is a lot of word/byte selection logic that can slow things down. With the upper speed of available z80s (z180s180) being 33mhz there is plenty of speed so that it can hurry up and wait for the disk, floppy and user. Allison

25 years, 9 months

Building your own Z80 / performance

by allisonp＠world.std.com

< I have never tho't of the Z80 as a slow processor. I interfaced a Z180 < running at 18 mhz to a VGA board recently and using a C compiler was abl < build a machine that acted essentially like a 80286 running at 10 mhz. Inthe early to most of the 80s I've spent a fair amount of time using 6 or 8mhz z80s to blow the doors off of x86 hardware. The z80 is a better IO machine! < Z80 is a pretty cool platform for building an extremely cheap and modern < network computer since it acts a lot like a 16 bit processor. There's < you can do with it with a relatively minimal effort. The z280 is even better, the enhancements it has are pretty decent and it's faster still. Doing your own design in a FPGA is harder than would first appear. Unless your using a known core things like gate delays (race conditions) can bite you. Also some routing paths can be slower than others so you may get far less than the rated Tpd. Allison

25 years, 9 months

64bit data 128Bit address...Re: Building a Z-80 (64bit!!!)

by allisonp＠world.std.com

< As for the suggestion of using a DEC alpha to emulate a Z-80... I thoug < about it but, those darned Alphas are way too expensive for us hobbist Some of the early slow ones (under 200mhz) are in the used market for under a couple hundred. The other approches witll cost that much or more for superfast memory. < < Tell me more... what are Verilog, Xilinx or Altera? FPGA, Field programable gate arrays. The fast ones are down in the 2.5ns range per gate. < Anyone know if Zilog is going to beef up the Zx80 line???? The z380 series is that. There is already a z382. Allison

25 years, 9 months

Model 4 I/O ports - found

by ard＠p850ug1.demon.co.uk

A non-text attachment was scrubbed... Name: not available Type: text Size: 382 bytes Desc: not available Url : http://www.classiccmp.org/pipermail/cctalk/attachments/19981022/9d484203/at…

25 years, 9 months

Hello Again

by Philip.Belben＠PowerTech.co.uk

Well, I'm back. And what a long time it took. Two weeks in the USA, seeing California and attending the VCF; nearly a week in bed with food poisoning after the flight home; and two weeks on this bl**** awful new e-mail software at work before I could get it to behave well enough to trust it with the volume of mail Classiccmp provides. (We're using Lotus Notes, BTW. Most of us now call it Lotus Not.) I really just wanted to say I thoroughly enjoyed the VCF. Well done Sam, a wonderful effort. In a recent telephone conversation with Tony Duell, he mentioned that someone (was it Kevan?) was thinking of doing a VCF in the UK. To that person: I am definitely prepared to support you. I can provide plenty of computers for exhibiting; I can talk on some suitable subject; I am prepared to chip in a fair amount of dosh to get this show on the road. Looking forward to a lot of fun again reading Classiccmp, Philip. <><><><><><><><><><><><><><> Philip Belben <><><><><><><><><><><><><><> Bloedem Volke unverstaendlich treiben wir des Lebens Spiel. Grade das, was unabwendlich fruchtet unserm Spott als Ziel. Magst es Kinder-Rache nennen an des Daseins tiefem Ernst; Wirst das Leben besser kennen, wenn du uns verstehen lernst. Poem by Christian Morgenstern - Message by Philip.Belben(a)powertech.co.uk

