Benchmarking different block sizes is absolutely useless. It's ridiculous that they didn't even do a full test of all the common (16, 32, 64, 128) block sizes. No empirical data is obtained here - no direct comparisons may be made of the tested devices because of the laziness of the reviewer. By leaving the defaults, he's assuming the user has no idea what their own data delivery needs are.
The only users who should even contemplate deploying a RAID array will certainly do the research to come up with the ideal stripe block size, given their usage patterns and requirements.
cmd line = 10 year old software. bash/pine/vim = old software. They do what they do well, but they were written during a time when ram/cpu were limited.
Perhaps the charges that Linux developers aren't innovating but simply trying to mirror microsoft are truer than I thought - they even bring in MS' feature-bloat..
Stop trying to be all things to all people. Choose a target market and go with it. I don't think there's anything wrong with trying to be a current-generation windows alternative. Just realize that this is what desktop linux development is aiming at. Windows isn't big because of crappy programming - far from it. Microsoft is in the business of hiring the best software developers money can buy. Windows gets fatter because they're putting more into it. In that vein, so does Linux. Simple enough concept.
Here's a guide for using old computers: match the hardware and software. A 10 year old computer is probably best paired with 10 year old software. If you wanna recycle 10 year old hardware with new software, be prepared for a lot of work, or a lot of suck. Either way the result won't be very pretty, or speedy.
If sysadmins are/will always be in high demand, why did Humber College (biggest tech college in Ontario) just fold its Information Technology Studies department, orphaning all its current students? Why is own Bachelor of Applied Computing program at the University of Guelph-Humber barely able to generate enough interest to get half of a full class admissions for 2004-2005 academic year?
While there may be demand and a decent marketplace for sysadmins, there sure as hell isn't interest in the field for the kids entering post-secondary.
Wow, that's a sign of desperation. I've heard tidbits of information at the major tech news sites that although IBM boasts it can get to 3GHz on the PPC970 architecture, and that a G5's power consumption is very nicely low at 2GHz, that changes very quickly as the speed starts to ramp. The fact that they need watercooling stock to get to 2.5GHz seems to confirm this.
This is quite disturbing. It confirms the overall signs that photolithography scale shrinks aren't working anymore. I had thought that perhaps Intel's problems with Prescott were an isolated incident, but it doesn't seem so now. AMD has only just begun experimenting with 90nm, and now it appears that IBM, the only company so far to have said anything positive about their progress at 90nm, is having to (it would seem) overclock their chips and watercool them to get to a stable and quiet 2.5GHz..
All this talk of crashing is so 5 years ago. In the 2 years I've been working as a sysadmin, in both an academic and corporate setting, with windows 2000, nt, and XP-pro as the standard user O/S, I have not seen a single crash or BSOD. Most of the issues I encounter are, in order:
-dead hardware
-spyware
-login problems (flaky domain servers)
-backup restorations
To be honest, I haven't had a problem with the Stability of a Windows O/S since 2000 came out.
Aww, poor muffin. Microsoft software isn't "cool" enough for you. Haha, you'll learn. Linux is great for those "set it and forget it" chores. Windows is great the vast majority of PC users who don't care to learn another O/S:)
Winamp 5 loads in under 3-4 seconds on my athlon 1.5GHz / 256MB sdram. iTunes, with no playlists at all, took well over 15 seconds to load, and took up more resources than Winamp. Not acceptable:)
I don't use a Mac. A significant majority of iTunes users don't use a Mac. A significant majority of people who do amateur mixing and DJ'ing at parties and the like use FLAC and vorbis. And MPC equals AAC in audio quality, with less processing overhead. Plus the itunes software itself is rather...slow.
MP3 and VBR MP3 are 1 format. AAC is the second. WAV doesn't even count. Apple lossless I've never heard of, AIFF I don't use, and I've never heard of Audible.
Most people I know who do real mixing and amateur party DJ'ing use Exact Audio Copy and either FLAC, ogg, or MPC. Cause MP3 sucks. AAC is slightly better but there are too many competing versions, and ITunes can't handle all of them (iTunes was fine for me playing music IT ripped, but not AAC stuff that Nero made for me...) MPC sounds the best overall and has good compression, and FLAC is needed when compression is NOT an option.
