Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

IRIX history?

121 views
Skip to first unread message

Khamba Staring

unread,
Apr 29, 2001, 7:59:43 AM4/29/01
to
Hi,

Does anyone know a good (online?) document about
the history of IRIX? I.e. what it was derived from &
advances over time?


Regards,

-- Khamba Staring

Gerhard Lenerz

unread,
Apr 29, 2001, 9:18:41 AM4/29/01
to
On Sun, 29 Apr 2001 13:59:43 +0200, Khamba Staring wrote:
> Does anyone know a good (online?) document about
> the history of IRIX? I.e. what it was derived from &
> advances over time?

Thats a good question! Searching google one gets the impression that
IRIX did not exist before 4.0.x. There are few sources that mention 3.x
and I found only one that states that the first IRIX version dates back
to the early 80s.


Gerhard:wq

Chris

unread,
Apr 29, 2001, 12:43:55 PM4/29/01
to
Didn't IRIX derive from MIPS OS? I have heard of IRIX 3.2, but no
earlier, I would presume that was from the late 80's because 4.02 came
out in '90, 5.1, 5.2 and 5.3 also came out I think, that was 1990-1995
sort of era, then the 6.xx things....

I really dont know many details, maybe someone could fill us in a bit
more...

Chris

Gerhard Lenerz wrote:


--
Chris Moxon
If you want to reply to this remove NOSPAM from the address.

Chris

unread,
Apr 29, 2001, 12:55:58 PM4/29/01
to
Here are all the irix's if anyone wants to add which year they were
released:
SGI IRIX 6.5.x
SGI IRIX 6.4
SGI IRIX 6.3
SGI IRIX 6.2
SGI IRIX 6.1
SGI IRIX 6.0
SGI IRIX 5.3
SGI IRIX 5.2
SGI IRIX 5.1
SGI IRIX 5.0
SGI IRIX 4.0
SGI IRIX 3.3
SGI IRIX 3.2

Walter Roberson

unread,
Apr 29, 2001, 1:37:06 PM4/29/01
to
In article <3AEC479E...@NOSPAM.orange.net>,
Chris <dr...@NOSPAM.orange.net> wrote:
:Here are all the irix's if anyone wants to add which year they were
:released:
:SGI IRIX 6.5.x
:SGI IRIX 6.4
:SGI IRIX 6.3
:SGI IRIX 6.2
:SGI IRIX 6.1
:SGI IRIX 6.0
:SGI IRIX 5.3
:SGI IRIX 5.2
:SGI IRIX 5.1
:SGI IRIX 5.0
:SGI IRIX 4.0
:SGI IRIX 3.3
:SGI IRIX 3.2

Those aren't all the IRIX's. IRIX 1.x and IRIX 2.x and IRIX 3.0 and
IRIX 3.1 all existed, but like I wrote in the Hardware FAQ
"What OS versions are supported on which platforms?",

Details up to IRIX 3.3 are largely lost in the
mists of time, as are details about when various platforms went out
of service.

I personally used IRIX 3.0 (but not, as I recall, IRIX 3.1), but
I haven't personally used IRIX 1.x or 2.x. There was a posting
a few weeks ago from someone hoping to find a copy of IRIX 2.2

If you are going to detail the histories, it is important to
distinguish the sub-releases of IRIX 4.0, as those occupied pretty
much the same historical status as if they were IRIX 4.1, 4.2 and so on.

Gerhard Lenerz

unread,
Apr 29, 2001, 1:15:33 PM4/29/01
to
On Sun, 29 Apr 2001 17:43:55 +0100, Chris wrote:
> Didn't IRIX derive from MIPS OS? I have heard of IRIX 3.2, but no
> earlier, I would presume that was from the late 80's because 4.02 came
> out in '90, 5.1, 5.2 and 5.3 also came out I think, that was 1990-1995
> sort of era, then the 6.xx things....

IRIX 3.2 is a version that turns up frequently^Won google when you
search for old irix versions trying things like "irix 3." as search
pattern.

> I really dont know many details, maybe someone could fill us in a bit
> more...

From what I have read recently 3.0 was shipped with the first IRISes and
as far as I remember the paper claimed that 3.1, 3.2 and 3.3 have been
released later with/for the 4d/1xx, 4d/2xx and 4d/3xx systems.

If that is true, than it might be safe to say that there is no IRIX <3.0
at last nothing like the IRIX/MIPS os we know today.

Gerhard

Gerhard Lenerz

unread,
Apr 29, 2001, 2:27:05 PM4/29/01
to
On Sun, 29 Apr 2001 17:55:58 +0100, Chris wrote:
> Here are all the irix's if anyone wants to add which year they were
> released:
> SGI IRIX 6.5.x
> SGI IRIX 6.4
> SGI IRIX 6.3

IRIX 6.2 All Impact - 1996 (August?)

> SGI IRIX 6.2

1996, 1st half (March?)

> SGI IRIX 6.1
> SGI IRIX 6.0
> SGI IRIX 5.3
> SGI IRIX 5.2

5.2 and 5.3 both 1994(?)

> SGI IRIX 5.1
> SGI IRIX 5.0

IRIX 4.0.xy
I am not sure but my guess would be somewhere 1991 +/- some
years because there have been various releases for new hardware
at that time (Indigo for example).

> SGI IRIX 4.0
> SGI IRIX 3.3
> SGI IRIX 3.2

IRIX 3.1
IRIX 3.0

Ger 'aaahh... history! *g*' hard:wq

Tomasz Korycki

unread,
Apr 29, 2001, 10:12:14 PM4/29/01
to
Gerhard Lenerz wrote:
>
> On Sun, 29 Apr 2001 17:55:58 +0100, Chris wrote:
> > Here are all the irix's if anyone wants to add which year they were
> > released:
> > SGI IRIX 6.5.x
> > SGI IRIX 6.4
> > SGI IRIX 6.3
>
> IRIX 6.2 All Impact - 1996 (August?)
>
> > SGI IRIX 6.2
>
> 1996, 1st half (March?)
>
> > SGI IRIX 6.1
> > SGI IRIX 6.0
> > SGI IRIX 5.3
> > SGI IRIX 5.2
>
> 5.2 and 5.3 both 1994(?)
>

Early 5.x came out rapidly one after another

> > SGI IRIX 5.1
> > SGI IRIX 5.0
>
> IRIX 4.0.xy
> I am not sure but my guess would be somewhere 1991 +/- some
> years because there have been various releases for new hardware
> at that time (Indigo for example).
>

4.0.5 came out in (methinks) late 1991. The 4.0.5IOP ("Indigo Only
Patch") was June 1992, at least that's when I got the tapes.

> > SGI IRIX 4.0
> > SGI IRIX 3.3
> > SGI IRIX 3.2

I first saw 3.2 at my College in 1990 - so it must have been released
earlier.

Dave Olson

unread,
Apr 29, 2001, 10:53:53 PM4/29/01
to
lene...@uni-trier.de (Gerhard Lenerz) writes:

| On Sun, 29 Apr 2001 17:43:55 +0100, Chris wrote:
| > Didn't IRIX derive from MIPS OS? I have heard of IRIX 3.2, but no
| > earlier, I would presume that was from the late 80's because 4.02 came
| > out in '90, 5.1, 5.2 and 5.3 also came out I think, that was 1990-1995
| > sort of era, then the 6.xx things....
|
| IRIX 3.2 is a version that turns up frequently^Won google when you
| search for old irix versions trying things like "irix 3." as search
| pattern.

