test-drb December 2002

test-drb@ccmp.vtda.org

250 participants
519 discussions

Recovering from disk failure - attn Adrian Vickers and Jim Austin

by t.myers＠protasis.co.uk

Hi, I've just had my disk holding my main mailbox fail on me (Yet another Fujitsu MPG fails, that's three in as many months.) Could Adrian and Jim please get back in touch with me, I've lost everything :'( Tim

23 years, 1 month

M200 interfacing

by pete＠dunnington.u-net.com

On Dec 30, 8:02, Sellam Ismail wrote: > On Mon, 30 Dec 2002 pete(a)dunnington.u-net.com wrote: > > That's an odd VIA. You mean a single 6522, right? > Sorry, it was a typo. I meant two 8-bit ports. Ah, I suspected it might be. > This is starting to make sense. > > But alll the documenation I've read so far indicates that the function of > the CA and CB lines is controlled by the PCR (Peripheral Control > Register). I'll go poke a 0 into the IER and then see if I can get bits > in the IFR to toggle by putting +5 on the CA1, CA2, CB1 or CB2 pins. The easiest way to set it up to play with, is first to pull each of CA1, CA2, CB1, and CB2 to +5V through four suitable resistors (anything from 1k to 100k will do for this). Then write 0x00 into the PCR, which makes the four handshake lines operate as independant inputs, with negative edge triggering. Write 0x7F into the IER to disable all the interrupts (so the Apple doesn't do any more resets), and 0x7F into the IFR to make sure it's cleared. Then keep polling the IFR while you ground one or more of the handshake pins. Look for any non-zero value. The relevant bits in the IFR should become set, according to which line(s) you ground. Note that this won't ever set bit 7 (MSB) of the IFR, because that only happens if interrupts are enabled, unfortunately. Also remember you have to manually clear the flags once they're set. If you want to use positive-going edges instead of negative-going, use pull-down resistors (don't rely on an unconnected input being low, it might float to any value) and set 0x55 in the PCR. If you are keen to use bit 7 in the IFR as a flag indicating "any input", stick an RTI instruction on the end of your code, and store the address of that RTI at IRQLOC ($03FE) first. That will prevent the interrupts upsetting the Apple. Alternatively, have the Apple execute an SEI instruction (0x78, sets the IRQ mask in the status register, to disable interrupts). Then write 0x64 into the IER to disable all the interrupts except from the handshake lines, and then write 0x9B into the IER to ensure all the handshake interrupts really are enabled (the default state, after a chip reset, is all disabled). -- Pete Peter Turnbull Network Manager University of York

23 years, 1 month

WWII fire control 'computers'

by mross666＠hotmail.com

Suggest you head over to usenet and ask on uk.rec.subterranea - the folks there are experts on most things concering British bunkers... Mike (the CoreStore is more than just ancient computers - see http://www.corestore.org/sub.htm ) http://www.corestore.org _________________________________________________________________ The new MSN 8: smart spam protection and 3 months FREE*. http://join.msn.com/?page=features/junkmail&xAPID=42&PS=47575&PI=7324&DI=74… http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_smartspamp…

