However, Microsoft's compiler is *damn* good (performance wise, not conformance wise). It makes fast code and compiles really, really fast. Watcom is a little slower in produced code speed, though I don't know about compile time. Still quite respectable, though.
aligning them to this 4k boundary. And don't forget Microsoft's other strategy for dealing with app startup time - to load all the required libraries for important apps at boot time so that they're all in-memory and ready-linked. Word doesn't start up nearly so quickly if you remove the Office Startup application from the Startup menu...
>>>>>>>>>
Actually, no it doesn't. Just tried this right now (rebooted, never had Office Startup in my menu). Its slightly slower, but still starts really fast. The main problem is that MS spends a lot of time polishing the code and the KDE guys don't. GUI responsiveness is *very* subjective, and you can do a lot of tricks to make the GUI seem faster. Take explorer for example. When you start it, its fully drawn. When I start Konqueror on my 300Mhz machine, most of the widgets are blank. It takes the better part of a second until they all get painted in. While this really doesn't effect the startup time, it makes the GUI SEEM much slower. Then take rezising. You can optimize resizing a lot to prevent the rubber-band effect. IE does this, Konq doesn't. Konq rubber-bands, IE doesn't. I wrote a simple GUI app for BeOS one using my own button-list widget. More than half the development time (of like three weeks) consisted of tweeking the widget to have absolutely no flicker when rezising the window or modifying items. It was a pleasure to use (for me anyway...) Honestly, I think that KDE-2 has achieved almsot a feature-parity with Windows. It doesn't have some of the more eosteric stuff, and internationalization isn't as good, but I think more development time really should be devoted to polish rather than adding spiffy features.
6) Smart. The basic prinicples of UNIX make it a joy to work with (see #7, below).
>>>>>>>>
Actually, this is a very good example of how newer OSs are better in some respects. The whole "everything is a file" thing has been streched *way* beyond practicality. My videocard really isn't a file. It doesn't make sense to treat it as a block device or a character device. Win2K, for example, one up's UNIX by treating everything as an object, something that is a much more general (and sane) abstraction. So, no, everything that can be done has *not* been done. What would really be interesting to me would be to see something stable and well-tested like UNIX used to try out some new OS concepts (like OO or new VM techniques, or presistant storage).
I have 256 megs of ram and it runs smooth as hell, much faster than Windows 2000.
>>>>
What's your config? I'm running a 300Mhz/256MB machine and Mandrake 8.0-RC1 runs a lot slower than Win2K. Unless you're comparing either GNOME or KDE-2 to Win2K, its really not a fair comparison. Its more comparable to NT-3.51!
For instance, I bzip2ing a large file, compiling a kernel, and hot-swapping partitions around (yes, I've done that, but it's not for the faint of heart), then yes, performance sucks.
>>>>>>>>>>>>&g t;
Win2K favors GUI processes over cpu-bound ones. Thus, a GUI thread will preempt a computing thread if applicable. The reasoning is that a 100ms delay in compile time is much preferable to a 100ms response time to clicking a button. Also, the Win2K scheduler does stuff like give temporary priority boosts to processes that have just gotten some input, and the memory system gives preference to the foreground GUI process. While this isn't UNIXy "clean" it does a damn good job of keeping the GUI responsive under load.
Also, the performance in Linux happens no matter whats running. KDE-2 is slow (not just in startup) in Gentoo Linux (which is really light as far as Linux distros go) and Mandrake. I have done the standard X reniceing tricks, but the GUI is still slow. And by slow I mean not it doesn't react to every command instantly. I know its a pretty high standard for a GUI, but if Microsoft can do it, surely the OSS community can!
Re:What do you expect from more powerful software?
on
Mandrake 8.1 Released
·
· Score: 1, Troll
I'm tired of software lagging drastically behind the hardware because some people are afraid of upgrading.