That's because the 3.0 and 3.1 series weren't the most stable possible
releases ;) 3.0 was, as I recall (possibly incorrectly) from my early
days at SGI (which started with 3.1e or f), the first IRIX MIPS
release. It was for the R2300 (aka 4D/50), whose CPU board was
OEMed from MIPS.

3.1e or f was the launch release for the 4D100 (first MP system) and
the 4D/20 Personal IRIS. The 4D/20 launched in Oct 88.

The other pre-3.2 releases were presumably for the 4D/60, 70, 80
and their various graphics options, but it's been too long, I
just don't remember.

| From what I have read recently 3.0 was shipped with the first IRISes and
| as far as I remember the paper claimed that 3.1, 3.2 and 3.3 have been
| released later with/for the 4d/1xx, 4d/2xx and 4d/3xx systems.

That matches up with my memory.

| If that is true, than it might be safe to say that there is no IRIX <3.0
| at last nothing like the IRIX/MIPS os we know today.

I seem to recall that the name IRIX wasn't used on the 68k based systems
that preceded the 3.0 release. There are still a few folks at SGI who
were there at the time, and perhaps they'll post more details.

Dave Olson
ol...@bengaltech.com
http://www.bengaltech.com/public/dave

Al Smith

unread,
Apr 30, 2001, 3:54:13 AM4/30/01
to
Hi folks,

For some reason I've got a copy of SGI's "Hardware developers'
handbook" which lists this kind of information. I didn't find it on
techpubs.sgi.com, and it also doesn't seem to have a part number.

Unfortunately it only lists the dates when the various hardware came
into being; and later on there is a table which lists which version of
IRIX was supported on which platforms. I'll see if I can summarise:

This is undoubtedly inaccurate in some details, and some platforms are
missing entirely (notable the Power Challenge R8000s). It may also be
worth pointing out that there are platforms where the IP number is
different from the IP listed in hinv, for example the Origin 200
claims to be an IP27, but is in fact an IP29.

Hope this helps,

Al.

Year Model CPU types CPU Freq System IRIX versions

1986 Twin Tower MIPS Board R2300 ? 4D/60 IRIX 3.0 - 5.2
IP4 R2000 8MHz 4D/50 IRIX 3.0 - 5.3
IP4 R2000 12.5MHz 4D/70 IRIX 3.0 - 5.3
IP4.5 R2000 16.7MHz 4D/80 IRIX 3.0 - 5.3
1987 Power Series Twin Tower IP5 R2000 16.7MHz 4D/120 IRIX 3.1 - 5.3
IP7 R2000 25MHz 4D/200 IRIX 3.2 - 5.3
1988 Personal IRIS IP6 R2000 12.5MHz 4D/20 IRIX 3.1 - 5.3
IP10 R2000 20MHz 4D/25 IRIX 3.1 - 5.3
IP14 R3000 30MHz 4D/30 IRIX 3.3.2 - 5.3
1989 Power Series/Power Centre IP13 R3000 33MHz 4D/300 IRIX 3.2 -
5.3
Power Series Single Tower IP15 R3000 40MHz 4D/400 IRIX 3.3 - 5.3
1991 Crimson IP17 R4000 100MHz IRIX 4.0.3 - 6.2
Indigo IP12 R3000 33MHz IRIX 4.0 - 5.3
IP20 R4000 100MHz IRIX 4.0.5E - 6.2
1993 Onyx/Challenge Rack IP19 R4400 various IRIX 5.0 -
Onyx/Challenge Deskside IP19 R4400 various IRIX 5.0 -
Indigo 2 IP22 R4400 75MHz IRIX 4.0.5H -
Indy IP24 R4x00 100-150MHz IRIX 5.1 -
1996 Origin 2000/Onyx 2 IP27 R10000 195- MHz IRIX 6.4 -
Origin 200 IP29 R10000 195- MHz IRIX 6.4 -
O2 IP32 R5k,R10k 150- MHz IRIX 6.3MR -
1997 Origin 200 Deskside IP27 R10000 195- MHz IRIX 6.4 -
Octane IP30 R10000 ?MHz IRIX 6.4 -

Marcus Herbert

unread,
Apr 30, 2001, 10:45:41 AM4/30/01
to
Khamba Staring <pur...@edoropolis.n.o.s.p.a.m.org> wrote:
> Does anyone know a good (online?) document about
> the history of IRIX? I.e. what it was derived from &
> advances over time?

Only some cites from various documents:

"Another major milestone in the 1980's for computer graphics was the
founding of Silicon Graphics Inc. (SGI) by Jim Clark in 1982. SGI
focused its resources on creating the highest performance graphics
computers available. These systems offered built-in 3D graphics
capabilities, high speed RISC (Reduced Instruction Set Chip)
processors and symmetrical (multiple processor) architectures. The
following year in 1983, SGI rolled out its first system, the IRIS 1000
graphics terminal."

"There are older workstations that use M68K processors that may be
available for low prices. The Sun 3/60 is very popular in the
education field and the HP 9000 310 uses a 68010 and the 320 uses
a 68020 CPU. The NeXT cube has a 68030 and the SGI Iris 3000 uses
a 68020."

"In 1982, a group of Stanford University Graduates formed a company to
develop their research work on the Stanford University Network (SUN)
computer. This company was called Sun Microsystems. Its first system
was based on Stanford's Motorola 68000-based board. Another company
which gained immediate success specialised in 3-D graphic workstations.
This company was called Silicon Graphics Inc. (SGI) and released its
IRIS workstation in April 1984."

"In March 1986 Silicon Graphics Inc. announced its intension to switch
from the Motorola 68000 processor to a RISC processor from MIPS Computer
Systems Inc. MIPS Computer Systems Inc. was formed in 1984 to make a
commercial version of the MIPS project at Stanford University. The
MIPS project (named for the key phrase microcomputer without interlocked
pipeline stages) showed that pipelining, although a well-known technique,
had been underexploited by earlier CISC architectures. The commercial
processor was unveiled late in 1985 as the R2000. The R2010 floating-
point accelerator (FPA) first shipped in mid 1987. "

"By 1992 Silicon Graphics, Inc. (SGI) was the leading use of MIPS
processors for computer systems. When MIPS Corporation's system business
collapsed in March 1993 SGI was willing to step in to rescue the company
and the MIPS architecture."

So obviously the first version of IRIX appeared in early 1984 when SGI
released its first workstation (between 1982 and end of 1983 SGI was
only producing a terminal called IRIS 1000). The first version of IRIX
that i know of is IRIX 3.2 which runs on the personal IRIS of a friend
of mine. So i guess that IRIX 1.x and 2.x ran on the motorola based IRIS
3000. Maybe IRIX 1.x on the IRIS 2000 and IRIX 2.x on the IRIS 3000? :-}

Ah yes.. the OBpicturelinks :-}

http://www.schrotthal.de/sgi/iris_1000.jpg
http://www.schrotthal.de/sgi/iris_3031.jpg
http://www.schrotthal.de/sgi/iris_3115_front.jpg

Heck! There are better pictures available from mars than from those old
SGI machines! I looked for quite some time to find at least these :-/

--
PGP2 Key-ID: 666/36540865 1997/06/09 <rho...@spam-filter.de>
GPG Key-ID: 1024D/2E2DAB44 2000-01-30 <rho...@spam-filter.de>
Geek-Code: GCS b O e+ h

Chris

unread,
Apr 30, 2001, 12:05:39 PM4/30/01
to
Here's the info I already have on the IRIS series machines:

I would presume that an early version of IRIX was running on these
boxes, as Marcus Herbert said.

