IRIX, or what?
Steve Kearney
(South West Scotland).
Is it Eye-rix, Ear-rix, i-rix, E-rix or what?
Steve
Brent Casavant
> Is it Eye-rix, Ear-rix, i-rix, E-rix or what?
^^^^^^^
Brent Casavant
--
Brent Casavant bcas...@sgi.com
Kernel Engineer http://reality.sgi.com/bcasavan
Silicon Graphics, Inc.
Steve Kearney
Well, us Brits (well, me anyway!) tend to say i-rix (short first
syllable).
Sorry!
Steve
Rob Warnock
+---------------
| Steve Kearney wrote:
| > Is it Eye-rix, Ear-rix, i-rix, E-rix or what?
| ^^^^^^^
But then, Brent's in Minnesota, you know. Most of us out
here in Mountain View (where it all started) pronounce
it IRE-ix (which Steve might like, actually).
-Rob
-----
Rob Warnock, 31-2-510 rp...@sgi.com
SGI Network Engineering <URL:http://reality.sgi.com/rpw3/>
1600 Amphitheatre Pkwy. Phone: 650-933-1673
Mountain View, CA 94043 PP-ASEL-IA
Scott Elyard
<s...@the-neuk.demon.co.uk> wrote:
> Well, us Brits (well, me anyway!) tend to say i-rix (short first
> syllable).
>
> Sorry!
>
> Steve
Do you also say i-ris (short first syllable)?
("Do you also say Froe-derik?")
--
.oO=----------------------------------------------------=Oo.
| Comic: http://www.archosaur.org/oscar/ |
| Life (in progress): http://www.archosaur.org/ |
`~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
Happy Chappy
eyericks.
"Steve Kearney" <s...@the-neuk.demon.co.uk> wrote in message
news:3B0C3581...@the-neuk.demon.co.uk...
Marcus Herbert
> Is it Eye-rix, Ear-rix, i-rix, E-rix or what?
Do you remember that first linux webpages/homepages? They all had that
this file linked:
http://www.kernel.org/pub/linux/kernel/SillySounds/english.au
Nobody knew about linux and so Linus had to tell them how to pronounce
it.
Is IRIX already down to some stage where nobody knows about it anymore?
Do we really need to record a file with the correct pronunciation? :O)
Who is going to record it?
--
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
Colin Anderson
>
> Steve Kearney <s...@the-neuk.demon.co.uk> wrote:
> > Is it Eye-rix, Ear-rix, i-rix, E-rix or what?
>
> Do you remember that first linux webpages/homepages? They all had that
> this file linked:
>
> http://www.kernel.org/pub/linux/kernel/SillySounds/english.au
>
> Nobody knew about linux and so Linus had to tell them how to pronounce
> it.
>
> Is IRIX already down to some stage where nobody knows about it anymore?
> Do we really need to record a file with the correct pronunciation? :O)
> Who is going to record it?
SGI's CEO, Bob Bishop, has a pretty nifty Australian + European
"international" accent and a nice pronunciation of "IRIX".
Or maybe we could have Cosmos record a few choice words.
- Colin
David Haworth
in turn, do you call people EYE-rritable? :)
I call it i-rix (short syllable) but hell, pronounce however it looks
good to you :)
dave
Chris Moxon
Steve Kearney wrote:
> Brent Casavant wrote:
>
>> On Wed, 23 May 2001, Steve Kearney wrote:
>>
>>
>>> Is it Eye-rix, Ear-rix, i-rix, E-rix or what?
>>
>> ^^^^^^^
>>
>
>
>
> Well, us Brits (well, me anyway!) tend to say i-rix (short first
> syllable).
>
> Sorry!
It's just you I'm afraid, I haven't heard anyone call it that!
--
Chris Moxon
If you want to reply to this remove NOSPAM from the address.
Thant Tessman
A long time ago, Silicon Graphics used to build a graphics terminal and
then a workstation called "Iris" like the flower or the part of your
eye. IRIX is wordplay on "Iris" and "Unix".
-thant
Colin Anderson
It all started with the IRIS 1000 terminal.
- Colin
Happy Chappy
"Colin Anderson" <co...@beyondboxes.com> wrote in message
news:3B0D5AA3...@beyondboxes.com...
> Integrated Raster Imaging System
>
> It all started with the IRIS 1000 terminal.
