The kernel "Linux", since that's what the kernel makers call it.
The distro whatever the distro maker calls it: "Red Hat Linux 8.0".
The POSIX utilities whatever their authors want to call them: "GNU grep".
Emacs whatever the emacs author wants to call it: "GNU emacs".
If everyone just named things what their *creator* wanted them called, life would be a lot simpler. It would also get rid of the stupid GNU/Linux debate.
Much as I hate to say it, the two are lightyears apart.
Hurd isn't really usable, unless you're doing it for the pure diehard ideology of it. It isn't technically notable (at least at the current time) and doesn't even begin to compare to Linux from a performance point of view.
When you see an announcement about Linux, it's usually something along the lines of "new major feature annouced that makes Linux better at foo than operating system baz". With Hurd, it's usually "Basic UNIX functionality foo now works, sometimes, assuming you don't push it too hard".
There are a lot of (better) kernels out there. BSD. Linux.
Frankly, the only people that get excited about Hurd are the uber hard core FSF folks.
With luck, in six years the Hurd might be up to where Linux is today.
Finally, one last point: the main reason Hurd exists is so that Stallman can slap the FSF's "GNU/" on top of an OS. It isn't to address some specific technical failure of Linux.
Outside of universities, and a few companies, how many non-techies used the Internet before the whole Information Superhighway thing that Clinton/Gore pushed as an agenda? It wasn't a tool that most people knew how to use or wanted. Sure, I could hop on a VAX terminal and cruise around, but I didn't see widespread desktop systems with Internet access around until after Clinton/Gore.
Clinton/Gore went for the schools, got people used to it, and funded library and other public access systems to try to get as many people able to use the Internet as a tool as possible, and gave incentives to telecom companies to build networks as fast as possible.
I suppose saying "build the Internet" is a bit overkill -- the procols were in place, and there was a large, working network. Building the infrastructure that lets the thing exist today, and makes it available to everyone, though....it isn't just an academic tool, or a tool used for a couple of UNIX geeks to chat via talk.
Also, I think defining it as building the Web is a little too harsh. The Web *did* happen to get popularized around the same time, but it certainly wasn't because the government was directly pushing Web browsing.
This isn't an issue of "lighting fast". The proposed solution, at least without serious modification, would massively fragment the hard drive. The only reason you don't *care* about fragmentation is because you enjoy the pleasant fruits of the fragmentation-resistant ext2, so you don't realize how bad fragmentation can get. The proposed system would fragment the filesystem so badly that a well-used FAT32 system would look contiguous as hell.
You could make a usable system that's somewhat similar...it could shift files around and use, say, a third of the free space for old files.
The Mac OS had a Trash Can from its introduct in '84. I believe that the Apple IIgs shipped with an OS with a Trash Can, and it wouldn't surprise me if the Lisa did the same.
Slashdot has had major software announcements for a long time. The idea is to provide an interesting medly of content that keeps nerds happy. When PHP forum bugfix notifications start coming up on Slashdot, then you can complain
There's been a lot of discussion about having a form of undelete in Linux on Slashdot. This is quite relevant.
Al (and Clinton) shoveled massive amounts of federal dollars into producing the Internet. If not for their strong pushing of spreading the Internet all over (starting in educational institutions), we wouldn't be anywhere near where we are today.
Yes, Al misspoke. But he was also crucial to the Internet being what it is today, so he gets some points.
Your mother is comfortable *installing* XP, or is this just hyperbole?
Linux really isn't hard to install any more, at least to the point of getting it up and running. I'd call the installation process on at least RH simpler than the Windows procedure.
That being said, configuring stuff not-out-of-box is where things get ugly. It's damn easy for an end user to just get a new video card, download their InstallShield program, and use it. And to *uninstall*, simple as that may seem.
Software packaged by your distro "just works" and at least with RPM is really easy to install and uninstall. However, a lot of drivers are not packaged in said manner. Sometimes you can't get a driver to compile, or instructions are written for another distro. Got a laptop with a wireless card, or an Nvidia card, or a weird USB device? If it works in Linux, the install procedure is not necessarily trivial.