IRIX 1.x on IRIS 1000s and IRIS 2000 Non turbos. The IRIS 2000 Turbos
and IRIS 3000 would I think run IRIX 2.X. When silicon graphics started
using MIPS for the Power Series /Professional series, MIPS OS, (which
was being developed by someone else) was bought off of them and
Intergrated into IRIX. (This is mere theory). This formed IRIX 3.X, and
note that MIPS OS was based a lot around BSD, so I would presume that
the original IRIX was based around SystemV, and thats what led to in
version 3.X SystemV with BSD extensions.

Specification:

1000/1200:

8MHz Motorola 68000 - PM1 (variant of Stanford UNiversity "SUN" board)
3 -4 MB RAM Micro Memory Inc. Multibus
No Hard Disk
Ethernet: Excelan EXOS/101
Graphics: GF1 Frame Buffer
UC3 Update Controller
DC3 Display Controller
BP2 Bitplane
1400:

10MHz Motorola 68010 - PM2
1.5Meg of RAM - PM2M1
72Meg ST-506 Hard disk
Ethernet: Excelan EXOS/101
Graphics: GF1 Frame Buffer
UC3 Update Controller
DC3 Display Controller
BP2 Bitplane
1500:

10 MHz Motorola 68010 - PM2
1.5Meg of RAM - PM2M1
474 Meg ST-506 Hard disk
Ethernet: Excelan EXOS/101
Graphics: GF1 Frame Buffer
UC3 Update Controller
DC3 Display Controller
BP2 Bitplane
The standard monitor shipped was 30Hz interlaced.

no user-accessible OS.

Support Ended: November 1989


I managed to compile some info about IRIS 2000s too.

2000 (Non Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: PM2M1
CPU: PM2 (68010)
FPU: SKYFFP-M-03
No HDD


2200 (Non Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: PM2M1
CPU: PM2 (68010)
FPU: SKYFFP-M-03
No HDD


2300 (Non Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: PM2M1
CPU: PM2 (68010)
FPU: SKYFFP-M-03
ST-506 Hard Drive


2400 (Non Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: PM2M1
CPU: PM2 (68010)
FPU: SKYFFP-M-03
ST-506 Hard Drive
2500 (Non Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: PM2M1
CPU: PM2 (68010)
FPU: SKYFFP-M-03
SMD Hard Drive


2300T (Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: IM1
IP2 (68020) 12MHz
FPU: FP1 (Optional)
ST-506 Hard Drive


2400T (Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: IM1
IP2 (68020) 12MHz
FPU: FP1 (Optional)
ST-506 Hard Drive


2500T (Turbo)

Ethernet: Excelan EXOS/201
Graphics:
GF2 Frame Buffer
UC4 Update Controller
DC4 Display Controller
BP3 Bitplane
RAM: IM1
IP2 (68020) 12MHz
FPU: FP1 (Optional)
SMD Hard Drive


NOTES:

Non-Turbo partitioning is unusual!

The only hardware difference between a non-Turbo system and its Turbo
counterpart is the CPU and RAM. For this reason, Turbo was also offered
as an upgrade path from the original 2400

The 2500 and 2500T are packaged differently as well. They live in a 6'
tall 19" EIA rack. There is room for two SMD drives in the bottom of the
rack (after all, they weigh about 150lb. each), and the electronics
chassis resides in the top half

Turbo and non-Turbo RAM cannot be mixed. The IP2 introduced a new
(ribbon-cable) local bus through which it communicated with the IM1(s)
and FP1. The PM2 simply used the Multibus.

In short, the Turbos have much more in common with the 3000s than they
do with the non-Turbo 2000s

The standard monitor shipped was 60Hz non-interlaced.

The rear panel has four (4) DB-25S connectors. The first (top) is
labelled "Control Panel / Port 1". This is the console connector, for
use with the IRIS keyboard. The remaining ports, labelled "Port 2",
"Port 3", and "Port 4", correspond to ttyd1, ttyd2, and ttyd3 respectively.

2000 keyboards have a 5-pin 180-degree DIN connector which plugs into
the keyboard junction box. The box may be free-standing or integrated
into the monitor case. The box is in turn cabled to the "Control Panel /
Port 1" connector on the rear panel with a male-to-female (DB-25P to
DB-25S) cable, pinned 1-1,2-2,...,25-25. The junction box also has a
DB-9S connector into which is plugged the mouse.

Controllers

Qualogy DSD-5217 ST-506/QIC-02

"DSD" stands for Data Systems Design, which apparently was Qualogy's old
name. This board is sometimes referred to as the "Midas". In the ST-506
systems, it is both the disk controller and cartridge tape controller.
It also has a connector for a 360K floppy.

Interphase 3030 Storager ESDI/ST-506/QIC-02

The Storager will support both ESDI and ST-506 drives, but in these
systems is used only for ESDI and cartridge tape. It too supports
floppies (3, 5, and 8 inch).

There have been reports of difficulty in using large ESDI drives with
the Interphase Storager, the suggestion being that there is a size limit
on the drive that may be used. In fact, the problem is not in the size
of the drive, but rather the higher data transfer rate that generally
accompanies larger drives. Different versions of the Storager exist,
capable of sustaining different data rates. According to Interphase.

Interphase 2190 SMD

This is the controller used with big (these days more a comment on
physical size than capacity) drives such as the Fujitsu Eagles. Unlike
the two previously described controllers, it does not have a QIC-02 tape
interface. Therefore, this board is used in conjunction with the Qualogy
5217, which provides the catridge tape I/O.

Let me Know if any of it is wrong, or if you have anything else to Add.

I have come to the conclusion that the IRIS 1000s and 2000 Non Turbos
are very similar, and therefore are likely to run the same OS, wheras
the IRIS 3000 and 2000 Turbo's are similar and they are also likely to
run the same OS.

Chris


Marcus Herbert wrote:


--

Chris

unread,
Apr 30, 2001, 12:33:31 PM4/30/01
to

Walter Roberson wrote:

There are a few sub releases of the 5.x and 6.x which might be important
here too

IRIX 5.0,
IRIX 5.0.1,
IRIX 5.1,
IRIX 5.1.1,
IRIX 5.2,
IRIX 6.0,
IRIX 6.0.1,
IRIX 5.3,
IRIX 6.1,
IRIX 6.2,
IRIX 6.3, and
IRIX 6.4.

That's the order they were released in! 6.0 was before 5.3 apparently!
Why? I don't know, anyone care to fill me in?

