Hi all,
I am attempting to get Helios running on a 64-core T805-based Transputer
array. This is a one-of-a-kind system with a custom case. It uses a
single-board 486DX with 8MB of RAM on an ISA card for the host system. The
PC front end runs Windows 3.1 or MS-DOS 6. There are eight T805s on eight
custom cards. It doesn't use TRAMs, but I have a feeling these are wired
like such on the board. Pictures of the system can be found here:
http://imgur.com/a/fCA3C
I've been getting all of my Helios editions from
http://www.classiccmp.org/transputer/helios.htm. I first tried version 1.1,
and after copying the contents over to the hard drive from a single floppy
disk, running the server results in an error:
ns : Incompatible version of Rmap 6e627573
ns : error is fatal, exiting.
I finally got 1.31 working on the system by downloading another DOS
executable of the server and playing with some configurations. I also used
ispy and mtest to see the structure of the system, which uses several C004
cross links. Unfortunately, ispy only reports a single T805-25 processor
connected to the host at address 0x150.
Does anyone have any ideas as to why the other 63 T805s and several C004s
aren't showing up?
Thanks in advance,
Kyle
On Tue, 28 Jan 2014 01:55:51 -0600, you wrote:
>From: Rick Murphy <rick at rickmurphy.net>
>To: cctalk at classiccmp.org
>Subject: Re: Using two RL02 drives on OS/8? (cont'd)
>Message-ID: <201401280235.s0S2Z0Es004934 at rickmurphy.net>
>Content-Type: text/plain; charset="us-ascii"; format=flowed
>
>At 04:13 PM 1/27/2014, Charles wrote:
>
>>More info. OS/8 is running on Drive 0. I am able to use RLFRMT (or
>>RL2FMT) to format the pack in Drive 1 (and it does not access or
>>overwrite Drive 0)... so at least SOME part of OS/8 recognizes that
>>there are two RL02 drives!
>
>Those programs access the drive hardware directly and don't use OS/8
>for anything.
>
>
>>BUILD defaults to DSK=SYS if SYS is not changed. BUILD showed
>>initially DSK=RL20:R2A0 which is indeed the same as SYS.
>>
>>However, if a SYS command is issued during BUILD, according to the
>>manual when issuing the BOOT command it is supposed to ask if you want
>>a new DSK or not. This does not happen - it just says "SYS BUILT" and
>>goes back to the dot prompt.
>>
>>I found that I can manually type DSK=RL21:R2A1 in BUILD and it accepts
>>that without error. Then a "SAVE SYS BUILD 0-7577, 10000-17577=0;200"
>>per the manual.
>>At that point, rubbing BUILD again shows DSK=RL21:R2A1. OK so far.
>>
>>But when I ask OS/8 for the directories of each partition on RL21:
>>(R2A1, R2B1, R2C1, R2D1) what is displayed is the corresponding
>>directory of RL20: (R2A0, R2B0, R2C0, R2D0)! The Drive 1 light never
>>flashes.
>
>You can't change DSK to a device that doesn't have an active handler.
>
>And, your unit convention is wrong - OS/8's device name mapping scheme
>is a pain, so you have to be careful how you reference things.
>
>What you should have in BUILD for this configuration is the following
>drivers:
>R0AB which gives devices R20A and R20B
>R0CD which gives devices R20C and R20D
>R01E which gives devices R20E and R21E
>R1AB which gives devices R21A and R21B
>R1CD which gives devices R21C and R21D
>
>If you don't have those, you'll have to LOAD them into build.
>
>So, you need to insert the drivers for drive one:
>$ RUN SYS BUILD
>
>$ IN R1AB:R21A
>$ IN R1AB:R21B
>$ IN R1CD:R21C
>$ IN R1CD:R21D
>$ IN R01E:R21E
>
>If you have enough driver slots, of course. Now device R21A: will go to
>the right place.
>The unit number is hard coded into the drivers so you need to use a
>driver binary built for the right unit number.
> -Rick
Thanks for the info, Rick! That sheds considerable light on my
puzzlement. Although I won't have enough slots to use the "wedgies"
(the small E partitions), 80% of two RL02's is more than enough!
However. Where do I find the R0AB, R0CD, R1AB, R1CD drivers? They
don't appear to be on the diagpack2.rk05 image where all the "good"
OS/8 stuff seems to be. All I see there is:
RL0 .BH 2 4-JAN-81
RL1 .BH 2 4-JAN-81
RL20 .BH 2 4-JAN-81
RL21 .BH 2 4-JAN-81
RL2E .BH 2 4-JAN-81
RL2SY .BH 2 4-JAN-81
RLC .BH 2 4-JAN-81
RLSY .BH 2 4-JAN-81
Of those, RLSY, RL0, RL1 and RLC are for an RL01 drive;
RL2SY, RL20, RL21 (which I have installed) and RL2E (which I omitted
as noted above) are for RL02.
