Intergraph Interact 340
ers...@electronic-recovery.com
running. The machine has a hard disk, but will not boot. From
what I've heard from the Intergraph techs, is that I will need a
set of "rebuild" floppies to boot off of. The only way to make
a set of rebuild floppies is to run a utiliy on another Clix
machine. Unfortunatly, I do not have another Clix computer to do
this. I also have a set of "Intergraph Systems Software" on CD-ROM.
Anyone see a way to get this Interact 340 running? This Interact
340 system is interesting in that it is dual headed and has a 36
inch by 24 inch digitising table.
Thanks for any help.
Chris Bailey
the speed of either a fast 486 or a slow pentium. Only thing is, it
will suck down the power like a hungry elephant.
If you're really dead set on getting this running, you'll need a set
of 5 1/4" rebuild floppies for your particular set of system cd's.
Are you sure the hard disk is working? It's probably either a 670 mb
or a 355 mb disk. There again, you would be better off with a 486.
If not, you could probably pick up a 1 - 2gb disk drive for this on
ebay for a fairly reasonable price. Not every disk drive will work on
this though, so you might want to check first.
You might want to post the dates on the cdroms so that someone who has
the facilities for building rebuild disks and is willing to build them
for you, can do so.
If you can't find anyone else to build you a set of disks, let me
know. If I get a chance, I'll try to hook up one of my Interpro 225's
and build you a set. That would at least get you started.
On Tue, 04 Jan 2000 11:13:49 -0600, ers...@electronic-recovery.com
wrote:
Nik Simpson
"Chris Bailey" <cbailey...@cander.net> wrote in message
news:3873e856.524269@news-server...
> Not really worth the effort. What you will get is something that has
> the speed of either a fast 486 or a slow pentium. Only thing is, it
> will suck down the power like a hungry elephant.
A 340 doesn't even come close to a fast 486, it's 1988 technology and uses
the original C100 CLIPPER processors.
--
Nik Simpson
Chris Bailey
:> But it still sucks power. I've uses an Interserve 300 for a
space heater before! Add the graphics boards and the dual monitors
and you get almost as much heat as your home furnace can put out!
On Thu, 6 Jan 2000 06:55:53 -0500, "Nik Simpson" <n...@hiwaay.net>
wrote:
Carl R. Friend
> Not really worth the effort. What you will get is something that has
> the speed of either a fast 486 or a slow pentium. Only thing is, it
> will suck down the power like a hungry elephant.
Sometimes that's not the issue. Perhaps he merely wants to run
the thing as an exercise or is interested in old iron.
> You might want to post the dates on the cdroms so that someone who has
> the facilities for building rebuild disks and is willing to build them
> for you, can do so.
What's the legality of doing this? I made a set of rebuild-floppy
images which I can use from any old box to write new ones. It'd be
trivial to gzip up a bunch and post 'em somewhere, but I'm not the
type who likes to step on toes....
--
+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:crfr...@ma.ultranet.com +---------------------+
| http://www.ultranet.com/~crfriend/museum | ICBM: N42:22 W71:47 |
+------------------------------------------------+---------------------+
Chris Bailey
months, the software license is built into the system. In other
words, if he has an Interact 340, then he has a license for CLIX.
Now, that does NOT mean that he has a license for ALL CLIX software,
but that he can run any of the "free" software that came on the System
cd. He also doesn't have any support from Intergraph itself unless he
purchases a maintenance contract of some sorts.
I don't think this will happen real soon, but it would be great if
Intergraph would release their licensing for all of their old CLIX os
& applications, but also release all of the source code. Since
Intergraph no longer sells this stuff, they wouldn't be hurt
financially if they released it all to Public Domain.
Carl R. Friend
>
> Well, unless Intergraph has changed their policy in the last couple of
> months, the software license is built into the system. In other
> words, if he has an Interact 340, then he has a license for CLIX.
I truly doubt that they've changed their policy on the matter.
Sometimes I wonder if they even know what CLIX is any more....
> Now, that does NOT mean that he has a license for ALL CLIX software,
> but that he can run any of the "free" software that came on the System
> cd. He also doesn't have any support from Intergraph itself unless he
> purchases a maintenance contract of some sorts.
Sadly, the pieces that the "base" version of CLIX lacks includes a
C compiler and development libraries. There is a _very_ old version
of gcc included in the UNIXCFG product, but I doubt that can do much
more than preprocessing and basic linking.
Getting a new version of gcc, or egcs, running on CLIX would be
very nice, but one heck of a lot of work.
> I don't think this will happen real soon, but it would be great if
> Intergraph would release their licensing for all of their old CLIX os
> & applications, but also release all of the source code. Since
> Intergraph no longer sells this stuff, they wouldn't be hurt
> financially if they released it all to Public Domain.
That has been suggested in the past, but there never was any motion
on it from Intergraph; they won't even release the programming specs
for the oldest of hardware. Given that, getting Intergraph to PD the
OS is highly unlikely. A hobbyist "license" (along the lines of what
DEC did with the 12, 18, and 36-bit software) would be a wonderful
thing, though, but still couldn't include a compiler as they licensed
the things from first Green Hills, then Apogee. And without a compiler
suite, a UNIX box for a hobbyist is pretty useless....
Nik Simpson
> Well, unless Intergraph has changed their policy in the last couple of
> months, the software license is built into the system. In other
> words, if he has an Interact 340, then he has a license for CLIX.
> but that he can run any of the "free" software that came on the System
> cd. He also doesn't have any support from Intergraph itself unless he
> purchases a maintenance contract of some sorts.
>
> Intergraph would release their licensing for all of their old CLIX os
> & applications, but also release all of the source code. Since
> Intergraph no longer sells this stuff, they wouldn't be hurt
> financially if they released it all to Public Domain.
The copyright owners of System V would be somewhat pissed at Intergraph if
they released the source for CLIX, it's the sort of things that gets you
into trouble with lawyers and Intergraph wouldn't want that, now would they!
Seriously, until AT&T or whoever holds the rights to SVR3.1 and SVR3.2
release those sources into the public domain Intergraph is powerless to
release the source code for CLIX. As to releasing the source of the
applications, I suspect Bentley might have a few words to say about that,
and without releasing Microstation it would be hard to release anything
else.
--
Nik Simpson
Carl R. Friend
>
> The copyright owners of System V would be somewhat pissed at
> Intergraph if they released the source for CLIX, it's the sort of
> things that gets you into trouble with lawyers and Intergraph
> wouldn't want that, now would they!
Actually, the source needn't be released, merely a license to use
the binaries for non-commercial purposes. I doubt that there would
be a whole lot of argument about that.
Once thing, though, that a source release would expose is whether
CLIX uses the AT&T or BSD date routines. (Hint: Why else does a
"date +%y" command return ":0"?)
> Seriously, until AT&T or whoever holds the rights to SVR3.1 and
> SVR3.2 release those sources into the public domain Intergraph is
> powerless to release the source code for CLIX.
