Reliving The Glory Days of SGI
devin15 writes "Remember in the '90's when the tech boom was in full swing and SGI was the darling of the 3D graphics industry, whatever happened to those days? Wired is running an article about a group for whom the glory days of SGI have not yet gone. From the article:" If the Mac community is dwarfed by the Microsoft horde, the number of SGI users amounts to a rounding error.""
One particular quote I found interesting is, ""In the SGI hobbyist world it's not six degrees of separation, it's three, often less. I recently met one of the industrial light and magic guys who worked on Star Wars: Episode II." I find that this happens all the time in the slightly larger Mac crowd. Easy to pick out the users and get an in-depth conversation started. Once you start you find any and all sorts of wierd and useful connections. Heck, thats mainly how I have the current job I have. Also while travelling overseas the other week I ran into a corporate Apple guy that used to work with my boss. Small world definitely, and being an active part of a small, but active community makes it even more personal.
Glad that there are opportunities for people to keep SGI going. I know I sure have looked at all of those eBay auctions at one time just to see what it was all about. At the current going price on some of the older hardware, I don't see what you have to lose.
The best thing about SG workstations was(is) that they came in funky blue or green boxes rather than beige. And this was years before Apple caught onto the idea and applied it to the iMac.
Oh, they were pretty good at their job, but perhaps that's just a coincidence.
I was at a confernce in orlando last week, and there was a parallel conference which seemed to be mostly military simulation stuff, they seemed to be pretty strong there. Guess they moved to the more lucrative stuff.
sorry officer, left my sig in my other computer.
I learned the power of u??x on an SGI workstation about ten years ago. Being stuck on a 386sx system running dos at home I longed for an Irix machine of my own.
I saw this article last week and enjoyed reading it, but at the end I was still left wondering "WHY?" I love old radios and stereo gear so I'm not unappreciative of the nostalgia aspect, but my linux desktop now is, in most ways, just as fulfilling as the old irix system I grew to love.
They're cool looking computers, but in the end that entire stack of SGIs shown in the fellow's home office probaby has about as much power as the Nvidia/AMD box sitting on my desktop. In the end I'd rather have something gorgeously deco that I could keep around for years and upgrade as needed.
I think the publicity SGI got from this end of the business helped the rest of their business. They'd probably disagree, at least at the point they got out of the business.
But via the publicity from this ariticle, /., and others talking, maybe SGI will re-think this. Heck any loss they get from low sales will be offset by the overall corporate business increase, I bet. It's worth the shot.
The whole 'UNIX workstation' market is gone.
Sun? SGI? HP? DEC?
Computers became powerful and inexpensive too fast. Clusters killed the big servers.
I have an Indy that I picked up free, and the real problem is support.
I'd like to get a more up-to-date version of Irix on it, but going from the 6.5.0 disks that I have to the most current releases is a pain. A big pain. A pain that makes the most b0rk3d RPM install look like a hot bath with a supermodel.
I don't want a full support contract from SGI - for a 150MHz machine that would be a total waste of time and money.
What I'd *love* would be a way to get a set of current disks for, say US$30, with the disclaimer "You are on your own. Don't call us, we won't call you."
I've been looking at putting Linux on it, just to have a bit more "support" on the machine. Now that the video subsystem is a bit better supported I may just do that.
www.eFax.com are spammers
When it comes to video, a $2,000 Mac still doesn't have the same capabilities as an SGI machine.
I thought Macs are known for their media handling capability. The fact that you can get one of those 10+ year old SGI machines for dirt cheap now and get better video editing is a bit shocking. Then again, the quote includes the word "capabilities", so perhaps that does not necessarily reflect performance/processing speed.
I find that inst, even in the newer releases of IRIX, makes installing an IRIX system a chore.
Using their latest release and overlays I still have dependencies that can not be met. It can be frustrating to anyone who is used to a sane installer, like the ones provided with Solaris, HP-UX and most Linux distros.
Filesystems were not recreated sometimes when I made the install, and configurations were left on the system. I'm not a Unix god, but that is not how most operating systems install, or how I think they should work.
v4sw6PU$hw6ln6pr4F$ck 4/6$ma3+6u7LNS$w2m4l7U$i2e4+7en6a2X h
Remember in the '90's when the tech boom was in full swing and SGI was the darling of the 3D graphics industry, whatever happened to those days?
