Well, I dug in to the prints for the RL02 and wrote up notes for myself on
all the signals feeding the fault lamp and exactly what pins of what chips
they were on (E1 and E2 of the drive logic board). Went downstairs and
pushed the BA11-K back into the rack (I had it out to check the ACLO & DCLO
signals). I pulled out the RL02 and lifted the logic board and got my notes
ready. Instead of a logic probe I decided to use a logic clip. Easier &
quicker in this case because I knew just what pins to look at on which
chips, and all the signals of interest were on two adjacent chips.
I powered up everything including the RX02. When I hit load, the drive spun
up and came ready. No fault at all. The RL02 and RX02 both boot & work fine.
I've never been quite so disappointed that a system worked, because I know I
did nothing to fix it. I tried to recreate the problem all afternoon - with
the machine cold, or with it running the instruction set excerciser for
hours, no hiccups. The odd part is, the "RL02 faulting upon spinup IFF the
RX02 is powered on" WAS completely reproducable, and it never once failed to
fail. It failed absolutely 100% of the time for the past few weeks. Not only
if it was on when the RL02 spinup was attempted, but if the RX02 was off and
the RL02 spun up successfully - the moment you would apply power to the RX02
the RL02 would immediately fault. Now I can't get it to fail, and all I did
between times was extend the BA11-K, hook up a probe and look at a few test
points, then put it back.
Perhaps it is cables. I'll be trying again tomorrow to reproduce the problem
and hope it reoccurs :\
Jay
I was given an Intellec 4, but unfortunately no documentation.
Also included was a smaller PCB intel 05-0436-000 which also includes
American Optical P/N 11761-450. The board is about 4.5 x 5" with a 44
pin edge connector at one end, a 40 pin socket connector, and a 50 pin
IDC header at the other end. It includes the usual 4004 chipset (4004,
4201, 4001x4 and 4002x2) with date codes from '74-'78
I'd like to get the Intellec 4 back to operation, anyone have
documentation they could share? Also if anyone knows anything about the
American Optical board I'd appreciate information.
I'm not looking to sell or Ebay the unit - any help would be so that I
could enjoy the system myself, and document it for the future. It's got
a s/n below 20 - just curious if anyone knows how many Intellec 4's were
actually produced. I had an Intellec 4-40 back in the mid 1980's that I
saved from a scrapyard and donated, but that's that last time I have
seen one personally.
Many thanks for any help!
- Gary
Jay,
----- Original Message ----
I distinctly remember two of the projects. One was on the
working of the heart. It was a board that tilted front to back on a
triangular wedge. Routed into the board was a logical diagram of the human
heart. The valves were one way metal flaps that opened by rotating on a thin
metal shaft or wire. The blood was a set of marbles. When you tilted the
board to the back, the marbles rolled through the valves in one direction,
then when you tilted it back the other way the flaps only let the marbles go
into the correct chambers.
---
I remember that!!! I wanted to build it, but for any number of reasons didn't. For a long time now I remember associating the marbly heart with my ESR Dr. Nim toy, which used marbles to play the game of Nim.
What I don't remember is which I came across first- Dr. Nim or the Marbly article.
The idea of using marbles, tilt boards, gates, has been tucked away in my brain to some day emulate **something**.
Scott
____________________________________________________________________________________
Be a PS3 game guru.
Get your game face on with the latest PS3 news and previews at Yahoo! Games.
http://videogames.yahoo.com/platform?platform=120121
I hope you don't mind me asking as this is not a tech quiestion per
se, but if anyone in the world will know the answer it will be this
group.
I'm looking for a book that I rented from a library in 1973 or 1974
on computers. In the back of the book it has plans for building a
computer out of basic components, including a telephone dial for
input. The output is shown on lights and I remember there being more
than a few bulbs. The "computer" probably does nothing more than add
numbers. I use the book to research and write my 8th grade science
paper and then I tried to build the computer for my 8th grade project
but could not complete it at the time. Now that I'm older and wiser I
figured I'd try again.... (mid life crisis?)
I don't remember any ICs. It was pretty simple electrical components.
Anyone know the title or author of this book?
Thanks.
