Slashdot Mirror
SGI Faces Bankruptcy
Richard Finney writes "
The stock chart tells the story: One time Silicon Valley high-flyer and contender for the Unix crown, SGI stock price dropped 20% on Friday ... deep into penny stock territory ... after releasing fiscal fourth quarter results. The Mountain View, California maker of high end computers is '
exploring financing alternatives with its lender and other sources.' With mounting losses and investors giving ol' Silicon Graphics the thumbs down, things aren't looking good."
It's a shame to see a company that had such interesting hardware and operating system going down. I used IRIX on an O2, and loved it. Was way ahead of its time.
But I have to ask, is there really any reason why to get an SGI today? I can see a company with an installed base of SGIs upgrading or what-not...but do they really offer anything new or different?
This is not a troll, it's an honest question. Back in the budding early days of the workstations sure, I could see getting these machines to work on 3D graphics etc etc. But now that 3D graphics cards are on regular PCs and Macs and both can run UNIX type operating systems, what does SGI or SUN for that matter have that you can't get elsewhere?
I'd be interested in knowing what others think about this or why they would keep going to SGI.
"Leo Fender was in a 'state of grace' when he designed the Stratocaster." -- Paul Reed Smith
I still can't figure out why anybody would buy a Sun box?
Because some people need big SMP systems with operating systems that have the features that big organizations need.
Linux-on-Opteron plus the OSS tool makers are getting there, but not yet.
"I don't know, therefore Aliens" Wafflebox1
If you're thinking about companies that'll buy up SGI rather than just kill them off, don't forget that other Jobs' company with a focus on software rendering products. They've been known to also be big consumers of SGI stuff when they're putting together a new demo reel every couple of years.
PowerPC in Apple, SPARC in Sun, and now MIPS in SGI...
Don't forget the DEC Alpha. Fastest CPU all thru the 90s.
"I don't know, therefore Aliens" Wafflebox1
owned an indigo2 for a while, nice r10000 mips. nice having a 64-bit cpu and operating system back in 1999. well designed too.
the problem with sgi is that it's been living in the year 1995 since 1990, which was working well for it for a while, but when commodity gear just starts killing your performance and cost there comes a point where you have to move on to a new platform. this is like sun, except sun seems a little farther along and willing to keep pushing forward, while sgi just keep digging bigger and bigger holes for themselves.
sad, but the dot-com boom which fed these companies also birthed the commodity pc boom which killed them. i actually want to lump apple in that same catagory, but unlike the rest which stayed in their path and carved themselves farther and farther from the mainstream, apple kept pushing to keep their market position, and in pc's managed to keep their niche. surprising, but their success in the last few years had very little to do with their core pc business, and everything to do with i*'s keeping their brand warm.
just hope these same market forces end up killing the ms monopoly they created, an good open sourced os (not necc. linux) would make a lot of the hardware innovation that stopped post-lintel possible again.
The first rule of USENET is you do not talk about USENET.
Oh I agree that cycles per second is not a useful measurement of performance. The point of my post was that even if your application performs two or three times better per clock on a given architecture you are going to get beaten if you competition does four mhz for every mhz you do at a fraction of the price.
And anyway I have a lower slashdot ID, so I win.
I am a lawyer and this constitutes legal advice and I shall indemnify you against any losses arising from taking it.
1. Make something that is X better than everything else.
2. Count on the fact that people will pay Y times the common average going rate for "the best".
3. Charge X*Y+Z where Z is an arbitrary high number chosen by management who are paying more attention to the stock prices than the computer science.
4. Neglect the fact that while many people will makes googly noises about "the best", they will go for "good enough" in proportion to the constant Z, and that this effect will increase over time.
"I may be synthetic, but I'm not stupid." -- Bishop 341-B
They are both assets. If you think drivers are so easy to write, why don't you try writing one? Here's a hint: the main difference between a professional card that sells for $2000 and a gaming card with the same chipset (which sells for $200) is the drivers.
NVIDIA has a very good and very fast OpenGL implementation, not to mention lots of optimizations and tricks. The driver is as much of an asset as the hardware; it's certainly just as important for performance. If you've ever used ATI's version of OpenGL (which is half-assed at best), you'll realize how much of an asset the driver really is.