25 years, 9 months

OT: Modern college educations

by franke＠sbs.de

>> A lot of discussion here on college educations, my 0.02. >> 1) I don't expect New College Grads (NCG's) to know a lot, I expect them to >> know how to find something out that they don't know and how to power >> through bull s**t type work. I also expect them to know the basic theory > Please note that the 2 people I was moaning about earlier (one wanted a > 362.83 Ohm resistor for his LED, the other couldn't grasp '5V across > 4.7kOhms is a little more than a milliamp') were EEs. Thats bad - he should at least have taken a class about resistor networks to build his needed one in a 3D configuration :) >> 2) I don't expect CS majors to be taught assembly, per se. In case you > Oh, IMHO all CS students should have some idea as to what the 'computer' > they are writing programs for actually is. And that means having some > idea of digital electronics and assembly language. I find using tools > that you don't fully understand is a darn good way to produce poor code > (or whatever). So, you're just talking about the usual way , and why languages like C++ are so popular ... >> A fun exercise (in a nerdly sort of way) is to presume you've been dumped >> onto some raw continent with nothing but your brains and underpants, now >> build a PDP-8. (You can assume that you will have food and shelter.) > Hmm... I'd rather build a relay logic machine. Drawing the copper wire > for the coils would be painful but possible. Similarly making soft iron > cores and armatures. A lot easier than trying to make transistors, anyway. In fact, I think I would try to go for a more mechanical device - so I don't have the problem to create a power source and _stable_ voltage wich again includes semiconductors or other kinds of very delicate equipment like mercury rectifiers. >> One of the things that struck me about a 'dead' PC I was attempting to fix >> was that the BIOS flash had been zorched and a) Not only was their no way >> to recover the bios but b) the chipset used was both non-standard and made >> by a now non-existent company who left behind no records. Talk about > Thankfully the BIOS on this PC is in OTP EPROMs, and I have the official > source listings anyway. Ditto schematics of everything but the hard disk. > This machine can be repaired. But independent of the chipset, one should be able to build a minimal start up BIOS to launch a real mode DOS and then a reprogramming is just some steps away. >> and get it working again, when the engine computer on your car breaks, your >> out of luck. > Why do you think that when I get a car I am _not_ having any electronics > anywhere enar the engine :-). Mechanical stuff I can understand and > repair. Electronics I can understand, but there's no way I could make a > custom chip at home. Thats why I still drive a 1988 Skoda Rapid - no electronics. Just imagine the finger print sensor of your new Mercedes S-class care goes wild ... (But on the other hand, a CX25 TRI, maybe as a 6 wheeler, is still a dream car ...) Gruss H. -- Ich denke, also bin ich, also gut HRK

25 years, 9 months

FS in UK - Intel & Cromemco Etc. circa 1974 & 1980

by headcase＠eclipse.co.uk

Thanks to all those of you who replied about the computer equipment I'm selling I read all the replies up to about second week in September but lost those together with a massive pile of un-read mail up to a few weeks back.. If you havent had a reply - sorry - over confidence or Bill's bugs - I still haven't figured out which I've sorted through all my computers and equipment and listed them on my web pages .http://www.eclipse.co.uk/great-gull/ together with photos etc The Intell, Addmaster, Roytron, Mesonix etc. stuff is on http://www.eclipse.co.uk/great-gull/sellintel.htm I have listed all the information I have, together with photographs of each item. Unfortunately I seem to have lost the folder containing all the original documentation but most of the stuff is circa 74/75. The Cromemcos and software stuff are on http://www.eclipse.co.uk/great-gull/sellcromemco.htm There are three working Cromemco Systems circa 1978 to 1984 - two System Three's and one Z-2H - together with a pile of instruction and technical manuals. I have listed some of the more interesting software on some of the one hundred and fifty (approx) 8" disks. There may be some software or manuals that are needed by subscribers to this list - that shouldn't be bundled for sale with the equipment. If having seen my web site - it's not bad for someone who never progressed past machine code - you think I have, or even might have, something you want e-mail me and I will try to reply promptly. Unfortunately most of the stuff is stored 150 miles north of here in Worcester, so I won't get the chance to delve deeper very often. It may offend some sensibilities that I will be listing most of the stuff on eBay just as soon as I have time. Those of you who think I should be less grasping should remember that I bought all this stuff new before most of you were born, when In the mean time I'm open to any offers and and will pay/arrange shipping to most countries Otherwise see it on eBay in due course. Jim Bunting - headcase(a)eclipse.co.uk M.V. Great Gull, Double Locks Hotel, Canal Banks, Exeter Ship Canal, Exeter, Devon, U.K. EX2 6LT. Phone No. 44 (0) 1392 493311 (On Board)

25 years, 9 months

Emulation or the real thing?

by allisonp＠world.std.com

< I'm still skeptical. I'd love to hear more first-hand reports of < the oldest code still running as-is. Come on, code from the 50s < that's never been replaced? Running on what? Under an emulator? Ok, howabout a Brigeport milling machine with the PDP-8E it was purchased with in 1975, still running the same code. code from the 50s would likely be fortran or Cobol and yes it would be portable enough to go from one machine to the next with only a compile (no edits). Code from the late 60s era machine could still be running oth either native hardware or later machines that had to support old code. What is missed is code is expensive and hard to maintain so if you have something good and known you run it for a long time. I'm still running 8080/z80 code from the late 70s early 80s! If it aint broke don't fix it and the system that it was genned on is also 20 years old last March. So it's very easy for old code to linger. Allison