I was wondering if anyone knows of a website with images of the above
mentioned machine. I'd be quite interested to see what it (and the other
pre-Mips machines) look like.
btw the IRIS 1000 used an 68K for it's processor, didn't it?
-Tom
Gerhard Lenerz
> I was wondering if anyone knows of a website with images of the above
> mentioned machine. I'd be quite interested to see what it (and the other
> pre-Mips machines) look like.
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
And that's it. I wonder if there are any other pictures around and if
there are some machines left...
> btw the IRIS 1000 used an 68K for it's processor, didn't it?
From an a recent post:
1000/1200:
8MHz Motorola 68000 - PM1 (variant of Stanford UNiversity "SUN" board)
1400:
10MHz Motorola 68010 - PM2
1500:
10 MHz Motorola 68010 - PM2
Gerhard:wq
David Haworth
> then a workstation called "Iris" like the flower or the part of your
> eye. IRIX is wordplay on "Iris" and "Unix".
that may well be, but I think i-rix sounds cooler and thats how I will
continue to pronouce it, after all, if thats what I think then it must
be correct :)
and one more thing... HOLY WAR! :)
dave
Colin Anderson
was a IRIS 2500, but I don't recall exactly. There were a few photos of
the rack, but no good board shots.
Most of the IRIS machines were deskside, though a few were rackmount
(which worked out quite well, as large drives at the time could easily
be mounted on the same rack).
- Colin
Colin Anderson
as the "IP2" which used a 68020.
I've heard of "IP1" in passing, but I've never read about it. Perhaps
"IP1" refers to PM1 or PM2??
- Colin
I R A Darth Aggie
<tgqoo1j...@corp.supernews.com> wrote:
+ I was wondering if anyone knows of a website with images of the above
+ mentioned machine. I'd be quite interested to see what it (and the other
+ pre-Mips machines) look like.
As I recall, the SGI 3000 (? I first saw them in 1989) was a rather
large, square box with wheels, roughly the height of a two-drawer file
cabinet, and perhaps a tad wider..
+ btw the IRIS 1000 used an 68K for it's processor, didn't it?
The aforementioned 3000 used a 68010...or was it a 68020? and running
on the order of 10 MHz... ;)
James
--
Consulting Minister for Consultants, DNRC
I can please only one person per day. Today is not your day. Tomorrow
isn't looking good, either.
I am BOFH. Resistance is futile. Your network will be assimilated.
Alex
speaking french I tend to pronounce sth like i-rix (short i), but I just
realized that when speaking english, I pronouce Eye-rix...
Anyway, nobody ever complained, so I guess everybody knows what I'm talking
about...
Alex
"Steve Kearney" <s...@the-neuk.demon.co.uk> wrote in message
news:3B0C2F4E...@the-neuk.demon.co.uk...
Chris Moxon
from peoples websites. I did a lot of research on the IRIS series of
workstations. I would serve it from here for a little while if people
were interested enough. The only thing was I took a few bits from Ian's
website (mainly about the I2 and Crimson), I wasn't going to show it to
anyone, but if Ian has no objection, then i'll put it up for a while.
Cheers,
Chris
Gerhard Lenerz wrote:
Matthew Hodgson
"Happy Chappy" <ind...@madasafish.com> wrote in message
news:tgqoo1j...@corp.supernews.com...
>
> "Colin Anderson" <co...@beyondboxes.com> wrote in message
> news:3B0D5AA3...@beyondboxes.com...
> > Integrated Raster Imaging System
> >
> > It all started with the IRIS 1000 terminal.
>
> I was wondering if anyone knows of a website with images of the above
> mentioned machine. I'd be quite interested to see what it (and the other
> pre-Mips machines) look like.
>
I've got 3 IRIS 1400s still running (reasonably) happily in pretty good
condition - early prototypes that SGI released back in 1984. I've been
meaning to take photographs for the last decade or so (in the absence of any
decent ones out there on the net) - but I have the problem that the room
that they're currently in is so cluttered with other kit that I can't get
far enough away to do them justice. I promise that over the next few weeks
or so I'll wheel them out and take some exceptionally high resolution
snapshots, screenshots, pictures of the boards, etc.
