test-drb October 1999

test-drb@ccmp.vtda.org

216 participants
748 discussions

by CLASSICCMP＠trailing-edge.com

In the past few months, I've been heavily involved with upgrading some CAT scanners that date from the late 70's. I've mostly been involved with the PDP-11 side, but bolted onto the PDP-11 (or, actually, the other way around) there's a custom piece of hardware (two dozen hex-height cards) that does all the heavy number crunching. Backprojection is what the process is called, and it's basically a deconvolution of the data measured by the individual detectors to make the pretty picture of your insides that a CAT scanner displays. The power and speed of this classic hardware is quite impressive, considering it's nearly 20 years old now. We've been timing other approaches to backprojection, and if we implement the same algorithm that the custom hardware is doing but on a modern 533 MHz Pentium III, we discover that the custom hardware is twice as fast as the Pentium. There are some other competing technologies, making use of DSP chips, which promise to be faster, but at a substantial development cost. Is this a fair comparison? Maybe not! The Pentium system costs roughly $2000, while the custom backprojection hardware was several hundred thousand dollars back in the late 70's. The PC boards that make up the backprojector make use of a unique technology: the PC boards themselves are just plain old two-layer boards, with the actual PC traces carrying only power and ground. All the actual wiring is done by fine machine-laid wiring between pads, with all the wire layers "potted" in a plastic compound. Obviously, reworking such a PC board is extremely difficult, but it gives impressive component densities for the time. The modern equivalent of this is multi-layer PC boards, naturally. Of course, there's another sort of power involved too: the custom backprojection hardware requires +5V at 350 Amps, and some installations have multiple backprojectors! -- Tim Shoppa Email: shoppa(a)trailing-edge.com Trailing Edge Technology WWW: http://www.trailing-edge.com/ 7328 Bradley Blvd Voice: 301-767-5917 Bethesda, MD, USA 20817 Fax: 301-767-5927

25 years, 2 months

Floppy disks again

by allisonp＠world.std.com

<> I have em. Both 10 sector NS* format (no even for $20 I dont give them <> up) and also some of my TI99/4a likely others that I'd have bought aroun < <Wow! What brand of diskettes were they? All the NS diskettes I've seen <have the standard windows, but I think they were just 3M and Verbatim <diskettes. Actually a mix of NS* hard sector and some soft sector, Shugart, dysan, Verbatum, memorex and 3M, looking at them. It's a standards thing. At the time 1978 35 track was the standard, 40 wasn't far behind but it was different in that the media had to have a longer opening to work right. Allison

25 years, 2 months

Classic power!

by allisonp＠world.std.com

<This was known by the trademark "Multiwire", and was often used for <quick-turn production. DEC used it on a few boards in the KL10-PV <processor; later revisions of the same boards were conventional PCBs. that was it. It's had two characteristics, high density and lines could cross (it was insulated). Since the "inner layers" were power and ground they were easy to modify (wires could be seen and removed or cut) and the transmission line characteristics were better than the then vogue 2 layer boards. All the lead holes were plated through as were any vias. Allison

25 years, 2 months

Ahhh!

by mbg＠world.std.com

I grabbed a copy of the picture... a little dark... I had to do some adjustments to it... What I see are: o Two H960 tall cabs o One cab has 4 RL series drives in it (pretty obvious) o The other has what appears to be a pdp-8/e/f and two pdp-8/A's, both with the programmer console. Megan Gentry Former RT-11 Developer +--------------------------------+-------------------------------------+ | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | | Unix Support Engineering Group | (home): mbg!world.std.com | | Compaq Computer Corporation | addresses need '@' in place of '!' | | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | | Nashua, NH 03062 | "pdp-11 programmer - some assembler | | (603) 884 1055 | required." - mbg | +--------------------------------+-------------------------------------+