A few other things that are nasty include:
* Networking. i swear to God that there's either a bug in the Linux kernel or in RH's networking scripts since time immemorable, since *every* system I've ever used will sometimes, despite the fact that the routing tables are correct, refuse to properly route information. I can pretty consistently get this on a wide variety of RH's distros by running/etc/rc.d/init.d/network restart and changing settings periodically. Of course, everything works beautifully after a cold boot...just not after restarting the network. I've seen this since at least the early 6.0 days, and up through 7.3.
* Windows has ZoneAlarm. Linux has the amazingly powerful iptables, with *no* really good, really solid front ends (though lots of half-finished freshmeat projects). If you want a personal firewall, Linux can give you an incredible amount of power...*if* you're willing to fight with iptables for a few days.
* Linux has *no* fully working, reliable ICQ program. This is an embarrassment. It isn't really Linux authors' fault -- trying to reverse engineer ICQ is not trivial -- but if I try to send a Windows user a file and can't, the only thing they learn is "Linux can't do IM properly". Yes, I know about Jabber -- which no one uses.
* Linux has, AFAIK, *no* finished, fully featured 3d modelling programs. Someone who likes to dabble with 3d work can run out grab lots of low end 3d modelers on Windows. There are *tons* of Pov front ends, none of which begin to compare to fully blown Windows modelling programs. Oh, and I'm not talking about multi-thousand dollar movie studio packages -- I mean stuff that a home user could use.
* Linux has *no* finished, fully featured vector graphics programs. Yes, lots of projects underway like sodipodi, sketch, kontour...and none of them are remotely usable for a real life production artist.
Good points. Here's a few more.
on
Red Hat 8.0 Reviewed
·
· Score: 5, Informative
Insightful.
I'd also like to point out that I've done a bit of benchmarking gcc, and optimizing for a particular processor makes almost no difference on the vast majority of software.
The biggest win comes from flipping on -O3. Then if you can get away with it, -fomit-frame-pointer, which helps the register-starved x86, but keeps you from looking at stack traces and debugging crashed programs (or sending in useful bug reports). -fexpensive-optimizations have also helped a bit too, and for certain packages, -ffast-math can be big. -march=pentium2 makes next to no difference on anything I've tried benchmarking. -DNDEBUG is potentially good...seems like most production software is compiled with assertions enabled, when they're really intended for debugging.
The Pentium 1 sucked at running code compiled for the 386/486. This is why you got compilers like pgcc, a Pentium-optimized Mandrake distro, and lots of talk about architecture optimizations. With the Pentium 2, Intel realized that all software was not going to be recompiled for each processor (at least in Windows land), and did a really solid job of running 386 code.
So, as far as architectures go, the Pentium 1 is the odd man out. If you have a Pentium 1, it sucks to run any code other than stuff compiled for your chip. If you have anything else, you'll generally get very minimal gains from compiling specifically for your processor instead of for the 386.
Finally, most people don't actually care about the maybe 10% speedup they can get by recompiling software using optimization flags other than just -O2. They care about interactive latency. Look at Mac OS X. OS X is *hideously* slow, but it *feels* pretty fast because it has good UI latency -- it jacks UI priority and puts a lot of emphasis on slapping something on the screen that's updated as soon as the user does anything.
On Linux, here are the big culprits.
Jack the nice value of X from 0 to -10, if your distro doesn't already do so (take a look in top and see what it's running at). The nice value doesn't make it much "faster", but it does significantly improve latency, so you can get crisp edge-flipping and updating.
Turn *on* DMA and umasked interrupts (insert usual warnings about potential problems with *really* old computers having these on). hdparm -u1 -d1/dev/yourharddrivenamehere. Significantly reduces "jerkiness" in X when doing disk access, including paging. For a long time, a lot of distros left this off by default.