In quick response to anyone's questions, they do indeed run 68010 chips on
the PM2 motherboard (which is different to the IP1 that was introduced for
the 3000 series machines) - the board itself is a slight modification of the
Stanford UNiversity "SUN" board that was around at the time. The boards
occupied a 20 slot Multibus card cage. My machines have 8Mb of RAM in them,
and 24-bit color (in single buffer mode - two sets of 4096 colormap in
double buffer mode). No Z-buffer (hardware or otherwise) was ever available,
although a hardware floating point board based on a Weitek chip was
available. They have 2 72Mb Vertex MFM discs each (which are easily the most
unreliable part of the system) - with System Software 1.6 and 1.7(beta)
installed. (ironically, in light of the name of this thread, this was way
before anyone ever thought of rewriting/rebranding the OS as IRIX - the
'System Software' was still the original 'System V with BSD Enhancements'
UNIX as provided originally for SGI by UniSoft. IRIX came slightly after the
transition to the MIPS architecture with the introduction of the 4D/60).
As the entire architecture is designed for wireframe manipulation and
rendering with no provision for bitblting between RAM and bitplanes,
interpolated shading is impossible (although I did port PovRAY to it in
years gone by, and write my own Phong shading renderer that used the
Geometry Engine pipeline for doing transforming model geometry to screen
coordinates and then writepixels() to rasterise). There is rather crippled
support in the GL for drawing solid filled polygons, but nothing
particularly usable. Nothing fancy like antialiasing - but none the less
they run a surprisingly creditable and complete implimentation of IRIS GL
(and C was even included as standard in the OS back then!). I still use them
for basic IRIS GL development work occasionally to keep them ticking over,
although the absence of TCP/IP (at the time of writing) makes file transfer
slightly nontrivial. Oh yes. And they run flight.
Anyhow, I'll write up as much information as I can on them sometime soon -
If anyone has any further questions in the interim I'd be happy to try to
field them.
Matthew.
Chris Moxon
Matthew Hodgson
"Chris Moxon" <dr...@NOSPAM.orange.net> wrote in message
news:3B0E4907...@NOSPAM.orange.net...
> It's up now on port 80 on 213.105.80.17
>
> http://213.105.80.17:80
>
>
> Chris
>
>
> Chris Moxon wrote:
>
> > I made an sgi hardware guide a while back incorporating the best details
> > from peoples websites. I did a lot of research on the IRIS series of
> > workstations. I would serve it from here for a little while if people
> > were interested enough. The only thing was I took a few bits from Ian's
> > website (mainly about the I2 and Crimson), I wasn't going to show it to
> > anyone, but if Ian has no objection, then i'll put it up for a while.
> >
> > Cheers,
> >
> >
> > Chris
Shouldn't you give credit to John Levine *somewhere* on your site
considering the amount of information on the IRIS xxxx series which you've
lifted verbatim from his excellent Silicon Graphics IRIS 2000/3000 FAQ?
Chris Moxon
Matthew Hodgson wrote:
Sorry about that - it's on there now.
I won't be running this server for much longer btw. So take a look while
you can!
Emmanuel Florac
>
> "Happy Chappy" <ind...@madasafish.com> wrote in message
> news:tgqoo1j...@corp.supernews.com...
> >
> > "Colin Anderson" <co...@beyondboxes.com> wrote in message
> > news:3B0D5AA3...@beyondboxes.com...
> > > Integrated Raster Imaging System
> > >
> > > It all started with the IRIS 1000 terminal.
> >
> > I was wondering if anyone knows of a website with images of the above
> > mentioned machine. I'd be quite interested to see what it (and the other
> > pre-Mips machines) look like.
> >
>
> I've got 3 IRIS 1400s still running (reasonably) happily in pretty good
> condition - early prototypes that SGI released back in 1984. I've been
> meaning to take photographs for the last decade or so (in the absence of any
> decent ones out there on the net) - but I have the problem that the room
> that they're currently in is so cluttered with other kit that I can't get
> far enough away to do them justice. I promise that over the next few weeks
> or so I'll wheel them out and take some exceptionally high resolution
> snapshots, screenshots, pictures of the boards, etc.
>
Oh yes, pleeaaase, I'd just love to see some screenshots...
--
------------------------------------------------------------------------------
Emmanuel Florac | Kreode technologies
------------------------------------------------------------------------------
Matthew Hodgson
"Emmanuel Florac" <efl...@kreode.com> wrote in message
news:3B0E7EAC...@kreode.com...