Source is, sometimes, not all it's made out to be. CLIX was a
fine OS for its day, and blended most of the best bits of AT&T UNIX
with the Berkeley release. It's still running; I have a CLIX box on
my desk at work and love it dearly (and have threatened people who
wish to replace it with bodily harm). CLIX is a _very_ interesting
environment in that it's frozen in time (about 1989 as a guess) and
hasn't evolved one bit since then (makes porting stuff "fun").
> As to releasing the source of the applications, I suspect Bentley
> might have a few words to say about that, and without releasing
> Microstation it would be hard to release anything else.
Bentley would, indeed, go after Intergraph in a big way if Inter-
graph released any source to Microstation (and rightfully, for the
more modern stuff -- the ethics of "hiding" the older code is, well,
questionable, in my mind). Microstation need not be released by
Intergraph (let Bentley do it if they desire); the thing here is that
the OS and platform are "usable" (for varying values of "usable").
This implies (at the very least) a meaningful release of the
development libraries so a decent port of gcc or egcs can get done!
Art Marriott
> Actually, the source needn't be released, merely a license to use
> the binaries for non-commercial purposes. I doubt that there would
> be a whole lot of argument about that.
Actually, there would probably be a whole lot or argument from SCO, or
whoever now owns the rights to System V Unix.
My resollection (anyone feel free to take pot shots at this) is that
only Sun, of all the original commercial Unix licensees, has a
grandfathered deal they negotiated with AT&T to provide per-machine
licenses with unlimited users. That's why CLIX (and other System V
derivatives like AIX, etc) come with limititations on the number of
simultaneous logins. The more users, the more expensive the license.
It's one of the reasons for the growing popularity of Linux.
>
> Once thing, though, that a source release would expose is whether
> CLIX uses the AT&T or BSD date routines. (Hint: Why else does a
> "date +%y" command return ":0"?)
If you check Intergraph's Web site, there's an explanation that the
Clipper systems have a hardware/firmware Y2K problem, which is
ostensibly corrected by the most recent CLIX nucleus. (There are
writeable control stores for not only the CPU but several subsystems
that get set at boot time.)
________________________________________________________________________
|
AXENHAMMER SYSTEMS | Art Marriott
Applied Technology | President and Janitor
| artma...@mailandnews.com
________________________________|_______________________________________
Nik Simpson
"Carl R. Friend" <carl....@prescienttech.com> wrote in message
news:387BE82A...@prescienttech.com...
> Nik Simpson wrote:
> >
> > The copyright owners of System V would be somewhat pissed at
> > Intergraph if they released the source for CLIX, it's the sort of
> > things that gets you into trouble with lawyers and Intergraph
> > wouldn't want that, now would they!
>
> the binaries for non-commercial purposes. I doubt that there would
> be a whole lot of argument about that.
I agree to some degree, but I was responding to a request that someday the
source for both CLIX and its applications might be released, that's simply
unrealistic. I can certainly see a good argument for Intergraph putting the
CLIX baseline and bootfloppy images up for download since they are of no use
to anyone without a CLIPPER machine, and anybody with one should be entitled
to access to these things that were bundled with the machine.
>
> Once thing, though, that a source release would expose is whether
> CLIX uses the AT&T or BSD date routines. (Hint: Why else does a
> "date +%y" command return ":0"?)
I'd be very surprised if the date routine was anything but stock SVR 3.1
unless it some BSDisms crept in to satisfy POSIX or FIPS compliance.
>
> > Seriously, until AT&T or whoever holds the rights to SVR3.1 and
> > SVR3.2 release those sources into the public domain Intergraph is
> > powerless to release the source code for CLIX.
>
> Source is, sometimes, not all it's made out to be. CLIX was a
> fine OS for its day, and blended most of the best bits of AT&T UNIX
> with the Berkeley release. It's still running; I have a CLIX box on
> my desk at work and love it dearly (and have threatened people who
> wish to replace it with bodily harm). CLIX is a _very_ interesting
> environment in that it's frozen in time (about 1989 as a guess) and
> hasn't evolved one bit since then (makes porting stuff "fun").
I don't think that's completely fair to CLIX, some stuff was done in the
early nineties, for example the entire networking stack was rewritten to
solve some serious performance issues. Also things like the Native X11
server were not introduced until the nineties.
>
> > As to releasing the source of the applications, I suspect Bentley
> > might have a few words to say about that, and without releasing
> > Microstation it would be hard to release anything else.
>
> Bentley would, indeed, go after Intergraph in a big way if Inter-
> graph released any source to Microstation (and rightfully, for the
> more modern stuff -- the ethics of "hiding" the older code is, well,
> questionable, in my mind). Microstation need not be released by
> Intergraph (let Bentley do it if they desire); the thing here is that
> the OS and platform are "usable" (for varying values of "usable").
> This implies (at the very least) a meaningful release of the
> development libraries so a decent port of gcc or egcs can get done!
I'm glad to say that my interest in this subject is now entirely academic. I
left Intergraph in August (or to be more accurate, they left me) and now
work for a startup doing Storage Area Networks in Florida. So if anybody
needs to implement a SAN anytime soon, just drop me a line :-)
--
Nik Simpson
Nik Simpson
"Art Marriott" <marr...@accessone.com> wrote in message
news:387C41...@accessone.com...
> >
> > Actually, the source needn't be released, merely a license to use
> > the binaries for non-commercial purposes. I doubt that there would
> > be a whole lot of argument about that.
>
> whoever now owns the rights to System V Unix.
>
> My resollection (anyone feel free to take pot shots at this) is that
> only Sun, of all the original commercial Unix licensees, has a
> grandfathered deal they negotiated with AT&T to provide per-machine
> licenses with unlimited users. That's why CLIX (and other System V
> derivatives like AIX, etc) come with limititations on the number of
> simultaneous logins. The more users, the more expensive the license.
> It's one of the reasons for the growing popularity of Linux.
That's a reasonable description of things as they stand today. Actually Sun
bought out it's licence completely from Novell during the period when Novell
were "the keepers of the flame." Right now I'm not sure who does own the
rights to System V and derivatives, but as you say, with the growing
acceptance of LINUX its becoming somewhat moot.
--
Nik Simpson
Chris Bailey
Welcome to the world of Ex-Intergraph Employees. I honestly
think we need to start a club or something. I was offered a job with
an unrelated industry and left Intergraph the first of November. I
really hate to say this, but Intergraph was a company that had
everything going for it and then just let it slip through it's
fingers.
On Wed, 12 Jan 2000 09:10:45 -0500, "Nik Simpson"
<nik.s...@datacoresoftware.com> wrote:
>
>"Carl R. Friend" <carl....@prescienttech.com> wrote in message
>news:387BE82A...@prescienttech.com...
>> Nik Simpson wrote:
>> >
>> > The copyright owners of System V would be somewhat pissed at
>> > Intergraph if they released the source for CLIX, it's the sort of
>> > things that gets you into trouble with lawyers and Intergraph
>> > wouldn't want that, now would they!