Brent L. Bates

unread,
Apr 30, 2001, 1:27:42 PM4/30/01
to
No, IRIX didn't come out until SGI switched from the Motorola 68000
processors to the MIPS processors. The OS on SGI's Motorola based systems was
not called IRIX, I don't believe it really had a name back then, and it was a
bit of a hodgepodge of SYS V and BSD, a bit of the worst of both. :-)
I've been using SGI's since we first got our IRIS 3130 back in 1986. I
believe it cost ~$60-80k list back then. Major `cool' factor back then, lots
of oh's, ah's, and wow's. :-)


--

Brent L. Bates (UNIX Sys. Admin.)
M.S. 912 Phone:(757) 865-1400, x204
NASA Langley Research Center FAX:(757) 865-8177
Hampton, Virginia 23681-0001
Email: B.L....@larc.nasa.gov http://www.vigyan.com/~blbates/

Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B)
This email address may not be added to any commercial mail list with out
my permission. Violation of my privacy with advertising or SPAM will
result in a suit for a MINIMUM of $500 damages/incident, $1500 for repeats.

Marcus Herbert

unread,
Apr 30, 2001, 4:05:48 PM4/30/01
to
Chris <dr...@nospam.orange.net> wrote:
> Here's the info I already have on the IRIS series machines:
[..]

Excellent information! Piece of cake!

We should set up some sort of "Old IRISes FAQ" or maybe a website. I can
even host it here for free on one of my machines.

Btw.. more pictures would be great. All you people out there having such
machines, make photos of them and send me :O)

Gerhard Lenerz

unread,
Apr 30, 2001, 4:26:04 PM4/30/01
to
On 30 Apr 2001 20:05:48 GMT, Marcus Herbert wrote:
> We should set up some sort of "Old IRISes FAQ" or maybe a website. I can
> even host it here for free on one of my machines.

IIRC there is one on Ians website.


Gerhard:wq

Chris

unread,
Apr 30, 2001, 4:55:26 PM4/30/01
to
A lot of that information is from that FAQ, I just put it in a more
digestible form for when it went in my SGI Hardware Handbook. The
information for these IRIS's is _very_ scarce, I might be able to get
some info from a few contacts I have at sgi though, I'll keep you posted.

Chris

Gerhard Lenerz wrote:

Steve Allen

unread,
Apr 30, 2001, 4:35:56 PM4/30/01
to
In article <GCM8I...@arl.army.mil>,

"Brent L. Bates" <blb...@vigyan.com> writes:
> No, IRIX didn't come out until SGI switched from the Motorola 68000
> processors to the MIPS processors. The OS on SGI's Motorola based systems was
> not called IRIX, I don't believe it really had a name back then, and it was a
> bit of a hodgepodge of SYS V and BSD, a bit of the worst of both. :-)
> I've been using SGI's since we first got our IRIS 3130 back in 1986. I
> believe it cost ~$60-80k list back then. Major `cool' factor back then, lots
> of oh's, ah's, and wow's. :-)

I still have in my files a quote and brochure from when we ordered a 3130
back in 9/87. It listed as follows:
IRIS 3130 -- $59,900
Add'l 170MB Disk -- $6,000
Workstation XNS S/W -- $1,200


The brochure speaks of:

Standard Software:
-UNIX System V operating system with Berkeley 4.2 and Silicon Graphics
enhancements
-C compiler and development environment
-IRIS Graphics Library
-IRIS Window Manager
-IRIS Programming Tutorial

Optional Software:
-FORTRAN, Pascal and Ada compilers
-EMACS text editor

There was a prototype graphical user interface on the system as well,
but I cannot for the life of me remember what it was called. NeWS
was the window system on the early 4D machines...


~Steve

--
Steven R. Allen - SGI Admin Weenie
http://www.eskimo.com/~wormey/

Contrary to popular belief, Unix is user friendly.
It just happens to be selective about who it makes friends with.

The charity that is a trifle to us can be precious to others. -- Homer

Brent L. Bates

unread,
Apr 30, 2001, 7:44:54 PM4/30/01
to
The windowing system on the old 3130's was called MEX. I'm almost
positive that is an acronym for something, but it has been probably 7 years
since I've seen that machine, let alone remember all those details. :-)

David Evans

unread,
Apr 30, 2001, 11:49:44 PM4/30/01
to
In article <3AED93D...@NOSPAM.orange.net>,

Chris <dr...@NOSPAM.orange.net> wrote:
>
>That's the order they were released in! 6.0 was before 5.3 apparently!
>Why? I don't know, anyone care to fill me in?

6.0 was a 64-bit version based on 5.2 code. 6.1 was a 64-bit version
aligned with 5.3.

--
David Evans (NeXTMail/MIME OK) dfe...@bbcr.uwaterloo.ca
PhD Student, Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/
University of Waterloo "Default is the value selected by the composer
Ontario, Canada overridden by your command." - Roland TR-707 Manual

Emmanuel Florac

unread,
May 1, 2001, 7:59:18 AM5/1/01
to
Dans l'article <GCM8I...@arl.army.mil>, blb...@vigyan.com a dit...

> No, IRIX didn't come out until SGI switched from the Motorola 68000
> processors to the MIPS processors. The OS on SGI's Motorola based systems was
> not called IRIX, I don't believe it really had a name back then, and it was a
> bit of a hodgepodge of SYS V and BSD, a bit of the worst of both. :-)
> I've been using SGI's since we first got our IRIS 3130 back in 1986. I
> believe it cost ~$60-80k list back then. Major `cool' factor back then, lots
> of oh's, ah's, and wow's. :-)
>


According to the IRIS FAQ ( found on futuretech web site ) m68k IRIS OS
was called GL2-W, not IRIX.

--
-------------------------------------------------------------------------
Emmanuel Florac | Kreode technologies
-------------------------------------------------------------------------

Claude Cantin

unread,
May 2, 2001, 11:33:47 AM5/2/01
to
If you want to see a few pictures of "older" (not antique) models, go to

http://www.sao.nrc.ca/imsb/rcsg/rcsg/computer_room/index.html

This is our computer room, pictures taken about 18 months ago.

Since then, the (R8000 Power Challenge L/4) has been replaced with a
Origin 200/4. Maintenance cost themselves for the Power Challenge will
pay for the Origin 200 in 3-4 years...

The oldest, still running system in those pictures is the now 4-CPU Power
Series 240 XL (started with 8, but CPU boards keep going belly up!), bought and
install in February 1989.

And yes, I'm looking after all of them (and many, many more). I started work
with SGIs with IRIX 3.2. Prior to that, I was looking after mainly SUNs.

I'll only add that the systems in that room put more power than my own
furnace at home...

Claude

Marcus Herbert (rho...@spam-filter.de) wrote:

--

-------------------------------------------------------------------------------
Claude Cantin Claude Cantin
Piece 2025, 100 Prom. Sussex Rm 2025, 100 Sussex Dr.
Soutien informatique a la recherche Research Computing Support
Services de gestion de l'information Information Management Services
Conseil National de Recherches National Research Council
Ottawa, Canada (K1A 0R6) Ottawa, Canada (K1A 0R6)

claude...@nrc.ca 1-613-993-0822 (FAX: 993-3127)
http://www.nrc.ca/imsb/rcsg
-------------------------------------------------------------------------------

Chris

unread,
May 2, 2001, 12:44:17 PM5/2/01
to
Some Info:

Pre IRIX 1983-1986 } SystemV with 4.3 BSD Extensions

IRIX was born in 1986 when the first MIPS processor was born, and IRIX
was derived from MIPS OS mainly, but a lot of the added code was used
from pre-IRIX releases.

