Slashdot Mirror

The Linux MIPS howto says that the Cobalt Raq and Qude are based on the IDT R5230. I don't know much about the innards of the WebTV Plus, but it you could hack together some kind of alternate boot device (or just replace the ROMS), you could presumably put Linux on it -- but there is no Ether, so it's actual usefulness is questionable.

You can obtain the specs on ICC profiling at http://www.color.org/ but the algorithms are proprietary...

Interestingly enough, Silicon Graphics is a member of ICC, and perhaps an inquiry can be made as to a open/closed-source Linux/*BSD port of their color calibration software (BTW - Photoshop 2.5 was excellent on Irix).

However, as both M$ and Apple are also members of ICC, I'm sure both would object to any OSS port :-/

~AC

It could also be because they want to address a perceived deficiency in Linux, and working on *BSD doesn't do that.

Hmm, I wonder why not...perhaps because *BSD doesn't have the perceived deficiency?

(Half ;-), since I'm still wondering why people are writing a GPL'ed Fortran 95 front end when there already is one that's part of SGI's Pro64 Project.)

To sum up my aborted post, Cray has been evolving from the single processor Cray-1, to the Multi-processor Cray YMP, to the massively distributed T3E. Seymour and Cray Computer Corp. (spun off of CRI in the late 80's or so) failed because they couldn't push as much performance through a smaller number of processors. Eventually the physical laws of silicon (Seymour even tried GaAs to get more performance) take over, and you must expand the number of processing units to get greater and greater performance.

The T3E is a 3D toroidal-constructed system. SGI's Origin uses the Hypercube. Sun uses whatever Cray's Business Unit did back in the day before SGI sold the Starfire, renamed the Sun UltraEnterprise 10000, to Sun (SMP I guess). The model works. That's not disputed.

The High-End market that Cray and Hitachi serves is fairly stagnant (growing slightly more than inflation) at around 1 Billion USD/year (IIRC). It doesn't grow 40% per year like standard PCs, handhelds, or the streaming video & porn market. The pie is only so big, so IBM and Sun choose the bigger market; they're exploiting the internet. ASCI projects don't make much money. They're done for the press they receive. I've heard of companies exploiting Cray's extreme I/O bandwidth for file archivers to tape robots, but that's about it for general purpose. You wouldn't buy an SV1 or a Hitachi to run Apache, that's for sure.

As Durinia pointed out elsewhere in this discussion, Ford and other auto companies still use Cray Vector machines, as well as other research labs, etc. Vector use isn't dead in the US. It's just not the centerpiece, I guess.

along with MIPSpro 7.2.1.3

http://support.s gi. com/colls/patches/tools/relstream/index.html

No major super-huge changes, lots of small fixes and improved support for Octane2 VPro gfx and Onyx/Origin 3000 hardware. Gotta love an OS like this that can tame a 512 processor Origin 3000 with 1 TB RAM yet still work great on my Indy!

What are your goals, how many concurrent jobs will you be running (and with what priorities), and do you know where the bottlenecks reside?

Clustering, high-performance computing in general encompasses a huge number of problems and solutions. There are literally gobs of different routes one could take. Beowulf and Benchmarks, while easy to remeber and look at, are not the solution to everything. Perhaps you need the vector performance of a Cray or maybe the cache-coherent shared-memory system of a Data General AViiON or Silicon Graphics Origin. It all depends on your needs. Do the research before assuming you need one exact solution.

FWIW, you may want to look at SGI's Advanced Clustering Environment for an all-inclusive, free, open-source solution. It's available for both SGI MIPS IRIX and IA-32/Intel Linux and works quite well with SGI's great Performance Copilot analysis software. They also know a thing or two about high performance computing. If you need more power you can build a warehouse of Linux boxes or a buy a 512-processor Origin 3000 (w/ 1TB RAM and 714 GByte/sec bandwidth)... or a cluster of those!

My $0.02

What are your goals, how many concurrent jobs will you be running (and with what priorities), and do you know where the bottlenecks reside?

Clustering, high-performance computing in general encompasses a huge number of problems and solutions. There are literally gobs of different routes one could take. Beowulf and Benchmarks, while easy to remeber and look at, are not the solution to everything. Perhaps you need the vector performance of a Cray or maybe the cache-coherent shared-memory system of a Data General AViiON or Silicon Graphics Origin. It all depends on your needs. Do the research before assuming you need one exact solution.

