Even if an upgrade is available, nobody actually upgrades every year.
Apple users do. Then again, maybe that's only because Apple makes sure its new products are worth the upgrade.
Re:Keep putting it off. Please !
on
Longhorn in 2006
·
· Score: 1
Windows Server 2003 had that one major exploit since it shipped.
Meanwhile, every major Linux distro has several each month
A theoretical remote shell vulneribility in an SSL library (used only on SSL servers) is a far cry from an exploitable remote root hole in the default install. True, Linux isn't perfect. Still, I could have left my Linux box unpatched for the last year and it'd still have no known remote holes. The same could not be said of a similarly configured Windows machine.
How does it compare to Crystal Space which runs on many platforms, and has a large amount of developers?
Unreal also runs on many platforms and has many developers, plus it actually looks half-decent, has good physics, nice scripting, and powerful dev utils.
I can imagine there are a small number of Linux families--multiple people to a single computer
Family boxes aren't the only multiuser Linux machines. I imagine there are quite a few Linux servers with hundreds or thousands of users each. That would more than make up for the discrepancy.
Linux media players have got to be the worst of the bunch
Admittedly, xmms, xine, and gmplayer suck gui-wise, but not all Linux players are bad. Look at totem, rhythmbox, and mplayer for some good UI design in Linux players.
No, it would just be a blowout by the human, if it were even possible to arrange such a thing (which it's not, the human brain has no clockspeed and works so differently from any computer that their speeds cannot be compared).
It'd be very difficult to make sure the returned data was legitimate: someone could poison the results and give Fritz bad moves. That's not really a problem for d.net: they can always just verify the results on another computer. For Fritz, time is critical, and the network overhead would already be eating a lot of it.
With the exception of a mirror link on a highly-loaded story, links from the/. comments section rarely get more than a few dozen clicks. It's really not a very big deal.
Yeah, but what game uses it as a second keypad and requires you to type numbers with it at the sme time? None? Then why not just map '8' to 'up, '2' to 'down', etc. in the game's control settings, and get rid of Num Lock?
In order to access the Internet, VMWare would have to go through the "real" os, which would have to have internet connectivity. It would be about as secure as running something in a chroot jail or under user-mode Linux - better than usual, but not perfect.
Maybe you could animate Toy Story in an afternoon, but could you write the script and do the voice acting in that timeframe? Could you make a Matrix, LOTR, or Monty Python film? No, you couldn't. There's a lot more to a good movie than the computer graphics.
I believe what Google is actually searching for is "to be" OR "not to be", which may or may not bea bug, depending on what you consider correct behavior to be.
Re:Won't work unless everyone implements this
on
Spoofed From: Prevention
·
· Score: 2, Insightful
True. But, if you implement it, you can be sure that no spammer will forge your domain, which can save a lot of headaches.
Obviously, a change in bitness is not the only thing that can make a computer obselete. However, 386's could still run almost all modern software for many years after the 486 was introduced. Meanwhile, 286's were virtually useless a few years after the 386 was introduced.
What I'm saying is that, in a few years, all Mac software will be compiled for the G5 and its successors. Most companies will continue to compile versions for 32-bit platforms, but eventually they'll stop. At some point, probably within four or five years, you're gonna have to upgrade your perfectly-usable computer (and it will be perfectly usable, just as P3 450Mhz boxes are perfectly usable now) to run some program that's not compiled for your architecture. For many people, that won't matter, as they would have upgraded long before that time anyway. But, for some people it'll be a big deal.
I never said that anyone should stop buying 32-bit computers. I merely mentioned that 64-bit computers do have at least that one advantage, which is quite true. There's no need to stop buying 32-bit computers, but, if you're planning on keeping your new box for a long time (as you would be with most top-end boxes), it's something you should consider.
Really? What would you say the fastest thing is on the Windows side? A top-of-the-line dual Xeon? They're neck-and-neck, and the G5 is cheaper. A top-of-the-line P4? The G5 smokes it, except maybe for gaming. An Opteron? Probably beats the G5 in a few cases, but costs more for a comparable system. The G5 is a far cry from being a far cry from the fastest x86 box.