> Matthew Hodgson a *crit :
> >
> > "Happy Chappy" <ind...@madasafish.com> wrote in message
> > news:tgqoo1j...@corp.supernews.com...
> > >
> > > "Colin Anderson" <co...@beyondboxes.com> wrote in message
> > > news:3B0D5AA3...@beyondboxes.com...
> > > > Integrated Raster Imaging System
> > > >
> > > > It all started with the IRIS 1000 terminal.
> > >
> > > I was wondering if anyone knows of a website with images of the above
> > > mentioned machine. I'd be quite interested to see what it (and the
other
> > > pre-Mips machines) look like.
> > >
> >
> > I've got 3 IRIS 1400s still running (reasonably) happily in pretty good
> > condition - early prototypes that SGI released back in 1984. I've been
> > meaning to take photographs for the last decade or so (in the absence of
any
> > decent ones out there on the net) - but I have the problem that the room
> > that they're currently in is so cluttered with other kit that I can't
get
> > far enough away to do them justice. I promise that over the next few
weeks
> > or so I'll wheel them out and take some exceptionally high resolution
> > snapshots, screenshots, pictures of the boards, etc.
> >
>
> Oh yes, pleeaaase, I'd just love to see some screenshots...
>
I'll do my best. During my POV-Ray days on the machines I wrote a little bit
of GL which used readpixels() to dump the bitplanes to disk at
(1024x768x24bit) to disk as a Targa file. The combination of readpixels(),
incredibly slow disk access and my reasonably flaky C skills at the time
meant that the program takes an amazingly long time to run. The end file
(without any rll compression) was several megabytes long and so took up a
considerable (i.e. 10%) of the /usr partition of the disk - hence making the
program something of a liability in the end.
I never tested it on taking screenshots of the console text-viewport display
mode (almost 100% analogous to the cyan-bordered text console of current day
IRISes) - but as far as I can remember ginit() doesn't touch the bitplanes
until a gconfig() or gclear() is called, which should be sufficient to run
readpixels() to grab a pretty picture of the text viewport to disk. (AFAIK,
no windowing system was ever available for the 1400s - although MEX may have
possibly run on the architecture. I have a demo for the 1500 running on the
machine which shows that the clipping & windowing architecture was certainly
in place for it to be possible. On my installation, however, you sit in text
mode all the time until you call a ginit() whereupon GL takes over the whole
screen until the appropriate gexit().). With a bit more intelligence, I
might also be able to intercept some of the demos in mid-graphics such as
'shuttle' (running the engine which eventually evolved into powerflip) and
take screenshots of those too - although it might some require ingenuity to
avoid corrupting the bitplanes with a return to the text viewport. Doing
such nonstandard things is liable to be slightly dubious, as GL was very
much under development back then - the list of bug fixes in the 1.7 Release
Notes is highly amusing.
I'm not going to get a chance to power the machines on and take a look for a
week or so, and I am painfully aware that the harddiscs in the machines are
on their last legs - and whilst I have the original distribution media of
the 1.6 and 1.7 System Software on QIC tape, I have a horrible feeling that
it may have degraded to the point of no return over the decades.
Does anyone have any tips on ressurecting elderly QIC tapes?
I'm going to try to read them on my Personal IRIS first in case the 1400s'
tape drives have decided to turn to the dark side - but I am fully expecting
much fun and games to ensue. So to conclude, it might take a bit longer than
expected to produce some pretty pictures, but I'll do my best.
Matthew.
Happy Chappy
"Emmanuel Florac" <efl...@kreode.com> wrote in message
news:3B0E7EAC...@kreode.com...
> Matthew Hodgson a *crit :
> >
>
> Oh yes, pleeaaase, I'd just love to see some screenshots...
Me Too! Especially if you have the OS up and running
Happy Chappy
>
Big thanks to the others that posted reply's to my question also.
:)
Steve Kearney
Aye, but, I'm originally from England, Lancashire in fact.
I though I had the only SGI's in Dumfries and Galloway!!
Matthew Hodgson
> Happy Chappy wrote:
> Aye, but, I'm originally from England, Lancashire in fact.
>
>
> I though I had the only SGI's in Dumfries and Galloway!!
>
> > Funnily enough I'm also from south-west Scotland and I pronounce it
> > eyericks.