If you're doing something that doesn't need low latency in the background, *nice* it. I run all compiles niced to 20. I can be compiling six or seven packages with no user-perceptable slowdown at all. Software that's always sucking down a little CPU in the background but still should be interactive (like lopster or gtk-gnutella) should be niced to 5 or so.
Make all your cron jobs run at nice 20 (crontab -e, edit command line to contain nice -n20). They have no reason to demand interactive latency, and you *do* need said latency for your UI.
If you run any servers on your workstation, they should run around nice 10. They need to get back to the user, but they shouldn't make your UI get unpleasant when they get hit.
Renice esd/artsd to -15. If these don't get CPU *right away* when they need it, your sound will break up. Frankly, I dumped esd/artsd, and got a sound card with hardware mixing (ALSA.9 + Sound Blaster Live! with the emu10k1). It's not worth it. 99.99% of Linux users will never, ever need network transparency or any other features that you get with a sound server. They *do* want sound that doesn't break up, and having hardware mixing does that for them. Ye gods, it'd be nice to see Linux have some architecture that does "opportunistic" mixing (hardware mixing if any channels are left, software if not).
Use a decent window manager. Sawfish is incredible if you're an edge-flipping maniac like me and like zero edge resistance. Why? Sawfish is actually not that *fast*, but they've compensated for that fact, which makes them beat any other window manager I've seen at edge flipping latency. Sawfish doesn't block other app redraws when edge flipping until it's redrawn its titlebars, as other WMs do, so you get much faster redisplay of app windows. Beautifl design.
Finally, I've had good experiences with redefining HZ in the kernel. Unfortunately, one of the side effects of using the X11 architecture is that anything going to the screen has to wait for a context switch -- first, the app tells X to display something, then we wait until X is active and actually display it. This isn't a huge deal unless you have a bunch of processes that all want CPU time, and you have an app or X that's blocking on I/O (say you've paged out an app). Then your ten compiles, and the lowly default 100 HZ in the x86 kernel mean that it takes a full tenth of a second just to move from the user app to X. If the app is displaying a big pixmap that has to be paged it, it has to draw a little bit, start paging the thing back in, draw a little more...it's I/O bound and yet it isn't gettting a chance to keep the ATA bus saturated. Jack HZ to 1000 or 1024 and recompile your kernel, and you should notice slightly better UI latencies (NOTE: at one point, this caused oddities in some libc call lke usleep or something, and made a couple games run too fast...I don't think this is an issue any more).
Other wins: Use mozilla 1.1 (much faster redraw than 1.0), use an up-to-date version of gtk2 (wow, the version RH is packaging is much faster at rendering aa text than the old snapshot I had from Ximian), use the blisteringly fast rxvt instead of the slow gnome-terminal or konsole. Use gnuserv mode in emacs/emacs -- that way, you open a *single* copy of emacs and then just open new windows in it. Opening files is about 50 times faster.
After following all these tips, you can play with Linux the way it was meant to be seen.
Anyway, point is that mplayer wins hands down as the best media player, and RH doesn't package it (and the mplayer guys strongly discourage binary packages, anyway)...so I always just download and install the thing. Not exactly that much pain.
I have to agree with the article author. For ages, I've been wanting a new mouse with *lots* of buttons to bind to interesting window manager functions (for example, I have my current fourth button bound to drag windows, which is a bit timesaver...I can also use it to bring windows to the front without "passing through" the click to the application...but I don't have to disable passthrough clicks, so I can still work when I'm copying and pasting from one window to another). The problem is that Logitech firmly refused to put out any new mice...just these little awful rounded four button mice. I loved their original "wedge" shapes, and I refused to buy an MS mouse (plus, the MS mouse button layout is really annoying...if you've used one, you know what I mean).
Then they come out with this thing. If I want a wireless optical many button mouse, I can get one. If I *don't* want the wireless bit (I, like some other people, am not a huge fan of wireless devices.), I can drop one model down and get the same mouse sans wireless bit.