FWIW, you may want to look at SGI's Advanced Clustering Environment for an all-inclusive, free, open-source solution. It's available for both SGI MIPS IRIX and IA-32/Intel Linux and works quite well with SGI's great Performance Copilot analysis software. They also know a thing or two about high performance computing. If you need more power you can build a warehouse of Linux boxes or a buy a 512-processor Origin 3000 (w/ 1TB RAM and 714 GByte/sec bandwidth)... or a cluster of those!

My $0.02

The 3D graphical file browser seen in Jurassic Park was fsn, a throwaway, proof-of-concept tool developed at SGI. It was real.

Yes, the program that she was using was a real program available for real unix systems--you can download it from SGI's web-site.

It's from Jurassic Park. The girl sits down at a console with a fake 3d window manager and says "This is unix! I know this!"

Actually, it wasn't a fake 3D window manager, it was a real 3D file manager. It is actually named fsn and you can read about it and download it from here.

-jfedor

It just makes more sense, rather than supporting their own OS.

Check their roadmap if you like. They are not saying that, but i think the writing is on the wall.

Look here.

KiboMaster wrote:
...in most movies they usually create a blue screen on the monitor...

That's because with a traditional CRT monitor the contrast would be terrible and if you could see the picture on the screen at all it would have scan shadows. In this case they used an SGI 1600SW digital LCD flat panel so they might not have had to bother blue screening it.

it's called fsn and was an experimental file manager developed by a team at sgi.

ccg

There isn't this huge supply of bad memory out there (Radio Shack jokes aside) because memory manufacturers are pretty clever. Bad memory is put into things like:

Audio storage devices, like answering machines and mp3 players, where a bit or two of failure will just end up as a teeny bit more noise.

Cheap digital cameras (once again, a bad pixel here or there....)

Toys. They actually call bad memory "toy memory" sometimes.

SIMMS. You take (for example) 4 bad chips and 1 good chip and get the equivalent of 4 good chips (by replacing bad io's on the bad chips with io's on the good chip). There are jillions of ways to do this, and companies have pretty much done them all.

Sell them at CompUSA to people who don't know any better. (Sorry, couldn't resist)

If I were you, I'd download memtest86 right now.

• Does it run every possible combination of CPU instructions on boot up? No!
• Does it check every single block on the hard drive? No!
• Does it check all the blocks of floppies, CDs, DVDs, etc to make sure they work? No!
• If the memory test is essential to the functioning of the system, why do they let you skip it?

I'm sorry, but I guess I agree with a previous poster that the SGI "Reality Center" desk displays are truly awesome!

--

I saw these a while ago, and have just been itching to post about them.

If you're a UNIX geek (such as myself), checkout SGI's Reality Center Walls. Be sure to checkout the large images of the Reality Center Walls and Reality Center Room. They're awesome, but the price will make you toss your cookies.

HP makes something similar, but I'm having trouble finding it again on their web page. Any idea what it's called guys (and gals)?

attr(1) - manipulate Extended Attributes on filesystem objects

rfc2624 - NFS Version 4 Design Considerations

Tux2 sounds interesting and all but the other horses are so far along, why bother? I suppose if you just can't stand to backup and reinstall; then this might be FM.

I suppose they are geek creds for the whole "Phase Tree" thing but how long is going to take to catch up?

That said, it does sound like a cool project and something fun to play with but it will be a long, long time before I stick something so new and groundbreaking on a box I rely on.

As soon as it get's outta beta, I'm jumping on XFS' bandwagon.

As the SGI guys say:

• Sub-second filesystem recovery after crashes or power failures (never wait for long fscks again)
• 64-bit scalability: millions of terabytes, millions of files, and a million files per directory (no more 2 GB limits)
• High reliability and performance from journaling and other advanced algorithms

C'mon this filesystem rocks and SGI is releasing it GPL; game over. SGI made on killing on this thing. I doubt if it wasn't for the expected storage projections and CXFS we would be seeing it but it's not like most of the machines I am messing with need that much storage.

Today I'm using Reiserfs, tommorow I am using XFS and if this one catches up in performance and reliabillity then maybe Tux2.

Tux2 sounds interesting and all but the other horses are so far along, why bother? I suppose if you just can't stand to backup and reinstall; then this might be FM.

I suppose they are geek creds for the whole "Phase Tree" thing but how long is going to take to catch up?