I used an SGI Iris 24 bit color workstation with a 21" monitor back in 1990. I still get misty thinking about it. We used them for computational chemistry and visualization. Shading, transparency, GL had it all even back then. Coming as I did from a Vax 750 background, this was pretty amazing. The workstation came with a flight simulator to show off GL graphic power. These were beautiful machines, solid, well engineered. The aethetics have not been surpassed to this day. Sadly, some business guy tried to turn SGI into a PC company, and they alienated their devoted scientific and engineering users. Same thing happened to Sun except they sold out to corporate IT and big iron.
an ill wind that blows no good
I remember when @home went belly up. the headend was packing up the SGI servers that @home had there and I pulled the SGI case badge off of one of them.
I still get funny questions from friends that notice it on my antec case at home and is the best looking company/equipment logo I have ever seen.
I always wanted an Octane, but they are still going for insane prices on ebay, and today it really is not worth tinkering with anymore.
Do not look at laser with remaining good eye.
-- Thou hast strayed far from the path of the Avatar.
#1. Their machines are still propietary. they are using their using Altix system but require an ATI FireGL card. ummm.. no thanks. which brings us to #2. #2. we are now using exclusively windows and linux. my machine(our machines) run faster, smoother and have the latest openGL libraries, functionality. when we want to get a new GPU we get one, take out the old card and plug the new one in. #3. $$$$.. and lots of it. lets say you want to get a cluster with 5 CPU's, along with a host node. each node has a Geforce 6800, 4GB of RAM, 3.6 Ghz CPU's, you buy the software for it, and all the outs and ins of the system. on average this system will cost you $80,000. to buy one SGI box that is inferior to this cluster, even a small SGI supercomputer would not outperform it plus just the MAINTENANCE on this SGI will cost you $80,000 or more per year. this is what it would cost to REPLACE your old cluster after just one year with the latest graphics cards, latest processors and you still have maintenance that costs nothing compared to that. i think we can all agree what the obvious choice of computing power is.
Or $10 for an adapter.
Make sure your monitor supports "sync on green" and an adapter will work. I had trouble with a Sun workstation and adapter with a KDS monitor because Sun uses composite sync and the PC monitor uses seperate H and V sync.
Anyway, last year I noticed SGI stuff going for bargain prices and since it had been a dream machine since 1992 I picked up an Indigo (teal) off ebay for $100, complete with a 19" monitor - shipping was $50, and then picked up a purple Indigo. It's a beautiful desktop with anime wallpaper, transparent aterm windows and is nice for working w/ blender stuff altho rendering jobs get sent to a newer machine.
try { do() || do_not(); } catch (JediException err) { yoda(err); }
I was working in a simulation firm when the times shifted for SGI. We had some SGI RE2's that cost us about 200k £, expencive stuff in other words. My boss gave me an assignment in 1996 to find a graphics card for PC's that we could run our simulator on, and I heard rumour about a company with ex SGI guys that had started to make graphics cards for the PC market. I got the stats for a new SLI card they had made, and was asoniced of what they had in fillrates and such. My co-workers frowned at the stats thinking it was a hoax, but I convinced my boss on a gut feeling to buy the 2k £ card. We actually got a bundle deal with a company called OpenGVS that made 3D API, so it was a good deal. The card lived up to our expectations. When talking to SGI at several occations I got a taste of their arroganse when it came to the PC graphics boards, they rightfully claimed that it was no match for their super-computers, since it was missing FSAA, AF. Still I was getting the idea that their machines were very overpriced, they were in 1996 selling desktops like the Indigo2 for 20k £ and these DID NOT EVEN HAVE TEXTURING ! Now we have PC cards that have FSAA and AF with higher resolutions for a fraction of the price. PC's are so cheap that simulator companes now use one PC per projector, where a SGI have to split its screen into one area for each projector. No wonder they failed to keep the market, they will have to blame themselfs for their arrogance.
Back when SGI's best people split and left to form 3DLABS (or NVIDIA - Forget which. I am sure someone out there will point out that I could have looked it up, but I don't care - my point is still valid), the heads at SGI didn't want to sell just a Video Card. So all those talented people decided to leave and make globs of money (and my 6800 and I thank them!). SGI only wanted to sell their overly priced 100% solution. And by the time they did sell PCs, it was overpriced and way too late.