>>>>>>>>
That's bullshit. Win2K runs faster than KDE2 on the same hardware and does *more*. Having more features doesn't necessarily make something slower. What makes things slower is having stupid core technologies that emphasize "coolness" over simplicity and usability. Pick a random Linux GUI technology. XUL, for example, is totally useless. Who needs so much extensive customizability? If you do, you shouldn't be using mainstream software. Take Bonobo. Who needs to use remote components on a desktop machine? Developers who work on "core" programs don't have the freedom to do what they want to. They have a responsibility to deliver a product that is usable and works well. If that means leaving out the eosteric features, then so be it.
Dude. Its called not running something unless the user requests it. Mandrake's default install has at, crond, and another time thingy running. I never have and never will use a scheduler, but I don't know if the system uses them for anything so I'm not comfortable turning them off. RedHat is even stupid. I don't know if it is still true, but with the 6.x series, they'd start sendmail automatically, even in the desktop install. WHY???
The Windows NT thread was somewhat better but each new version required new hardware because of the bloat.
>>>>>
Dude, these days Linux requires *more* hardware than a comparable Win2K system. I've been playing with KDE 2.2.1 on Mandrake 8.1 for awhile, and the damn thing just crawls. Whenever I start up Konqueror it not only takes several seconds, but I get an ugly partially drawn window before the damn thing finally completes. While IE resizes smoothly, Konq (actually, every single KDE 2 program, I think its a problem in the widget set) rubber-bands like anything. Everytime something happens onscreen (like when a webpage gets rendered) my mouse "sticks" a little. When kwrite takes longer to load than Word, you know your system has a problem. GNOME isn't much better, it loads faster but the interface problems remain. Of course, I could always downgrade to something lighter like XFCE, but that doesn't really have the "desktop" feel you get in Win2K and even then the performance isn't that great. My mouse sticks even when using something like Window Maker. Anyway, sorry for the long rant, but people who claim that Linux has some sort of performance edge for desktop use are seriously deluded. Linux's only advantages on the desktop are stability, power of management, and freeness. On a desktop, managability's usefullness is fairly limited, and with Win2K's very stable nature, the stability arguement isn't that strong either. Until the desktop environment developers stop adding so many useless features and focusing on usability, Linux will really not be an upgrade for Windows NT users.
Which would be correct. Linux is a kernel; a kernel's job is to run programs.
>>>>>
But those words carry a far larger meaning than the sentence implies. DirectX does only do sound, graphics, and input, but those words understate its power.
Just a tad, and DirectX Graphics's superiority is in areas that require hardware that most consumers (early adopters excluded) do not yet own. Even so, I hope Microsoft is enjoying its fleeting precious moments [gocollect.com] of superiority because they will end very soon.
>>>>>>>>
Why? The link you provide doesn't have anything to do with DirectX or OpenGL, so I surmise that you have no proof that OpenGL will suddenly stage a comeback. Even if OGL 2.0 gets rolling right now (it won't) it will quite awhile before the glacial-paced ARB will produce something useful.
Or "GLUT is infinitely weaker than DirectInput." Could you elaborate?
>>>>
Read the GLUT API, then read the DirectInput API. Juding from the GLUT 3.7 API reference GLUT supports three button mice, keyboards, and spaceballs. DirectInput can be programmed to handle anything, from mice with dozens of buttons buttons to 6 axis force-feedback joysticks to full-body cybersex suits. Also, GLUT's callback-based mechanism doesn't exactly scream "performance" in a game setting.
The sound quality of the General MIDI support on most consumer-grade sound cards sucks, and General MIDI has never been good for many genres of electronic music.
>>>>>>>
That's why DirectX doesn't use standard MIDI support anymore. On cards that have poor MIDI support, MS uses its (pretty good) software synth.
Point me to a.mid file doing a good impression of a tb-303 to convince me otherwise. Why else did Unreal Tournament use S3M, XM, and IT tracked music instead of MIDI, and most newer games like q3a use streaming MP3 or ogg anyway?
>>>>>
The music requirements of Q3 or UT aren't exactly terribly stressing. MIDI is uniquely suited to all sorts of things, such as dynamic adaptation to the game environment, that are hard to do with pre-recorded music. With DirectMusic, you can program scores that follow a general patterns, but fluctuate depending on conditions or at random times. The power is there, it is only a matter of time (DirectMusic only became mature fairly recently) until some clever developer decides to use it.
