Sorry can't help with the 11/70.
On a similar thread, for those interested in HP stuff - I did find a
schematic for an
HP2116 32-bit floating point CORDIC co-processor. It's all in US patent
3766370
dated 1973. Patent contains complete schematic, flow charts and microcode.
Uses some logic family I've never heard of, SL159XX series ?
Chris
In a message dated 12/10/2001 5:22:47 PM Eastern Standard Time,
cdrice(a)pobox.com writes:
<< Among them are several Apollo computers, and stack of misc. hardware
(token
ring parts and cables, misc. video and i/o cables, etc). and a very large
stack of documentation. >>
wish I had the space; I'm within 25 miles of the computers.
As previously posted, the H89 which Joe Rigdon kindly gave me has a problem
with the keyboard input: sometimes it works properly, but sometimes double
characters or wrong characters are produced. This behavior can change
while using the machine -- one moment it's okay, the next moment I get
wrong or double characters.
Anyone have a set of docs they can reproduce for me? Or any experience
with the keyboard controller circuitry on this machine?
TIA --
Glen
0/0
> Let's say a DEC PDP-1 comes up for auction. The economy is
> good and times are stable.
>
> What would you be willing to pay for a DEC PDP-1?
>
> This is strictly a hypothetical.
I'd be willing to pay everything I've got, which would
*not* be enough to be the winning bidder.
So, the real question should be, what would an auction
of a PDP-1 top out at?
-dq
> -----Original Message-----
> From: Don Maslin [mailto:donm@cts.com]
> > It also has (...and this is the reason I rescued it, even
> though it is an
> > intel machine) a really odd graphics adaptor with a 15-pin (two row
> > D-shaped) connector that drives a full-page monochrome
> (white) EGA monitor.
> Sounds Mac like.
You're not kidding. The installation of MS-DOS on the disk resides in the
"SYSTEM" directory.
That aside, I get the impression they used GEM desktop (there are small
components stuck in with the installation of ventura publisher -- either it
came with ventura or was part of the system...)
There is also a more "standard" monitor plug next to the strange one (on the
same card)...
> Thanks, Chris, I appreciate the info. Sorry I cannot help you.
Thanks for the attempt. If they made an 8000, maybe I should keep an eye
out.
Regards,
Chris
Christopher Smith, Perl Developer
Amdocs - Champaign, IL
/usr/bin/perl -e '
print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");
'
Hi everyone;
It's been a while since I last coded stuff for robot control. The last
time I did that, the DSP56K (dsp56001 in particular) chips were
readily available. Now I need to develop a new platform for
research purposes and I am finding that the once ubiquitous dsp56k
chips have been eol'd (end-of-line'd in motorola parlance).
The suggested replacements are the dsp56301 or 303 chips, which
are code compatible, have 24 bit addressing (instead of 16 bit),
run at 80MIPs instead of 20, and have all the glue logic
for PCI or ISA interfacing built-in (dsp56301, which makes sense in my
application because the robot supervisor will be a Linux server
receiving commands from robot application clients over tcp/ip;
the dsp will be a slave that does the low level yet massive
number crunching stuff). So why don't I just go and choose
the newer parts? Several reasons:
1) I'd prefer to go with unix-based free software tools. There is
a56k and gcc56k for the dsp56k. While the dsp56.3k is supposedly
code-compatible, I am sure that some tweaking would be required to
make these tools work with the new family, and compilers are not
my area of expertise.
2) The older parts have roughly 100 pins and can be wirewrapped, the
new parts have 192 pins and require modern pcb design and production
techniques, which are outrageously expensive in this corner of the
world. I am not comfortable wire-wrapping a design that runs at
80MHz instead of 20.
3) From experience, I know that 20MIPS is enough for the task at hand.
So, I am faced with a familiar problem; an old part that will do the
job in a simpler design is no longer (readily) available; there is
a new, better, faster part, but there aren't as many software goodies
to go with it, and the hardware design tools and fabrication are more
expensive. The key issue is the present availability of the older
parts. I've checked some places and they seem to actually be out
of these chips. So my question to the list is: do you know where
to get dsp56001A chips? Or better yet, do you have some that you don't
plan to use?
carlos.
--------------------------------------------------------------
Carlos E. Murillo-Sanchez carlos_murillo(a)nospammers.ieee.org
Carlos
Sorry I got carried away. If all you need is a slave processor and 20Mips,
then the 56303-EVM is ideal. You get 64k x 8 flash, 32k x 24 fast SRAM,
a 16-bit stereo codec and a pin header for the host port. Also included is a
56002 which acts as an RS232 to JTAG interface for Domain Technologies
Debug-56k debugger, also free from Domain Technologies, www.domaintec.com.
Chris
> -----Original Message-----
> From: Don Maslin [mailto:donm@cts.com]
> Are you sure that it is not a CPT 8000? That one I am a bit familiar
> with and find ample references by a google search. On 9000, I draw a
> blank.
Well, it says 9000 on the front, I believe. :)
> If it is truly 9000, what floppy disk size does it use?
It was shipped, I think, with both 3.5 and 5.25" high-density floppy drives.
I just found another 3.5" floppy that works in it on Saturday night.
I got no manuals, no software (other than what was extremely messed up and
left on the drive), and a disassembled machine.
The machine is now in more-or-less good shape. I need to pick the tumbler
keyboard lock. For now I've just disconnected it. (Or find a key that
works) I should also replace the power adaptor I've got plugged into the
floppy drive, since it's also got a signal cable adaptor that's not being
used. :)
As I was saying, the installation was pretty botched by the time I got it.
It boots -- thinks it runs MS-DOS 3.2 -- and that's really about it. I
believe the CPU was intel 286. It has a "Tall Tree Systems" JRAM card and
JLaser 3 (I think) daughterboard. a 20MB MFM (or RLL?) 3.5" half-height
hard disk. (Miniscribe, I believe)
It also has (...and this is the reason I rescued it, even though it is an
intel machine) a really odd graphics adaptor with a 15-pin (two row
D-shaped) connector that drives a full-page monochrome (white) EGA monitor.
That's pretty much all I know about it.
Regards,
Chris
Christopher Smith, Perl Developer
Amdocs - Champaign, IL
/usr/bin/perl -e '
print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");
'
Back in September I wrote:
>I've got a lead on a full and working electron microscope.
>It's on-topic because it's old and has some sort of
>computerized digitizing unit. :-)
>I've lined up a big Diesel stake (flat-bed) truck.
>It'll be about a 2-3 hour drive. We've got straps,
>boards, tarps. I'm most concerned about getting it
>*off* the truck and into my office or home basement.
>Any other thoughts from the group?
An update: it's in my garage now. The x-ray spectrometer
is a 11/23-based system running custom apps over RT-11.
The SEM has an 8085-based system for automating various
aspects of the electron microscope column. It's an
AMRAY 1610T, circa 1983:
http://www.threedee.com/jmosn/microscopes/amray/index.html
Even at 1,200 pounds, the column rolled easily on a
pallet jack, off the loading dock ramp and onto the
truck. To get it all off the truck, the farmer next-door
helped with his power-tilt front-loader bucket.
The next trick will be to levitate it down the stairs
and into the basement.
If I can get it all back together again and happy,
it'll magnify down to about 50,000 x.
- John