23 years, 1 month

M200 interfacing

by pete＠dunnington.u-net.com

On Dec 28, 19:53, Sellam Ismail wrote: > The next step is to wire in the data signals and start to read data. The > VIA gives me two 16-bit ports and an additional 4-bit port. That's an odd VIA. You mean a single 6522, right? It has two bidirectional 8-bit ports, each of which has 2 associated handshake lines. Each of the bidirection lines can be set to be an input or an output, so it might seem like you have 16 possible inputs and 16 possible outputs -- but you only have 16 connections (plus 4 handshake). > Anyway, I'm in the process of figuring it out. Everything on the card is > accessed through it's I/O addresses, being C080 + (slot * 16). So I have > it in slot 4 which makes all it's I/O available at C0C0-C0CF. I've found > the timer locations and some 16-bit registers. The only 16-bit parts of a 6522 are the timers. Timer 1 is a 16-bit counter accessed as two adjacent 8-bit bytes, and two 8-bit latches (used to reload it when it counts down to zero). Timer 2 is a 16-bit counter accessed as two adjacent 8-bit bytes, without latches (so useful only for counting pulses, or as a one-shot interval timer). > I also found by accident a > timer that generates a RESET interrupt when it elapses--at least that's > the theory I'm going under since the Apple reset itself after I was > playing with the registers. I also believe I remember there being this > feature on the 6522, but I haven't had a chance to read that part of the > docs yet. A 6522 doesn't have any pin that would normally go to the RESET line, but it does have an interrupt output. If you set something in register 14 (the Interrupt Enable Register) and a timer timed out, or you created a transition on one of the handshake lines, the 6522 would generate an interrupt. The interrupts on a standard Apple ][, ][+ or //e aren't used -- it does all normal operations by polling, which is why it's so slow for some things -- and an interrupt causes the processor to jump to a location in the ROM. Just where depends on the ROM, but in an original Monitor ROM or an Autostart ROM, it ends up either going directly to the same routine as BRK does (which will perform the equivalent of a soft reset) or jumps through a vector which you're supposed to set up first. If you've not preset that, I think it ends up in the BRK routine (it certainly does in an Autostart ROM). > One thing I'd like to investigate after I'm done with this is to see if > there are enough inputs on the Apple itself to forgo the necessity of the > 6522. The Game I/O port has 7 inputs, if you count the 4 joystick inputs > that can be used as simple TTL inputs, plus the 3 push-button inputs. The > other inputs would be the keyboard. It has 10 Y-inputs and 6 X-inputs, > plus SHIFT and CONTROL. Er, not exactly. As Tony pointed out, the keyboard includes a decoder (called a ROM in the manual), and the key matrix is decoded by scanning. Either the X or Y lines are outputs (the manual doesn't say which, but IIRC it's the Y lines). Only SHIFT, CTRL, and SHIFT LOCK are plain inputs, and they only serve to modify the 7-bit code sent to the motherboard. > If so, then there should be enough combinations of inputs on the keyboard > port to allow the data signals to go through that, and then the GAME I/O > has just enough inputs to cover all the status signals (HOPPER CHECK, > MOTION CHECK, ERROR, BUSY, INDEX MARK, and READY). I would just use the 6522. You need 12 data, plus 6 status listed above. Is there also a strobe of some sort (or is that READY?)? 8 lines on port A, plus 4 on port B for the data, CA1 for the READY or strobe, and some mixture of the remaining 4 data lines and the other three handshake lines should do it. Don't bother with interrupts to start with. Leave the IER all zeros, and just poll the IFR to look for the flag for CA1. Or just poll it looking for a negative number (which means bit 7 is set, which means one of the other flags is also set). -- Pete Peter Turnbull Network Manager University of York

23 years, 1 month

Can't figure out CA and CB ports on 6522

by foo＠siconic.com

I'm having a heck of a time figuring out how to use the CA and CB ports on the 6522 as inputs. From the documentation I have, it seems the CA port can be used for one TTL input, and CB can be used as two. But how do you set them up as such? The documentation is confusing. I'll sleep on it and see if either a) I wake up refreshed and figure it out or b) wake up and find a nice helpful message waiting here ;) Sellam Ismail Vintage Computer Festival ------------------------------------------------------------------------------ International Man of Intrigue and Danger http://www.vintage.org * Old computing resources for business and academia at www.VintageTech.com *