25 years, 2 months

Heathkit H-11 restoration questions

by ethan_dicks＠yahoo.com

--- allisonp(a)world.std.com wrote: > > The fact that the machine hangs solid when the card is where it's > > supposed to be may be a clue that the it's permanently stuck interrupting. > > Depending on backplane the grant may not end-up where you guess it is. I was assuming a sinusoidal AB-AB chain starting from the slots labelled "A" and "B" (out of A, B, C, and D). Now... the one twist is that the cards are installed with the CPU facing the bottom of the cage (solder side up), but the keys on the backplane connectors are a dead giveaway. > My memory is fuzzy on this but I think Ht-11 (heath os) did not like RTC > <line event> interrupt and would hang on that. Hmm... I don't have any HT-11 disks, only RT-11. But the LTC *was* enabled. > Yes, also DLV11 or the heath serial card both should have HALT on BREAK > enabled. <both support it as do the 11/03> I didn't even try that. For now, because of bootstrap considerations, I'm running with an 11/23plus CPU and a VT-220 (because it's close). Speaking of bootstraps, what did the H-11 come with stock? BDV-11? Custom Heath board? -ethan ===== Infinet has been sold. The domain is going away in February. Please send all replies to erd(a)iname.com __________________________________________________ Do You Yahoo!? Bid and sell for free at http://auctions.yahoo.com

25 years, 2 months

A GT40 thought

by ard＠p850ug1.demon.co.uk

A non-text attachment was scrubbed... Name: not available Type: text Size: 1306 bytes Desc: not available Url : http://www.classiccmp.org/pipermail/cctalk/attachments/19991004/6a849db1/at…

25 years, 2 months

Heathkit H-11 restoration questions

by mbg＠world.std.com

>Yes it was or in a later time frame m8186 (kdf11, 11/23) with m8044s. >that workes better than 11/23+ (m8189) as it was a dual solution and the >hath backplane was rather cramped. Mine actually has an 11/53+ board in it... :-) I do have the H27 board and drives and a heath serial line board, though I use a DLV11-J and and -E in it. I also have an RQDX3 connected to a table-top RD54 nowadays... Megan Gentry Former RT-11 Developer +--------------------------------+-------------------------------------+ | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | | Unix Support Engineering Group | (home): mbg!world.std.com | | Compaq Computer Corporation | addresses need '@' in place of '!' | | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | | Nashua, NH 03062 | "pdp-11 programmer - some assembler | | (603) 884 1055 | required." - mbg | +--------------------------------+-------------------------------------+

25 years, 2 months

DEC GT40 original list price?

by mbg＠world.std.com

>By the way, I am curious where you got your lander program... I ask >because a few years back, I contacted Jack (via a pointer from the >Computer Museum in Boston), and what he had was a 9 -Track tape that he >had gotten from someone who had preserved a copy while he was at the >University of Wisconsin-Milwaukee. He sent me the tape, and I read it >and returned the contents to him, including the lander program, some demo >programs, etc. The version I have is the one I obtained back in 1975 or so... one that Jack and I worked on when I was a student at WPI and he was visiting the college for some reason... The ME department had a GT44 (11/40 based system with a VT11 and VR14) and we did some work on it to fix some intensity problems with the thrust and directional control bars as well as getting it to work with a programmable clock. Megan Gentry Former RT-11 Developer +--------------------------------+-------------------------------------+ | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | | Unix Support Engineering Group | (home): mbg!world.std.com | | Compaq Computer Corporation | addresses need '@' in place of '!' | | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | | Nashua, NH 03062 | "pdp-11 programmer - some assembler | | (603) 884 1055 | required." - mbg | +--------------------------------+-------------------------------------+

25 years, 2 months

Any good dumpsters or surplus stores in Philly?

by bobstek＠ix.netcom.com

I will be in Philadelphia all day Friday with nothing but time on my hands. Can anyone recommend any good hunting grounds for 1975 - 1985 micro's or micro magazines? Bob Stek bobstek(a)ix.netcom.com Saver of Lost SOLs

25 years, 2 months

Old Apple II, Commodore Amiga & Vic 20 Computers available

by dheap＠metz.une.edu.au

The above may be of some interest to list subscribers in Australia: ------------------------------------ The School of Psychology has some old computer gear which is of no commercial value. The items include: Apple II with various interface cards, MicronEye video camera, printer Commodore Amiga computers and accessories Commodore Vic 20 Computers and accessories These items may or may not be in working order, but may be of interest to someone who collects ancient computers. I'd hate to have to throw them in the big green bin. If interested please contact me, or one of the Psychology Technical staff. -- Dave Heap mailto:dheap@metz.une.edu.au School of Psychology, <= postal address University of New England, phone: + 61 2 6773 2587 fax: + 61 2 6773 3820 Armidale NSW 2351, Australia http://metz.une.edu.au/~dheap/

25 years, 2 months

← Newer
1
...
65
66
67
68
69
70
71
...
75
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 October 1999