With this thing, hooking up your PC to your stereo system is no longer a chore. Up till now I've been toying with the idea of building a small via Epia box that would plug into my stereo and take music from my network share. Nuts to that, I'll just buy one of these things! It's too bad it has to use iTunes though...Hey Apple! how about making a media player that doesn't take forever to load, and can handle more file formats than...2! I have 300CDs that I archived to mpc format, and winamp/xmms handles them just fine. Confining the player to just mp3 and aac is pathetic. You can do better.
Like in any industry, there will be the ultra-highend enthusiast niche. Alienware, VoodooPC, Falcon NW, and others have been catering to these kind of users for years. Any commentary about pricing is pointless: these people pay big bucks to get bragging rights to the fastest, most tricked out, and beautiful (damn, that alienware case is gorgeous) machines money can buy.
It's the same in many industries, especially the automotive industry. Any commentary about how "it's different with cars, they aren't obselete in 3 years" is pointless: the automotive industry's pace of improvement and innovation is much, much slower than the PC industry's.
And just like with cars, we have nerds who buy honda civics and rice them up with neon lights, big, loud heatsink fans, awesome paintjobs, spoilers, etc etc. (case modders if you're dense).
In Frank Herbert's "Dune", there exist a sisterhood of women called the "Bene Gesserit". With mnemonic devices, hypnotic therapy, advanced meditation, and hundreds of centuries of a selective human breeding program, these women had mastered biofeedback to the point of being able to alter their metabolism, and possibly grant themselves immortality. Thought it echoed this article somewhat.
Anyway, that whole "overpopulation" argument that's cropping up here is bunk: the richest countries in the world have a population growth rate so small that it couldn't sustain their populations without immigration. Now also consider that the only people who could afford this methuselah treatment are the super rich - not too many are there.
I don't really get what you're saying...then again, I'm the guy who yells "take it off!" when the theatre dude comes in to say his shpiel before the movie starts.
If I had any intention of going to the Harry Potter movie, I would go. Regardless of whether there's a ripped copy available online. People don't go to movie theatres because it's their only way to see a flick - they go for the theatre experience: big screen, big sound, greasy food.
It's nice to get a little history lesson from someone who was there during the "birth" of the industry as it exists today. His words strike a chord with me (bad pun, not intended) and I am saddened by how they don't don't seem to offer any hope for our generation of music. Even if the music industry dies, it's not going to happen overnight, and we have to sit here and wade our way through the sludge that sells for "music" these days.
discrete parallel graphics processing has been around for a while. The most notable example of it is probably 3DFX and their Voodoo-2 cards. However, there's a problem with this tactic, namely, in the "diminishing gains" department.
So here's the question:
-How is pixel processing going to work? For a given frame, there is vertex, texture information, as well as the interesting little shader routines that work their magic on these pixels. How are you going to split up this workload between the 2 GPUs? you can't split a frame up between the GPUs, that would break all texture operations and there would be considerable overhead with the GPUs swapping data over the PCI bus. *MAYBE* having each gpu handle a frame in sequence would do the trick, but, again, it's a dicey issue.
It would appear to me that this dual-card graphics rendering is quite similiar to dual-gpu graphics cards. Except, where in a graphics card you can handle cache/memory coherency and logic arbiting easily due to the proximity of the GPUs, with this discrete solution you run the problem of having to use the PCI Express bus, which, as nice as it is, is certainly not that much faster than AGP.
So I say, power to you Alienware. If you can pull it off with Nvidia, ATi et all, great. It's too bad the cynical side of me thinks this idea reeks of those blue crystals marketing departments love:)
Remember the cool CG Link/Gannondorf battle Nintendo showed us before Gamecube came out? And then they created that annoyingly cute cell-shaded Zelda game instead?...I do
check out this new modification of itunes i'm working on!
*OSX_admin bans amateur_developer*
check out this new app I have for osx!
that app now belongs to Apple computer systems, and since you're not old enough to legally sign our dev agreements, you've lost rights to your code you helped with too
Slashdot Mirror
While no-one may restripe an array, they will certainly take the cluster size into consideration when first building the array, right?
