Basically, my question is what are the major obstacles (besides the breadth
of the project) to implementing the VAX in an FPGA, as compared to any
other ISA?
Is anyone aware of the existance of a maintanance / service manual for
the old TI "portable" data terminal, model 725?
Note that is is one of the very early (and very large) thermal printing
terminals from TI,
http://terminals.classiccmp.org/wiki/index.php/File:Silent700_725ad.jpg
I have several here that I am trying to keep running but have never
had technical information that would include schematics or other
parts information.
Chris
--
Chris Elmquist
Am 19.07.2014 23:08, schrieb Enrico Lazzerini:
> I managed to find a bios for an IBM 5170 AT 286 Type1 Mainboard. It is
> here: 30aprile1989 (list here:
> http://ibm-pc.org/firmware/ibm/5170/5170.htm)
> In practice I have manually added the number of cylinders, heads, and sect
> / trk that allow the machine to recognise a disk IBM 250MB. At this point
> regularly boots DOS.
> But I wish to program these data permanently in the bios. Does anyone know
> what program to use and give me the procedure? I tried with BIOSUTL, but
> the new EPROM do not boot the machine.
> With BIOSUTL i made what follow: read actual BIOS, you can add new disk
> geometry parameters at free 47 position, then you have to recalculate bios
> checsum, then BIOSUTIL devides BIOS into EVEN and ODD file so i can finally
> program them into two 27256 150nS eprom.
> Thank you
> Enrico
>
>
Hi..
the biosutil maybe one way. The version I use told me:
Biosutil V1.1
InfoMatrix Bios utilities for the AT BIOS.
Brad Gibson Copyright (C) 1990 by Secret Software
and I believe there's no other version.
*----------------------------------------*
Other idea
It maybee better to add a enhanced bios to your PC.
There is the xtide bios at
https://code.google.com/p/xtideuniversalbios/
The xtide bios works on most xt/at.
You can easily ad an enhanced bios if you have an free rom place in a
network card like a ISA 3com etherlink.
As Im from germany and my english isn't as good as it should be here is
an explanation from
http://flint.cs.yale.edu/feng/research/BIOS/BIOS-report.pdf
The IBM BIOS gains much of its versatility by being an extendable BIOS.
That is, the full extent of the BIOS is not cast forever in the silicon
of the single PROM chip holding the firmware. The IBM BIOS can accept
additional code as its own into one integrated whole. Hence additional
PROM chips containing BIOS routines can be added to the PC.
The BIOS will incorporate these new routines.
The key for making BIOS extendable is a Firmware routine that enables
the BIOS to look for add-in code. During the boot up, BIOS code reads
through the address range that is set aside for firmware looking for
codes stored on add-in boards. If a valid section of code is found, the
instructions are added to the BIOS repertory. For instance a new
interrupt routine can be added or the functions of existing routines can
be changed.
During POST after interrupt vectors have been loaded into RAM, the
resident BIOS code instructs the computer to check its ROM memory for
the occurrence of the special preamble bytes, that mark the beginning
of add-in BIOS routines. The BIOS searches for these preamble bytes in
the absolute address range 0C8000 - 0F4000.
Here the XTIDE BIOS is found and added so you can use it.
Greetings
fritz
Anybody out there older than me whose memory works better than mine or
younger but smart enough to figure out what these were used in? More
digging...
Any interest?
Y048 Bell tape controller with cable
Y051 Bell Binary to ASCII converter
Y064 Card reader
Y185 Stare 3 ???
Y197 DMA interface
W742 Universal digital input
W743 Universal digital input
W950 Wire wrap module- unused
W968 Collage mounting board- unused
Thanks, Paul
At 12:00 AM 7/21/2014, Eric Smith wrote:
>On Sun, Jul 20, 2014 at 8:10 PM, Rick Murphy <rick at rickmurphy.net> wrote:
> > I don't know anyting about a PDP-14,
>
>The machine is question is a VT14, not a PDP-14. The VT14 is a
>specialized terminal used to develop ladder logic programs for a
>PDP-14.
>
> > but it it used a VT8-E,
>
>It didn't.
Full circle. There seem to be two sets of assertions here.
One that a VT14 used the VT8-E board set, and another that it used a
different set for video.
You're asserting that it's not a VT8-E, which means that many of my
assumptions were wrong. In one of it's modes, which would allow it to
be a good display device, the VT8-E uses DMA from main memory to drive
a bitmap display. That memory has to be on the Omnibus, and has to be
read/write to be of any value. That causes the following assumption:
> > it had to be an Omnibus machine with Omnibus memory,
>
>It was.
>
> > so it could easily be turned into a PDP-8/E
>
>That doesn't necessarily follow, any more than it would follow that
>because some sort of embedded device contains a Z80 CPU chip, it can
>easily be turned into a CP/M system.
You have a KK8-E cpu and the standard Omnibus which can accept a range
of peripherals. You can't usually plug extra hardware into an embedded
system.
> > with the right CPU card set.
>
>It had a KK8-E CPU card set. To get it to run general-purpose PDP-8
>software, you're going to have to replace the memory, as the MR8-F
>isn't too useful for that.