These mice are *rechargeable* with a cradle, and no batteries, so if you're a fan of wireless mice, you can't get much better.
NT4's policies are pretty bad. It defaults to a blank administrative password, an administrator username of "Administrator" (and there are ways to obtain the administrator username if this is changed, anyway), sharing all the drives as hidden "administrative shares", *resharing* them at the next boot if you disable sharing...
The best thing to do is just axe the Server service. I've seen so many remotely exploitable boxes (probably ~70% of home NT4 users had this open) that it would blow your mind.
Then if you upgraded from NT4 to 2k, it would keep the same configuration...
I wrote that post specifically saying that the approach that UC Irvine was using (trying to detect P2P traffic) had holes, and should be moved to a quota/priority system.
I *also* listed some of the ways to bypass such a system, which, despite your claim, was not the question in the article.
Finally, I wasn't responding to the article directly. I was responding to another post, which was *also* talking about detecting and limiting P2P traffic, making my post quite relevant.
The convention for a *long* time, which, I believe, was the first big convention as regards major version number upgrades, related to application software. The idea was to use major version number to represent the file format version -- if the application changed file formats, the major version got bumped.
Dammit, we have a kickass I/O subsystem, cool. But without disk-access priorities attached to processes, it's kind of hard to fully take advantage of it.
I still play zangband on a regular basis. I don't play any graphical game on a regular basis. Oh, and I do have a certain taste for text-based interactive fiction.
Now, I will grant you that angband has gotten prettier over the past years (color, dynamic lighting as you walk around your text-based dungeon), but a graphics-fest it's not.
All the games you've mentioned -- Half Life, Counter-Strike, Starcraft, Baldur's Gate 2, NOLF, Soul Calibur, AoE 2, Sims, Black and White -- all of those have fallen behind, and aren't played much, but angband players still play angband.
And I liked Total Annihilation much more than Starcraft, though Starcraft was definitely prettier.
Err...the reason console games are popular right now is because a *new generation of consoles just came out*. So right now, the consoles are technically impressive, and consoles are popular. It's a cycle though -- as the systems age, computers will appear more and more appealing and consoles be less interesting...until the next generation of consoles comes out. This cycle has happened since the Nintendo era when people first started getting personal computers in their home at a reasonable rate, and isn't going to stop any time soon.
Neither computer nor console gaming is going to go anywhere any time soon.
Plus, there are genres of games that have evolved for each platform that aren't nearly as good on the others. The side-scroller will forever be a console game, as will the fighting game (a la Street Fighter). The FPS and the turn-based-strategy game will always be computer games. Games where expandability/editibility/conventional input devices are helpful will always have a plus on the computer.
That being said, I find it monumentally frusterating how little memory ships with most consoles. Programmers can *always* pull off good new tricks to improve performance and graphics given more RAM to play with. Cut into the fancy graphics hardware a bit, and give the machines more memory, and they'd be a lot more competitive with some of the things done in the PC world.
I mean, a 64MB of RAM total X-Box looks unpleasantly sparse when compared to a PC with twice that on the video card alone, and half a gig of system memory. Coders *can* come up with cool precomputation techniques if you give them the chance!
They have a interdisciplinary class which is immensely popular called Building Virtual Worlds. You grab a programmer, a musician, a modeler, and someone to script (rough outline, as tasks move around a bit and aren't very formalized), and make a virtual world. You get to play with neat VR hardware like headsets. In the past, notable efforts have included a Godzilla game with a breath input (a flap in front if your mouth), an Akira game, and so on. Modeling software that's cutting edge research stuff used for fast prototyping and building with some neat UI work is available, since it's produced at the university.
Slashdot Mirror
Why not call:
The kernel "Linux", since that's what the kernel makers call it.
The distro whatever the distro maker calls it: "Red Hat Linux 8.0".
The POSIX utilities whatever their authors want to call them: "GNU grep".
Emacs whatever the emacs author wants to call it: "GNU emacs".
