30 Sep
2009
30 Sep
'09
11:19 a.m.
I think I asked about this several years ago, but I thought I might
try my luck again.
Is anyone aware of any surviving ARPAnet IMP System Software, in any
form?
-Seth
30 Sep
30 Sep
12:39 p.m.
Seth Morabito wrote:
I think I asked about this several years ago, but I
thought I might
try my luck again.
Is anyone aware of any surviving ARPAnet IMP System Software, in any
form?
I'm working on a radio with some folks from BBN. I'll ask them.
Any idea what is was built on? i.e. I assume there was a cross assembler
which ran
somewhere.
-brad
--
-----------
Brad Parker
Heeltoe Consulting
http://www.heeltoe.com
781-483-3101
12:47 p.m.
From: Brad Parker
Sent: Wednesday, September 30, 2009 12:40 PM
Seth Morabito wrote:
> I think I asked about this several years ago, but I thought I might
> try my luck again.
> Is anyone aware of any surviving ARPAnet IMP
System Software, in any
> form?
I'm working on a radio with some folks from BBN.
I'll ask them.
Any idea what is was built on? i.e. I assume there was
a cross assembler
which ran somewhere.
I would assume that it was built on a Honeywell 316, since that was the
basis of the IMP. A former Honeywell employee tells me that that had its
own environment.
Rich Alderson
Vintage Computing Server Engineer
Vulcan, Inc.
505 5th Avenue S, Suite 900
Seattle, WA 98104
mailto:RichA at vulcan.com
mailto:RichA at livingcomputermuseum.org
(206) 342-2239
(206) 465-2916 cell
http://www.pdpplanet.org/
http://www.livingcomputermuseum.org/
1:10 p.m.
Brad Parker wrote:
Seth Morabito wrote:
DDP-516
http://en.wikipedia.org/wiki/Interface_Message_Processor
even surviving binary downloads would be interesting.
I have been told that BBN handled the downloads over the net, and that
local nodes didn't maintain copies of the software.
I think I asked about this several years ago, but
I thought I might
try my luck again.
Is anyone aware of any surviving ARPAnet IMP System Software, in any
form?
I'm working on a radio with some folks from BBN. I'll ask them.
Any idea what is was built on? 
2:45 p.m.
Hi,
even surviving binary downloads would be interesting.
Leonard Kleinrock still has his IMP sitting in his office at UCLA. I
tried to talk him into making a dump to paper tape, but he seemed a bit
reluctant - perhaps anybody in the US could go there, fire up the
machine and get the dump. I'm quite confident that the 516 just works.
That's at least my experience with several H316 machines.
I have been told that BBN handled the downloads over
the net, and that
local nodes didn't maintain copies of the software.
Very modern!
BTW, Leonard Kleinrock does not have the software.
I also tried to get the H316 implementation of FORTH (must be one of the
very first) from the author of FORTH - but he has no copy left :-(
Best wishes,
Philipp
1 Oct
1 Oct
11:48 a.m.
Seth Morabito wrote:
I think I asked about this several years ago, but I
thought I might
try my luck again.
Is anyone aware of any surviving ARPAnet IMP System Software, in any
form?
After talking with a person from BBN it might be possible to read
some old 7-track tapes with some code on them. Apparently the
earliest versions were written in PDP-1 assembler with macros.
Later a cross compiler was written for TENEX.
I'll dog it for a while and see what happens. I suppose the next
question is, "does anyone have docs on the Honeywell 516"? and
is there a simulator?
I'm game to write one for simh if there isn't one (but I'll hold
off if someone else has their heart set on it)
it might be more fun to write the pdp-1 assembler :-)
-brad
-----------
Brad Parker
Heeltoe Consulting
http://www.heeltoe.com
781-483-3101
1:06 p.m.
On Oct 1, 2009, at 11:48 AM, Brad Parker wrote:
Seth Morabito wrote:
That is some potentially very good news! Thanks for asking about this!
I think I asked about this several years ago, but
I thought I might
try my luck again.
Is anyone aware of any surviving ARPAnet IMP System Software, in
any form?
After talking with a person from BBN it might be possible to read
some old 7-track tapes with some code on them. Apparently the
earliest versions were written in PDP-1 assembler with macros.
Later a cross compiler was written for TENEX. I'll dog it for a while and see what happens. I
suppose the next
question is, "does anyone have docs on the Honeywell 516"? and
is there a simulator?
There is not (yet) a 516 simulator, but Bob Supnik's SIMH does have an
H316 simulator. See: http://simh.trailing-edge.com/h316.html
Later IMPs were H316s, so possibly some of the code on those 7-tracks
is written for them?
I have no idea how much work would be involved in adding a 516 mode to
SIMH, nor how hard it would be to simulate the bit-serial IMP-host
interface hardware, but I'd certainly be game to help with that.
-Seth
1:28 p.m.
I have no idea how much work would be involved in
adding a 516 mode to
SIMH, nor how hard it would be to simulate the bit-serial IMP-host
interface hardware, but I'd certainly be game to help with that.