I see. Perhaps it's the case that the VT14 doesn't have Omnibus RAM,
and only has a MR8-F which the KK8-E CPU uses to run the code that
manages the display. This is the first I've heard of the MR8-F, which
apparently has onboard RAM to provide the read/write memory needed.
(The older Ombibus ROM devices shadowed core.)
>The power supply isn't suitable for core
>memory, but it might be for semiconductor. The keyboard and display
>are going to be useless. The backplane, CPU, power supply, and
>teletype interface are the only parts of the VT14 that you're going to
>end up using. I wouldn't call that "easy", nor would I call the
>result a "PDP-8/E".
Yes, and the power supply isn't useful unless you're using something
other than core. And how many slots does the backplane have? Yes, it's
possible, but given what I know now it's not easy. It's a source for a
donor CPU board set, and maybe a TTY interface (assuming it's KL8
compatible) and that's about it. Possibly workable with the recent
semiconductor memory boards. Oh, and of course, no front panel, making
it hard to do anything useful unless you reprogram the MR8-F.
-Rick
Hello Mark,
well I suppose it would be very useful for this kind of operations.
And as the whole read operation would be in the control of the PC,
you could retrieve the whole disk, or a part, or a single sector..
whatever you want.
The first step is to debug it and let it work with SIMH.
Then I will try it with my 8" double floppy unit.
At this time, adding support for any other drive would be a matter of
knowing the peripheral
control registers and producing a set of scripts from these informations.
I will keep you informed about it.
Thanks
Andrea
On 7/19/2014 2:08 PM, Enrico Lazzerini wrote:
> I managed to find a bios for an IBM 5170 AT 286 Type1 Mainboard. It is
> here: 30aprile1989 (list here:
> http://ibm-pc.org/firmware/ibm/5170/5170.htm)
> In practice I have manually added the number of cylinders, heads, and sect
> / trk that allow the machine to recognise a disk IBM 250MB. At this point
> regularly boots DOS.
> But I wish to program these data permanently in the bios. Does anyone know
> what program to use and give me the procedure? I tried with BIOSUTL, but
> the new EPROM do not boot the machine.
> With BIOSUTL i made what follow: read actual BIOS, you can add new disk
> geometry parameters at free 47 position, then you have to recalculate bios
> checsum, then BIOSUTIL devides BIOS into EVEN and ODD file so i can finally
> program them into two 27256 150nS eprom.
> Thank you
> Enrico
>
Enrico,
I suspect you do not have a good checksum on your image, unless you took
pains to change that as well as the table Once you have modified your
table, you have to bring the sum which will be visible in the bios space
at F000:0 thru F000:ffff back to 0.
A simple checksum of the data is all that the bios does. See below. I
have patched numerous bios images, but always have to add a fudge in
some byte somewhere to bring the simple sum below back to 0 overall.
Jim
I used a copy from bitsavers for the listing below. Had to clean up the
OCR a bit :-)
http://bitsavers.trailing-edge.com/pdf/ibm/pc/at/1502494_PC_AT_Technical_Re…
If you look at the PC At tech reference, bios listing you can see what
they do:
On page 5-36, during startup ROS_CHECKSUM is called to check the roms.
On page 5-67 (PDF file page 205 on bitsavers)
EXTRN ROM_ERR: NEAR
;----------------------------------------
; ROS CHECKSUM SUBROUTINE ;
;----------------------------------------
ASSUME CS;CODE, DS:ABSO
POST3:
ROS_CHECKSUM PROC NEAR ; NEXT ROS MODULE
SUB CX,CX ; NUMBER OF BYTES TO ADD IS 64K
ROS_CHECKSUM_CNT: ; ENTRY FOR OPTIONAL ROS TEST
XOR AL,AL
C26:
ADD AL,DS:[BX]
INC BX ;POINT TO NEXT BYTE
LOOP C6 ; ADD ALL BYTES IN ROS MODULE
OR AL,AL ; SUM = O?
RET
ROS_CHECKSUM ENDP
At 02:34 PM 7/20/2014, Eric Smith wrote:
>On Sun, Jul 20, 2014 at 5:46 AM, Jerome H. Fine
><jhfinedp3k at compsys.to> wrote:
> > My question was not intended to be limited to the VT14 AS SUPPLIED!
>
>Nor was my answer.
>
> > Rather, could the VT14 have been relatively easily modified (i.e.
> without
> > major changes AND without having to reverse the changes OTHER than
> > to removing them with ONLY a screwdriver or other similar tool within
> > about 30 minutes) to be able to run PDP-8 software.
The VT8-E was a set of modules that turned a PDP-8 into a terminal.
In one mode, it DMA'd a chunk of memory and displayed the contents.
Interesting to fire it off when running memory checks, but that's about
it. It also had a bitmap font generator.
You could run an OS with the VT8-E running, but you couldn't run
anything like OS/8 with the VT8-E as the console as it ate too much
memory and wasn't in any way KL8 compatible.
I don't know anyting about a PDP-14, but it it used a VT8-E, it had to
be an Omnibus machine with Omnibus memory, so it could easily be turned
into a PDP-8/E with the right CPU card set.
-Rick
