Slashdot Mirror
A Look at IRIX 6.5.17
XFS writes "OSNews got their hands on the latest version of IRIX, 6.5.17 (released in August), and they have published an interesting article about it and they explain why IRIX was and still is, one of the best workstation Unices out there. Especially when it comes to multimedia/GL performance. I hope SGI will do something with IRIX though, as they seem to have let it fall behind and be one of these great technologies that get lost through various corporate focus shifts..."
I think it is pretty interesting that the benchmark that they used measured memory throughput of the graphics CPU, as opposed to, say, an actual workload-handling of the OS. In other words, this is a synthetic benchmark, versus a real-world benchmark. They say, "Look! We can do memory transfers really really fast!"
Unfortunately, memory transfers are not the world when it comes to multiprocessor multimedia boxes. The overhead comes in when you're trying to synchronize a large number of threads/CPUs to do a large task. For example, an Oracle database.
Sun has proven that it scales up the tree very well with large numbers of processors. But from my understanding, Linux is more efficient with a low processor count, and less and less efficient with more processors.
I question its ability to do anything with a real workload. And I've even more suspicious because they use a benchmark I've never heard of to push its superiority on a single-aspect synthetic benchmark.
Not sure what this is implying, but it seems to be a surprisingly common misconception that MacOSX has vector based artwork. Not so. GNOME can do, and I think KDE3.1 can as well, via SVG. MacOS icons though are just bitmaps in a variety of sizes, with some scaling/blending algorithms applied.
The SGI desktop is of course based on a heavily modified commercial X Server. And here I will stop for a second, get a big breath and say: 'wow'. I have never seen an X server being so fast, on a 5-year old machine (no matter if this is an SGI machine or not).
I'd kind of expect this given that IRIX comes as a bundle with the hardware. When you choose the hardware as well as the software you can of course optimize the drivers a lot, so you will get good speeds out of it. XFree has to deal with a lot of different hardware, and the driver manufacturers are sometimes less than helpful. Probably worth remember that IRIX won't have some of the newer X extensions like XRender.
That's funny.
AIX has more (and growing) marketshare than Irix has had in years. The Power4 and Power5 chip is attracting alot of business away from Sun.
SGI has been obsolete since 1996. A $2,500 Dell Workstation can do as much as a $25,000 SGI workstation.
Wake up McFly! It's 2002 calling!
Conformity is the jailer of freedom and enemy of growth. -JFK
IRIX might have enterprise-level performance, but it suffers enormously in the usability department. I work in a lab where IRIX is standard, because 64-bit memory addressing and extreme graphics performance using ImageVision is a must. However, I keep running into issues with the development tools. Most impotantly, SGI's cc (c compiler) is slow and hard to customize flags on, especially for debugging. Furthermore, frequently, if my program commits a memory fault, it receives a SIGKILL rather than a segfault which makes it very difficult to debug (this usually happens if the malloc pool gets corrupted or while using ImageVision).
The ImageVision library (an OpenGL-based image processing system) hsa great performance and features. However, it refuses to link with programs not built with cc (thus, no gcc!). Furthermore, programs that seem to follow spec mysteriously die with a SIGKILL during deallocation. I certainly realise that I might be doing something wrong in the way I call the library, but it does not provide any error
message, exception, or fault.
Finally, IRIX standard header files are a colossal mess and almost impossible to use. Standard C and C++ objects are casually redefined throughout the header structure.
crash-resistant, high-performance file system. Ever heard of "XFS"??? It's journaled and has been around almost longer than the FreeBSD project.
First multithreaded kernel: Um.... Right... Multithreaded kernels have been around for probably a decade if not more. FreeBSD is hardly the first. Irix has had kernel threads for ages. The first reference I can find to them is in '95 (and I suspect they have been around longer than that) when FreeBSD didn't even run on multiprocessor systems.
First "compact" kernel: What is a "compact" kernel? The FreeBSD kernel is a monolithic BSD kernel. Irix is a monolithic System V kernel. Even Linux is a monolithic kernel (of Linus + other's design). Microkernels haven't lived up to their initial hype (though MacOS X uses one), but neither they nor monolithic kernels are "obsolete".
Now don't get me wrong, FreeBSD is a great OS. I have run it in the past and regularly use it. But it doesn't run on 1024 processors, have multiple tens of terabytes of storage in a single filesystem, and manage a terabyte of RAM. It's not designed for that. Irix is.
Go Badgers! -- #include "std/disclaimer.h"
Minor quibble, but it mentions it's different than Linux because it's UCB BSD based. It's not (well, it is up to teh point that SVR4 took a lot of BSDisms) but it's SVR4 machine. Linux distros generally take a bit from classic BSD, a bit from SVR4, and a bit of whatevehell else they want, so they're all a bit different.