25 years, 9 months

OT: Modern college educations

by maxeskin＠hotmail.com

> >> >> A lot of discussion here on college educations, my 0.02. >> >> 1) I don't expect New College Grads (NCG's) to know a lot, I expect them to Don't. They're here to make the college money, not to get ahead in the real world. This is already starting to bite back, and will get worse. >I would also expect them to have some practical knowledge as well. And >how to make order-of-magnitude calculations/reliable guesses for results in >their subject. > >Please note that the 2 people I was moaning about earlier (one wanted a >362.83 Ohm resistor for his LED, the other couldn't grasp '5V across >4.7kOhms is a little more than a milliamp') were EEs. It's reasonable >that a chemistry student, or a classics student, or something like that >wouldn't have a clue about resistor values, but for an EE? It worries me. >It wories me a lot. But, why? Most of them will be getting a job in which they either follow flowcharts, or in which they figure out the prettiest way to put U1-49 on a PCB and attach a 6-pin connector to pins 1-6 of each U. If the world is unlucky enough to have one at a real circuit without a manual, the few comptetent engineers will find a way to make this guy's function unnecessary. At the end, of course, this raises unemployment and decreases skill. It's clear a crisis will occur <100y >Oh, IMHO all CS students should have some idea as to what the 'computer' >they are writing programs for actually is. And that means having some >idea of digital electronics and assembly language. I find using tools >that you don't fully understand is a darn good way to produce poor code >(or whatever). The thing that annoys me terribly is that people think they can do 'interesting' things on a computer, like web pages, and graphics, and so forth. IMHO, an artist should play with Photoshop and POVRay, while a person interested in computers should learn assembly. It's not mutually exclusive, of course, but I think there are too many people who could contribute a lot to technical stuff, while they sit lazily clicking away at the canvas. >Of course. Things you've taught yourself are rarely forgotten. If you >want to understand something then you will. Unlike the student who learnt >the book to pass the exam and has no real clue as to what it means. > Here's something I read in a ham radio book: "There are two ways to prepare for any test. One way is to study to learn the material, and the other is to pass the test. I suggest you study to pass the test" He was, of course, talking about the Amateur Radio certification test. Clearly, this book was made not for people interested in radio, but those who want to send messages about their private life all over the universe. > >> One of the things that struck me about a 'dead' PC I was attempting to fix >> was that the BIOS flash had been zorched and a) Not only was their no way >> to recover the bios but b) the chipset used was both non-standard and made >> by a now non-existent company who left behind no records. Talk about > >Thankfully the BIOS on this PC is in OTP EPROMs, and I have the official >source listings anyway. Ditto schematics of everything but the hard disk. >This machine can be repaired. > >> unfixable! When my PDP-8 breaks I can always go back to first principles > >Exactly. I've fixed my 8/e, 11s and PERQs to component level when necessary. > >> and get it working again, when the engine computer on your car breaks, your With a car, it's not that bad. You could order replacement chips sometimes, or you could try bypassing the messed-up function. >Why do you think that when I get a car I am _not_ having any electronics >anywhere enar the engine :-). Mechanical stuff I can understand and >repair. Electronics I can understand, but there's no way I could make a >custom chip at home. > >> Graphics card? "I'm sorry Mr. McManis but that information is only >> available under NDA to qualified customers who can prove a market of at >> least 1K units/month." Talk about self defeating! > >More people should insist on proper documentation. On a couple of >occasions I've returned a device to the shop that sold it under the UK >sale of goods act. The reason? It was not fit for the purpose that I >bought it for because important documentation (register maps, connector >pinouts, etc) was not available. Alas I doubt if the company ever >realised or cared... The thing is, the documents often don't exist. My father once worked for a HV PSU manufacturing firm, where the blueprints were just barely enough to manufacture the device. > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com

25 years, 9 months

Building a Z-80 (64bit!!!)

by allisonp＠world.std.com

< But just tell me where 32 Bit instructions can be usefull ? < In fact, I can't think of any part inside a BASIC. But the < higher clockrates are a nice thing - and maybe the z180 for < 3.0 style bankswitching. They can address the full 32bit address space (without banking) and the z380 adds a host of addressing modes not in the z80 like 32bit indirect, PC relative and stack relative with long pointers. Also MUL, DIV and a handfull of other nice instructions that would sharply speed on the math functions. Also there are 4 sets of registers compared to the dual set of the z80/z180. Allison

25 years, 9 months

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

test-drb October 1998