If everyone just named things what their *creator* wanted them called, life would be a lot simpler. It would also get rid of the stupid GNU/Linux debate.
Much as I hate to say it, the two are lightyears apart.
Hurd isn't really usable, unless you're doing it for the pure diehard ideology of it. It isn't technically notable (at least at the current time) and doesn't even begin to compare to Linux from a performance point of view.
When you see an announcement about Linux, it's usually something along the lines of "new major feature annouced that makes Linux better at foo than operating system baz". With Hurd, it's usually "Basic UNIX functionality foo now works, sometimes, assuming you don't push it too hard".
There are a lot of (better) kernels out there. BSD. Linux.
Frankly, the only people that get excited about Hurd are the uber hard core FSF folks.
With luck, in six years the Hurd might be up to where Linux is today.
Finally, one last point: the main reason Hurd exists is so that Stallman can slap the FSF's "GNU/" on top of an OS. It isn't to address some specific technical failure of Linux.
What, if any, enhanced form of QoS would you have included in TCP/IP?
Outside of universities, and a few companies, how many non-techies used the Internet before the whole Information Superhighway thing that Clinton/Gore pushed as an agenda? It wasn't a tool that most people knew how to use or wanted. Sure, I could hop on a VAX terminal and cruise around, but I didn't see widespread desktop systems with Internet access around until after Clinton/Gore.
Clinton/Gore went for the schools, got people used to it, and funded library and other public access systems to try to get as many people able to use the Internet as a tool as possible, and gave incentives to telecom companies to build networks as fast as possible.
I suppose saying "build the Internet" is a bit overkill -- the procols were in place, and there was a large, working network. Building the infrastructure that lets the thing exist today, and makes it available to everyone, though....it isn't just an academic tool, or a tool used for a couple of UNIX geeks to chat via talk.
Also, I think defining it as building the Web is a little too harsh. The Web *did* happen to get popularized around the same time, but it certainly wasn't because the government was directly pushing Web browsing.
This isn't an issue of "lighting fast". The proposed solution, at least without serious modification, would massively fragment the hard drive. The only reason you don't *care* about fragmentation is because you enjoy the pleasant fruits of the fragmentation-resistant ext2, so you don't realize how bad fragmentation can get. The proposed system would fragment the filesystem so badly that a well-used FAT32 system would look contiguous as hell.
You could make a usable system that's somewhat similar...it could shift files around and use, say, a third of the free space for old files.
And what percentage of "normal users" install their own software instead of getting someone else to do it for them on Windows?
Umm...I've had decidedly nasty experiences with failed installs of IE that definitely did *not* fully roll back.
I'm not saying the technology isn't there, but even MS doesn't use it, at least some of the time.
mplayer is a bitch to install, but awfully simple to use once it's in. And in terms of performance and flexibility, it can't be beat.
I always mount with noatime. Nothing I hate worse than constant background writing on my hard drive.
The Mac OS had a Trash Can from its introduct in '84. I believe that the Apple IIgs shipped with an OS with a Trash Can, and it wouldn't surprise me if the Lisa did the same.
Slashdot has had major software announcements for a long time. The idea is to provide an interesting medly of content that keeps nerds happy. When PHP forum bugfix notifications start coming up on Slashdot, then you can complain
There's been a lot of discussion about having a form of undelete in Linux on Slashdot. This is quite relevant.
Al (and Clinton) shoveled massive amounts of federal dollars into producing the Internet. If not for their strong pushing of spreading the Internet all over (starting in educational institutions), we wouldn't be anywhere near where we are today.
Yes, Al misspoke. But he was also crucial to the Internet being what it is today, so he gets some points.
Your mother is comfortable *installing* XP, or is this just hyperbole?
/etc/rc.d/init.d/network restart and changing settings periodically. Of course, everything works beautifully after a cold boot...just not after restarting the network. I've seen this since at least the early 6.0 days, and up through 7.3.
Linux really isn't hard to install any more, at least to the point of getting it up and running. I'd call the installation process on at least RH simpler than the Windows procedure.
