8 Aug
2012
8 Aug
'12
2:17 p.m.
On Wed, 8 Aug 2012, Richard wrote:
I think you are thinking of the 1100 and 1200. The 1400 used the larger
20-slot deskside chassis (same as your 3100) and was a complete
workstation with a hard drive for local storage. Historically speaking, it
is an important system in the timeline of both Silicon Graphics and 3D
graphics, but strictly for demonstration purposes, a 3x00 is going to have
more memory and graphics processing to work with.
Well, even SCSI drives are getting harder and harder to find (in working
condition and at sane prices). Eventually we are going to have to find a
cost-effective way to reliably replace such SCSI drives with currently
available parts. [Yes, CF-IDE + IDE-SCSI solutions have been used by some,
but this tends to be prohibitively expensive for most people.]
Well, by repairs I'm even talking about the failed ring/fork terminal
connections I had to replace for the DC power supply wiring. Some of them
had been poorly crimped and had failed (we've all see this). The "preserve
everything exactly how it is" mentality from most "museums" would mean
that they wouldn't cut off the burned terminals and reterminate the wiring
(with exact replacements even).
Tothwolf <tothwolf at concentric.net> writes:
Considering the size of the 4D series machines I have, that isn't that
surprising. Each deskside machine would pretty much have to be shipped on
its own pallet.
To me at least, the Octane seems more
"uncommon" than the 4D series... I
own 5 different models of 4D series machines, including one which is
housed in a large brown rack (all 5 machines came from the same place),
yet I've never seen an Octane in person.
Octanes are offered up regularly on ebay. Personal Iris a little less
often, but they still have been showing up with some frequency. What
is harder to find is the larger machines. Anything that can be easily
sent by a parcel carrier shows up with more regularity than something
that requires freight shipment. It's been a long while (multiple
years) since I've seen a deskside Onyx for sale on ebay. I lucked out
that a whole bunch of stuff was being surplussed from Boeing in St.
Louis all at the same time: Onyx 2 Reality Monster, Challenge XL, Onyx
XL, Crimson, Onyx deskside, spaceballs. The 1x00, 2x00
and 3x00 used the same deskside chassis, so the average
person just looking at the exterior of the computer wouldn't see much
difference between them. You 3100 being complete and fairly well populated
with boards might actually make a better working demonstration than an
earlier machine such as the 1400.
Yes, it's nice to have the "high end" of the line for demonstration
purposes, but it's also useful to have a low-end member of the line to
represent the range of the product at that time. Also, the 1x00
models were more like terminals and the 3x00 models were more like
workstations, so there are other differences besides performance and
capacity. The real
challenge with my 1400 I think will be the hard drive. I pretty
much put my 1400 project aside when I discovered that the V170 hard
drive's head lock solenoid was stuck because I just don't have the means to
service the drive (oddly enough, the drive shows signs of previous service
work).
Yeah, archiving any perishable bits (PROMs, hard drives, floppies,
etc.) is always the most challenging part of any restoration, I think.
It's always what's on my mind first before I start doing anything else
with a machine. With later SGIs, the hard drives are commodity SCSI
drives, so archiving them isn't difficult. Most museums
would never even consider making the repairs that this IRIS
1400 requires to get it fully functional.
Even the CHM has done restoration projects, but compared to the ones
they've done so far I expect an IRIS 1400 would be pretty far down on
their priority list. For me, it would be high up there as I'm focused
on graphics. LCM might do it, but even for them the IRIS 1400 is
getting a little far afield of their primary focus (minicomputer
timesharing).