Slashdot Mirror
Yes.
Even if an upgrade is available, nobody actually upgrades every year.
Apple users do. Then again, maybe that's only because Apple makes sure its new products are worth the upgrade.
Meanwhile, every major Linux distro has several each month
A theoretical remote shell vulneribility in an SSL library (used only on SSL servers) is a far cry from an exploitable remote root hole in the default install. True, Linux isn't perfect. Still, I could have left my Linux box unpatched for the last year and it'd still have no known remote holes. The same could not be said of a similarly configured Windows machine.
I thouhg he was busy uniting the Triforce and saving Princess Zelda...
Unreal also runs on many platforms and has many developers, plus it actually looks half-decent, has good physics, nice scripting, and powerful dev utils.
The only problem being that it doesn't actually work...
I doubt Apple has the production capacity for that many G5s.
Why not?
Family boxes aren't the only multiuser Linux machines. I imagine there are quite a few Linux servers with hundreds or thousands of users each. That would more than make up for the discrepancy.
The GameCube is already $99, and has been for a week or two. The rumor is that the PS2 may soon also be $99.
Most of the nonstandard bittorrent clients include support for bandwidth throttling. Hopefully, the official client will too in the near future.
Admittedly, xmms, xine, and gmplayer suck gui-wise, but not all Linux players are bad. Look at totem, rhythmbox, and mplayer for some good UI design in Linux players.
No, it would just be a blowout by the human, if it were even possible to arrange such a thing (which it's not, the human brain has no clockspeed and works so differently from any computer that their speeds cannot be compared).
It'd be very difficult to make sure the returned data was legitimate: someone could poison the results and give Fritz bad moves. That's not really a problem for d.net: they can always just verify the results on another computer. For Fritz, time is critical, and the network overhead would already be eating a lot of it.
lie - to present false information with the intention of deceiving. As Saige's comment was not intended to decieve, he was not lying. Q.E.D.
With the exception of a mirror link on a highly-loaded story, links from the /. comments section rarely get more than a few dozen clicks. It's really not a very big deal.
Some of us would give a lot to be able to hit 25KB/sec.
Yeah, but what game uses it as a second keypad and requires you to type numbers with it at the sme time? None? Then why not just map '8' to 'up, '2' to 'down', etc. in the game's control settings, and get rid of Num Lock?
In order to access the Internet, VMWare would have to go through the "real" os, which would have to have internet connectivity. It would be about as secure as running something in a chroot jail or under user-mode Linux - better than usual, but not perfect.
Maybe you could animate Toy Story in an afternoon, but could you write the script and do the voice acting in that timeframe? Could you make a Matrix, LOTR, or Monty Python film? No, you couldn't. There's a lot more to a good movie than the computer graphics.
I believe what Google is actually searching for is "to be" OR "not to be", which may or may not bea bug, depending on what you consider correct behavior to be.
True. But, if you implement it, you can be sure that no spammer will forge your domain, which can save a lot of headaches.
So would the Lexus, unless they're really enormous.
What I'm saying is that, in a few years, all Mac software will be compiled for the G5 and its successors. Most companies will continue to compile versions for 32-bit platforms, but eventually they'll stop. At some point, probably within four or five years, you're gonna have to upgrade your perfectly-usable computer (and it will be perfectly usable, just as P3 450Mhz boxes are perfectly usable now) to run some program that's not compiled for your architecture. For many people, that won't matter, as they would have upgraded long before that time anyway. But, for some people it'll be a big deal.
I never said that anyone should stop buying 32-bit computers. I merely mentioned that 64-bit computers do have at least that one advantage, which is quite true. There's no need to stop buying 32-bit computers, but, if you're planning on keeping your new box for a long time (as you would be with most top-end boxes), it's something you should consider.
Really? What would you say the fastest thing is on the Windows side? A top-of-the-line dual Xeon? They're neck-and-neck, and the G5 is cheaper. A top-of-the-line P4? The G5 smokes it, except maybe for gaming. An Opteron? Probably beats the G5 in a few cases, but costs more for a comparable system. The G5 is a far cry from being a far cry from the fastest x86 box.