At 03:01 PM 10/31/2002 -0500, Bob Shannon wrote:
Ok, the corosion of the PCB etches does not scare me,
I've got a full set
of spare boards. A few of them have bad chips, but in most cases these
have already been isolated to the failed component.
I've never seen the schematics for a long-vector Imlac, and I wonder how
much work would be envolved in adding this capability to a 'stock' PDS-1.
A partial set of PDS-1D schematics (and other documentation) are available
here:
http://www.spies.com/~aek/pdf/imlac/
I say partial, because they don't contain the pages for the long vector
option, which
appear on sheets 853 and 854 and comprise boards 301, 302, and 303 of the
PDS-1D.
Does anyone (Tom U.) know what method the long vector
hardware uses to
generate long vectors? I'd guess this is done with binary rate
multipliers in much the same way as the DEC VT-11.
Yes. I think that you are correct in assuming that the long vector control
is built around
a BRM. Without some extensive research and comparison between the PDS-1 and
PDS-1D
schematics, I cannot estimate how much additional hook logic is required on
other boards
in order to support the addition of the long vector option.
Just how much circuitry is on this long-vector board?
How different is
the vector processor circuitry on a PDS-1D versus a PDS-1?
And ae we sure that Mazewar needs long vectors?
As quoted from the Mazewar source:
; THIS VERSION REQUIRES A GRAPHICS IMLAC WITH LONG VECTOR HARDWARE
; MULTI-LEVEL SUBROUTINING, AND 8K DISPLAY ADDRESSING MOD.
I think that the 8K display addressing assumes an 8K machine, so Bob needs
to get the
other 4k of memory working on his machine before working on the 8K display
addressing
mod.
The comment also indicates that the Multi-nested Display Subroutine feature
is required.
This is the 8 Level DT register stack option, which if it were present in
the document on
Al's site would be on sheet 858 (board 304) for the PDS-1D. I have all of
these prints, but
when I made the copy for Al, I made it from an incomplete set which did not
include these
options. I will put together the missing pages and send them to Al for
scanning... Again,
without some research, I don't know the level of difficulty and support by
other boards to
allow the addition of the 8 Level DT stack option.
If these options were to be added to Bob's machine, there might be a
question of space
in which to plug the additional hardware. It is less of a question of the
density of the logic
and more one of connector pin count. Also, I'm not sure that Bob want's to
modify his
machine to this extent.
It should be noted that Sellam has both a PDS-1 and a PDS-1D in his
collection (or at
least I think that this is the case -- further pictures of his two machines
would confirm
or deny this).
--tom
Sellam Ismail wrote:
On Wed, 30 Oct 2002, Bob Shannon wrote:
I'd be really interested in learning more
about this!
Here is a page about the Imlac I have (it was given to me by Doug Salot,
who developed this page):
http://www.blinkenlights.com/classiccmp/imlac/
It doesn't say anything about the long-vector option, although I could've
swore he did (maybe it's there and I just didn't see it in my quick
refresh perusal). You can see the board in the upper-right corner of this
picture (or maybe it's the middle right?):
http://www.blinkenlights.com/classiccmp/imlac/imlac6.jpg
Also, I've got lots of parts to get your PDS-1
in running order.
Cool!
Sellam Ismail Vintage Computer
Festival
------------------------------------------------------------------------------
International Man of Intrigue and
Danger
http://www.vintage.org
* Old computing resources for business and academia at
www.VintageTech.com *