Also the only OS/8 Extensions manual I can locate is the one that
shows how to install and boot an RL01, not a pair of RL02's.... I
think most of the RL02's ended up on PDP-11 systems ;)
-Charles
Could I interest anyone in my collection of BYTE magazines, dating from
issue #16 (DEC 76) to Vol 12 No 2 (FEB 87), missing Vol 11 No 11, for
pickup in the San Francisco Bay area in California?
--
Jeff Woolsey {woolsey,jlw}(a){jlw,jxh}.com first.last(a){gmail,hp,jlw}.com
Spum bad keming.
Nature abhors a straight antenna, a clean lens, and unused storage capacity.
"Delete! Delete! OK!" -Dr. Bronner on disk space management
"Card sorting, Joel." -me, re Solitaire
Quote : >> Could I interest anyone in my collection of BYTE magazines, dating from
issue #16 (DEC 76) to Vol 12 No 2 (FEB 87), missing Vol 11 No 11, for
pickup in the San Francisco Bay area in California? <<
Hello, Did this collection find a good home ??
Else, maybe I can offer one.
Regards, Gerard
---
Ce courrier ?lectronique ne contient aucun virus ou logiciel malveillant parce que la protection avast! Antivirus est active.
http://www.avast.com
All this talk about RL01/02s and RK05s has reminded me that it would be
nice to get the device handler for my dual Xebec hard sector floppy drives
connected to my -8/M. I have only a couple of bootable disks with a couple
of versions of OS/8. A lot of them are corrupted from best I can tell.
I've developed a way to dump all of the readable blocks from disk over the
serial port, which, as you can imagine, does take a while. I've also
developed a little program that can parse the data into RIM format to be
disassembled.
>From the OS/8 Software Support Manual, I can see that the handlers should
be located within blocks 16 through 25, octal. Should I be attempting to
dump those blocks and try to reverse engineer it that way, or is there a
way in OS/8 to directly dump each individual handler?
Any tips would be much appreciated.
Thanks,
Kyle
I quit fiddling around with TME and went to QEMU like I should have in the
first place. Following this page exactly:
http://en.wikibooks.org/wiki/QEMU/SunOS_4.1.4, I got barfage that looked
like this:
Extracting the sunos 4.1.4 sun4m 'uucp' media file.
Extracting the sunos 4.1.4 sun4m 'Games' media file.
Extracting the sunos 4.1.4 sun4m 'Versatec' media file.
/usr/etc/install/tar/export/exec/sun4_sunos_4_1_4/versatec: cannot
extract file.
add_services: 'tar xpfb
/usr/etc/install/tar/export/exec/sun4_sunos_4_1_4/versatec 32k 2>>
/etc/install/suninstall.log' failed.
Please check local media device
/usr/etc/install/tar/export/exec/sun4_sunos_4_1_4/versatec
Press <return> to continue
You have sunos 4.1.4 sun4m release media volume -1 mounted
Please mount sunos 4.1.4 sun4m release media volume 1
Press <return> to continue
I upgraded QEMU. That didn't fix the problem. Then I looked at my SunOS
4.1.4 iso. The file export/exec/sun4_sunos_4_1_4/versatec is a corrupted
tar. I can manually exclude that package, but I'd like to have something
that really works all the way, particularly for when I get my hands on my
real Sun hardware again.
Does anyone know where I can find a known-good ISO? The md5sum of my iso
is this:
927fa22042a70bf21a80be9393b15770 sunos_4.1.4_install.iso
Second question...
With the resulting install of SunOS 4.1.4, I'm having trouble getting the
emulation talking to the outside world. Through trial-and-error, I
figured out that QEMU is assigning the guest the IP of 192.9.200.1 and the
host as 192.9.200.2. Plugging these into what I see at the wikibooks.org
page, I get results that seem like things went okay, but pings and telnets
timeout. What did I miss?
--
David Griffith
dgriffi at cs.csubak.edu
The thread on using two RL02 drives with OS8 made me think about this, but I didn't want to confuse that thread with my tangential topic.
Is there any difference in RL02 drives and/or disk packs for PDP-8 systems vs. PDP-11 systems (other than the obviously different controller cards for OMNIBUS/UNIBUS/etc.)? If so, how do I evaluate and identify the specialized bits and pieces?
I ask this because I have a few RL02 drives waiting for my PDP-11/44 project, as well as a pile of RL02 disk packs. If there are differences for PDP-8 vs. PDP-11 applications, i.e. due to different word sizes, then I'll need to keep my eyes open for mismatched parts in my pile. Conversely, if and when I get a PDP-8 system with an RL02 controller card, I'll wonder whether I can use any of the pieces I already have.
I seem to recall reading in recent months about some DEC drive and/or disk pack that would have different hard sector notches or some such thing for PDP-8 vs. PDP-11 applications, but I don't recall which type of drive it was.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
On Jan 27, 2014, at 7:17 PM, Grumpy ol? Fred wrote:
> "REAL computers make their own heat."
Hah! Another proposal for computer classification!
Microcomputer : can keep a bread-box warm.
Minicomputer: can keep a basement warm.
Server: can keep a warehouse warm.
Supercomputer: can keep a time zone warm.
Fred, you da man!