IRIX 3.0 1986
IRIX 3.1 1987 (early)
IRIX 3.2 1987 (late)
IRIX 3.3 1988
IRIX 4.0.2 1990
IRIX 4.0.3 1991 (early)
IRIX 4.0.5 1991 (late)
IRIX 5.0 1993 (early)
IRIX 5.1 1993 (late)
IRIX 5.2 1994 (early)
IRIX 6.0 1994 (late)
IRIX 6.0.1 1994 (late)
IRIX 5.3 1995 (early)
IRIX 6.1 1995 (Challenge/Indigo2)
IRIX 6.2 1996 (Universal - up to '96)
IRIX 6.3 1996 (O2)
IRIX 6.4 1996 (Origin/Onyx2/Octane)
IRIX 6.5 1998 (Universal - Thank God!)
IRIX 6.5.11 2001 (Continuation of the 6.5.x release series)

So here is the information I have tried to put together, if anything is
wrong (which it could well be), let me know - or post amendments here.

If anyone has exact dates for sub 4.0 releses etc, add them, or any
other info for that matter

Thanks,

Chris

Khamba Staring wrote:

Claude Cantin

unread,
May 2, 2001, 1:38:03 PM5/2/01
to

Correction: please look at

http://www.sao.nrc.ca/imsb/rcsg/rcsg/computers/index.html

instead. That one will work.

The original URL I mentionned contains information that may be considered
"confidential". I have copied the original location to another directory,
where you all have access to, and will leave it there until at least
June 1.

Sorry for the inconvenience (I had forgotten I had closed off access to
the original location for all but internal NRC people),

Claude


Claude Cantin (can...@nickel.sao.nrc.ca) wrote:
: If you want to see a few pictures of "older" (not antique) models, go to

Sevo Stille

unread,
May 2, 2001, 7:16:45 PM5/2/01
to
"Brent L. Bates" wrote:
>
> The windowing system on the old 3130's was called MEX.

Right, at least on GL2-W3.x. I vaguely remember MEX replacing something
else.

> I'm almost
> positive that is an acronym for something, but it has been probably 7 years
> since I've seen that machine, let alone remember all those details. :-)

If I manage to find a framegrabber with suitable specs (or if somebody
knows how to switch a 3120's screen resolution to PAL or NTSC), I could
provide some screenshots...

Sevo

--
Sevo Stille
se...@ip23.net

Emmanuel Florac

unread,
May 3, 2001, 4:34:02 AM5/3/01
to
Dans l'article <3AF0955D...@ip23.net>, se...@ip23.net a dit...

That would be just plain GREAT... Oh pleeease, even a bad photograph of
the CRT....

Colin Anderson

unread,
May 3, 2001, 5:14:41 AM5/3/01
to
I too, would be happy with a bad photograph.

I'd like to suggest that perhaps someone at SGI outta dig thru the old
dusty boxes of 1980's brochures and whatnot... but I think that can wait
until after a profitable quarter, Lots of work to be done before then!

- Colin

Brent L. Bates

unread,
May 3, 2001, 8:53:39 AM5/3/01
to
No, at least on the old 3130's, MEX was the only windowing system SGI had
on those old machines.
If you check the very old SGI email list archives, assuming they are
still around, that ftp site (ftp.arl.mil) has/had a utility I posted for doing
screen dumps on those old 3130's. If you don't use that, you could always do
what we did back then, use a camera in front of the monitor with a light
shield around the monitor to cut down the glare.
That site also had archives of all the old posts. May be google could
download those and add them to their news group archives.

--

Brent Casavant

unread,
May 3, 2001, 11:05:03 AM5/3/01
to
On Wed, 2 May 2001, Chris wrote:

> Some Info:
>
> Pre IRIX 1983-1986 } SystemV with 4.3 BSD Extensions
>
> IRIX was born in 1986 when the first MIPS processor was born, and IRIX
> was derived from MIPS OS mainly, but a lot of the added code was used
> from pre-IRIX releases.
>
> IRIX 3.0 1986
> IRIX 3.1 1987 (early)
> IRIX 3.2 1987 (late)
> IRIX 3.3 1988
> IRIX 4.0.2 1990
> IRIX 4.0.3 1991 (early)
> IRIX 4.0.5 1991 (late)
> IRIX 5.0 1993 (early)
> IRIX 5.1 1993 (late)
> IRIX 5.2 1994 (early)
> IRIX 6.0 1994 (late)
> IRIX 6.0.1 1994 (late)
> IRIX 5.3 1995 (early)
> IRIX 6.1 1995 (Challenge/Indigo2)
> IRIX 6.2 1996 (Universal - up to '96)
> IRIX 6.3 1996 (O2)
> IRIX 6.4 1996 (Origin/Onyx2/Octane)
> IRIX 6.5 1998 (Universal - Thank God!)

IRIX 6.5.1 08/24/1998
IRIX 6.5.2 11/17/1998 (Introduction of Maintenance and Feature streams)
IRIX 6.5.3 02/09/1999
IRIX 6.5.4 05/11/1999
IRIX 6.5.5 08/06/1999
IRIX 6.5.6 11/03/1999
IRIX 6.5.7 02/10/2000
IRIX 6.5.8 05/09/2000 (Introduction of Octane2)
IRIX 6.5.9 08/09/2000 (Introduction of Origin/Onyx 3000)
IRIX 6.5.10 11/08/2000

> IRIX 6.5.11 2001 (Continuation of the 6.5.x release series)

IRIX 6.5.11 02/07/2001

Extrapolating from the above trend, you can expect 6.5.12 "any day now" :)

Brent

--
Brent Casavant bcas...@sgi.com
Kernel Engineer http://reality.sgi.com/bcasavan
Silicon Graphics, Inc.

Frank Everdij

unread,
May 30, 2001, 9:48:57 AM5/30/01
to
Khamba Staring <pur...@edoropolis.N.O.S.P.A.M.org> wrote:
> Hi,

> Does anyone know a good (online?) document about
> the history of IRIX? I.e. what it was derived from &
> advances over time?


try: http://perso.wanadoo.fr/levenez/unix/history.html
--
drs Frank Everdij Email:F.Ev...@citg.tudelft.nl Tel:88202 Room:6.08
System Administrator for Structural Mechanics
Dept. of Civil Engineering TU Delft

Walter Roberson

unread,
Jun 8, 2001, 10:15:21 PM6/8/01
to
In article <3AED8D53...@NOSPAM.orange.net>,
Chris <dr...@NOSPAM.orange.net> wrote:
:Here's the info I already have on the IRIS series machines:

:IRIX 1.x on IRIS 1000s and IRIS 2000 Non turbos. The IRIS 2000 Turbos

:and IRIS 3000 would I think run IRIX 2.X. When silicon graphics started
:using MIPS for the Power Series /Professional series, MIPS OS, (which
:was being developed by someone else) was bought off of them and
:Intergrated into IRIX. (This is mere theory). This formed IRIX 3.X, and
:note that MIPS OS was based a lot around BSD, so I would presume that
:the original IRIX was based around SystemV, and thats what led to in
:version 3.X SystemV with BSD extensions.

I see here, and saw in Dave Olson's posting, an indication that
the 2.x series never ran on the MIPS based machines. I have, however,
found historical evidence otherwise.