Atlas Shrugged : Thematic Story
IBM and Sun are hanging in the workstation business because they are making CPUs still. IBM's POWER architecture is thriving, especially with Mac and soon XBox variants giving them mass market reach. Sun, well, I don't know how they do it.
This is my sig.
"This is a Unix system. I know this." - Lex.
http://www.csse.monash.edu.au/~lloyd/tildeImages/F ilm/JPark/
Here a Sig There a Sig Everywhere a Sig Sig...
In 1997, Clemson University spent a couple hundred thousand dollars on a state of the art network of SGI-O2's and a very expensive 8 processor server.
By 2001, my PIII/500+Voodoo 4+RH 7.3 was smoking the O2's. People with new Athlon+(GeForce||Radeon) systems were putting mine to shame. The new cheap-ass Dell workstations in the computer labs would have been better than the O2's at that point.
Spending that much money on hardware that is obsolete in less than 5 years is not a good investment.
The next year they switched to a Linux/MacOS X setup.
SGI faced the innovator's dilemma big-time; it was tricky to cannabalize their $2 billion workstation business for a $300 million graphics card market. And to move from being a full-system vendor to being a graphics card vendor. And even with all the management and business-issue problems, I noticed three problems their engineering effortsg never overcame:
- trouble with quality and shipping on time (see IMPACT)
- couldn't match/switch from 3-4-year development cycles of the workstation business to 6-month product cycles of the PC graphics card business
- engineers were loath to give up control of the chipset/box/OS in order to settle for just controlling the graphics subsystem. They tried to be a full-system player in a PC world. Given that Compaq couldn't really do it (something that was at least semi-obvious at the time), its not a surprise they, coming from the workstation space, couldn't do it with their integrated NT workstations.
- The engineers were delivering product that was differentiated but not in the areas that the biggest customers cared the most about. The benefits of UMA (unified memory architecture) graphics just weren't in sync with what the market most wanted: the fastest 3D at the cheapest price. And in the classic workstation space, polygon-pushing was what was most needed. Half their business was CAD workstations and in the end they lost that to Sun/HP/IBM who didn't have the sexy texture mapping stuff but could render polygons "good enough".
SGI also benefitted from many years from the other workstation vendors under-investing in 3D graphics. When that era ended, even the workstation business they were in got a heck of a lot more competitive.
Anyway, that's what comes to mind when I remember back to SGI in the mid-90s. In hindsight, I don't know of any silver bullets that would have gotten them out of the situation; it was death by a thousand cuts. At the time, I wondered if a merger with Apple would have made sense but it wasn't clear that the disfunctionality of the two organizations at the time would have melded into something better. Maybe a damn good CEO could have helped them carve out a more defensible role in the industry; that's the only thing that got Apple through as far as I'm concerned.
Inside was a collection of workstations all running very impressive (at the time) GL demos with realtime "twist this knob and rotate the champagne glass" kind of stuff.
We have at least three Origin 2000 systems, one is 96 node...so you know the demos must have helped at least some :)
If it wasn't for our Origins running Matlab I probably would not have tried linux until much later. The only reason I tried linux was to use X and run Matlab remotely.
An SGI workstation is about equal to the graphics power of a PS2. SGI learned the hard way that if you need to ride the crest of Moore's law then you need massively large capital investment to do it. Niche 'power' workstations is a dead business.
That was probably the Interservice/Industry Training, Simulation and Education Conference.
I wouldn't look to military simultion for an example of a growth area. Some of the simulators are as old as the planes themselves, 30 years and older, with upgrades every three to five years to keep them up to date. FORTRAN is still the universal language, or at least the F77 dialect. C is starting to take over, but slowly, and Ada still has a sizable presence. In general, technologies and practices lag five to ten years behind the rest of the commerical world.
On the other hand, it is fairly secure work if you can get it. Lots of people can start in simulation and retire in it, which isn't true of a lot of industries. If you can get a security clearance, you are in even better shape.
So, don't worry about international outsourcing - just become a military contractor!
"Hey, I know this. It's UNIX!"