Paul
Anyone know if NeXT monochrome slabs have dual SCSI buses (one internal and
one external) or not? I've got one here that resolutely refuses to talk to any
internal SCSI device [1], but will happily talk to (and boot from, if
necessary) external devices. In fact, if anything is plugged into the internal
bus at all then the system won't even boot from an external device.
I've ruled out internal cabling, internal power connector, something shorting
the PCB underneath, and an actual device fault - which means it's possibly a
bad solder joint somewhere or a cracked PCB track. However, *if* the slabs
have two separate SCSI buses then it's possible that there's a fault with the
actual controller too which is knocking out the internal bus.
[1] Actually it "talks" in that there seems to be some device activity, but
always falls over with a (useful!) "SCSI error" message whenever there's an
internal device present.
I'm not too fussed as the colour slab works and this one works too (albeit
hanging the drive off the external connector), plus we've already got a good
mono slab at the museum. It'd just be nice to get to the bottom of what's
actually broken!
cheers
Jules
On 22 Jan 2007 at 16:49, Fred Cisin wrote:
> The AT started in 1984. Some areas (both geographic, and social)
> immediately went for it, and some put it off as long as they could, since
> the IBM/MICROS~1 software didn't provide any real incentive to upgrade
> other than high density drives and a little faster;
> until Windoze 3.1 and OS/2.
More than a little faster, at least to my recollection. Something
like 3 times as fast. 16 bit disk I/O and a CPU with nearly 4 times
the transistor count of the 8088. A lot of folks who bought the
original 6 MHz PC AT discovered overclocking.
While you could find 8 and 9 MHz 8088/8086 systems, neither came
close to a 6 MHz AT in terms of performance. And if you were
adventuresome and clocked that PC AT at 12 or (I've heard it was
done) 20 MHz, the gains were breathtaking--and you had a convenient
place to cook lunch.
One thing that IBM did that really toasted me back then was messing
up on the 8237 DMA controller hookup such that memory-to-memory DMA
didn't work. It could have made the whole business of extended
memory use a lot simpler.
Cheers,
Chuck
Oops - After I sent that long message, I realized I was confusing the
DN10000 (custom bitslice CPU emulating a 68K) with the DN300 (68K desktop
workstation). My description applies to the DN300, not the DN10K...
Cheers,
jp
>Missing cards? They just might be empty, unused slots in the
>backplane.
That sounds like you've got the basic compliment. As I recall from my days
running a network of them, the base DN10K has four cards (Network, video,
CPU and memory). Extra slots were available for an FPU, more memory (and
maybe a disk controller?). The CPU should be easy to identify by the
presence of a MC68010 chip.
The DN10K's typically did not have their own disk drives; instead they
booted (and paged!) off of a "partner" server node having the disk. The
network adapters (in a plastic box that screwed onto the back) were
spectacularly flaky on those machines.
The DN10K's had a switching power supply with a subtle audible
hissing/buzzing noise corresponding closely to the machine's
activity. With experience, the power supply noise was almost as
informative as a blinking-light front panel. For example, it'd make a
distinctive rhythmic chirp when the network was interrupted and the paging
algorithm was frantically re-trying the operation to clear the page
fault. If not fixed ASAP, the screen would break apart with an ugly
"PARTNER NOT FOUND" message.
Between the flaky network adapters and the one-fails-they-all-die ring
network topology, keeping a DN10K network running was challenging.
Cheers,
jp
Today I started stripping out good parts from a Convex C240
minisupercomputer - a machine I had purchased from a scrap dealer
almost ten years ago. It was a disaster of a deal - one of the rare
times I feel I was screwed my a junkman. The machine turned out to be
missing many parts, including processors and power units, cables were
chopped, memory was swiped, plus there was mouse crap evident - yuck.
For years I have been threatening the part it out, and now is the
time.
A few parts will end up on Ebay, the metal will go to the scrapyard,
and I am keeping the boards that I have for now. I know Convex
machines are not very common, but if there are others that need spares
for 200 series machines*, please let me know, as the boards and
(hopefully) backplane will be for sale.
* A guy I know that worked at Convex said the processors should mostly
interchange - although a rev level problem might pop up.
--
Will