The sockets API is also protocol independent and can support any protocol for which your sockets implementation (such as BSDsock or Winsock2) has a backend.
>>>>>
True, but DirectPlay is a *much* more integrated solution. It does lobbies, transport, all sorts of things.
Why can't you use the game engine (built on d3d+dsound or opengl+openal) to do cut scenes, as Metal Gear Solid and Zelda 64 do? Or do you have some other reason for wanting to play movies?
>>>>>>>>
DirectX isn't necessarily just for games. You can use DirectX to get hardware acceleration for DVD playing and other multimedia uses. Also, many games use video quite well. The Final Fantasy series, for example, uses full motion video to a great effect. Also, game engines still aren't up to the quality of good FMV yet.
All three libraries have DirectX backends, but they also have backends for operating systems not controlled by Single Point of Failure Corp. [google.com]
>>>>>
Yes, but their functionality is a fairly limited subset of DirectX's features. Also, it is not fair to criticize DirectX just because MS sucks. DirectX is a quality technology, Microsoft or not.
What really bothers me about Gentoo (which is great, don't get me wrong) is the lack of good binary packages. The official policy seems to be compile-based. However, my poor 300MHz machine really can't handle compiling KDE every time a new release comes out, and sometimes you just want to type rpm -i *.rpm instead of waiting for a 15 minute compile to complete. If Gentoo had RPM integration (or maybe dpkg or something) then it could really be a force in the distro world.
In Win2K you'll get a segfault with that code. Or are you referring to Win9x, whose design goal wasn't stability but compatibility? Qualify correctly what you say.
Actually, bonobo's are a species in the family Pongidae, which also includes chimps, gorillas and orangutans. While they were regarded as a subspecies of chimpanzee until 1933, they are actually a seperate species entirely. As for being new, I never said Bonobos were new. I said that the knowledge that Bonobos are hypersexual gives a new meaning to the Bonobo component framework.
DirectX does only three things: graphics, sound, and input.
>>>>>
That's like saying that all Linux does is run programs!
OpenGL handles graphics
>>>>>
Well, but Direct3D is a tad ahead at the moment.
GLUT handles input,
>>>>>
But if infinately weaker than DirectInput
and OpenAL [openal.org] can do sound.
>>>>>
But does it do MIDI like DirectMusic?
Then what about protocol-independant multiplayer (DirectPlay), and multimedia (DirectShow)?
Other multimedia programming libraries include Allegro [sourceforge.net], SDL [libsdl.org], and ClanLib [clanlib.org], all of which can coexist peacefully with OpenGL graphics.
>>>>>
They can do the same with DirectX, but because of all of DirectX's features, you don't need extra libs.
Kinda like how Glide won? Or MeTaL? Or PowerSGL? Direct3D is not extendible (game developers stopped MS from making it extendible) and it owns the majority of the consumer 3D market. All the graphics cards makers are actively in support of it, while most offer less support for OpenGL. The fact that a standard interface like DirectX allows manufacturers to reach *everyone* with a 3D card, not just those with particular 3D cards outweighs any advantages of proprietory APIs. The game developers like it because it simplifies development for them. The hardware manufacturers like it because it broadens the use of 3D hardware. Consumers like it because they aren't stuck buying unsupported crap (generally...)
I don't know how much many of you know this, but bonobo's are actually a species of monkey with hyper-sexual behavior. Gives a whole new meaning to the idea of interfacing bonobo components with each other...
That's a stupid comment. If someone has been using the 2.4.10-preX series and can volunteer some information, isn't it easier if he just talks about it rather than making everyone go through the processes of trying it out?