Apple's moving much more in a consumer electronics direction than in a computing direction, but I still think it would have been interesting if Apple had bought SGI while they were developing OS X.
Both companies had a solid niche in computer graphics; SGI's in 3D visualization, and Apple's in 2D design. Apple was going to introduce a UNIX based operation system, IRIX is a UNIX based operating system. Both companies are involved in computing, but not so much in the transactional data processing side that HP/IBM/Sun are involved in, and neither one was ever in the position to make meaningful moves in that market. Both had clientelle willing to spend more on their products than the products of their more direct competitors to get either their specialized hardware or software.
I think it would have benefitted Apple by giving their products more industrial/data center credibility, in addition to general upward mobility for hardware and software, especially in the 3D visualization realm. SGI on the other hand would have gotten access to more mainstream applications (in their late 90s heydey you COULD get stuff like Photoshop for the SGI) and easier integration with a desktop-priced computer.
In the end if it was done right, I think you could have had a really cool computing environment based on a common operation system. Research departments or other entities with uniqure requirements could have been "all Apple" with desktop Macs and machine-room servers all sharing the same user interface and capable of running the same applications (think fat binaries with MIPS and PPC, instead of PPC and 68K).
It might have led to some interesting clustering concepts integrating the desktops and the big boys for shared/distributed computing, NUMA, and other stuff.
Anyway, I think there was an interesting business case for such a merger. Most Apple fans (often rudely) disgree, and think of Apple as perpetually a personal computer/consumer electroncis company when I thought they could have been and done more. Oh well, it's too late now.
I have to agree. I read this and was a bit mistified. I work with and have worked with SGI stuff for years and years. Even their current stuff. They even still write some pretty cool software too, (which I can't talk about due to Gov. Classification). In fact I see more SGI stuff now than I did 10 years ago. They are definetly firmly in the Fed's back pocket. I am sure they are in maore areas, just by the stuff I see, I know it has to be applied elsewhere. In fact there are several Slashdot articles I have read where I think, "If they only knew what was really out there". SGI rocks BIG TIME dudes, (Today more than ever)!
Back before they shrunk their name, Silicon Graphics Inc. had some fun stuff to play with but even in the glory days I wouldn't believe how overpriced it was.
I've used the green boxed machines (their name escapes me), the Iris, the Indy, the O2, and a whole bunch of "oven" machines. All of them very nice to play with but all of which were very expensive. These where the guys who came up with IrisGL which was the forerunner to OpenGL. They went "64-bit" early too although they did it the wrong way (changing the OS moniker to "IRIX64" broke many Makefiles). All was right and good...as long as there was no one else in the same product space.
It was around the mid 90s when several new things started to pop up. Sun and HP noticed how SGI was a "darling" and wanted in on the action and tried to create their own "graphics workstation" both of which weren't as nice and often times a lot cheaper. Around this time, as well know, a little OS known as Linux started to get some steam and a little project known as Mesa started to actually conform to OpenGL.
So now they had pressure from the top and the bottom. I also viewed their buying Cray as a bad move because it didn't make their technology any cheaper to compete against Sun and HP let alone the cheap Windows or Linux workstation with a semi-decent AGP card.
The last SGI machine I saw ran Windows 2K. Such a shame because it was still way overpriced from what you could buy "off the shelf". Maybe things would have been different if they embraced Open Source to cut down the overhead. I honestly don't know. Retreating into the supercomputer product space made me notice how much they were the Amiga/Commedore of the 90s. They were too pretty, too expensive, too early.
Although not actually Acorn.
SGI helped grow (accidently - probably by being too short termist) MANY graphics firms. 3dfx had a good number of ex-SGI staff, nVidia has oodles of them, some are at MS working on D3D (when SGI dropped the ball on OpenGL - it didn't keep up the the HW), 3DLabs has a couple but 3DLabs was always a competitor of SGI (and 6000 miles away!). Most famously is ArtX who I _think_ did the GPU for the Gamecube but are now wholey owned by ATi. Many of the ArtX team had worked on the RIP in the N64 then split away as SGI seemed to drop the ball on that one too.
There's probably more than that. Sorry to be so down on SGI but they REALLY let things go badly wrong....
Time flies like an arrow. Fruit flies like a banana.