The 516 and the 316 are 100% code compatible. Just the cycle time
differs (0.96us for the 516, 1.3us for the 316 IIRC)
Adrian
http://www.series16.adrianwise.co.uk
2 Oct
2 Oct
5:14 a.m.
Adrian Wise wrote:
The 516 and the 316 are 100% code compatible.
Just wanted to write that...
Just the cycle time
differs (0.96us for the 516, 1.3us for the 316 IIRC)
Have you tuned up your H316?
My machine is running on 1.6us! In that
case I would be glad to get your CPU and memory timing configuration :-)
(Otherwise it would be an interesting experiment to see if and how much
faster a H316 could be made. I'm quite confident that Honeywell/3C
delivered the machines in a very stable configuration with big reserves
with respect to reliability)
Best wishes,
Philipp :-)
3 Oct
3 Oct
12:31 a.m.
Hi Philip
Old age must be creeping up on me since I clearly don't recall
correctly! The H316 is 1.6us as you say.
Adrian
Philipp Hachtmann wrote:
Adrian Wise wrote:
The 516 and the 316 are 100% code compatible.
Just wanted to write that...
Just the cycle time differs (0.96us for the 516,
1.3us for the 316 IIRC)
Have you tuned up your H316? My machine is running on
1.6us! In that
case I would be glad to get your CPU and memory timing configuration :-)
(Otherwise it would be an interesting experiment to see if and how much
faster a H316 could be made. I'm quite confident that Honeywell/3C
delivered the machines in a very stable configuration with big reserves
with respect to reliability)
Best wishes,
Philipp :-)
5 Oct
5 Oct
5:45 a.m.
Hey, is the IMP software also what was used to make a TIP or a TAC, or
were those different?
On Oct 3, 2009, at 2:31 AM, Adrian Wise wrote:
Hi Philip
Old age must be creeping up on me since I clearly don't recall
correctly! The H316 is 1.6us as you say.
Adrian
Philipp Hachtmann wrote:
> Adrian Wise wrote:
>> The 516 and the 316 are 100% code compatible.
> Just wanted to write that...
>> Just the cycle time differs (0.96us for the 516, 1.3us for the 316
>> IIRC)
> Have you tuned up your H316? My machine is running on 1.6us! In that
> case I would be glad to get your CPU and memory timing configuration
> :-)
> (Otherwise it would be an interesting experiment to see if and how
> much faster a H316 could be made. I'm quite confident that
> Honeywell/3C delivered the machines in a very stable configuration
> with big reserves with respect to reliability)
> Best wishes,
> Philipp :-)
4:55 p.m.
Daniel Seagraves wrote:
Hey, is the IMP software also what was used to make a
TIP or a TAC, or
were those different?
It was different. TIPs were terminal servers.
I just talked to the two people who brought up the first IMPs at SRI and
UCLA. The earliest ones booted from paper tape. Once there was critical mass
BBN added a boot rom board, and the software was remote loaded from them
or from a neighboring IMP (they all ran the same version of the software).
5:45 p.m.
Does anyone still have the TIP or TAC software?
Also, did the TIP/TAC require an IMP of its own?
On Oct 5, 2009, at 6:55 PM, Al Kossow wrote:
Daniel Seagraves wrote:
Hey, is the IMP software also what was used to
make a TIP or a TAC,
or were those different?
It was different. TIPs were terminal servers.
I just talked to the two people who brought up the first IMPs at SRI
and
UCLA. The earliest ones booted from paper tape. Once there was
critical mass
BBN added a boot rom board, and the software was remote loaded from
them
or from a neighboring IMP (they all ran the same version of the
software).
1 Oct
1 Oct
1:47 p.m.
On Thu, 1 Oct 2009, Brad Parker wrote:
After talking with a person from BBN it might be
possible to read
some old 7-track tapes with some code on them. Apparently the
earliest versions were written in PDP-1 assembler with macros.
Later a cross compiler was written for TENEX.
My question would be, what do they have for TENEX, and would they be willing
to release that. It would be nice to see TENEX preserved. Though on the
PDP-10 we already have TOPS-10, TOPS-20, and ITS. ISTR, that bitsavers has
a few TENEX files.
Zane
1:58 p.m.
New subject: TENEX (was Re: IMP system software)
Zane H. Healy wrote:
On Thu, 1 Oct 2009, Brad Parker wrote:
Mark Crispin has some stuff. I have some from the SRI archives. Dan
Murphy should have some things, but it may have been lost when they
wiped out his tapes with PDP-1 TECO on it.
There is also the TENEX variant used on the Foonlys (FOONEX).
After talking with a person from BBN it might be
possible to read
some old 7-track tapes with some code on them. Apparently the
earliest versions were written in PDP-1 assembler with macros.
Later a cross compiler was written for TENEX.
My question would be, what do they have for TENEX, and would they be
willing
to release that.
5325
days inactive
5331
days old
14 comments
9 participants
participants (9)
-
adrian＠adrianwise.co.uk -
aek＠bitsavers.org -
apw＠adrianwise.co.uk -
brad＠heeltoe.com -
dseagrav＠lunar-tokyo.net -
hachti＠hachti.de -
healyzh＠aracnet.com -
RichA＠vulcan.com -
sethm＠loomcom.com