test-drb August 2016

test-drb@ccmp.vtda.org

175 participants
281 discussions

by rodsmallwood52＠btinternet.com

Hi Guys I sent out a photo of the new PDP-8/i with a real one for comparison to everybody who might be interested. If you did not get the email and would like one please let me know Rod (Panelman) Smallwood

7 years, 11 months

Differences between the PDP-11/15 and PDP-11/20

by jnc＠mercury.lcs.mit.edu

> From: Paul Popelka > I was wondering how much delay the KT-11B introduces. That's a _very_ interesting question; AFAIK, the documentation doesn't say. If there's a cache miss, of course, there's one memory cycle delay to load it. If the cache hits, though, there's still added gate delays going through the KT11 - perhaps 20-30 or so (to make a complete guess) - at ~10nsec each, that would be an extra 200 nsec per memory cycle. Not insignificant... > There is still the actual implementation of the KA11 changes that can > be reverse engineered if someone is so inclined. Well, that might not be trivial - if the boards have ECOs, it may be hard to tell them from the KT11 changes. The sheet which gives the wiring changes for the KA11 backplane _is_ still there - although there are indications on it that the actual machine differs from the prints! Wheee! :-( Noel

7 years, 11 months

BEL constant in MAINDEC-8E-D0AB-PB

by pontus＠Update.UU.SE

Hi I've been running and looking at the PDP-8/E INSTRUCTION TEST 1. And it is supposed to write the BEL (07) character after each round of passed tests. However the code loads and outputs the constant "0207", which is 07 with the eigth bit set. I think the 8/E serial interface outputs all eight bits in this case? I suppose a real ASR-33 would ignore the eight bit? Why does the code not load 07? Thank you, Pontus.

7 years, 11 months

DEC Indicator Panels page

by jnc＠mercury.lcs.mit.edu

> Oooh, thanks ever so much for turning that up!!! OK, I have added them to the page - I lost a little resolution rotating them to be level, but there's still more than enough to recogize them, and mostly read them. > So that mystery panel seems to be a general panel, more associated with > the CPU than anything else; one line does seem to be the reader, but > one quadrant is the KT15 memory management, one is the KA15 priority > interrupt system, and there's general CPU/system stuff throughout With that, I think we have most of the PDP-15 panels (although the VT15 image is still pretty crummy). That leaves only these two: http://www.simulogics.com/nostalgia/DEC/15_05.jpg although that looks like early marketing material, so perhaps those panels neve made it into production machines? I wonder if one of them is a BD15 (whatever that might be) - or if that's the name for the CPU panel (above)? Noel

7 years, 11 months

Differences between the PDP-11/15 and PDP-11/20

by paulp1＠gmail.com

> From: Noel C. > Was the Computer History Wiki thing of any use? Yes. It answered a little of what I was wondering about. It mentioned that there is a cache of page table entries. I was wondering how much delay the KT-11B introduces. > > Does anyone know if schematics for it were included in the auction? > > If the schematics are available, are there plans to get onto bitsavers? > The answers apparently are yes, and yes - although there are two parts to the prints (the KT11-B itself, and the KA11 changes), and it seems that unfortunately there's (at least) one page missing from the 'KA11 changes' part. Oh well. There is still the actual implementation of the KA11 changes that can be reverse engineered if someone is so inclined. I'm just glad the winner of the auction is willing to make available the information they did get.

7 years, 11 months

X server for original PC (8088/8086)

by seefriek＠gmail.com

From: tony duell <ard at p850ug1.demon.co.uk> > On Fri, Aug 19, 2016 at 5:56 PM, Steven M Jones <classiccmp at crash.com> wrote: >> >> Before anyone gets too excited about the blistering speed of the 60 >> MHz TMS34010, ... However, since it has a >> graphics-optimized instruction set, it was still able to do some >> things noticeably faster than the 16 MHz 80186 would have. Price/performance for the tms34010 was terrible; it was somewhat faster at graphics (bit-oriented) ops than an 80186 (or other contemporary processor), but it was several times more expensive. TI tried to sell around that by claiming it was a complete general purpose processor in addition to graphics processor so you could build a whole system using the tms34010 as the brains. Unfortunately, if you actually did that, you found that it could manage kbd/mouse/net *or* do graphics, but not really both. It was also integer-only, and had a slow, 16-bit memory interface that killed performance unless you used expensive VRAMs (this was before VGA made VRAM cheap). And TIGA never really took off. Bonus: the development tools were pretty awful. One of the weirder C compilers I've used. Intel came up with the i82786 around the same time that was cheaper, and it looked like you could cook up a cheap 80186+82786 X Term setup that would be competitive. However, I never saw a product like that, just a couple of PC/AT plugin cards (Belltech BLIT). > Somewhere I have a thing badged 'Princeton Ultra-X'...It uses an 80188 for I/O > (including 10Mbps ethernet). The Xserver is in EPROMs and appears to run on > the TMS34010 graphics processor. Yup...I worked for the company that designed those. Good times (really...I learned a *lot* about a lot of things), but glad I was an ops guy and not an engineer or developer. There were probably 20 other shops making X Terminals at the same time, 'cause that was the future. I recall having stacks to play with because the market for them evaporated far sooner than marketing predicted. KJ

