test-drb September 2015

test-drb@ccmp.vtda.org

198 participants
312 discussions

by jnc＠mercury.lcs.mit.edu

> From: Mike Stein > I don't think I can scan the print set; IIRC the pages were longer > than 14". How much longer? My A3 scanner will take up to 17". I'd be happy to scan them for you (and return them afterwards) if you send them to me. (BTW, this offer is open to everyone/anyone - although if I get 23 sets of prints, please don't expect instant service! Please contact me first.) Noel

7 years, 10 months

piggybacked TSOP memory chips

by dave＠661.org

Someone on #classiccmp showed pictures of a DDR SDRAM module with piggybacked TSOP memory chips. I've never heard of doing this with surface-mounted devices. http://imgur.com/a/CGk8h -- David Griffith dave at 661.org A: Because it fouls the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail?

7 years, 11 months

Announcing TCP/IP for RSX-11M-PLUS

by bqt＠softjar.se

Well, it's been a long time project, but I'm happy to finally announce a more public initial release of TCP/IP for RSX-11M-PLUS. This is the result of over 20 years of development. Needless to say, I've been doing a lot of things over the years, and this code have been through four reimplementations over the years. What I now release is something that I believe is a nice and useful piece of software. I am aware of the fact that most people do not use these machines any longer, but if someone actually wants to talk to me about support for this or other RSX software, let me know. Also, feel free to spread this information to anyone who might be interested, anywhere. So - what is in this release? It is a complete implementation of ARP, IP, UDP, and TCP for RSX-11M-PLUS. It has been tested on RSX-11M-PLUS V4.6, but should work on any V4 release. There might be some small tweaks or fixes required, but nothing major. It do require a system with split I/D-space, or else at least the TCP part will not fit. For Unibus machines, it should be possible to run without any additional software except what is in a base RSX distribution. For Q-bus machines, DECnet is required for ethernet networking. The TCP/IP stack can co-exist with DECnet. Some utilities also utilize RMS for file access. A bunch of tools, utilities and libraries are also included. These include: . IFCONFIG network configuration tool. . NETSTAT network information tool. . PING . TRACEROUTE . DNS client . FTP daemon . FTP client . HTTP server . TELNET client (rudimentary) . TFTP client . TFTP server . INET server that can do SINK, ECHO, DAYTIME, QUOTE, and IDENT . NTP client . LPR client that sits in the queue manager (rudimentary) . FORTRAN-77 library . BASIC+2 library . PDP-11 C library The implementation fulfills most of the requirements put forth in RFC 1122. There are a few limitations because of restrictions in the PDP-11, but none of them should really cause any problems. Documentation is still on the thin side, but example configs are also provided, along with installation scripts. A bunch of test programs and example programs are also included, as well as the sources of all tools and libraries. The TCP/IP stack itself only comes in binary form. All tools are also included precompiled in the distribution, so an installation only have to build the stack itself for your system, and then you should be ready to go. The API only have a slight resemblance to the Unix sockets API. However, if someone sits down to write code to use TCP/IP under RSX, I'm sure they will discover that it is extremely easy to use the libraries, or the basic functions. The TCP/IP implementation is mostly written as device drivers. This also have some other interesting implications, such as it is possible to access TCP as a normal file. You can, for instance do something similar to the Unix netcat command by issuing the MCR command: > PIP TI:=TC:"foo.com";4711 which would open a connection to foo.com, on port 4711, and any data sent from that machine will be shown on the terminal. The resources used by TCP/IP are modest. A memory area (size selectable at generation/startup) is used internally. The amount of memory in the private pool limits the amount of data that can be buffered. Normal pool is used in a small quantity for each TCP port that is open. People are welcome to play around with this, and make improvements. Contributions of code is most welcome. There are still lots of things to do. The programs marked as rudimentary should be rewritten. The most obvious thing still missing is a telnet daemon, which probably is my next step. However, the reason for now announcing the release is that it can finally be distributed natively from an RSX host. The main locations to download the TCP/IP for RSX are: Madame.Update.UU.SE (anonymous ftp). This is one of my development systems for this software. It runs under E11, and if things are down, I blame E11. :-) When connected, you are already in the right directory. There is both an RL02 disk image there, which can be downloaded by anyone. If you happen to have an RSX system which you are conneting from, you can also try getting the BQTCP.TAP tape image. Such an image will not transport cleanly to a non-RSX system, however. Sorry. ftp.Update.UU.SE (anonymous ftp) - /pub/pdp11/rsx/tcpip The disk image is normally duplicated to ftp.update.uu.se as well, so the same file can be found there. I hope some people will find this useful/amusing. :-) Johnny Billquist -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol

7 years, 11 months

Opening a DECserver 90M External PSU

by robert.jarratt＠ntlworld.com

One of my DECserver 90M PSUs got dropped and stopped working as a result, possibly because it got pulled by the cable. That sounds like the kind of damage that might be repairable. I tried to open the enclosure and I found a hole under one of the labels, expecting it to be a screw hole, but it isn't. I am guessing you just have to pry the halves apart, but without knowing where the clips are inside it is easy just to break the thing. Does anyone know how you open these PSUs? Thanks Rob PS Yes I know these PSUs are notoriously unreliable, and that you can make a PSU with a DIN plug and a wall wart, but I would still like the original to work if possible.

7 years, 11 months

PDP-12 Restoration at the RICM

by michael.99.thompson＠gmail.com

We did a lot more debugging on the TC12 LINCtape controller. We saw a 500ns glitch in the LMU MOTION signal that corresponded to a short slowdown in tape speed. We will investigate this next week. We entered the LINC instruction to check a single block (0707) in the left switches and a block number (0777-0000) in the right switches. When we pressed the DO key it should go to that block on the LINCtape. With large block numbers (07xx) and with the tape positioned half way through the tape it worked OK. With lower block numbers it sometimes could not find the block and searched back-and-forth on the tape. The logic analyzer showed that the block numbers were correct in a sequence of several blocks, and then it will read a bad block number. The TC12 would tell the TU55 to turn around, it would read a good block number, realize that it was going the wrong direction, and turn the tape around. It would then read a good block number, and then a bad block number, and turn around. At this point we don't think that we are working with bad tapes, but the problem might be in either the TU56 tape drive, or the TC12 LINCtape controller. We see bad behavior in both devices so we will do as Charles Lasner suggested and swap a TU55 and TU56 between the PDP-12 and the PDP-8/I. This will let us test the TU56 with a known good TC01 LINCtape controller, and test a known good TU55 with a questionable TC12 LINCtape controller. We ran the A-to-D converter test and were rewarded with a display on the VR14 that showed correct operation of the knobs and A-to-D converters. -- Michael Thompson

7 years, 11 months

Calcomp 1039 plotter docs?

by simski＠dds.nl

Hi All, we have a 1039 in our space with the user guide, but without any service docs. Our specimen does not react to buttons except the reset and test buttons. the four statusleds light up on a reset and after a second the center two leds start blinking in sequence. paper and pens are loaded as per the user guide. anyone with documentation on this thing? -- Met vriendelijke Groet, Simon Claessen drukknop.nl

7 years, 11 months

Looking for RX01/02 rack slides

by charlesmorris800＠centurytel.net

Thinking I had an easy solution to the missing outer slide, I ordered a $40 pair of new heavy-duty rack slides from eBay. Beautiful pieces and ball-bearing slides, decent quality hardware too. Only had to drill a couple of holes in them so they'd match the drive chassis, mounted everything up, and then discovered that they were 1/4" too narrow, no matter how I arranged the brackets, and the drive wouldn't slide between the rack rails. Crap! So I bought two 24" pieces of 2 x 2 x 1/8" angle iron at the local steel place for a whopping $1.40, drilled four holes in each, sanded off the scale/surface rust, bolted them to the rack and slid the drive right in. You can't even see the "homemade" part unless you're really looking, too. Should've done that to start with. If I ever need to service the drive, I'll just slide it back out and set it on the bench... http://s1181.photobucket.com/user/DrCharlesMorris/media/PDP-8/P09-29-15_19.… http://s1181.photobucket.com/user/DrCharlesMorris/media/PDP-8/P09-29-15_19.… However, if anyone should stumble over a set of stock RX sliding rails I'd be interested for the sake of originality ;) Jay, let me know exactly what measurements you would need. There's a pretty good view of the inside rail on the second pic. thanks Charles