That said, it does sound like a cool project and something fun to play with but it will be a long, long time before I stick something so new and groundbreaking on a box I rely on.

As soon as it get's outta beta, I'm jumping on XFS' bandwagon.

As the SGI guys say:

• Sub-second filesystem recovery after crashes or power failures (never wait for long fscks again)
• 64-bit scalability: millions of terabytes, millions of files, and a million files per directory (no more 2 GB limits)
• High reliability and performance from journaling and other advanced algorithms

C'mon this filesystem rocks and SGI is releasing it GPL; game over. SGI made on killing on this thing. I doubt if it wasn't for the expected storage projections and CXFS we would be seeing it but it's not like most of the machines I am messing with need that much storage.

Today I'm using Reiserfs, tommorow I am using XFS and if this one catches up in performance and reliabillity then maybe Tux2.

And then vi x.log to see what error messages the X server might be spitting out. I used this trick to configure my SGI 1600SW and Number Nine Revolution IV. Turns out that I had to put in the exact frequency settings for the monitor definition (couldn't leave them out or use a range).

Other choice is, oh horrors!, to visit Xi Graphics and pay money for working, well developed drivers...

Once upon a time, a programmer couldn't figure out why an SGI Origin IPC app was slower with 4 meg buffers than with 1 meg buffers. Buffer size is directly related to I/O performance, right? An impatient SGI development engineer said, "Cache blowout! Next!" A slow-witted tech writer couldn't quite follow what the engineer was saying, and also had his breed's supestitious aversion to undefined jargon. "If a programmer thinks you can increase buffer size forever, he's as ignorant about 'cache blowout' as I am, right?" After a friendly shouting match and a few mild death threats, the engineer finally explained the concept. And they all became friends again -- until the next bug.

You can read the result at support .sg i.com (free registration required; ignore the request for an SGI serial number).

__________

> Making Incredible Profits from MisAplied Patents, perhaps

Hehe. Nice try. Funny, but no.

Here's a short summary.

Mipmap comes from latin: "multum in parvo" meaning many things in a small place.

Since a picture is worth a thousand words, just look at the pictures here:
Gamasutra article on mipmapping, and Mipmapping pictures and theory

I won't discuss Isotropic or Anisotropic texture filtering, since the pictures can convey the concept clearer then my mangled descriptions. :-)

Although I will leave with:
- Tri-linear filtering is bilinear filtering (2x2 texels) applied between 2 mipmaps.
- You will also (rarely) see the term 'ripmaps' which I beleive the RGB components are stored seperately (instead of interleaved.)

Hope that helps.

Get the MultiLink Adapter and that problem goes away. You get all modes full screen, and it does an excellent job zooming them as well.

...which makes me think that Apple's flatpanels are just colored-plastic covered SGI flat panels.

Unless the colored plastic can turn a 17.3-inch (diagonal) display into a 22-inch (diagonal) display, I doubt that the Apple Cinema Display is a colored-plastic-covered SGI 1600SW.

The 1600SW is fantastic but the Number Nine card could be better

It's also supported by the 3DLabs Oxygen VX1-1600SW. I don't know whether that card's better than the #9 Revolution IV-FP or not. (Presumably one reason SGI switched is that #9 is out of business....)

(and doesn't have a driver in XFree86 4).

One has been checked into the XFree86 CVS tree, so some future 4.x release will probably support it.

I don't think there's currently any XFree86 4.x support for the VX1-1600SW, but it may appear in the future.

They now supply a MultiLink adapter which allows the monitor to accept many types of video input.

...although, as I read SGI's FAQ on the MultiLink Adapter and, in particular, the answer to "What happens if my card is not SuperWide savvy?", you don't get 1600x1024 unless you have a "SuperWide Savvy" adapter - and you may need driver support for that; see the SuperWide Savvy page.

The 1600SW is fantastic but the Number Nine card could be better

It's also supported by the 3DLabs Oxygen VX1-1600SW. I don't know whether that card's better than the #9 Revolution IV-FP or not. (Presumably one reason SGI switched is that #9 is out of business....)

(and doesn't have a driver in XFree86 4).

One has been checked into the XFree86 CVS tree, so some future 4.x release will probably support it.

I don't think there's currently any XFree86 4.x support for the VX1-1600SW, but it may appear in the future.

They now supply a MultiLink adapter which allows the monitor to accept many types of video input.