If you look on ftp://ftp.sgi.com/comp.sys.sgi/list1-999.Z
you will find,

==== begin quote ===
From: je...@norge.SGI.COM (Jeff Doughty)
Message-ID: <17...@sgi.SGI.COM>
Date: 16 Jul 88 17:51:06 GMT
[...]
There was a bug in the 2.0 operating system for 4D turbo (aka IP4, single board
computer) systems that caused this behavior. It has been fixed in release 3.0.
=== end quote ===

The IP4 was MIPS R2000 system at 8.0 or 12.5 MHz.

The very next message in the archive says,

=== begin quote ===
From: ml...@dale.acc.Virginia.EDU (Michael L. Johnson)
Message-ID: <10...@virginia.acc.virginia.edu>
Date: 18 Jul 88 19:21:46 GMT
[...]
I have been trying to get the Korn shell "ksh" to run on my 4D/70
(software release 2.2) [...]
=== end quote ===

Further down there is a message <5...@voodoo.UUCP> refering to version 2.0
for a 4D/70.

A little further down, <4...@sdrc.UUCP> of 7 Dec 88 talks about
the 1988 SIGGRAPH,m commenting,
=== begin quote ===
"This was while SGI was apologizing for all of the problems people were having
with 4D1-3.0 prompting the quick release of 4D1-3.0 Rev. B."
=== end quote ===

This gives us a timeframe for the 3.0 release which is noticably later
than the references in this thread, which indicated that IRIX 3.2
was out in 1988. If you follow down further...

=== begin quote ===
From: ki...@warp.SGI.COM (Kipp Hickman)
Message-ID: <24...@sgi.SGI.COM>
Date: 10 Jan 89 18:02:30 GMT
[...]
Release 3.0 of the IRIX 4D system supports [...]
[...]
For the 3.2 software release, the program bindkey(1G) will be provided
which gives a simple user interface to the simple keyboard re-mapping
that 3.2 will support. [...]
=== end quote ===

This message is key in that it directly ties the phrase "Release 3.0"
with "IRIX" -- though it doesn't call the OS itself IRIX, it establishes
IRIX and 3.x sequencing numbers in the early 1989 timeframe, and
thus acts to confirm that earlier references to "release 3.0" and
"release 2.0" and that 4D1-3.<whatever> are all part of a continuing
development stream for the MIPS series.]

The message also establishes that IRIX 3.2 was not out until at least 1989,
not the 1988 claimed earlier. Later references in the message establish
that 3.1 *was* out by early 1989. (It also indicates that release 4.0
was already being planned in that timeframe.)

When I head further down in the list, I see a reference to release 3.1D .
I had forgotten completely that the 3.1 series had lettered sub-releases.
No-one has been mentioning these. "release 4D1-3.1 Rev D which was
released to manufacturing on Monday February 27th [1989]" A later early 1989
message indicates that the 3.2 release was due mid-year.


Now, just to confuse things, there are references in the archive to
things like the 3.5 and 3.6 release. These are releases for the 3000
series, and do NOT appear to be part of the same numbering sequence.
[That's why the tie between 3.0 and IRIX was so crucial a moment ago.]
The SGI 3000 series Unix version 3.6 appeared to be newish around 1
July 88. If I read correctly, development of the OS for the 3000
series continued after the 4D series was launched. I, for one,
previously had developed the impression that when the MIPS series came
out, the transition was pretty sharp with essentially no further
development on the 3000 series.


I see a reference to "IRIX4.1" on 8 Apr 89, but I am sure that that is
spurious, or a reference to an IRIX release corresponding to System V
4.1. For more on the correspondance to System V releases, look down to
<30...@sgi.SGI.COM> 10 Apr 89.

Walter Roberson

unread,
Jun 8, 2001, 11:58:29 PM6/8/01
to
In article <9fs0rp$3q9$1...@canopus.cc.umanitoba.ca>,
Walter Roberson <robe...@ibd.nrc.ca> wrote:
:If you look on ftp://ftp.sgi.com/comp.sys.sgi/list1-999.Z
:you will find,

Something else ancient of interest to historians:

Look in that file, and analyze <30...@sgi.SGI.COM>

=== begin quote ===
From: j...@patton.SGI.COM (Jim Barton)
Date: 10 Apr 89 18:24:36 GMT
[...]
Why worry? The SGI ExtentFileSystem is >faster< than the BSD FFS. For
instance, on the Jim Barton extra-special-whizzy single-and-multi-process
blow-out-the-buffer-cache benchmark (substantiated by the AIM II disk
benchmark and other tests), UMIPS 3.0 FFS on the M/120 runs about 15% slower
than IRIX 3.0 EFS on the >exact same hardware<. [...]
[...]
[...] In my lab, we use
a 4D/120 with 2 extra processors (an "unofficial" 4D/140). [...]
[...]
Maddog 4D/140S IRIX 4D-3.2A (Alpha 7)
=== end quote ===

Combined with

=== begin quote ===
From: j...@patton.SGI.COM (Jim Barton)
Message-ID: <27...@sgi.SGI.COM>
Date: 28 Feb 89 16:11:46 GMT
[...]
For everything else, the machines really are binary compatible. In fact,
UMIPS binaries will run on SGI hardware if they stay away from special
functions.
=== end quote ===


If we put together the pieces, the 'M/120' must be the 4D/120, the
project code name of which was 'Maddog'.

Jim's ">exact same hardware<" implies that one of two things was true:

A) That he was able to run the UMIPS operating system itself on the 4D120;
or

B) That he had access to the code for the UMIPS 3.0 FFS (Fast File System)
and was able to run the FFS under IRIX 3.0.

The second possibility is far more boring than the first, as the first
would imply that there was more than one operating system that -could-
have run on the 4D120 (even if it was never marketed.)

It appears from other references that the UMIPS operating system was
named RISC/os.

Dave Olson

unread,
Jun 9, 2001, 7:31:16 PM6/9/01
to
robe...@ibd.nrc.ca (Walter Roberson) writes:
| If we put together the pieces, the 'M/120' must be the 4D/120, the
| project code name of which was 'Maddog'.

Nope. The M/120 was a true MIPS (the company) box, not a different
name for the 4D/120. He was comparing them because they had the
same processor, and the rest of the hardware wasn't wildly different.

The personal IRIS intro'ed in October 88 with 3.1[e|f]. 3.2 was
sometime in 89, as your previous posting noted. If I said otherwise,
I was confused ;)

| A) That he was able to run the UMIPS operating system itself on the 4D120;
| or
|
| B) That he had access to the code for the UMIPS 3.0 FFS (Fast File System)
| and was able to run the FFS under IRIX 3.0.

Reasonable from the postings, but not correct ;)

| It appears from other references that the UMIPS operating system was
| named RISC/os.

Yeah. I seem to recall that one of the uname fields was UMIPS.

Dave Olson

unread,
Jun 9, 2001, 7:36:26 PM6/9/01
to
robe...@ibd.nrc.ca (Walter Roberson) writes:
| I see here, and saw in Dave Olson's posting, an indication that
| the 2.x series never ran on the MIPS based machines. I have, however,
| found historical evidence otherwise.
|
| If you look on ftp://ftp.sgi.com/comp.sys.sgi/list1-999.Z
| you will find,

Jeff should know. I suppose there's a posssiblity that some early
development was done and called 2.x, but I'm pretty sure that SGI never
released an IRIX 2.anything (or anything 2.anything!) for MIPS.