<tries to detangle top/bottom posting meleé, gives up, and opts to middle
post>
I'm definitely British and based in Oxford - yet I pronounce it Eyericks,
which seems to make to sense considering (as pointed out elsewhere on the
thread) that IRIX is a blatant pun on IRIS - which I've never heard
pronounced as anything but Eyeris. I know someone who hails from Leeds,
though, who pronounces it 'Irricks' - which sounds more like some kind of
eczema cream than a rather funky OS imo. As regards calling it I.R.ix - that
seems just as bizarre and perverse as referring to IRISes as I.R.I.Ses. Also
has vague terrorist overtones. (/me waves to the FBI). And 'E-rix' sounds
like like it should belong to someone called Eric. but I digress. :-)
M.
Matthew Hodgson
> The CPU boards I've heard of for IRIS are the two you mentioned, as well
> as the "IP2" which used a 68020.
>
> I've heard of "IP1" in passing, but I've never read about it. Perhaps
> "IP1" refers to PM1 or PM2??
>
To correct my post from further down the thread (and answer the question),
IP1 never officially existed. If anything, it was a relabelling of the PM2
board as used in the IRIS 1400, 1500, 2000, 2200, 2300, 2400 and 2500. The
first actually used IP series board was the IP2 as used in the 2300T, 2400T,
2500T and 3000 series machines. The main distinction between the PM and IP
boards were that the PM were based on reference boards developed at Stanford
University whereas the IP boards were (and still are, i hope ;) developed in
house by SGI themselves (although often based on reference boards from
MIPS - explicitly the case for the original IP4 boards used in the 4D/50;
the first of the MIPS machines).
Or perhaps the IP1 was a prototype of the "Turbo" 68020-based motherboard
that eventually became the IP2.
The antics of the IP numbering system are quite interesting, though - I've
got a Personal Iris 4D/20 which is labelled up as a IP6.5 - and appears to
be an IP12 complete with R3000@20MHz - but underclocked to 12MHz to render
it functionally equivalent to a IP6. Dave Olson states in the past that at
the end-of-life of the 4D/20 SGI ended up underclocking (possibly slightly
defective) 4D/25s and selling them as 4D/20s - although makes no reference
to changing the IP number on those boards. When I get a chance I intend to
replace the crystal with a 20MHz one and see what happens.
Does anyone know anything more about these IP6.5s before I blow up my
mainboard?
Matthew
Colin Anderson
Though the most notable pronunciation I've ever heard came from a
Scottish professor...
Imagine if you will a fellow that looked like a cross between
Groundskeeper Willie and Mr. Burns with a strong but slow Scottish
accent mixed with a wheeze.
"So you use AhhhwwwwReeeekksssss?"
He scared the living daylights outta me. It was pure evil... I could
picture a 20D11 granite monitor with flames shooting up from behind, two
red eyes atop a black background on the display, and his voice saying
"awwwreeeksss"...
Yikes!
Needless to say, I avoided that scary dude.
- Colin
Rob Warnock
+---------------
| To correct my post from further down the thread (and answer the question),
| IP1 never officially existed. If anything, it was a relabelling of the PM2
| board as used in the IRIS 1400, 1500, 2000, 2200, 2300, 2400 and 2500. The
| first actually used IP series board was the IP2 as used in the 2300T, 2400T,
| 2500T and 3000 series machines. The main distinction between the PM and IP
| boards were that the PM were based on reference boards developed at Stanford
| University whereas the IP boards were (and still are, i hope ;) developed in
+---------------
FWIW, somewhere I once heard (though a web search failed to confirm)
that "IP" had originally stood for "Internal Processor"...
-Rob
-----
Rob Warnock, 31-2-510 rp...@sgi.com
SGI Network Engineering <URL:http://reality.sgi.com/rpw3/>
1600 Amphitheatre Pkwy. Phone: 650-933-1673
Mountain View, CA 94043 PP-ASEL-IA
Rob Warnock
| the PM2 motherboard (which is different to the IP1 that was introduced for
| the 3000 series machines) - the board itself is a slight modification of the
| Stanford UNiversity "SUN" board that was around at the time.
I think you meant "Stanford University Network (SUN) board". Those were
originally designed to be Ethernet-connected bit-mapped terminals, to be
sprinkled around the campus providing access to all the various big servers.
Matthew Hodgson
"Rob Warnock" <rp...@rigden.engr.sgi.com> wrote in message
news:9emplr$bah6r$1...@fido.engr.sgi.com...