I was an SGI employee during the "glory days', and got to watch the company go downhill until I got laid off in the umpteenth wave of "rightsizing". But in spite of the layoff, I love SGI. The only problem with SGI was that they were just too d@mn good at everything. They treated their employees like kings. Their pay was 10-20% above their competitors. They had free sodas and gourmet coffee for employees *before* the dot.com boom. Their machines were always the best-of-the-best. Most powerful CPUs, best graphics, most user friendly OS. Their suppot staff were highly trained degreed EEs who actually knew how the comuters worked down to the circuit level, not fresh-out-of-highschool dweebs with a 3-month certificate in micros@ftology. What happended to SGI is an allegory for what has happened to America in general. Cheap mass-produced commodity junk has taken the profit out of the market, and forced everyone to lower their standards. Veyr much like the SouthPark episode "Something Wallmart this way comes." Ultimately we will all end up buying $100 dollar commodity computers, not because they are good or powerful, but because they will be all we can afford on our $10/hr jobs as janitors of the Microsoft plumbing.
"Sic Semper Path of Least Resistance"
Upon re-reading my comment, I think I was a little harsh on SGI's support. We've always gotten what we asked for. Sometimes is was a little late in coming. And sometimes we wondered why we even needed to ask (e.g. DVD-ROM support in mid-2004). And sometimes it's taken a lot of back-and-forth to really get a solution. But, in their favor, SGI has always been decent about supporting their platform. They just can't keep up with an army of talented volunteer pros and hobbyists.
4-star general in a one-man army.
Then Microsoft bought Softimage, and made them come out with an NT version. The first serious OpenGL graphics cards (DirectX was stil in the future) came from vendors like Fujitsu and Dynamic Pictures. They didn't work very well. Installation required direct cooperation with the board developers. But they did have the 4x4 matrix multiplier for geometric transforms and a hardware Z buffer, just like an SGI machine.
That's when the studios started gettting NT-based animation systems. They weren't standard desktop PCs at first, though. Intergraph sold "high end NT workstations", and it was worth it simply because they could make the graphics board play nice with the motherboard. Softimage on NT on the DEC Alpha had a following.
One real issue for a few years was that it was seen as "unprofessional" to be using a PC for animation. At one point I had a Pentium Pro in a black rackmount case, and industry people asked me where they could get one like that, so their shop would look "professional".
Then came mainstream motherboards with AGP slots, and finally, the graphics board had enough memory bandwidth to work right. Then serious graphics boards went mainstream, and it was all downhill for Silicon Graphics after that.
> something as simple as printer usage was a PITA.
:)
They should have used Impressario. Printing & Scanning made easy - only from SGI
1% APY, No fees, Online Bank https://captl1.co/2uIErYq Don't let your $$$ sit in a no-interest acct.
A few years later, after the SGI had fallen apart and was long since replaced by a far cheaper NT Workstation running the newly ported-to-NT Softimage|3D, I went out and bought a PDA to assist me with my meeting and contact organization.
I was flipping through the technical specs in the manual when I ran across:
Casio Cassiopeia E-100/105 Info
Processor: Mips R4000
ROM CPU: 131 MHz
I'm Rick James with mod points biatch!
I don't think you know the difference between editing and compositing for vfx (which is what inferno does).
Avid is still the standard editor in Hollywood, although FCP has been making major inroads.
It's not offtopic, dumbass. It's orthogonal.
The US Postal Service has thousands of SGI O200 and 1100 computers in use as backend processors for image recognition. Any time you send a letter, an image of the mail piece is sent to a system with racks of them, to be recognized on custom software from Lockheed-Martin. The O200s are actually not bad computers, they have a lot of ram and fast scsi drives, and quad Mips processors running between 200 and 400 Mhz, although parts for them are fantastically expensive. Of course they are running IRIX. The 1100s are just 1U rackmount dual proc Pentium IIIs running linux. One of the main reasons IRIX was used was the availability of an OSI networking stack, which is used to communicate to some of the ancient-but-still-working-well sorting machines. The strange thing about all this is that I am usually the first one to evangelize the networking abilities of Linux, but I've never seen an OSI stack for it.
Did you mount a military-grade, variable-focus MASER on an unlicensed artificial intelligence?