Still, Jeff was there before I was... Maybe it wasn't an official
release, but rather a beta?

Walter Roberson

unread,
Jun 9, 2001, 10:57:19 PM6/9/01
to
In article <olson.9...@redtail.bengaltech.com>,
Dave Olson <ol...@bengaltech.com> wrote:
:The personal IRIS intro'ed in October 88 with 3.1[e|f]

The earliest reference I can see to the Personal IRIS was
10 Oct 88, at which time several customers had already called with
complaints about unreadable tapes. By 19 Oct 88, U. Toronto already
had some on hand. This suggests that the unit might have been
introduced shortly before the beginning of October 1988.

<890215213...@adt.uucp> of 15 Feb 89 has a clear reference
to a 4D/20 running '4D-3.1 4D/20 REV. B', the context of which
implies that the 4D/20 had a seperate Rev. B release. This
version format is anachronistic, as version letters were not
displayed in release strings until Rev. D! Still, I do see
clear references in the archives to PIs running release 3.147xxx
(aka Rev C.). I also see a clear indication that for 3.1D there was
an IP6 version distinct from the other.


Walter Roberson

unread,
Jun 9, 2001, 11:27:44 PM6/9/01
to
:Jeff should know. I suppose there's a posssiblity that some early

:development was done and called 2.x, but I'm pretty sure that SGI never
:released an IRIX 2.anything (or anything 2.anything!) for MIPS.

I finally found irrefutable proff that the 2.0 series was released.
If you examine the list2000-2999 message series from ftp.sgi.com
you will see the IRIX 3.2 release notes, message <44...@sgi.sgi.com>.
Section 3.2.1 of those notes contains a backwards compatibility
matrix that explicitly discusses 4D1-2.0 executables
(which were compatible if graphics was not used.)


Those old list* messages have a quite a few references to
people running 2.0 (mostly) or 2.2 software on 4D70's, or occasionally
on 4D60's. Also on one 4D50 that I've noticed.

====

For those of us who are completeness freaks ;-) I found references
in the archives to the following pre-3.0 versions of IRIX for MIPS machines:

2.0, 2.2, 2.2b
3.0, 3.0B, 3.0.1 (once)
3.1, 3.1B, 3.1C, 3.1D, 3.1D IP6 Version, 3.1F, 3.1G
3.2, 3.2.1, 3.2.2, 3.2.3 (rare!), 3.2C (mistake?), 3.2E (mistake?)

Also of historical interest is the following from list6000-6999 :

=== begin quote ===
From: ro...@graphics.upenn.edu (Charlie Root)
Subject: IRIS 3030
Message-ID: <ROOT.90Oc...@graphics.upenn.edu>
Date: 23 Oct 90 14:38:44 GMT
[...]
The Computer Graphics Research Laboratory
at
the University of Pennsylvania

We have a Silicon Graphics IRIS 3030 with the following :

* IP 2 Processor / Revision B * running IRIX 3.0.1
=== end quote ===

The interesting bit about this is that it seems to claim that
there was a version of IRIX supported on the IRIS 3030 (68020 machine)!

There was one reference to IRIX 3.0.1 for the 4D/70GT in the
archives, so it did seem to exist. Possibly, though, the poster
made a mistake when determining the software version for their 3030??
All the other 3030 references I've seen talk about GL2 version 3.5
or 3.6.


I did not try to catalogue the 3.3 releases, and there is still
room for people to go in and figure out more precisely when
each of the 3.x releases came out.

Daniel Widyono

unread,
Jun 9, 2001, 11:53:49 PM6/9/01
to
Interesting. I think we threw that piece of equipment out a LONG time ago
(and we're no longer named the CGRL). Therefore, I can't help prove whether
or not 3.0.1 ran on it :) We may still have the 4D's around but I don't think
they power up at all anymore.

Dan W.

: The Computer Graphics Research Laboratory


: at
: the University of Pennsylvania

: We have a Silicon Graphics IRIS 3030 with the following :

: * IP 2 Processor / Revision B * running IRIX 3.0.1
: === end quote ===

: The interesting bit about this is that it seems to claim that
: there was a version of IRIX supported on the IRIS 3030 (68020 machine)!

: There was one reference to IRIX 3.0.1 for the 4D/70GT in the
: archives, so it did seem to exist. Possibly, though, the poster
: made a mistake when determining the software version for their 3030??
: All the other 3030 references I've seen talk about GL2 version 3.5
: or 3.6.

--
-- Daniel Widyono # http://www.cis.upenn.edu/~widyono
-- wid...@cis.upenn.edu # CIS Dept., SEAS, University of Pennsylvania
-- Mail: Rm 556, CIS Dept # 200 S 33rd St # Philadelphia, PA 19104

Walter Roberson

unread,
Jun 10, 2001, 2:56:56 PM6/10/01
to
In article <9fupfg$ri7$1...@canopus.cc.umanitoba.ca>,
Walter Roberson <robe...@ibd.nrc.ca> wrote:
:Also of historical interest is the following from list6000-6999 :

: We have a Silicon Graphics IRIS 3030 with the following :


: * IP 2 Processor / Revision B * running IRIX 3.0.1

I cross-checked 100 or so of the old messages that had a
1000 or 2000 or 3000 series number along with the word IRIX, and
this was the *only* one that claimed IRIX running on a 3000 series.

There was another message from the same time frame (from Brent)
saying specifically that the operating systems were different
between the two, when someone was asking what the latest IRIX version
for their 3000 series was.

I now think it quite likely that the above quoted message was
a mistake on the poster's part.

Keith Huff

unread,
Jun 10, 2001, 10:17:25 PM6/10/01
to
It must have been a mistake as I had an old IRIS 3030 with a full
manual set. The manuals made no mention of IRIX but rather stated the
operating system as GL2-W2.x. I can confirm that the 3030 did not run
IRIX, I tried loading the IRIX 3.1 and 3.3.2 tapes I have, no go. Since
I could not find the original OS I scrapped the 3030.

Keith Huff

ksh...@fast.net

Walter Roberson

unread,
Jun 10, 2001, 11:13:13 PM6/10/01
to
In article <9fupfg$ri7$1...@canopus.cc.umanitoba.ca>,
Walter Roberson <robe...@ibd.nrc.ca> wrote:
:I finally found irrefutable proff [sic] that the 2.0 series was released.

:If you examine the list2000-2999 message series from ftp.sgi.com
:you will see the IRIX 3.2 release notes, message <44...@sgi.sgi.com>.
:Section 3.2.1 of those notes contains a backwards compatibility
:matrix that explicitly discusses 4D1-2.0 executables

I have been going over the old messages again, and I see a pattern
emerging that has the added bonus of explaining why there was
no IRIX 1.x .

Consider this old version string:

IRIX System V Release 4D1-3.13809261636

I correlated references to System V releases with timings of
IRIX releases. What I found was that the 4D1 release number
corresponded very well to the available System V release numbers
[allowing for time to do a port.] Thus, if one elides the '4D1'
trademark from the version string and re-parses, one gets,

IRIX(TM) System V Release 3r1 (internal release #3809261636)

This correspondance tracks well for 3.0, 3.1, 3.2, and 3.3.