7 years, 11 months

A new Lisp-based OS that hearkens back to the old days of comprehensible computers

by jnc＠mercury.lcs.mit.edu

> From: Liam Proven I am _very much_ in sympathy with the complaints here; I too feel that modern computers are too complex, etc. (Although some of it, like the entire computer turning into a single chip, were/are inevitable/unavoidable.) I like the functionality of modern system, but I feel they are _more complex than they need to be_ to generate that level of functionality. However, one thing I am going to quibble with: > This is a nice explanatory quote: > The main reasons TempleOS is simple and beautiful are because it's > ring-0-only .. Linux wants to be a secure, multi-user mainframe. ... > It was simple, open and hackable. It was not networked. ... It was ^^^^^^^^^^^^^ > simple and unsecure. If you don't have malware and you don't have > bugs, protection just slows things down and makes the code complicated. Note the part I highlighted. If you want to have a system that's network-capable, which is pretty much mandatory for a _really_ usable system in this day and age, i) that means Web-capable, and ii) if it's Web-capable today, it has to be able to handle what I dub 'active content' (JavaScript, etc) - i.e. content coming off the network which contains code, which runs in the local machine. To paraphrase a certain well-known SF work, IMO active content is probably the worst idea since humans' fore-fathers crawled out of the mud. It's _potentially_ a giant, gaping security hole - one that in today's OS's is responsible for a huge share of security issues. (There _is_ a way to have systems which aren't as vulnerable, but it means having military-grade security on everyone's machine - and no, I don't mean crypto; probably not likely, alas.) I mourn the early days of the Web, when there was no active content - just text, images, etc, etc. But no, they had to add all sorts of flashy eye candy - and did so in a way that makes basically all modern machines horribly insecure. But let me dispense with the soap box... Anyway, the inevitable consequence is that if you want a networked machine, it's _not_ going to be simple. Alas. You're basically sharing the machine with _lots_ of other people - effectively, every Tom, Dick and Jane out there in the Internet. In other words, you need everything one normally saw/sees in a time-sharing machine. (And I'm not talking about wimpy ones like Unix/Linux. I mean industrial strength ones like Multics.) Noel

7 years, 11 months

Thoughts on manual database design?

by mark＠markesystems.com

>> Not sure why you have VARCHARs for primary keys, why not use the >> conventional auto-increment int so you can dispense with >> the LastGeneratedArtifactID table. >> > Because my artifact ID's are not always just numbers. In some cases > they may already be marked on an artifact (though typically not for > manuals - but this is just the first of a set of such projects, and they > *are* marked on many of my computer boards). You can still force the artifact ID VARCHARS to be unique, and index them as well, of course. There are at least a couple of reasons to have primary keys that are independent of the "visible" key. First, if the user-visible key ever changes (what if the inventory tag falls off and is lost?), that will break all the links that refer to that record (or else you'll need extra code to handle this). Also, there have been many times when some aspect of a key that's directly tied to an external bit of information needs to change format - numeric to character, or length change, or... This too will break things. > No, I don't need made up primary keys. The other tables have the keys > they need to guarantee uniqueness - in some cases the PK is made of up > two or more columns. I seriously dislike the current fad of inventing > such keys when they are not needed. I too used to develop new databases this way, figuring that since a certain bit of information is guaranteed to be unique (or that I want to guarantee its uniqueness), I'd use that for the primary key. After getting bitten more times than not, I now almost always create an auto-number key whose only purpose may be for internal linkage. (I recently developed a project using Zoho Creator, which was a learning experience to say the least. It's worth noting that an explicit ID field is part of *every* data table that you can create there - there's no way around it. And, it turned out that it was pretty darn helpful a lot of the time, too.) >> Another thing, although MySQL is fine but for this I think SQLite might >> be a better choice of db. Its access methods are all in-process ie. no >> external >> dbms service to bother with, just a library to link in and the physical >> database is a disk file (.s3db extension). It has a much 'lighter' db >> footprint. >> > As I mentioned in another response, I truly dislike SQLite, based on my > experience with it on my Garmin GPS. I'm still not sure why - my experience has been very good. What bad experiences have you had? ~~ Mark Moulding

7 years, 11 months

Interlisp-D (Medley) "display font" disks for the 1186?

by derschjo＠gmail.com

Hi all -- I've been working on getting a Xerox 1186 workstation up and running again, using the floppy images on Bitsavers. I have the "Medley" Interlisp-D software installed (after writing out and installing from ~25 floppies) and running and I'm attempting to load in the related libraries and software (another 10-20 floppies or so). Some of these libraries have dependencies on various font files, which I do not seem to have and haven't been able to track down. I see vague references in the documentation to a floppy disk set labeled "Display Fonts" but these do not appear to be on Bitsavers. From writing out a few floppies and looking at their contents on the 1186, I do not believe that these have any relationship to the Viewpoint Font disks (though if anyone knows differently, do let me know). Anyone out there have any experience with this? Anyone happen to have these floppies and/or images of them? Thanks as always, Josh

7 years, 11 months

Jump to page:

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 September 2015