That being said, configuring stuff not-out-of-box is where things get ugly. It's damn easy for an end user to just get a new video card, download their InstallShield program, and use it. And to *uninstall*, simple as that may seem.
Software packaged by your distro "just works" and at least with RPM is really easy to install and uninstall. However, a lot of drivers are not packaged in said manner. Sometimes you can't get a driver to compile, or instructions are written for another distro. Got a laptop with a wireless card, or an Nvidia card, or a weird USB device? If it works in Linux, the install procedure is not necessarily trivial.
A few other things that are nasty include:
* Networking. i swear to God that there's either a bug in the Linux kernel or in RH's networking scripts since time immemorable, since *every* system I've ever used will sometimes, despite the fact that the routing tables are correct, refuse to properly route information. I can pretty consistently get this on a wide variety of RH's distros by running
* Windows has ZoneAlarm. Linux has the amazingly powerful iptables, with *no* really good, really solid front ends (though lots of half-finished freshmeat projects). If you want a personal firewall, Linux can give you an incredible amount of power...*if* you're willing to fight with iptables for a few days.
* Linux has *no* fully working, reliable ICQ program. This is an embarrassment. It isn't really Linux authors' fault -- trying to reverse engineer ICQ is not trivial -- but if I try to send a Windows user a file and can't, the only thing they learn is "Linux can't do IM properly". Yes, I know about Jabber -- which no one uses.
* Linux has, AFAIK, *no* finished, fully featured 3d modelling programs. Someone who likes to dabble with 3d work can run out grab lots of low end 3d modelers on Windows. There are *tons* of Pov front ends, none of which begin to compare to fully blown Windows modelling programs. Oh, and I'm not talking about multi-thousand dollar movie studio packages -- I mean stuff that a home user could use.
* Linux has *no* finished, fully featured vector graphics programs. Yes, lots of projects underway like sodipodi, sketch, kontour...and none of them are remotely usable for a real life production artist.
Insightful.
/dev/yourharddrivenamehere. Significantly reduces "jerkiness" in X when doing disk access, including paging. For a long time, a lot of distros left this off by default.
.9 + Sound Blaster Live! with the emu10k1). It's not worth it. 99.99% of Linux users will never, ever need network transparency or any other features that you get with a sound server. They *do* want sound that doesn't break up, and having hardware mixing does that for them. Ye gods, it'd be nice to see Linux have some architecture that does "opportunistic" mixing (hardware mixing if any channels are left, software if not).
I'd also like to point out that I've done a bit of benchmarking gcc, and optimizing for a particular processor makes almost no difference on the vast majority of software.
The biggest win comes from flipping on -O3. Then if you can get away with it, -fomit-frame-pointer, which helps the register-starved x86, but keeps you from looking at stack traces and debugging crashed programs (or sending in useful bug reports). -fexpensive-optimizations have also helped a bit too, and for certain packages, -ffast-math can be big. -march=pentium2 makes next to no difference on anything I've tried benchmarking. -DNDEBUG is potentially good...seems like most production software is compiled with assertions enabled, when they're really intended for debugging.
The Pentium 1 sucked at running code compiled for the 386/486. This is why you got compilers like pgcc, a Pentium-optimized Mandrake distro, and lots of talk about architecture optimizations. With the Pentium 2, Intel realized that all software was not going to be recompiled for each processor (at least in Windows land), and did a really solid job of running 386 code.
So, as far as architectures go, the Pentium 1 is the odd man out. If you have a Pentium 1, it sucks to run any code other than stuff compiled for your chip. If you have anything else, you'll generally get very minimal gains from compiling specifically for your processor instead of for the 386.
Finally, most people don't actually care about the maybe 10% speedup they can get by recompiling software using optimization flags other than just -O2. They care about interactive latency. Look at Mac OS X. OS X is *hideously* slow, but it *feels* pretty fast because it has good UI latency -- it jacks UI priority and puts a lot of emphasis on slapping something on the screen that's updated as soon as the user does anything.