>>
>> Actually, the source needn't be released, merely a license to use
>> the binaries for non-commercial purposes. I doubt that there would
>> be a whole lot of argument about that.
>
Ziphius
>everything going for it and then just let it slip through it's
>fingers.
>
As a longtime Intergraph customer, I whole-heartedly agree. We have to beg out
local Intergraph office to sell us software. When we went from CLIX to Wintel,
I had to explain to the local rep what the differences in Intergraph hardware
models was.
I miss my 6800, grep, awk, scpio -C63480 - z20, and the time when the Blue
Screen meant success !
Dan
Art Marriott
>
> Nik,
> Welcome to the world of Ex-Intergraph Employees. I honestly
> think we need to start a club or something. I was offered a job with
> an unrelated industry and left Intergraph the first of November. I
> everything going for it and then just let it slip through it's
> fingers.
>
the harder you try to hold onto a snowball, the faster it melts. There
was a tremendous effort put into winning the Navsea/Navair/Navfac
military contracts in the early '90s, only to have the Cold War come to
an end. If that huge chunk of business had actually happened, it could
have given them some breathing room in which to re-create their business
model to match what the marketplace was turning into. Then again, that
might have simply enabled them to stay in their ponderous,
government-contractor mode even longer.
H-P and IBM have certainly shown that there's business to be had
providing high-end WinTel systems that are (1.) well designed; (2.)
decently constructed; and (3.) backed up with professional field
support.
Then again, the "Windows Revolution" has caused even worse things to
happen to other companies. Consider DEC, for instance......!
Nik Simpson
> providing high-end WinTel systems that are (1.) well designed; (2.)
> decently constructed; and (3.) backed up with professional field
> support.
>
Actually, I think you could make a case that Intergraph proved that before
either HP or IBM even thought of the idea. But then, Don Quixote like, they
started tilting at Windmills and sued Intel, which if you want my honest
opinion (and one I had at the time) was a bonehead move of enormous
proportions.
--
Nik Simpson
Carl R. Friend
>
> Nik,
> Welcome to the world of Ex-Intergraph Employees. I honestly
> think we need to start a club or something.
It's probably not the forum to ask in, but what the hey, this is
USENET!
How many of us here are ex-Intergraph?
Nik I knew about, and now Chris comes to the fore. I'm ex-Ingr
myself, having yanked on the ejection-lever (of my own volition) in
1989 when the challenge left the Field Engineering arena (11s were
fun, VAXen usually told you what board to swap (usually), but the
handwriting was on the wall with the single-board machines!).
Nik mentions:
> I don't think [a 1989 "freeze" date is] completely fair to CLIX,
> some stuff was done in the early nineties, for example the entire
> networking stack was rewritten to solve some serious performance
> issues.
Perhaps I'm being cruel, but that's how the environment "feels";
getting modern stuff to compile on it is usually a major hassle if
it can be done at all. And even if you can get something to compile,
there's no guarantee it'll run (Apache comes to mind).
Yes, native X was a product of the '90s but by that time the
foundation had been pretty much frozen. At what release did the
network bits change? It might be fun to seriously "retro-port"
Apache to see if the kernel bug was in the "old" stack.
"Ziphius" comments later on in the thread:
> I miss my 6800, grep, awk, scpio -C63480 - z20, and the time when
> the Blue Screen meant success !
They're available out there, usually for the asking. That's why
it'd be nice to see a general hobbyist license granted. It'd be
such a shame to see the architecture die totally.
Art Marriott
> Nik mentions:
>
> > I don't think [a 1989 "freeze" date is] completely fair to CLIX,
> > some stuff was done in the early nineties, for example the entire
> > networking stack was rewritten to solve some serious performance
> > issues.
>
> Perhaps I'm being cruel, but that's how the environment "feels";
> getting modern stuff to compile on it is usually a major hassle if
> it can be done at all. And even if you can get something to compile,
> there's no guarantee it'll run (Apache comes to mind).
Some of this has to do with CLIX being based on SVR3. Most "modern"
Unices (Solaris, Irix, etc.) are derived from SVR4. My understanding
was that INGR didn't want to invest in licensing and porting SVR4,
probably because the decision to move to Windows was made much earlier
than most of us could have imagined.
Actually, AIX is to this day an SVR3 derivative, but IBM made some
really massive changes to its innards, adding things like long
filenames, infinite inodes, volume management, their own TCP/IP stack,
and so on. My guess is that they didn't like the Berkeley stuff and
they didn't want to wait for AT&T to do it.
>
> Yes, native X was a product of the '90s but by that time the
> foundation had been pretty much frozen. At what release did the
> network bits change? It might be fun to seriously "retro-port"
> Apache to see if the kernel bug was in the "old" stack.
>
> "Ziphius" comments later on in the thread:
>
> > I miss my 6800, grep, awk, scpio -C63480 - z20, and the time when
> > the Blue Screen meant success !
>
> They're available out there, usually for the asking. That's why
> it'd be nice to see a general hobbyist license granted. It'd be
> such a shame to see the architecture die totally.
AS I mentioned in my earlier post, they *can't* put CLIX in the public
domain because it's not really theirs to do that with. On the other
hand, at this point there's certainly not much justification for them
not releasing the details of the system hardware/firmware. Then someone
could hope to make an intelligent attempt to port Linux or some such OS
to the platform.
________________________________________________________________________
|
AXENHAMMER SYSTEMS | Art Marriott
Applied Technology | President and Janitor
(206) 522-8606 | marr...@accessone.com
________________________________|_______________________________________
I'd rather have government on my back than Tim Eyman
in my face.
________________________________________________________________________
Nik Simpson
> > some stuff was done in the early nineties, for example the entire
> > networking stack was rewritten to solve some serious performance
> > issues.
>
> Perhaps I'm being cruel, but that's how the environment "feels";
> getting modern stuff to compile on it is usually a major hassle if
> it can be done at all. And even if you can get something to compile,
> there's no guarantee it'll run (Apache comes to mind).
To be honest most of the changes after about release 3 of CLIX were under
the hood so to speak, rip out the old networking stack and build a new one,
but no major changes to the APIs, hence the porting issues that occur more
frequently as time goes by. I think the last major API stuff would have been
making CLIX POSIX 1003.1 compliant for the Navy NAVSEA contract.
>
> Yes, native X was a product of the '90s but by that time the
> foundation had been pretty much frozen. At what release did the
> network bits change? It might be fun to seriously "retro-port"
> Apache to see if the kernel bug was in the "old" stack.
>
The work was going on C1991 when I moved from the UK office to Huntsville, I
think that corresponds with the release 5 kernel. IIRC, the work was being
done in order to reach the Navy's FTP throughput requirements of the NAVFAC
contract, certainly FTP performance was a big issue during those tests.
--
Nik Simpson
Nik Simpson
> Carl R. Friend wrote:
>
> > Nik mentions:
> >
> > > some stuff was done in the early nineties, for example the entire
> > > networking stack was rewritten to solve some serious performance
> > > issues.