7 years, 11 months

cctalk Digest, Vol 26, Issue 20

by seefriek＠gmail.com

From: Zane Healy <healyzh at aracnet.com> > > I tend to think that X11 over serial would be nothing short of nightmarish. > I ran X over an ISDN 64K link with 19.2Kbps backup for years in the late 80s/early 90s, and it was pleasantly usable enough to be my daily remote work access. Admittedly, Framemaker was a bit of a pig over it, and we weren't web browsing pages filled with 500KB of JS and images. KJ

7 years, 11 months

What is the KH11-A? (was Re: Differences between the PDP-11/15 and PDP-11/20)

by ethan.dicks＠gmail.com

On Sun, Aug 21, 2016 at 1:53 PM, Noel Chiappa <jnc at mercury.lcs.mit.edu> wrote: > http://gunkies.org/wiki/KA11_CPU Interesting stuff on the KT11, but tangentally, I've been going over my pile of parts (that I've described a few times over the years) from an 11/20 I pulled from the dumpster at work 30 years ago. I finally got around to photographing the module handles and inventorying many of the boards and I've learned a thing or two and have generated a question or two. First, the manufacture date for most of the parts is in 1972. The front panel says "11/20", so that all jibes. I appear to have a CPU (more below), an M7800 DL-11 console interface (not the older multi-board KL-11), a lot of MM-11 4K core sets, and a happy surprise (again, more below) and not much else from that extraction (no bootstrap card, for example, or peripheral adapters). Second, many/most of the CPU modules are -YA variants, and the "PDP-11 Field Guide" mentions that they would be part of a KH11-A. About the only description I can find for what that is, is a single mention (in PDP-11_PeripheralsHbk_1973.pdf) of the "large-system capability option" related to the DT03-F Unibus switch requirements. Anyone know what this is? I'm reasonably certain it doesn't matter for my needs, but I'm curious. Third, the happy surprise is a full set of boards for a KE11-A. I have been sniffing around for a KE11-B (single-board version), and now, I don't need to (as long as this one works or can be repaired). This excellent for me since my goal all along has been to fire up enough hardware for Warren Toomey's reconstruction of ~1972 UNIX (v1 kernal plus v2 utilities). Yes, I know I can run that on SIMH (I have), even wrapped up in a Docker (I will), but I've wanted to run it on the real hardware for 11 years and I'm one step closer. I have a RK11-D that works, and an RK11-C that will likely require much attention (never tried to use it), so the only thing I'm really missing (besides enough time to repair/debug all of this) is something to emulate an RF11 for the swap disk. I'm now doubly sad that I lost my Diablo 30/RK03 in a flood 25 years ago, but I'll be functional with an RK05 or two. Fourth - I don't think I got the H720 power supplies with this (or the fans) - that was what I think was removed prior to this unit hitting the dumpster. I may have the G772 power cards (and then again, I may not), but those wouldn't be hard to reproduce. I am entirely satisfied with modernish switching supplies, but I see the H720 puts out about 15A of +5V and -15V, plus -22V for the core and +8V for the front panel bulbs. The bulb voltage is probably no big deal to come up with, but -22V is a bit of an oddball. I'm also willing to forego 100% core in favor of MOS memory, including any modern repros that come out in the next year or two. To start, finding even one H720 to run the CPU box (I have 2 more boxes for the rest of the memory) would be enough to get going. So... for now, as I continue to pull the parts together, my real question is, what is a KH11-A? Thanks for any enlightenment. -ethan

7 years, 11 months

What is the KH11-A? (was Re: Differences between the PDP-11/15 and PDP-11/20)

by jnc＠mercury.lcs.mit.edu

> I recall reading _somewhere_ about some early PDP-11 memory management > thing used on early PDP-11 Unix that supported the KE11 ... by having a > small window that allowed user code access to the KE11. ... > Does this ring any bells for anyone? Never mind - found it, it was in "Odd Comments and Strange Doings in Unix": https://www.bell-labs.com/usr/dmr/www/odd.html The memory management thing in question was the KS11; I looked, but could not find anything about it. Does anyone know of anything? Noel

7 years, 11 months

What is the KH11-A? (was Re: Differences between the PDP-11/15 and PDP-11/20)

by jnc＠mercury.lcs.mit.edu

> From: Ethan Dicks > Second, many/most of the CPU modules are -YA variants, and the "PDP-11 > Field Guide" mentions that they would be part of a KH11-A. About the > only description I can find for what that is, is a single mention .. > of the "large-system capability option" related to the DT03-F Unibus > switch requirements. Anyone know what this is? I've been told that it's a revision to the CPU timings to reduce NPR latency, although I have no details; I had previously looked online, and was not able to locate anything about it. > a full set of boards for a KE11-A. I have been sniffing around for a > KE11-B (single-board version), and now, I don't need to (as long as > this one works or can be repaired). Doesn't it need a custom backplane too? Do you have that? > my goal all along has been to fire up enough hardware for Warren > Toomey's reconstruction of ~1972 UNIX Maybe someone here can scratch a mental itch for me; I recall reading _somewhere_ about some early PDP-11 memory management thing used on early PDP-11 Unix that supported the KE11 (which is a memory-mapped device) by having a small window that allowed user code access to the KE11. IIRC, it was a DEC thing, not something home-rolled at Bell. Does this ring any bells for anyone? Noel

7 years, 11 months

← Newer
1
...
5
6
7
8
9
10
11
...
29
Older →

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 August 2016