Benchmarking different block sizes is absolutely useless. It's ridiculous that they didn't even do a full test of all the common (16, 32, 64, 128) block sizes. No empirical data is obtained here - no direct comparisons may be made of the tested devices because of the laziness of the reviewer. By leaving the defaults, he's assuming the user has no idea what their own data delivery needs are.
The only users who should even contemplate deploying a RAID array will certainly do the research to come up with the ideal stripe block size, given their usage patterns and requirements.
cmd line = 10 year old software. bash/pine/vim = old software. They do what they do well, but they were written during a time when ram/cpu were limited.
Perhaps the charges that Linux developers aren't innovating but simply trying to mirror microsoft are truer than I thought - they even bring in MS' feature-bloat..
Stop trying to be all things to all people. Choose a target market and go with it. I don't think there's anything wrong with trying to be a current-generation windows alternative. Just realize that this is what desktop linux development is aiming at. Windows isn't big because of crappy programming - far from it. Microsoft is in the business of hiring the best software developers money can buy. Windows gets fatter because they're putting more into it. In that vein, so does Linux. Simple enough concept.
Here's a guide for using old computers: match the hardware and software. A 10 year old computer is probably best paired with 10 year old software. If you wanna recycle 10 year old hardware with new software, be prepared for a lot of work, or a lot of suck. Either way the result won't be very pretty, or speedy.
There may be an unexpected upside to lower tech enrollment: less competition for employment?
If sysadmins are/will always be in high demand, why did Humber College (biggest tech college in Ontario) just fold its Information Technology Studies department, orphaning all its current students? Why is own Bachelor of Applied Computing program at the University of Guelph-Humber barely able to generate enough interest to get half of a full class admissions for 2004-2005 academic year?
While there may be demand and a decent marketplace for sysadmins, there sure as hell isn't interest in the field for the kids entering post-secondary.
Wow, that's a sign of desperation. I've heard tidbits of information at the major tech news sites that although IBM boasts it can get to 3GHz on the PPC970 architecture, and that a G5's power consumption is very nicely low at 2GHz, that changes very quickly as the speed starts to ramp. The fact that they need watercooling stock to get to 2.5GHz seems to confirm this.
This is quite disturbing. It confirms the overall signs that photolithography scale shrinks aren't working anymore. I had thought that perhaps Intel's problems with Prescott were an isolated incident, but it doesn't seem so now. AMD has only just begun experimenting with 90nm, and now it appears that IBM, the only company so far to have said anything positive about their progress at 90nm, is having to (it would seem) overclock their chips and watercool them to get to a stable and quiet 2.5GHz..
Sir, I find you so intelligent and so witty that we must now fight, with kives! Yeah, to the death!
All this talk of crashing is so 5 years ago. In the 2 years I've been working as a sysadmin, in both an academic and corporate setting, with windows 2000, nt, and XP-pro as the standard user O/S, I have not seen a single crash or BSOD. Most of the issues I encounter are, in order: -dead hardware -spyware -login problems (flaky domain servers) -backup restorations To be honest, I haven't had a problem with the Stability of a Windows O/S since 2000 came out.
Aww, poor muffin. Microsoft software isn't "cool" enough for you. Haha, you'll learn. Linux is great for those "set it and forget it" chores. Windows is great the vast majority of PC users who don't care to learn another O/S :)
Winamp 5 loads in under 3-4 seconds on my athlon 1.5GHz / 256MB sdram. iTunes, with no playlists at all, took well over 15 seconds to load, and took up more resources than Winamp. Not acceptable :)
I don't use a Mac. A significant majority of iTunes users don't use a Mac. A significant majority of people who do amateur mixing and DJ'ing at parties and the like use FLAC and vorbis. And MPC equals AAC in audio quality, with less processing overhead. Plus the itunes software itself is rather...slow.
MP3 and VBR MP3 are 1 format. AAC is the second. WAV doesn't even count. Apple lossless I've never heard of, AIFF I don't use, and I've never heard of Audible.
Most people I know who do real mixing and amateur party DJ'ing use Exact Audio Copy and either FLAC, ogg, or MPC. Cause MP3 sucks. AAC is slightly better but there are too many competing versions, and ITunes can't handle all of them (iTunes was fine for me playing music IT ripped, but not AAC stuff that Nero made for me...) MPC sounds the best overall and has good compression, and FLAC is needed when compression is NOT an option.