23 years, 1 month

DIY Equipment Racks

by pete＠dunnington.u-net.com

On Dec 29, 22:06, R. D. Davis wrote: > While thinking about racks to use for mounting my PDP-11/44 > components, as well as other equipment from test equipment to audio > and synth equipment, something just occured to me: why bother with > hunting down steel racks when some 2x4s and lag bolts may suffice just > as well? I was thinking that one can just run 2x4s from the basement > floor up to the heavy wooden rafters, attach them to the rafters, and > then add horizontal supports at the bottom to space the vertical 2x4s > apart properly. Any thoughts on this? I guess the museum-type > equipment purists won't like the idea, but it would be a cheap and > functional solution for many of us. :-) Has anyone else here tried > this? Not exactly this, but in addition to (and adjacent to) my racks, I have some adjustable shelving which uses cantilever brackets, fitted to upright rails that are screwed to the 2"x4" vertical "studs" in the wall. The brackets are 610mm (about 24") long and so the shelves are 750mm (~30") deep. The brackets and rails are Spur "Steel-Lok" type and each pair of brackets is rated to hold 75kg (175lb). I tested their security by climbing up them before I entrusted my precious computers and VR14 to them, so I'm sure a 2x4 frame will hold any reasonable amount of rackmount equipment *providing* you have appropriate fixings. http://www.spurshelving.com/html/tech/steel.htm will show you the stuff I used, I'm sure there's something similar in the States. I used a baby version of the same stuff for shelves on a couple of steel-framed trolleys. HOWEVER, the real utility of a rack is that you can mount things on rails and slide them out whilst they're still connected; and that you can adjust the position of things as more interesting equipment comes along (thanks, James! [1]). In fact, it's very hard to work on some things unless you can slide them out part-way, like my PDP-8/E (which is slightly poorly ATM, but that's another story) or most of my PDP-11s. More than that, it's practically impossible to *use* certain things unless you can slide them out -- you need to slide an RL01/RL02 halfway out of the rack to change the disk pack, for example. Those b***s are heavy, and you really do want to use the proper rails. With the ball-bearing rails, changing an RL02 pack is an operation you can do without thinking about it; without, it becomes a "put it off until next week" job. The first catch is that rails and other (static) rackmount equipment are intended to fit precisely-spaced holes. Despite what some people think, they're neither random nor equally spaced (forgive me if you know this). They're in groups of three with one hole in the centre and one each 5/8" above and below, leaving 1/4" between the centre of the upper (or lower) hole and the start of the next U. In other words, the spacing goes 5/8", 5/8", 1/2", 5/8", 5/8", 1/2", ... and 1 rack unit (1U) is 1_3/4". American-style racks (RETMA?) use round holes, sized to be a good clearance for No.10 machine screws (or M5 m/screws) and are used with Tinnerman nuts. European racks use square holes, to which you fit cage nuts (usually M6 but sometimes M5), and the centre hole of each 1U group has, by convention, a little notch at the side, so it's easy to see where the centre of each 1U space is. For that reason alone, it would be much easier to buy the upright flanges which have the holes pre-punched in the right places. They look like overgrown Dexion angle, or steel angle with a lot of holes (front face) and slots (side, for mounting to the uprights). You can get them here in sizes >from 3U to 48U (common sizes are 3U, 6U, 12U, 18U, 32U, 42U, 48U). Price depends on quality (ie, strength) but they're not expensive. I bought a 24U strip and cut it up for my "hub of the universe" wiring rack. That's a 14" deep 6U high frame with sides of 1" square steel tube and flat-strip cross members at the back, mounted over the workshop door. It holds my main Ethernet switch, and several patch panels for UTP, coax, and fibre. Another reason to buy the proper uprights is that some rails, notably those for an RX02, PDP-8/E, and various other devices, fit to the back surface of the flange, not the front (as most "static" equipment does). If your wooden uprights are the right distance apart to directly fit the sort of equipment that has mounting "ears", ie screwing the ears straight into the front of the wood, you'll never be able to fit most types of slide rails, unless you do a bit of woodwork to provide a rebate. Ditto for most shelves, both static and sliding, except cantilever shelves (which fit from the front surface only, but cost more than ordinary ones). > Lastly, has anyone on this list tried retrofitting non-rack-mount > equipment into racks? E.g., welding (or "JB Weld"ing) rack-mount tabs > onto systems like PCs and Kaypros, as well as making rack-mountable > shelves to hold the Macintoshes, etc.? Rackmount shelves are good for that -- just make sure you have enough, or you end up with systems on top of each other again. Cantilever shelves are particularly good because you can always get at the mounting screws, which are only on the front. Conventional shelves are much cheaper but you need to be able to get your hand and screwdriver into the gap between the shelves to reach the screws on the underside of the shelf you're moving -- oh, and BTW they screw into the sides, which will be a little awkward with big woodscrews. We use lots of rackmount shelves at work, for PCs (mostly running Linux) and Sparcs that weren't designed to rackmount. Otherwise, just screw a bracket onto each side of the case. The screws needn't be very large, as most of the forces are in shear. Self-tapping screws will do for most metal cases, machine screws with washers and nuts on the inside for plastic. I wouldn't weld things. Cases are thin, usually zinc plated (nasty fumes), and hard to weld. I certainly wouldn't use epoxy or anything like that. If you ever want to remove the brackets, two or three small holes in each side is much better than the mess a broken weld or epoxy will leave. A few other thoughts: As someone else pointed out, having the side panels makes a big difference to the sound level. It might also make a big difference to the airflow, though, which is why proper racks have fans of their own, normally in the top :-) It's obvious that the width between the uprights is important, but just in case you're wondering, the depth is not (much). A lot of things are front-fixing-only, and most rails are adjustable. Anyway, most flanges have slots rather than holes for mounting, so some adjustment is possible. I can only think of two items where the exact depth matters: my SGI Origin 2000 (which has two slightly superfluous brackets at the back), and an RL01/2 drive, where the shipping bracket is supposed to be fitted to the rear flange (which therefore has to be a particular distance from the front flange). Nevertheless, there are standard depths, like 600mm, 800mm, and 950mm (in Europe) but they're more a matter of convenience. FYI, a standard DEC rack measures exactly 25" from front surface of front flange to back surface of back flange. To be honest, I think you'd be better off finding a secondhand rack. It will already have all the right holes, possibly the side panels, and if it has casters, it might be more versatile than a fixed frame. Providing the casters are strong enough to take the total weight (ie not intended to be supplemented or replaced by jacking-screw feet) it's useful to be able to move a rack a small amount from time to time. Don't forget you need access to the back as well as the front! [1] James is a list member who has occasionally helped me shuffle a PDP-11/40 up and down a rack. No need to remove the power supply if he's around :-) -- Pete Peter Turnbull Network Manager University of York