On Linux, here are the big culprits.
Jack the nice value of X from 0 to -10, if your distro doesn't already do so (take a look in top and see what it's running at). The nice value doesn't make it much "faster", but it does significantly improve latency, so you can get crisp edge-flipping and updating.
Turn *on* DMA and umasked interrupts (insert usual warnings about potential problems with *really* old computers having these on). hdparm -u1 -d1
If you're doing something that doesn't need low latency in the background, *nice* it. I run all compiles niced to 20. I can be compiling six or seven packages with no user-perceptable slowdown at all. Software that's always sucking down a little CPU in the background but still should be interactive (like lopster or gtk-gnutella) should be niced to 5 or so.
Make all your cron jobs run at nice 20 (crontab -e, edit command line to contain nice -n20). They have no reason to demand interactive latency, and you *do* need said latency for your UI.
If you run any servers on your workstation, they should run around nice 10. They need to get back to the user, but they shouldn't make your UI get unpleasant when they get hit.
Renice esd/artsd to -15. If these don't get CPU *right away* when they need it, your sound will break up. Frankly, I dumped esd/artsd, and got a sound card with hardware mixing (ALSA
Use a decent window manager. Sawfish is incredible if you're an edge-flipping maniac like me and like zero edge resistance. Why? Sawfish is actually not that *fast*, but they've compensated for that fact, which makes them beat any other window manager I've seen at edge flipping latency. Sawfish doesn't block other app redraws when edge flipping until it's redrawn its titlebars, as other WMs do, so you get much faster redisplay of app windows. Beautifl design.
Finally, I've had good experiences with redefining HZ in the kernel. Unfortunately, one of the side effects of using the X11 architecture is that anything going to the screen has to wait for a context switch -- first, the app tells X to display something, then we wait until X is active and actually display it. This isn't a huge deal unless you have a bunch of processes that all want CPU time, and you have an app or X that's blocking on I/O (say you've paged out an app). Then your ten compiles, and the lowly default 100 HZ in the x86 kernel mean that it takes a full tenth of a second just to move from the user app to X. If the app is displaying a big pixmap that has to be paged it, it has to draw a little bit, start paging the thing back in, draw a little more...it's I/O bound and yet it isn't gettting a chance to keep the ATA bus saturated. Jack HZ to 1000 or 1024 and recompile your kernel, and you should notice slightly better UI latencies (NOTE: at one point, this caused oddities in some libc call lke usleep or something, and made a couple games run too fast...I don't think this is an issue any more).
Other wins: Use mozilla 1.1 (much faster redraw than 1.0), use an up-to-date version of gtk2 (wow, the version RH is packaging is much faster at rendering aa text than the old snapshot I had from Ximian), use the blisteringly fast rxvt instead of the slow gnome-terminal or konsole. Use gnuserv mode in emacs/emacs -- that way, you open a *single* copy of emacs and then just open new windows in it. Opening files is about 50 times faster.
After following all these tips, you can play with Linux the way it was meant to be seen.
Instructions for "media distro":
Step 1: Install Red Hat
Step 2: Install mplayer and xmms
Has RH ever shipped with a built in media player?
Anyway, point is that mplayer wins hands down as the best media player, and RH doesn't package it (and the mplayer guys strongly discourage binary packages, anyway)...so I always just download and install the thing. Not exactly that much pain.
I have to agree with the article author. For ages, I've been wanting a new mouse with *lots* of buttons to bind to interesting window manager functions (for example, I have my current fourth button bound to drag windows, which is a bit timesaver...I can also use it to bring windows to the front without "passing through" the click to the application...but I don't have to disable passthrough clicks, so I can still work when I'm copying and pasting from one window to another). The problem is that Logitech firmly refused to put out any new mice...just these little awful rounded four button mice. I loved their original "wedge" shapes, and I refused to buy an MS mouse (plus, the MS mouse button layout is really annoying...if you've used one, you know what I mean).