With this thing, hooking up your PC to your stereo system is no longer a chore. Up till now I've been toying with the idea of building a small via Epia box that would plug into my stereo and take music from my network share. Nuts to that, I'll just buy one of these things! It's too bad it has to use iTunes though...Hey Apple! how about making a media player that doesn't take forever to load, and can handle more file formats than...2! I have 300CDs that I archived to mpc format, and winamp/xmms handles them just fine. Confining the player to just mp3 and aac is pathetic. You can do better.
Like in any industry, there will be the ultra-highend enthusiast niche. Alienware, VoodooPC, Falcon NW, and others have been catering to these kind of users for years. Any commentary about pricing is pointless: these people pay big bucks to get bragging rights to the fastest, most tricked out, and beautiful (damn, that alienware case is gorgeous) machines money can buy.
It's the same in many industries, especially the automotive industry. Any commentary about how "it's different with cars, they aren't obselete in 3 years" is pointless: the automotive industry's pace of improvement and innovation is much, much slower than the PC industry's.
And just like with cars, we have nerds who buy honda civics and rice them up with neon lights, big, loud heatsink fans, awesome paintjobs, spoilers, etc etc. (case modders if you're dense).
Patents themselves are held in secret until granted. That's the whole "protecting good ideas from being ripped off" spirit of the process.
In Frank Herbert's "Dune", there exist a sisterhood of women called the "Bene Gesserit". With mnemonic devices, hypnotic therapy, advanced meditation, and hundreds of centuries of a selective human breeding program, these women had mastered biofeedback to the point of being able to alter their metabolism, and possibly grant themselves immortality. Thought it echoed this article somewhat.
Anyway, that whole "overpopulation" argument that's cropping up here is bunk: the richest countries in the world have a population growth rate so small that it couldn't sustain their populations without immigration. Now also consider that the only people who could afford this methuselah treatment are the super rich - not too many are there.
I don't really get what you're saying...then again, I'm the guy who yells "take it off!" when the theatre dude comes in to say his shpiel before the movie starts.
If I had any intention of going to the Harry Potter movie, I would go. Regardless of whether there's a ripped copy available online. People don't go to movie theatres because it's their only way to see a flick - they go for the theatre experience: big screen, big sound, greasy food.
It's nice to get a little history lesson from someone who was there during the "birth" of the industry as it exists today. His words strike a chord with me (bad pun, not intended) and I am saddened by how they don't don't seem to offer any hope for our generation of music. Even if the music industry dies, it's not going to happen overnight, and we have to sit here and wade our way through the sludge that sells for "music" these days.
My birthday's on the 15th of May. What a better present than a dissappointing end to a 30 year long franchise. Please Lucas...don't screw it up...
discrete parallel graphics processing has been around for a while. The most notable example of it is probably 3DFX and their Voodoo-2 cards. However, there's a problem with this tactic, namely, in the "diminishing gains" department.
:)
So here's the question:
-How is pixel processing going to work? For a given frame, there is vertex, texture information, as well as the interesting little shader routines that work their magic on these pixels. How are you going to split up this workload between the 2 GPUs? you can't split a frame up between the GPUs, that would break all texture operations and there would be considerable overhead with the GPUs swapping data over the PCI bus. *MAYBE* having each gpu handle a frame in sequence would do the trick, but, again, it's a dicey issue.
It would appear to me that this dual-card graphics rendering is quite similiar to dual-gpu graphics cards. Except, where in a graphics card you can handle cache/memory coherency and logic arbiting easily due to the proximity of the GPUs, with this discrete solution you run the problem of having to use the PCI Express bus, which, as nice as it is, is certainly not that much faster than AGP.
So I say, power to you Alienware. If you can pull it off with Nvidia, ATi et all, great. It's too bad the cynical side of me thinks this idea reeks of those blue crystals marketing departments love
oh man, imagine a beowulf cluster of those...;)
Remember the cool CG Link/Gannondorf battle Nintendo showed us before Gamecube came out? And then they created that annoyingly cute cell-shaded Zelda game instead?...I do
check out this new modification of itunes i'm working on! *OSX_admin bans amateur_developer* check out this new app I have for osx! that app now belongs to Apple computer systems, and since you're not old enough to legally sign our dev agreements, you've lost rights to your code you helped with too