23 years, 1 month

WWII fire control 'computers'

by julesrichardsonuk＠yahoo.co.uk

Hi, >> Does anyone know anything about the mechanical fire control computers >> as used in second world war for UK coastal defence batteries? > > You might try posting this question to the guys that know, on the > coastal-defense mail list (yes, there is a list for geeks of all kinds). > I can post the question to the list, if you wish. hey, that'd be much appreciated if you can forward my posting on my behalf. It's probably worth me joining at some point anyway - I've visited most of the coastal defence sites in the UK, but those have all been stripped of nearly all metal. The site in New Zealand seems quite unique just because it is largely intact - power generation and hydraulics are all still present and we hope to have engine room systems running again one day. The 9.2" guns went for scrap years ago, but there are still a few examples left around the world. Having a gun moving around on the hillside under its own power would be pretty cool! > Getting any information or artifacts will be very difficult - about as > close to impossible as you can get. The reason for this is post-war, the > fire control computers were of little use (they do exactly one function) and > have large scrap values (for example, a Mk IV TDC has about a _pound_ of > gold inside). When the mechanisms were scrapped, the manuals were simply > thrown away. that is a lot of gold! A lot of other documentation relating to the site has survived in city archives, including manuals, site plans, the fortress record book etc. but for some reason information relating to the fire control systems vanished - almost as though someone had a paranoia attack and purposely destroyed it. There's a rumour that there is one photo of the computer somewhere - I'm trying to trace its wherabouts at the moment. >> It's possible that the unit shared a lot of commonality with ship-based >> systems, in some cut-down form or other, but I haven't yet located anyone >> who can confirm this. > > Possibly. They probably would be cut down, as there are less variables to > deal with - not being on a ship means that you can eliminate the stable > element and gyrocompass inputs. yes, that's what I figured. Logically there are common elements between the systems, and probably enough to justify sharing of parts. Of course, whether this sort of foresight happened or whether the Army took the Navy's equipment and simply trimmed redundant functionality out is debatable! They may well have been completely different systems... (the 9.2" guns were ex-navy and I believe the hydraulic systems were shared too, but of course that doesn't mean to say that fire control systems were at all related) > I am very interested in these things (I think the fire control collection > now outweighs the computer collection). literally, I imagine - those mechanical systems must be pretty heavy :-) cheers Jules __________________________________________________ Do You Yahoo!? Everything you'll ever need on one web page >from News and Sport to Email and Music Charts http://uk.my.yahoo.com