...although, as I read SGI's FAQ on the MultiLink Adapter and, in particular, the answer to "What happens if my card is not SuperWide savvy?", you don't get 1600x1024 unless you have a "SuperWide Savvy" adapter - and you may need driver support for that; see the SuperWide Savvy page.

The 1600SW is fantastic but the Number Nine card could be better

It's also supported by the 3DLabs Oxygen VX1-1600SW. I don't know whether that card's better than the #9 Revolution IV-FP or not. (Presumably one reason SGI switched is that #9 is out of business....)

(and doesn't have a driver in XFree86 4).

One has been checked into the XFree86 CVS tree, so some future 4.x release will probably support it.

I don't think there's currently any XFree86 4.x support for the VX1-1600SW, but it may appear in the future.

They now supply a MultiLink adapter which allows the monitor to accept many types of video input.

...although, as I read SGI's FAQ on the MultiLink Adapter and, in particular, the answer to "What happens if my card is not SuperWide savvy?", you don't get 1600x1024 unless you have a "SuperWide Savvy" adapter - and you may need driver support for that; see the SuperWide Savvy page.

(Someone please mod this information up as it's fairly important to combat the misinformation raised in the extended posting.)

The SGI 1600SW is definitely still sold an supported and still winning awards. With the fairly new MultiLink Adapter it is no longer necessary to have special video cards -- absolutely any VGA signal capable of doing 60 Hz refresh can drive the screen.

Keep in mind however that a standard VGA signal will go through an analog stage so you lose some of the theoretically possible crispness of pure digital. To keep pure digital you would indeed need one of the supported digital cards.

Here's a set of questions and answers about the 1600SW and the MultiLink Adapter.

I just wish I had one on my desk!

(Someone please mod this information up as it's fairly important to combat the misinformation raised in the extended posting.)

The SGI 1600SW is definitely still sold an supported and still winning awards. With the fairly new MultiLink Adapter it is no longer necessary to have special video cards -- absolutely any VGA signal capable of doing 60 Hz refresh can drive the screen.

Keep in mind however that a standard VGA signal will go through an analog stage so you lose some of the theoretically possible crispness of pure digital. To keep pure digital you would indeed need one of the supported digital cards.

Here's a set of questions and answers about the 1600SW and the MultiLink Adapter.

I just wish I had one on my desk!

Troll eBay looking for either an SGI 1600SW or a Radius Artica ($1500-$1900) - both are the same device: 1600x1024 resolution, wide screen, awesome image quality, uber geek. Only two cards that I know of drive it directly; the now defunct Number Nine Revolution IV or the current 3Dlabs Oxygen VX1-1600. Looks like Xi Graphics have decent X support for both. Otherwise you can get the SGI multi-link adapter ($495) that will take analog DB-15 or DVP/DVI digital inputs and drive it that way. Though, whatever you do, get 100% digital from video card to display. DVI is the current standard in the PeeCee world with support from Matrox, nVidia, ATI on the video card side and more flat panels are coming out that have a DVI-D connector (i.e. Philips 150P) - see Tom's Hardware for a good write up.

Troll eBay looking for either an SGI 1600SW or a Radius Artica ($1500-$1900) - both are the same device: 1600x1024 resolution, wide screen, awesome image quality, uber geek. Only two cards that I know of drive it directly; the now defunct Number Nine Revolution IV or the current 3Dlabs Oxygen VX1-1600. Looks like Xi Graphics have decent X support for both. Otherwise you can get the SGI multi-link adapter ($495) that will take analog DB-15 or DVP/DVI digital inputs and drive it that way. Though, whatever you do, get 100% digital from video card to display. DVI is the current standard in the PeeCee world with support from Matrox, nVidia, ATI on the video card side and more flat panels are coming out that have a DVI-D connector (i.e. Philips 150P) - see Tom's Hardware for a good write up.

Maybe the 1600SW? There's information about that one here. Summary:
* 17.3" display
* 1600x1024 resolution
* 0.23mm DP (110 DPI)
* US$2475 @ buy.com

The SGI 1600 SW was/is a stunning display. SGI completely dropped the ball when it came to marketing and distributing the thing. I have one and a Number Nine Rev IV FP to drive it, but since #9 went out of business, support is impossible. SGI has a matrix of supported graphics cards and the flat panel goes for about $2100.00. It's a truly awesome display. I have never seen anything like it.