Re:Comprehensive, but contains a spurious assertio
on
Linux on the Desktop
·
· Score: 2
Not necessarily. Say you mmap a 2GB database file. Each process that maps the file uses about 2MB of page tables to maintain the mapping (even more in FreeBSD). If only some parts of the file are accessed often (which could very well be possible in a case like this) then the page tables mapping the less-used regions would be sitting in memory doing nothing. Sure you could have all sorts of complex logic in the program that tries to only map parts of files at a time, etc, but it is cleaner just to mmap the whole thing and let the OS deal with it. Page tables should be treated no differently from regular pages. The swap-out daemon should make no distinction between types of pages (other than locked pages of course!) It should choose which pages to page out based on how often they are touched. If a page table doesn't get touched often, it should be paged out with the anonymous memory that doesn't get touched often.
If someone came out with an easy way to make universal installers for linux programs, every developer out there would snap it up in a heartbeat.
>>>>>>>>>>>
I think its a reluctance of developers to standardize on anything. Some people are vehmently against the LSB, even though it tries to solve exactly the problems you mention.
I know this horse has been beat down to the ground, but it still bugs me. How can Linux get anywhere on the desktop without good performance? Win2K runs great on my relatively old 300MHz machine, but Linux/KDE seems much less responsive. It just seems to me that the Linux desktop is too much of a compromise. You can have cool features with KDE2, but you abandon performance. You can have a super-fast desktop with E17, but you have to put up with a tiny software base that actually uses E17's features natively. Win2K manages to get lots of features with good performance, so there is no reason to have to choose the lesser of several evils on the Linux desktop!
Re:Comprehensive, but contains a spurious assertio
on
Linux on the Desktop
·
· Score: 2
Linux's memory footprint is a lot larger because of several things. First, Linux can't page out kernel memory. On my Win2K machine, of the 36MB of kernel memory, only 8MB is unpagable. Since most of the OS's features aren't being used all the time, this can lower the memory footprint of a running machine. Also, Linux can't page out the page tables, while Windows 2000 can. This becomes a larger problem in something like XCOM (used in Mozilla) where lots of sizable files are mapped in at the same time, but not necessarily used all the time.
Slashdot Mirror
However, Microsoft's compiler is *damn* good (performance wise, not conformance wise). It makes fast code and compiles really, really fast. Watcom is a little slower in produced code speed, though I don't know about compile time. Still quite respectable, though.
Also, Intel C++ for Linux is free for non-commercial use, so the licensing isn't *that* draconion.
Actually, Intel's C++ compiler is the most optimizing. Both VisualC++ and ICL overtook watcom around version 10.5.
Since I am not a kernel
>>>>>>>>
Hopefully not! Oh, you mean kernel developer, don't you?
The Linux kernel is really not geared to be portable between compilers. It uses a lot of GCC extensions that other compilers just don't have.
aligning them to this 4k boundary. And don't forget Microsoft's other strategy for dealing with app startup time - to load all the required libraries for important apps at boot time so that they're all in-memory and ready-linked. Word doesn't start up nearly so quickly if you remove the Office Startup application from the Startup menu...
>>>>>>>>>
Actually, no it doesn't. Just tried this right now (rebooted, never had Office Startup in my menu). Its slightly slower, but still starts really fast. The main problem is that MS spends a lot of time polishing the code and the KDE guys don't. GUI responsiveness is *very* subjective, and you can do a lot of tricks to make the GUI seem faster. Take explorer for example. When you start it, its fully drawn. When I start Konqueror on my 300Mhz machine, most of the widgets are blank. It takes the better part of a second until they all get painted in. While this really doesn't effect the startup time, it makes the GUI SEEM much slower. Then take rezising. You can optimize resizing a lot to prevent the rubber-band effect. IE does this, Konq doesn't. Konq rubber-bands, IE doesn't. I wrote a simple GUI app for BeOS one using my own button-list widget. More than half the development time (of like three weeks) consisted of tweeking the widget to have absolutely no flicker when rezising the window or modifying items. It was a pleasure to use (for me anyway...) Honestly, I think that KDE-2 has achieved almsot a feature-parity with Windows. It doesn't have some of the more eosteric stuff, and internationalization isn't as good, but I think more development time really should be devoted to polish rather than adding spiffy features.
6) Smart. The basic prinicples of UNIX make it a joy to work with (see #7, below).