Then they come out with this thing. If I want a wireless optical many button mouse, I can get one. If I *don't* want the wireless bit (I, like some other people, am not a huge fan of wireless devices.), I can drop one model down and get the same mouse sans wireless bit.
These mice are *rechargeable* with a cradle, and no batteries, so if you're a fan of wireless mice, you can't get much better.
Only NT4.
NT4's policies are pretty bad. It defaults to a blank administrative password, an administrator username of "Administrator" (and there are ways to obtain the administrator username if this is changed, anyway), sharing all the drives as hidden "administrative shares", *resharing* them at the next boot if you disable sharing...
The best thing to do is just axe the Server service. I've seen so many remotely exploitable boxes (probably ~70% of home NT4 users had this open) that it would blow your mind.
Then if you upgraded from NT4 to 2k, it would keep the same configuration...
I wrote that post specifically saying that the approach that UC Irvine was using (trying to detect P2P traffic) had holes, and should be moved to a quota/priority system.
I *also* listed some of the ways to bypass such a system, which, despite your claim, was not the question in the article.
Finally, I wasn't responding to the article directly. I was responding to another post, which was *also* talking about detecting and limiting P2P traffic, making my post quite relevant.
Redundant my foot.
Umm...no.
The convention for a *long* time, which, I believe, was the first big convention as regards major version number upgrades, related to application software. The idea was to use major version number to represent the file format version -- if the application changed file formats, the major version got bumped.
Dammit, we have a kickass I/O subsystem, cool. But without disk-access priorities attached to processes, it's kind of hard to fully take advantage of it.
If you really want marketing.
I still play zangband on a regular basis. I don't play any graphical game on a regular basis. Oh, and I do have a certain taste for text-based interactive fiction.
Now, I will grant you that angband has gotten prettier over the past years (color, dynamic lighting as you walk around your text-based dungeon), but a graphics-fest it's not.
All the games you've mentioned -- Half Life, Counter-Strike, Starcraft, Baldur's Gate 2, NOLF, Soul Calibur, AoE 2, Sims, Black and White -- all of those have fallen behind, and aren't played much, but angband players still play angband.
And I liked Total Annihilation much more than Starcraft, though Starcraft was definitely prettier.
Err...the reason console games are popular right now is because a *new generation of consoles just came out*. So right now, the consoles are technically impressive, and consoles are popular. It's a cycle though -- as the systems age, computers will appear more and more appealing and consoles be less interesting...until the next generation of consoles comes out. This cycle has happened since the Nintendo era when people first started getting personal computers in their home at a reasonable rate, and isn't going to stop any time soon.
Neither computer nor console gaming is going to go anywhere any time soon.
Plus, there are genres of games that have evolved for each platform that aren't nearly as good on the others. The side-scroller will forever be a console game, as will the fighting game (a la Street Fighter). The FPS and the turn-based-strategy game will always be computer games. Games where expandability/editibility/conventional input devices are helpful will always have a plus on the computer.
That being said, I find it monumentally frusterating how little memory ships with most consoles. Programmers can *always* pull off good new tricks to improve performance and graphics given more RAM to play with. Cut into the fancy graphics hardware a bit, and give the machines more memory, and they'd be a lot more competitive with some of the things done in the PC world.
I mean, a 64MB of RAM total X-Box looks unpleasantly sparse when compared to a PC with twice that on the video card alone, and half a gig of system memory. Coders *can* come up with cool precomputation techniques if you give them the chance!
They have a interdisciplinary class which is immensely popular called Building Virtual Worlds. You grab a programmer, a musician, a modeler, and someone to script (rough outline, as tasks move around a bit and aren't very formalized), and make a virtual world. You get to play with neat VR hardware like headsets. In the past, notable efforts have included a Godzilla game with a breath input (a flap in front if your mouth), an Akira game, and so on. Modeling software that's cutting edge research stuff used for fast prototyping and building with some neat UI work is available, since it's produced at the university.