> >
> > Perhaps I'm being cruel, but that's how the environment "feels";
> > getting modern stuff to compile on it is usually a major hassle if
> > it can be done at all. And even if you can get something to compile,
> > there's no guarantee it'll run (Apache comes to mind).
>
> Unices (Solaris, Irix, etc.) are derived from SVR4. My understanding
> was that INGR didn't want to invest in licensing and porting SVR4,
> probably because the decision to move to Windows was made much earlier
> than most of us could have imagined.
That's not really the case, Intergraph never seriously considered doing
SVR4. Along with most vendors (other than Sun/AT&T) nobody wanted to do SVR4
in the late 80's because it was so controlled by Sun (can anybody say Java
:-). As a result the big effort to produce a unified UNIX created a new
schism with the formation of the Open Software Foundation and its b*****d
child OSF/1 (cries of "splitters" a-la The Life Brian sprung to mind.)
Intergraph was on the OSF/1 side of the schism.
We did actually start working on a OSF/1 port for CLIPPER, but it never got
very far as the OSF was collapsing under it's own wieght by the time it had
anything approaching a kernel source tree to port. The big OSF/1
announcement was made a IGUG in 1992, I know because I did the technical
presentation to the Systems and Networking SIG. What they didn't know at the
time (and I did :-) was that Intergraph's commitment to an OSF/1 port was
all but dead in the water as we were in advanced stages of negotiating the
NT deal with Microsoft. Infact, by the fall of 1992 I had to stand up in
front of much the same audience at FALLGUG and tell them that everything
they heard in my presentation at Spring GUG was a crock :-)
> Actually, AIX is to this day an SVR3 derivative, but IBM made some
> really massive changes to its innards, adding things like long
> filenames, infinite inodes, volume management, their own TCP/IP stack,
> and so on. My guess is that they didn't like the Berkeley stuff and
> they didn't want to wait for AT&T to do it.
Again, IBM was on the OSF/1 side of the schism and much of what you see in
the modern AIX sprang out of development work started for OSF/1. Saying that
AIX is a SVR3 derivative is like admiring the longevitiy of the proverbial
Hillbilly's axe, it's had the blade replaced twice and the handle three
times :-)
--
Nik Simpson
Tim Anderson
Nik Simpson <nik.s...@datacoresoftware.com> wrote in message
news:4PFf4.29$AY6.85@client...
>
> >
> > Yes, native X was a product of the '90s but by that time the
> > foundation had been pretty much frozen. At what release did the
> > network bits change? It might be fun to seriously "retro-port"
> > Apache to see if the kernel bug was in the "old" stack.
> >
>
> The work was going on C1991 when I moved from the UK office to Huntsville,
I
> think that corresponds with the release 5 kernel. IIRC, the work was being
> done in order to reach the Navy's FTP throughput requirements of the
NAVFAC
> contract, certainly FTP performance was a big issue during those tests.
>
The network core was replaced for NAVSEA. I know this because I managed
those parts during the benchmark and acceptance tests. The point to point
transfer rate went from ~0.6Mbs to ~2.0Mbs. There was another round of
revamping the code and by the time NAVAIR/SPAWAR rolled around the point to
point transfer rate was IIRC ~6.0Mbs. Overall the early 90's saw the
transfer speed on the CLIX boxes improve by nearly an order of magnitude.
Tim Anderson
Carl R. Friend
>
> My resollection (anyone feel free to take pot shots at this) is that
> only Sun, of all the original commercial Unix licensees, has a
> grandfathered deal they negotiated with AT&T to provide per-machine
> licenses with unlimited users. That's why CLIX (and other System V
> derivatives like AIX, etc) come with limititations on the number of
> simultaneous logins.
Interestingly, Intergraph's policy changed mid-stream in the Clipper
years fron an unlimited number of simultaneous users to a "licensed"
number. Sadly, the number picked, and the way it shook out in practise,
caused all sorts of problems in administration of the systems. The
break happened when the first 2000 (pizza box) machines came out.
To Intergraph's credit, they did "grandfather" the older CLIX
machines. However, the change was a source of infuriation for any
sysadmin who had to deal with CLIX machines.
> It's one of the reasons for the growing popularity of Linux.
Yep.
> If you check Intergraph's Web site, there's an explanation that the
> Clipper systems have a hardware/firmware Y2K problem, which is
> ostensibly corrected by the most recent CLIX nucleus. (There are
> writeable control stores for not only the CPU but several subsystems
> that get set at boot time.)
Clippers don't have writable control stores, at least in the micro-
code sense. Some of the later ones _may_ have had flash boot-code
memories (I was long gone by that time), but none of the ones I've
seen do.
My comment on ":0" returns for a "date +%y" is valid for the
release that Intergraph certified as "Y2k compliant". I smell BSD
in the family tree as all the AT&T variants return "00" and all the
_known_ BSD variants return ":0". The "fun thing" with ":0" is that
it sorts correctly _unless_ one does the _right_thing_ and sorts
numerically, in which case :0 breaks horribly. Go figure.
Carl R. Friend
>
> The network core was replaced for NAVSEA. I know this because I
> managed those parts during the benchmark and acceptance tests.
This is going to sound truly sick, but at what kernel release and/
or date did the stack replacement take place?
I'm still curious about my Apache experiments (which even with a
clean compile generate a kernel panic with the last released kernel)
and if an older stack would exhibit such a problem.
I'm just ill enough to try and back off to a 1991, or earlier,
kernel to try it....
Chris Bailey
clipper technology, why would Intergraph NOT release the programming
info into the public domain? At one time, field service was basically
supporting Intergraph through maintenance contracts. Can you imagine
how much money they could make by putting even a fraction of the
existing machines back on contracts?
Someone might be able to finally answer this question for me as well.
Rumors had it that there were two projects that never saw the light of
day. One was a port of the CLIX OS and API / Applications to the Intel
platform. The other was a version of Winnt that would run on the CLIX
platform. Does anyone here have more info on this? If there is any
truth to the rumors, wonder what it would take to entice Intergraph
into releasing some of this out into Public Domain? Just think, even
the source for even a start at porting Winnt for CLIX would help out
TONS for the porting of LINUX to CLIX.
Off topic here, but does anyone know where I can get rid of about 15
fully operational IP2700's?? Hate to dumpster them, but nobody seems
to want to even come get them for nothing. Picked up about 35 systems
from a State Surplus auction and already made my money back. Now just
want to clean out the garage so I can buy more stuff from the next
auction :>
On Fri, 14 Jan 2000 08:23:17 -0600, "Tim Anderson"
<moc.snoitulosgu@srednart> wrote:
>
>Nik Simpson <nik.s...@datacoresoftware.com> wrote in message
>news:4PFf4.29$AY6.85@client...
>>
>> >
>> > Yes, native X was a product of the '90s but by that time the
>> > foundation had been pretty much frozen. At what release did the
>> > network bits change? It might be fun to seriously "retro-port"
>> > Apache to see if the kernel bug was in the "old" stack.