The 1600SW is fantastic but the Number Nine card could be better (and doesn't have a driver in XFree86 4). They now supply a MultiLink adapter which allows the monitor to accept many types of video input.

Just the other day, I was browsing developer.intel.com, and found that HP was giving away an IA-64 development kit (including simulator), also for Linux. The URL is: http://developer.intel.com/design/ia-64/linux.htm. Also, SGI is giving away their Pro64 Linux tool suite, apparently GCC & friends with SGI enahancements for IA-64: http://oss.sgi.com/projects/Pro64/. Now I'm no IA-64 enthusiast, but the thought of extending x86 junk to 64-bits is just... chilling.

VRML is a dumbed down OpenInventor with committee add-ons.

Open Inventor is Open Source (buzz, ding) which breaths new life into it. No more big SGI boxes for me. It's installed on a Linux box I have access to, seems fairly fast with a TNT2 class accelerator, I know people who are having fun with it, I'm about to once more. You can hack the ASCII file format like VRML and/or you can code it in C++, unlike VRML, mix and match ASCII/binary/C/C++, lots of funky User Interface widgets off-the-shelf (unlike VRML)...

...another comment from Michael Tandy.

XFS is optimised for dealing with streaming media, and so deals well with high IO and large files.

This is not entirely true. XFS will stream media if you want using GRIO (guaranteed rate IO) features if you set your file system up that way, though I am not sure the Linux version will have that.

I would refer users to the website for a more comprehensive view of XFS. Basically XFS has been running for about 6 years on IRIX machines. It is a 64 bit file system, end to end. It is journaled. It is designed for speed, both at the OS level and at the hardware level (you can hit and sustain in excess of 97% of theoretical max drive performance in various cases with SCSI systems, and large block IO). It is designed so you can have millions of files in directories, with files in the petabyte size if needed.

Basically XFS is really one of the best file systems out there.

Saying it does streaming media is like saying Linux can be used via a vt100 emulator to edit files. It can do so much more, and it does it very well.

Combine XFS with a well designed volume manager, and you can have your file system saturate your IO bus. This is nice if you need lots of IO capability. XFS based filesystems (atop XLV) sustained 7 GB/s (thats gigabytes, not gigabits, per second) several years ago in a test, reading and writing to a single file. The limiting factor was the number of spindles one could attach to the machine. We used 864 if I remember the number correctly.

XFS scales provided the LVM scales.

> What they don't realise is that neither NT actually has a Journal File System

Bullocks. NTFS is a Journaling File System, albeit a crippled one, due to the fact that NTFS only journals META-DATA, not DATA.

http://oss.sgi.com/projects/xfs/papers/xfs_white/x fs_white_paper.html

http://www.executive.com/whats-new/whitepaper.asp

--
"We don't need no stinkin Karma" - 3 Amigos

This is not, properly, a function of the file system.

Instead, you want a file-monitoring daemon, independant of the many fs's you may be accessing. This will include monitoring the changes across NFS, etc., and removable-media.

SGI Irix does this beautifully with its File Alteration Monitor (FAM) and Inode Monitor (IMON).

If you've ever used the 4Dwm filemanager, you'll know what I mean.

Fortunately, SGI has added these projects to its OSS roster, and documented, portable sources are available at: http://oss.sgi.com/projects/fam/

Jeremiah Cornelius

It was SGI's File System Navigator - a rather cool program for doing 3d visualizations of a file system. It's fun to play with, and actually suprisingly usable for normal work.
--

I've also seen some of the really weird stuff done with massive numbers of paper "cutouts" pasted together.

Pretty neat.
----------

You may be interested in the work being done over at SGI: OpenGL® ABI for Linux

can Intel hardware currently run 200-prosessor SMP-systems?

The XFS FAQ recommends marking partitions which contain XFS as type 83. This may screw up ext2 aware utilities such as Partition Magic and Ghost as well as possibly making a mess when doing distribution upgrades! What the heck is SGI thinking?!

from the FAQ:

Q: Can I run XFS on top of LVM?
Yes you can but you need to use a patch for LVM because the XFS kernel contains stephen tweedies sard patches which change the format of /proc/partitions. Also for using XFS on LVM you should not enable the kiobuf option in the kernel XFS compile options. You may get a patch (created by William L Jones) to start from at

http://innominate.org/~graichen/projects/xfs/lvm .patch

See the head of it for details and a list of contributors.