Also, it appears to me that IRIX 4 was just about the same time
as SVR4. The Unix FAQ claims, though, that IRIX 4 was SVR3r2,
and I see from the archives that indeed SVR4 ABI compliance
wasn't until IRIX 5.

I hypothesize, then, that *in the IRIX 3 timeframe*, the
dot numbers were intended to correspond strongly to SVR release numbers.
I would even suggest that "IRIX" was considered more just a name
in that timeframe, and not a "brand" like it is now -- that the
emphasis was on SVR, not on the extensions.

Under this hypothesis 4D1-2.0 and 4D1-2.2 get explained as being
releases for the 4D MIPS-based architecture prior to 3.0 that
corresponded to SVR2.0 and SVR2.2 respectively, and the lack of
any evidence for 4D1-1.x can be understood as being because SGI
entered the market at the time of SVR2.

=====

Now the bit that I'm curious to track down is Vernon's mysterious
references to V-kernel and BSD4.2 experiments ;-)
[ <27...@sgi.SGI.COM> and <29...@sgi.SGI.COM> ]

Martin.Knoblauch

unread,
Jun 11, 2001, 2:46:56 AM6/11/01
to
"Brent L. Bates" wrote:
>
> The windowing system on the old 3130's was called MEX. I'm almost
> positive that is an acronym for something, but it has been probably 7 years
> since I've seen that machine, let alone remember all those details. :-)
>


MEX - MultipleEXposure

Martin
--
------------------------------------------------------------------
Martin Knoblauch | email: Martin.K...@TeraPort.de
TeraPort GmbH | Phone: +49-89-510857-309
C+ITS | Fax: +49-89-510857-111
http://www.teraport.de | Mobile: +49-170-4904759

Martin.Knoblauch.vcf

Martin.Knoblauch

unread,
Jun 11, 2001, 2:58:24 AM6/11/01
to
Walter Roberson wrote:
>
>
> I see here, and saw in Dave Olson's posting, an indication that
> the 2.x series never ran on the MIPS based machines. I have, however,
> found historical evidence otherwise.
>
Walter,

your mailer did not like my mail-setup at home. So, to answer your
question: my memories are dimm wrt the topic. If I wrote that we
upgraded from 2.0 to 3.0, it must have been true though :-) In any case,
3.0-3.2 were not the most remarkable releases wrt. stability. The later
3.3 incarnations were pretty good and probably still the best releases
for some of the earlier, non X-optimized, graphics adaptors.

Martin

"Martin.Knoblauch" wrote:
>
> Walter Roberson wrote:
> >
> > See my latest hypothesis in the newsgroup... ie, that in those
> > early years, 4D1-x.y designated SVRx.y (System V Release x.y),
> > and that it started with 2.0 because it started with SVR2.0 .
> >
> > walter
> Hi Walter,
>
> I had problems replying to you from home. Apparently your mailer did
> not like the mismatch between my home machine and the provider - so I
> must be a bad spammer :-) In any case, I am pretty sure from my posting
> that there was some version 2.0 for IRIX/MIPS. I would not have written
> otherwise. It might very well be that IRIX was born with 2.0 and 1.x was
> the stuff for the 68K based systems (although not called IRIX at those
> times).
>
> Martin
> -----------------------------------------------------------------------------------------------
> -----------------------------------------------------------------------------------------------
> Walter Roberson wrote:
> >
> > Hi,
> > I wonder if you could dig through your memories with respect to
> > the following comp.sys.sgi entry you once made:
> >
> > === begin quote ===
> > From: XBR2...@DDATHD21.BITNET (Knobi der Rechnerschrat)
> > Subject: Tmesh routinen in 4D/70G 3.0 Software
> > Message-ID: <880714070...@SMOKE.BRL.ARPA>
> > Date: 14 Jul 88 05:48:34 GMT
> > Organization: The Internet
> >
> > Hallo,
> >
> > we have a small problem with the 3.0 Software on a 4D/70G (not GT).
> >
> > We are using tmesh structures in our molcad software. on a 4D/70G with
> > our own tmesh-emulation (running 2.0 Software) and on a GT with 3.0
> > Software everything is okay (thats why I don't look for the error in
> > our code). Yesterday we upgraded our 70G to the 3.0 Software and it
> > seems that the first triangle after a bgntmesh is not drawn correctly.
> >
> > Has anybody seen this before? Solutions? Suggestions?
> >
> > Thanks in advance
> >
> > Martin Knoblauch
> >
> > TH-Darmstadt
> > Dept. Physical Chemistry 1
> > Petersenstrasse 20
> > D-6100 Darmstadt
> > West-Germany
> >
> > BITNET: <XBR2D96D@DDATHD21>
> > === end quote ===
> >
> > The question at hand is whether there was ever an operating system
> > release for the SGI MIPS machines with a version number lower than
> > 3.0; and secondarily, whether that operating system was called IRIX.
> >
> > The quoted message implies that you had direct experience in
> > running with IRIX 2.0 on a 4D/70G -- what needs to be confirmed
> > is that the 2.0 and 3.0 you refered to were operating system
> > versions ? And if they were, do you remember anything else about
> > what early versions were available?
> >
> > Thanks!
> >
> > Walter Roberson robe...@ibd.nrc.ca
> Hi Walter,
>
> that was an awful long time ago :-) I definitely can confirm that I
> worked with very early versions of the 4D systems. Both the G and GT
> based stuff. Basically we had the first 70G in Germany and I had first
> hand experience with pre-Beta GTs while we collaborated with SGI at that
> time.
>
> I cannot really remember the post you qoute, but I when I said we
> upgraded from 2.0 to 3.0, I must have been true. IIRC, the 1.0 and 2.0
> versions were neither long lived nor very stable though. Only 3.3 on its
> later incarnations (G and H) was really good. Not sure whether the
> 1.0/2.0 versions were called IRIX. Dave, are you listening? :-)
>
> Hope this helps
> Martin
> PS: Funny. It is the second time on less than a week that my Molcad past
> is coming back to me ...

Scott Henry

unread,
Jun 11, 2001, 6:32:02 PM6/11/01
to
>>>>> "D" == Dave Olson <ol...@redtail.bengaltech.com> writes:

D> robe...@ibd.nrc.ca (Walter Roberson) writes:
D> | I see here, and saw in Dave Olson's posting, an indication that
D> | the 2.x series never ran on the MIPS based machines. I have, however,
D> | found historical evidence otherwise.
D> |
D> | If you look on ftp://ftp.sgi.com/comp.sys.sgi/list1-999.Z
D> | you will find,

D> Jeff should know. I suppose there's a posssiblity that some early
D> development was done and called 2.x, but I'm pretty sure that SGI never
D> released an IRIX 2.anything (or anything 2.anything!) for MIPS.

D> Still, Jeff was there before I was... Maybe it wasn't an official
D> release, but rather a beta?

Hmmmm, I do remember installing 3.0alpha over whatever was already
there on a 4D70G in my early days. Whatever was the first release
with 4DDN... (aka DecNet).

--
Scott Henry <sco...@sgi.com> / Help! My disclaimer is missing!
IRIX/Linux MTS, / http://reality.sgi.com/scotth/
Silicon Graphics, Inc / SGI-One step ahead

0 new messages