>> >
>>
>> The work was going on C1991 when I moved from the UK office to Huntsville,
>I
>> think that corresponds with the release 5 kernel. IIRC, the work was being
>> done in order to reach the Navy's FTP throughput requirements of the
>NAVFAC
>> contract, certainly FTP performance was a big issue during those tests.
>>
>
>The network core was replaced for NAVSEA. I know this because I managed
Art Marriott
>
> Someone might be able to finally answer this question for me as well.
> Rumors had it that there were two projects that never saw the light of
> day. One was a port of the CLIX OS and API / Applications to the Intel
> platform.
That's news to me.
The other was a version of Winnt that would run on the CLIX
> platform. Does anyone here have more info on this?
Yeah, that was at one time at least under consideration. At least it
was *possible*...Microsoft had ordained that NT would only be ported to
big-endian processor architectures. That excluded Sun but left the
RS6000, the MIPS chip (SGI), the DEC Alpha, and presumably the Clipper.
At the time, Microsoft didn't want the world to think that NT was meant
"only for PC's". Since then, as the Intel chips have gotten faster and
the marketplace has come to accept hardware as a commodity, they've
pulled the plug on the other ports one at a time.
Actually, at the time Intergraph made public their intention to move
everything to NT, there was a plan put in motion have Howard Sachs and
the rest of the Clipper processor group go to work for Sun (they just
moved to a Sun facility across the parking lot from their old digs in
Palo ALto) and take with them some design stuff to go into the
UltraSparc, one of which was a "byte-sex switch". This would have
allowed the next generation of Sun hardware to handle an NT port.
Somewhare along the way, that fell by the wayside.
Just think, even
> the source for even a start at porting Winnt for CLIX would help out
> TONS for the porting of LINUX to CLIX.
That's probably even less likely than releasing CLIX for free.
Microsoft has retained proprietary rights to each and every detail of
NT's design. Anyone who gave that away would no doubt be dragged away
in the middle of the night to some dungeon in Redmond and subjected to
all manner of tortures.
>
Carl R. Friend
>
> > The other was a version of Winnt that would run on the CLIX
> > platform. Does anyone here have more info on this?
>
> Yeah, that was at one time at least under consideration. At least it
> was *possible*...Microsoft had ordained that NT would only be ported
> to big-endian processor architectures. That excluded Sun but left the
> RS6000, the MIPS chip (SGI), the DEC Alpha, and presumably the
> Clipper.
In fact some of Microsoft's own doco which was issued with NT 3.5
and 3.5.1 actually listed the Intergraph Clipper as one of the
"compatible" architectures. Whether they ever saw it run or not is
open to speculation.
I'll see if I can dig up some of the really old stuff to verify
that memory.
> > Just think, even the source for even a start at porting Winnt
> >for CLIX would help out TONS for the porting of LINUX to CLIX.
>
> That's probably even less likely than releasing CLIX for free.
> Microsoft has retained proprietary rights to each and every detail of
> NT's design. Anyone who gave that away would no doubt be dragged away
> in the middle of the night to some dungeon in Redmond and subjected to
> all manner of tortures.
Or, more properly, be dragged to the deepest dungeon in Maynard,
MA and have unthinkable things done to him. Oh, wait a minute,
they already blew that chance. :-)
Nik Simpson
> Just my two cents here, but other than the lawsuit with Intel over the
> clipper technology, why would Intergraph NOT release the programming
> info into the public domain? At one time, field service was basically
> supporting Intergraph through maintenance contracts. Can you imagine
> how much money they could make by putting even a fraction of the
> existing machines back on contracts?
What on earth makes you think that there are customers out there desperate
to put CLIX boxes back on maintenance:-) HEck if they want a UNIX box to
play with, they can pick up a LINX/x86 box for less than a years maintenance
on a CLIPPER/CLIX box. In addition, I'd strongly suspect that field service
has reduced the capability to support CLIX boxes (in terms of spare parts
and expertize) back to a minimum level consistent with the number of boxes
still in active use by customers. The last thing they would want would be a
bunch of people asking to put such boxes back on maintenance. I know it's
hard to accept, but as far as Intergraph is concerned CLIPPER and CLIX are
something from the dim and distant past, like InterAct 68Ks.
>
> Someone might be able to finally answer this question for me as well.
> Rumors had it that there were two projects that never saw the light of
> day. One was a port of the CLIX OS and API / Applications to the Intel
> platform.
To the best of my knowledge this was never even discussed and I was privy to
some pretty harebrained discussions in the first half of the nineties :-)
> The other was a version of Winnt that would run on the CLIX
> platform.
This has an element of truth in it. Part of the original deal between Ingr
and MS was that Ingr could put NT on a CLIPPER platform. The platform it
would run on, bore no resemblance to any CLIX machine ever shipped. The
prototypes of the CLIPPER/NT platform were publically demonstrated at AEC
tradeshow in Atlanta in 1993. They were very "alpha", NT 3.5 didn't so much
run on them as stumble drunkenly from BSOD to BSOD. That was about as
advanced as the port ever got. The hardware had a PC bus structure (EISA)
and was packaged inside a 2xxx series chassis. There were a number of issues
that were never satisfactorily resolved, these included:
1. Hardware paging management, the low level paging algorithms of NT at that
time matched quite closely the hardware architecture of the MIPS 3000/4000
CPU, this was radically different to CLIPPER and lead to serious performance
issues.
2. Microsoft had no intention of porting any of its applications to the
CLIPPER/NT platform (just like they didn't for other RISC platforms)
3. Intergraph by that time had recognised that it did not make financial
sense to continue throwing money at CPU development for a platform that
would, in a good year, sell maybe 20K CPUs.
4. The CLIPPER C400 was rapidly being overtaken in performance by Intel x86
processors like the 66MHz Pentium launched in 1993.
With these and other issues in mind, the CLIPPER/NT port was inevitably
stillborn. However at the time the need for a RISC based platform was still
something that seemed important. So when the CLIPPER development on C500
architectures was canned, the engineers from APD (Intergraph's Advanced
processor Division) were effectively sold to Sun. The original intent was to
develop a byte sex switchable SPARC architecture that could host a SPARC/NT
port. As part of the deal Intergraph would then sell these SPARC/NT boxes.
The problems with this included:
1. Doing anything at Sun with NT was a political minefield, a situation that
didn't improve with time.
2. The work was being done on a 32bit SPARC architecture that was rapidly
being abandoned in favor of the 64bit UltraSPARC by Sun.
3. Assuming everything went perfectly, at least problems 2&3 from the
CLIPPER/NT port still existed and arguably problem 4 was still an issue,
particularly for a 32bit SPARC platform.
4. By the time anything started to happen on this, Intergraph Computer
Systems was aggressively pursuing the idea of x86 based workstation
products.
With both Sun and Intergraph lukewarm about the whole thing, it came as no
great surprise when Sun canned the project before it even got as far as the
CLIPPER/NT platform.
>Does anyone here have more info on this? If there is any
> truth to the rumors, wonder what it would take to entice Intergraph
> into releasing some of this out into Public Domain?