> Matthew Hodgson <mj...@cam.ac.uk> wrote:
> +---------------
> | In quick response to anyone's questions, they do indeed run 68010 chips
on
> | the PM2 motherboard (which is different to the IP1 that was introduced
for
> | the 3000 series machines) - the board itself is a slight modification of
the
> | Stanford UNiversity "SUN" board that was around at the time.
> +---------------
>
> I think you meant "Stanford University Network (SUN) board". Those were
> originally designed to be Ethernet-connected bit-mapped terminals, to be
> sprinkled around the campus providing access to all the various big
servers.
>
Of course, you are completely right. I must have guessed the acronym at some
point and never thought of reevaluating it - or perhaps I was trying to
minimise parallels with the name of a certain other well known computer
company...
From what I've read, the original IRIS 1000 was essentially just one of
those machines hooked up to the boards that homed the Geometry Engine ICs
that had resulted from Jim Clark's research at Stanford (together with
bitplanes, rasterising hardware, DACs etc.).
Are there *any* of the original 1000 terminals still around?
Do SGI themselves own any terminals or workstations of this vintage?
Matthew.
____________________________________________________________
I seem to be having tremendous difficulty with my lifestyle.
Gerhard Lenerz
> FWIW, somewhere I once heard (though a web search failed to confirm)
> that "IP" had originally stood for "Internal Processor"...
"Inhouse Processor" is what I have read.
Gerhard:wq
Rob Warnock
+---------------
| Rob Warnock wrote:
| > FWIW, somewhere I once heard (though a web search failed to confirm)
| > that "IP" had originally stood for "Internal Processor"...
|
| "Inhouse Processor" is what I have read.
Ah, yes, you're correct!! (I was searching for the wrong term.)
Marcus Herbert
>> Who is going to record it?
> SGI's CEO, Bob Bishop, has a pretty nifty Australian + European
> "international" accent and a nice pronunciation of "IRIX".
> Or maybe we could have Cosmos record a few choice words.
Hey! Jolly good idea.. set up a website with some contest "How to
pronounce IRIX". Will do that later the day :O)
--
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
Halford Dace
>> Do you also say i-ris (short first syllable)?
> in turn, do you call people EYE-rritable? :)
> I call it i-rix (short syllable) but hell, pronounce however it looks
> good to you :)
You do? What would you call, say, a Personal Iris?
While on the subject of our glorious Irix, does anyone know
the current price for a 6.5.x RTU license?
Hal
Harri Haataja
>Colin Anderson <co...@beyondboxes.com> wrote:
>>> Who is going to record it?
>
>> SGI's CEO, Bob Bishop, has a pretty nifty Australian + European
>> "international" accent and a nice pronunciation of "IRIX".
>
>> Or maybe we could have Cosmos record a few choice words.
>
>Hey! Jolly good idea.. set up a website with some contest "How to
>pronounce IRIX". Will do that later the day :O)
Ogg some samples as well.
(My personal: eye-ricks and eire-ish =)
--
"It's seems to be spreading faster as Anna Kournikova"
-- Mikko Hyyppönen on VBSWG.X (fsecure.com)
Dave Olson
| The antics of the IP numbering system are quite interesting, though - I've
| got a Personal Iris 4D/20 which is labelled up as a IP6.5 - and appears to
| be an IP12 complete with R3000@20MHz - but underclocked to 12MHz to render
| it functionally equivalent to a IP6. Dave Olson states in the past that at
| the end-of-life of the 4D/20 SGI ended up underclocking (possibly slightly
| defective) 4D/25s and selling them as 4D/20s - although makes no reference
| to changing the IP number on those boards. When I get a chance I intend to
| replace the crystal with a 20MHz one and see what happens.
|
| Does anyone know anything more about these IP6.5s before I blow up my
| mainboard?
As I recall it, IP6.5 was the moderate re-layout for speed (and
probably some cost-reduction) version of the IP6, and was used for both
4D/20 and 25 (20 mostly to fulfil longer term contracts, and as
replacement boards when necessary). However, it's been 12 years, so I
may be remembering wrong. I guess when I said "under-clocking", I
probably meant that the board was *designed* to run at 20MHz, but if
intended to be sold as a 20, wasn't necessarily tested at full speed
(since their wasn't an upgrade path from the 20 to the 25).
Dave Olson
ol...@bengaltech.com
http://www.bengaltech.com/public/dave