23 years, 1 month

(no subject)

by fm.arnold＠gmx.net

>Message: 3 >Date: Fri, 27 Dec 2002 20:46:01 -0600 >Subject: DEC 3000 300X/300 >From: Bradley Slavik <fire(a)dls.net> >To: <cctalk(a)classiccmp.org> >Reply-To: cctech(a)classiccmp.org > >I finally am almost within reach of getting my DEC 3000 300 working. A few >years ago I foolishly purchased it without memory, thinking it would be a >simple matter to inexpensively pick up some memory later. How wrong I was. >Now I have purchased a DEC 3000 300X with 160MB memory with the intention of >lending 32MB to the 300. > My 300LX uses 8 or 32 mb 72 pin FPM-DIMM's (with Parity, that's important!). Did not try 16 or 64 mb modules, anyone else? >Now I look at the back of the machine and my main >question is what cable do I need to hook up my VT220, or my VT420, or what >model monitor with keyboard and mouse is this thing supposed to work with? > >In a typical DEC fashion, in the manual they do not mention model numbers of >monitor to go with it, but there are rough instructions that I may be able >to follow to get dumb terminal console working. I would ideally like to have >real monitor on 300X, Read the manual at page 1-3 for supported monitors. I think you are aware of Keyboard-mouse Cable 17-02640-01 and the DEC-protocol of the mouse. Otherwise see: http://www.netclique.net/oldmouse/Hawley/DEC-VAX.html or http://www.cs.utk.edu/~shuford/terminal/dec_mouse_news.txt http://groups.google.com/groups?q=author:mzthompson%40aol.com&hl=de&lr =lang_en&ie=UTF-8&selm=20021016091543.01470.00001267%40mb-mb.aol.com&rnum=4 (on one line). > >VT420 on 300, leaving my VT220 free for my MicroVax II. > Any ascii-terminal will work, use 9600 baud setting. I just tried to connect a VT220 to my 300LX to the 25 pole D-connector (plug 4 on page 3-6) and was successful with a BC22A-25 cable. (Dont forget to detach the keyboard from 300LX!) It takes about 1-2 minutes before the first text appears to the serial terminal. Output comes just after the 3-tone melody. Build-in CRT-output comes much faster. I have a VR297-DA, and tried to connect it to the 300LX to no success. (There is some picture and synchronisation, but the information appears 5 times adjacent, and is not readable.) Either this monitor does not match the 300LX or the video-setup of the internal adapter has the wrong parameters. Anyone some clues about this? To what workstation will the VR297-DA fit? I hope this helps you. Frank Arnold

23 years, 1 month

Sun workstation and HP 98789A monitor?

by rdd＠rddavis.org

Tonight while looking for a wall wart for an old Hayes modem, I discovered something that I didn't know I had: some sort of monitor cable adapter. At one end is a Sun workstation style monitor connector from which four cables are connected to BNC connectors with reg, green, blue and white labels. Does anyone on this group knows if it will work with an HP (Sony) 98789A fixed frequency monitor which has R, G, B and HD and VD sync connectors (can I use just one of these syncs?). I'd hook it up to see if it works without asking, but I'd prefer not to have to carry the monitor down to the basement if it isn't going to work. :-) -- Copyright (C) 2002 R. D. Davis The difference between humans & other animals: All Rights Reserved an unnatural belief that we're above Nature & rdd(a)rddavis.org 410-744-4900 her other creatures, using dogma to justify such http://www.rddavis.org beliefs and to justify much human cruelty.

23 years, 1 month

WWII fire control 'computers'

by julesrichardsonuk＠yahoo.co.uk

Hi, A bit (er, lot!) of a long shot this, but it's probably vaguely on-topic... Does anyone know anything about the mechanical fire control computers as used in second world war for UK coastal defence batteries? I did some restoration work out in New Zealand earlier in the year on a wartime coastal battery which was built to a British design, and used British systems. Unfortunately all information on 'out' plotting-room computer mysteriously vanished in the 1950's, and the unit itself sadly was cut up for scrap value many years ago. It'd be nice to get some information on possible units though just for something to show on site open days. It's possible that the unit shared a lot of commonality with ship-based systems, in some cut-down form or other, but I haven't yet located anyone who can confirm this. I seem to recall a few posts about US equivalent systems a few years back, so there's probably a few current list members who are interested in this sort of system despite the units being non-electronic in nature! cheers, Jules __________________________________________________ Do You Yahoo!? Everything you'll ever need on one web page >from News and Sport to Email and Music Charts http://uk.my.yahoo.com

23 years, 1 month

← Newer
1
2
3
4
5
6
7
8
9
...
52
Older →

Jump to page:

2026

2025

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 December 2002