The source for a CLIX/x86 or CLIPPER/NT port would have exactly the same
copyright ownership problems as the CLIPPER/CLIX port.
> Just think, even the source for even a start at porting Winnt for CLIX
would help out
> TONS for the porting of LINUX to CLIX.
Unfortuantely, it wouldn't! Because even if it were something that could be
released, the hardware architecture of the target for the CLIPPER/NT port
was radically different to any CLIX/CLIPPER platform. About the only thing
they had in common was the processor, the bus structure and everything else
was totally different.
--
Nik Simpson
Chris Bailey
wrote:
>
>"Chris Bailey" <cbailey...@cander.net> wrote in message
>news:387fc37f.949660@news-server...
>> Just my two cents here, but other than the lawsuit with Intel over the
>> clipper technology, why would Intergraph NOT release the programming
>> info into the public domain? At one time, field service was basically
>> supporting Intergraph through maintenance contracts. Can you imagine
>> how much money they could make by putting even a fraction of the
>> existing machines back on contracts?
>
>What on earth makes you think that there are customers out there desperate
>to put CLIX boxes back on maintenance:-) HEck if they want a UNIX box to
>play with, they can pick up a LINX/x86 box for less than a years maintenance
>on a CLIPPER/CLIX box. In addition, I'd strongly suspect that field service
>has reduced the capability to support CLIX boxes (in terms of spare parts
>and expertize) back to a minimum level consistent with the number of boxes
>still in active use by customers. The last thing they would want would be a
>bunch of people asking to put such boxes back on maintenance. I know it's
>hard to accept, but as far as Intergraph is concerned CLIPPER and CLIX are
>something from the dim and distant past, like InterAct 68Ks.
The cost / support factor is certainly right on target. As an
ex-Intergraph field engineer, I should know. I'm honestly surprised
that the existing clix customers don't sue Intergraph over the lack of
support that they're getting. But on the other hand, the boxes are
dead solid compared to the Intel stuff. I've had customers forget
where there clix servers were located simply because nobody had to go
over and reboot it every day!
>
>>
>> Someone might be able to finally answer this question for me as well.
>> Rumors had it that there were two projects that never saw the light of
>> day. One was a port of the CLIX OS and API / Applications to the Intel
>> platform.
>
>To the best of my knowledge this was never even discussed and I was privy to
>some pretty harebrained discussions in the first half of the nineties :-)
>
>> The other was a version of Winnt that would run on the CLIX
>> platform.
>
>This has an element of truth in it. Part of the original deal between Ingr
>and MS was that Ingr could put NT on a CLIPPER platform. The platform it
>would run on, bore no resemblance to any CLIX machine ever shipped. The
>prototypes of the CLIPPER/NT platform were publically demonstrated at AEC
>tradeshow in Atlanta in 1993. They were very "alpha", NT 3.5 didn't so much
>run on them as stumble drunkenly from BSOD to BSOD. That was about as
>advanced as the port ever got. The hardware had a PC bus structure (EISA)
>and was packaged inside a 2xxx series chassis. There were a number of issues
>that were never satisfactorily resolved, these included:
>
The Clipper to Intel conversion was finally accomplished. It was
called a Phoenix. It allowed the customer to take either an IP2400,
2700, 2800, 6400, 6700, or 6800 and convert to a Pentium 100. About
the only nice thing about this was it allowed you to use your existing
peripherals, memory and monitor to run Winnt 3.51. The IP2900 / 6900
was never certified for NT 4, so it died REAL fast. Most people that
have these critters wish they'd never spent the money on them.
What I was thinking of was more along the lines of just the hardware
routines and not the OS routines. This would help out with the basic
op codes and i/o routines that seem to be the sticking points in
getting LINUX up and running on the clix boxes.
Nik Simpson
> >
> >This has an element of truth in it. Part of the original deal between
Ingr
> >and MS was that Ingr could put NT on a CLIPPER platform. The platform it
> >would run on, bore no resemblance to any CLIX machine ever shipped. The
> >prototypes of the CLIPPER/NT platform were publically demonstrated at AEC
> >tradeshow in Atlanta in 1993. They were very "alpha", NT 3.5 didn't so
much
> >run on them as stumble drunkenly from BSOD to BSOD. That was about as
> >advanced as the port ever got. The hardware had a PC bus structure (EISA)
> >and was packaged inside a 2xxx series chassis. There were a number of
issues
> >that were never satisfactorily resolved, these included:
> >
> The Clipper to Intel conversion was finally accomplished. It was
> called a Phoenix. It allowed the customer to take either an IP2400,
> 2700, 2800, 6400, 6700, or 6800 and convert to a Pentium 100. About
> the only nice thing about this was it allowed you to use your existing
> peripherals, memory and monitor to run Winnt 3.51. The IP2900 / 6900
> was never certified for NT 4, so it died REAL fast. Most people that
> have these critters wish they'd never spent the money on them.
>
Yeah, it would have been cheaper to give every customer a replacement Intel
box, Phoenix was a marvellous feat of engineering but it was a dumb idea.
> >>Does anyone here have more info on this? If there is any
> >> truth to the rumors, wonder what it would take to entice Intergraph
> >> into releasing some of this out into Public Domain?
> >
> >The source for a CLIX/x86 or CLIPPER/NT port would have exactly the same
> >copyright ownership problems as the CLIPPER/CLIX port.
> >
> >> Just think, even the source for even a start at porting Winnt for
CLIX
> >would help out
> >> TONS for the porting of LINUX to CLIX.
>
> What I was thinking of was more along the lines of just the hardware
> routines and not the OS routines. This would help out with the basic
> op codes and i/o routines that seem to be the sticking points in
> getting LINUX up and running on the clix boxes.
The point I'm trying to make is that the box that NT was ported to had no
relationship hardware wise (with the exception of the CPU) to a CLIX/CLIPPER
platform that a customer would have seen. There was never any intent to
offer NT on 2xxx or 6xxx CLIPPER platforms. Had an NT box been shipped it
would have a radically different hardware architecture, so even if the
routines you speak of:
1. Still exist, and I sincerely doubt that they do, after all there would
have been no point in keeping them
2. Could be released without bringing Microsoft lawyers down on Intergraph's
head.
They would not provide any useful pointers to someone trying to port LINUX
to a conventional CLIPPER platform since they had a different memory
architecture, different I/O bus, different disk I/O, networking & graphics
implementations.
--
Nik Simpson
Carl R. Friend
>
> I'm honestly surprised that the existing clix customers don't sue
> Intergraph over the lack of support that they're getting. But on
> the other hand, the boxes are dead solid compared to the Intel stuff.
My take on this is that long-time Intergraph customers came to
expect this sort of thing from Intergraph. Ingr did _exactly_ the
same thing in the pdp11 to VAX cutover (it wasn't a migration) and then
did it again in the VAX to Clipper jump. That there was a decisive
"disconnect" between Clipper and Intel surprised nobody. At least
nobody who had studied the history of the company.
> I've had customers forget where there clix servers were located
> simply because nobody had to go over and reboot it every day!