I worked at Videologic when 3dfx were in their ascendency (comprising a lot of SGI engineers). We were producing some fine graphics chips (yup, the dreamcast STILL looks damn good to me ;-) and so were they, nVidia were giving us the TNT and TNT2 and _STILL_ SGI were trying to charge mega-bucks for performance that could be got straight from the shelf at a fraction of the price & AGP was just around the corner. 3DLabs (worked there too!) whose chips _are_ very good at geometry - a corner stone of 3D rendering - started making serious efforts on windows drivers... and the game was over for SGI
.com boom is over a lot of people still made a lot of money in share option trading at that time.
Yup, they should've done graphics cards. At one time they had all the knowledge they needed but i guess someone high up the company didn't like the competition or cut throat margins so decided not to. A lot of engineers jumped ship to nVidia or 3dfx, I guess they realised the money was goning to be elsewhere.
3D software ended up running OGL or DX under Windows using cheap 3D hardware. Since then few have considered SGI.
So... who killed SGI? Lack of "vision" really causing engineers to jump ship when they soppted opportunities elsewhere. Let's not forget that althhough 3dfx are gone and the
Time flies like an arrow. Fruit flies like a banana.
I've been working with Altix boxes for the past six months. They rock. The Numalink stuff works really well. I particularly like their FPGA boxes. The only competition for them in that arena is Cray, strangely enough. Nobody else can stream data into FPGAs at 6.5Gb/s straight out of the box. Nallatech, Starbridge, and the others are just wannabes in that arena. If SGI can get their FPGA boxes into the mainstream market they may have a chance for the Altix line to save them (or at least the engineers working on those ;-).
Unfortunately Sun
- Charges about as much as a low-end PC for the SunRay thin client
- Charges about as much as a mid-range business PC for the SunRay thin client if you want little frills like, y'know, a monitor.
- More or less requires the use of Solaris to use Sunray, which makes quite a bit of sense when you consider SunRay necessarily requires a hugeass multiproc server stowed somewhere, but which, seriously, is not something many people would want to use as a desktop OS. You could maybe sell the end user on Linux, if you set it up quite specifically. Solaris, um, that's a lot harder. The upshot of this is that SunRay probably only appeals to that small number of companies where everyone is or can expected to be a UNIX user.
So between these things, the only places I've aware of in the entire world using SunRay in a way that demonstrates its potential are large universities with big Sun contracts, and, um, Sun itself. If there's another business using this system I don't know what it is.I think this is kind of representative of Sun as a whole right now. They've got a WHOLE bunch of promising ideas and services and products. But they're not quite where they can be useful in a real world situation-- there's just those two or three simple-but-difficult-to-solve issues that hold it back from people buying it. In every case Sun could probably address these issues if they thought really hard about exactly who they wanted to buy this and why-- that is, they've got the neat tech but they don't have a clear picture of exactly what (not "it could be used in a multimedia telecommunications infrastructure!", an actual exact product) this tech should be used for in the real world.
In the meantime, the energy that could be used on figuring out how to leverage or market the things that Sun offers but no one else does (SunRay-ish stuff) is all being diverted into fighting uphill battles, mostly trying to keep a market presence for Sun's not-so-unique products-- for example, the Solaris vs. Linux fight-- which are still the cornerstone of Sun's business, but aren't necessarily the company's strength anymore now that similar or interchangeable products have become more commonplace.
I'm sure they're trying to figure this out also, and I'm sure there's some way Sun can change this situation, but I don't know when or if it will happen.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
Completely untrue. I was at SGI at the time. The grpahics workstation business wasn't great, but not collapsed.
The original poster was completely correct, the Microsoft deal burned $300 million of much needed cash.
The Farenheight debacle was another aspect of it. *DONT* deal with Microsoft. Just don't. Ever. No matter how attractive it looks on the surface.
But greed keeps people thinking "but it'll be different for *me*. They won't screw over *me*. I'm different....). Wait until Microsoft pulls the plug on the Microsoft/NetApp agreements for more of the same.
Jeremy.