But that's what a server is supposed to do. No, wait a minute - if
it doesn't continually crash you don't need all the admin staff to
look after it. It's all about _employment_, for crying out loud.
(I'll remove my tongue from my cheek now....)
> The Clipper to Intel conversion was finally accomplished. It was
> called a Phoenix.
I've heard rumours about these things. Did they really exist in
any substansive sense? NT 3.5 and 3.5.1 were absolutely horrid
things which noone sane would willingly run, so why would anyone
trade in a good functional CLIX server (or WS) for such a thing?
> What I was thinking of was more along the lines of just the hardware
> routines and not the OS routines. This would help out with the basic
> op codes and i/o routines that seem to be the sticking points in
> getting LINUX up and running on the clix boxes.
At least in the earliest systems, there was a lot of extra code
in the ROP and the IOP. Certainly Intergraph could release those
bits as they were completely proprietary to them. However, I doubt
that anyone would even try to do a Linux port to such a system; the
target would, almost inevitably, be a 2xxx or 6xxx series machine.
The actual Clipper instruction set and memory-management bits
were released quite a while ago (in a book with an ISBN) which was
originally done up by Fairchild (I have a copy). Of course it doesn't
include the instructions added in the C3 and C4 chipsets, but is a
decent source of information; what is completely lacking is the doco
on the workings of all the glue logic and controllers on the Intergraph
motherboards. I suspect that with full disclosure on the hardware
workings, a Linux port could get done (or a port of most anything
else written in a HLL). Worth it? It depends on who you are.
Nik Simpson
> > called a Phoenix.
>
> any substansive sense? NT 3.5 and 3.5.1 were absolutely horrid
> things which noone sane would willingly run, so why would anyone
> trade in a good functional CLIX server (or WS) for such a thing?
They certainly existed at Intergraph, I also beleive there were some
customers foolish enough to buy the "upgrade", go figure!
--
Nik Simpson
Tim Anderson
> Tim Anderson wrote:
> >
> > The network core was replaced for NAVSEA. I know this because I
> > managed those parts during the benchmark and acceptance tests.
>
> or date did the stack replacement take place?
I haven't a clue. It was fun to be involved with those projects, but those
little details have long since exited my cranium ;-)
Tim
paras...@my-deja.com
eye...
An Interact 340? Now there's a step into the
wayback machine Sherman. From what I recall it
would still be one of the ones with an IO Processor
running a 80386, with a seperate C100 CPU board (or
maybe even C300 CPU). Graphics would be on another
board still, two I believe for a dual screen
interact. Alot of the low level networking stack
would have run off the IOP also. Had we given up
on booting DOS off the IOP by then like you could
on the early IPros? You could use the Interact to
run DOS 3 if it did.
That IOP also had a parallel port that was used for
driving plotters. Don't suppose you have a 1988
vintage pen plotter laying around you need it to
drive?
I'm ex Intergraph. I worked on CLIX from 1986 to
1990 before leaving for the West Coast. Worked on
the CLIX kernel: Boot code and startup Blue Screen
code, UFS, NFS, RFS & some driver work and security
stuff.
It was fun work, but the whole IPro/Interact/
Interview setup was a wierd architecture. A
novelty item on a road littered with failed
workstation vendor wannabes. A bootable IO
processor on a workstation? Early versions of
CLIX, I can't recall if this one did or not, loaded
CLIX out of seperate disk partition. Thats where
the bootloader, IOP, & GP/Frame processor code
loaded from. It was 5.3 based, with a whole bunch
of 4.[23] BSD thrown in. They worked fine if you
didn't swap or page, but almost every INGR app
Why did Intergraph do anything the way they did
(and maybe still do)?
IMO, It was fun being way in the middle of North
Alabama doing those wacky NIH designs, but it lead
to a feeling of insulation and invulnerability that
was pure fantasy. I recall one particular meeting
w/a VP (in the test room where the Interact 340s
were kept in B11) where he boasted of all the West
Coast nerd blood that we were going to be letting
in the upcoming battle with Sun!
Heh. The blood flowed, but it wasn't out West.
At the same time, Intergraph never seemed like it
wanted to be in competition with anyone. Just earn
high margins off fat contracts w/the gov't and
fortune 100 companies. Hence the plethora of
expensive machines w/wierd architectures.
Anybody remember the Object File Processor? Or
FMU?
Roger
Sent via Deja.com http://www.deja.com/
Before you buy.
Chris Bailey
<carl....@prescienttech.com> wrote:
>Chris Bailey wrote:
>>
>> I'm honestly surprised that the existing clix customers don't sue
>> Intergraph over the lack of support that they're getting. But on
>> the other hand, the boxes are dead solid compared to the Intel stuff.
>
> My take on this is that long-time Intergraph customers came to
>expect this sort of thing from Intergraph. Ingr did _exactly_ the
>same thing in the pdp11 to VAX cutover (it wasn't a migration) and then
>did it again in the VAX to Clipper jump. That there was a decisive
>"disconnect" between Clipper and Intel surprised nobody. At least
>nobody who had studied the history of the company.
Yeah, but it's the same customers who got shafted by the VAX to Clix,
that got shafted the worst on the Clix to NT "conversion". Part of
this was Bentley not always honoring the existing Clix versions of
Microstation when it came time to "trade up". The worst is that
Intergraph is still charging the same exact rates for software
maintenance to the remaining Clix customers that they did when the
Clix was first sold. What remains of their support is nothing like it
first was. _THAT_ is why I'm surprised that there's been no lawsuit.
>
>> I've had customers forget where there clix servers were located
>> simply because nobody had to go over and reboot it every day!
>
> But that's what a server is supposed to do. No, wait a minute - if
>it doesn't continually crash you don't need all the admin staff to
>look after it. It's all about _employment_, for crying out loud.
>(I'll remove my tongue from my cheek now....)
>
>> The Clipper to Intel conversion was finally accomplished. It was
>> called a Phoenix.
>
> I've heard rumours about these things. Did they really exist in
>any substansive sense? NT 3.5 and 3.5.1 were absolutely horrid
>things which noone sane would willingly run, so why would anyone
>trade in a good functional CLIX server (or WS) for such a thing?
Yes, they do exist. I had one customer that paid to upgrade 5 of
their systems to the IP2930 (phoenix). They are definitely NOT happy
with the systems because they were promised support for them as well
as future upgradeability. This has NOT happened and they're about
ready to trash them and buy IBM pc's.
>
>> What I was thinking of was more along the lines of just the hardware
>> routines and not the OS routines. This would help out with the basic
>> op codes and i/o routines that seem to be the sticking points in
>> getting LINUX up and running on the clix boxes.
>
> At least in the earliest systems, there was a lot of extra code
>in the ROP and the IOP. Certainly Intergraph could release those
>bits as they were completely proprietary to them. However, I doubt
>that anyone would even try to do a Linux port to such a system; the
>target would, almost inevitably, be a 2xxx or 6xxx series machine.
Mainly what I was considering was a Linux port to the Server series
and not the workstation. The code dealing with the graphics would be
tough. Just Linux running on a server would be great. At least I'd
have some kind of justification for not dumpstering all of my old
stuff.
>
> The actual Clipper instruction set and memory-management bits
>were released quite a while ago (in a book with an ISBN) which was
>originally done up by Fairchild (I have a copy). Of course it doesn't
>include the instructions added in the C3 and C4 chipsets, but is a
>decent source of information; what is completely lacking is the doco
>on the workings of all the glue logic and controllers on the Intergraph
>motherboards. I suspect that with full disclosure on the hardware
>workings, a Linux port could get done (or a port of most anything
>else written in a HLL). Worth it? It depends on who you are.
I also have a copy of this book. It's old and not very detailed. It
also doesn't cover all of the other stuff like I/O and so on. So,
theoritically someone could use it to code routines for Linux, NT or
whatever, but it still wouldn't help tremendously.
Nik Simpson
<paras...@my-deja.com> wrote in message
news:86ea2a$482$1...@nnrp1.deja.com...
> Hey there, just passing thru when this caught my
> eye...
>
> An Interact 340? Now there's a step into the
> wayback machine Sherman. From what I recall it
> would still be one of the ones with an IO Processor
> running a 80386, with a seperate C100 CPU board (or
> maybe even C300 CPU).
Yup, that's the puppy. The C300 version though would have been a 3000
series. Came in two flavors, the Interpro version that looked like a large
Samsonite suitcase and the dual screen Interact that used something very
close to the Interact 68K chassis. Both of them were built like tanks, I
remember taking the side of my 3050 once and being stunned by the 1/4"
armour plate covering the internals held down by about 50 fasteners. I think
these things were used as impromptu tornado shelters in Huntsville.
>Graphics would be on another
> board still, two I believe for a dual screen
> interact. Alot of the low level networking stack
> would have run off the IOP also. Had we given up
> on booting DOS off the IOP by then like you could
> on the early IPros? You could use the Interact to
> run DOS 3 if it did.
All the I/O functions were implemented on the IOP board IIRC, networking and
disk I/O and yes I think the 340 was still supposed to be able to run DOS 3,
a neat party trick, but of little practical value in my opinion.
>
> It was fun work, but the whole IPro/Interact/
> Interview setup was a wierd architecture.
It got less weird after you left, to 2xxx and 6xxx families were relatively
mundane from an architectural standpoint.
> A novelty item on a road littered with failed
> workstation vendor wannabes. A bootable IO
> processor on a workstation? Early versions of
> CLIX, I can't recall if this one did or not, loaded
> CLIX out of seperate disk partition. Thats where
> the bootloader, IOP, & GP/Frame processor code
> loaded from. It was 5.3 based, with a whole bunch
> of 4.[23] BSD thrown in. They worked fine if you
> didn't swap or page, but almost every INGR app
The paging problems were inherent with SVR 5.3.1 since it had no concept of
demand pagaing and did everything by swapping. These problems were largely
solved when the SVR 3.2 demand paging code was bolted onto CLIX around the
time you left.
>
> Why did Intergraph do anything the way they did
> (and maybe still do)?
>
> IMO, It was fun being way in the middle of North
> Alabama doing those wacky NIH designs, but it lead
> to a feeling of insulation and invulnerability that
> was pure fantasy. I recall one particular meeting
> w/a VP (in the test room where the Interact 340s
> were kept in B11) where he boasted of all the West
> Coast nerd blood that we were going to be letting
> in the upcoming battle with Sun!
That would have to have been Bruce Imsand or Howard Sachs, both of whom were
capable of hubris of quite enormous proportions, a character trait inherited
by their successor Wade Patterson.
>
> Anybody remember the Object File Processor? Or
> FMU?
Yup, remember them all, although I have to say that a few years of therapy
and I'm almost OK now :-)
--
Nik Simpson
Chris Bailey
<nik.s...@datacoresoftware.com> wrote:
>
><paras...@my-deja.com> wrote in message
>news:86ea2a$482$1...@nnrp1.deja.com...
>> Hey there, just passing thru when this caught my
>> eye...
>>
>> An Interact 340? Now there's a step into the
>> wayback machine Sherman. From what I recall it
>> would still be one of the ones with an IO Processor
>> running a 80386, with a seperate C100 CPU board (or
>> maybe even C300 CPU).
>
>Yup, that's the puppy. The C300 version though would have been a 3000
>series. Came in two flavors, the Interpro version that looked like a large
>Samsonite suitcase and the dual screen Interact that used something very
>close to the Interact 68K chassis. Both of them were built like tanks, I
>remember taking the side of my 3050 once and being stunned by the 1/4"
>armour plate covering the internals held down by about 50 fasteners. I think
>these things were used as impromptu tornado shelters in Huntsville.
>
Nope, the _REAL_ story here is that some VP's brother-in law owned a
screw factory. That's why there were so many screws. As far as sheet
metal, that was Intergraph's unique brand of "Theft Deterrent". If
something turned up missing, then all you had to do was to look for
the person at the emergency room with a double hernia. But if that
person had an Intergraph ID, then they were a Field Engineer and not a
thief. Field Engineers knew better than to try to steal one of these
critters.
Arthur R. Marriott
> >remember taking the side of my 3050 once and being stunned by the 1/4"
> >armour plate covering the internals held down by about 50 fasteners. I think
> >these things were used as impromptu tornado shelters in Huntsville.
> >
> Nope, the _REAL_ story here is that some VP's brother-in law owned a
> screw factory. That's why there were so many screws.
That, I'm ABSOLUTELY willing to believe!!!
> As far as sheet
> metal, that was Intergraph's unique brand of "Theft Deterrent". If
> something turned up missing, then all you had to do was to look for
> the person at the emergency room with a double hernia. But if that
> person had an Intergraph ID, then they were a Field Engineer and not a
> thief. Field Engineers knew better than to try to steal one of these
> critters.
I've never tried to weigh them, but I'm sure the 3040 I have in the basement
wiegh more than the washing machine in the next room. For that matter, the
latter on spin cycle is quieter than those damn Maxtor disk drives.
Chuck Rheault
kit that allowed you to switch between a 2000/6000 and an Intel, using
the Intergraph monitors and digitizers for both CPU's. They also had a
version that converted a 200/300 series InterAct or InterView into an
NT/DOS box that used the Intergraph monitors and digitizers (but
scrapped any other Clix hardware. When I was at Md. SHA, we looked into
them but could not get procurement to go along with the idea. They sent
out a demo which we used for almost 3 months without any problems. I
never found a better dual screen setup with a large backlit digitizer
than the old InterView's. The "skyhook" monitor brackets brought the
screen to just the right distance from the eyes. Now, even with my
glasses I'm having difficulty seing the menu's and dialog boxes at our
digitizer, where the monitors are completely behind the tablet.
--
Chuck Rheault
Greenman-Pedersen, Inc.
Laurel, Maryland Branch
(Formerly with Md. DOT SHA (18+ Yrs.))