>>>>>>>>
Actually, this is a very good example of how newer OSs are better in some respects. The whole "everything is a file" thing has been streched *way* beyond practicality. My videocard really isn't a file. It doesn't make sense to treat it as a block device or a character device. Win2K, for example, one up's UNIX by treating everything as an object, something that is a much more general (and sane) abstraction. So, no, everything that can be done has *not* been done. What would really be interesting to me would be to see something stable and well-tested like UNIX used to try out some new OS concepts (like OO or new VM techniques, or presistant storage).
I have 256 megs of ram and it runs smooth as hell, much faster than Windows 2000.
>>>>
What's your config? I'm running a 300Mhz/256MB machine and Mandrake 8.0-RC1 runs a lot slower than Win2K. Unless you're comparing either GNOME or KDE-2 to Win2K, its really not a fair comparison. Its more comparable to NT-3.51!
For instance, I bzip2ing a large file, compiling a kernel, and hot-swapping partitions around (yes, I've done that, but it's not for the faint of heart), then yes, performance sucks.
>>>>>>>>>>>>&g t;
Win2K favors GUI processes over cpu-bound ones. Thus, a GUI thread will preempt a computing thread if applicable. The reasoning is that a 100ms delay in compile time is much preferable to a 100ms response time to clicking a button. Also, the Win2K scheduler does stuff like give temporary priority boosts to processes that have just gotten some input, and the memory system gives preference to the foreground GUI process. While this isn't UNIXy "clean" it does a damn good job of keeping the GUI responsive under load.
Also, the performance in Linux happens no matter whats running. KDE-2 is slow (not just in startup) in Gentoo Linux (which is really light as far as Linux distros go) and Mandrake. I have done the standard X reniceing tricks, but the GUI is still slow. And by slow I mean not it doesn't react to every command instantly. I know its a pretty high standard for a GUI, but if Microsoft can do it, surely the OSS community can!
I'm tired of software lagging drastically behind the hardware because some people are afraid of upgrading.
>>>>>>>>
That's bullshit. Win2K runs faster than KDE2 on the same hardware and does *more*. Having more features doesn't necessarily make something slower. What makes things slower is having stupid core technologies that emphasize "coolness" over simplicity and usability. Pick a random Linux GUI technology. XUL, for example, is totally useless. Who needs so much extensive customizability? If you do, you shouldn't be using mainstream software. Take Bonobo. Who needs to use remote components on a desktop machine? Developers who work on "core" programs don't have the freedom to do what they want to. They have a responsibility to deliver a product that is usable and works well. If that means leaving out the eosteric features, then so be it.
Dude. Its called not running something unless the user requests it. Mandrake's default install has at, crond, and another time thingy running. I never have and never will use a scheduler, but I don't know if the system uses them for anything so I'm not comfortable turning them off. RedHat is even stupid. I don't know if it is still true, but with the 6.x series, they'd start sendmail automatically, even in the desktop install. WHY???
How the bloody hell to you polarize a metal?
The Windows NT thread was somewhat better but each new version required new hardware because of the bloat.
>>>>>
Dude, these days Linux requires *more* hardware than a comparable Win2K system. I've been playing with KDE 2.2.1 on Mandrake 8.1 for awhile, and the damn thing just crawls. Whenever I start up Konqueror it not only takes several seconds, but I get an ugly partially drawn window before the damn thing finally completes. While IE resizes smoothly, Konq (actually, every single KDE 2 program, I think its a problem in the widget set) rubber-bands like anything. Everytime something happens onscreen (like when a webpage gets rendered) my mouse "sticks" a little. When kwrite takes longer to load than Word, you know your system has a problem. GNOME isn't much better, it loads faster but the interface problems remain. Of course, I could always downgrade to something lighter like XFCE, but that doesn't really have the "desktop" feel you get in Win2K and even then the performance isn't that great. My mouse sticks even when using something like Window Maker. Anyway, sorry for the long rant, but people who claim that Linux has some sort of performance edge for desktop use are seriously deluded. Linux's only advantages on the desktop are stability, power of management, and freeness. On a desktop, managability's usefullness is fairly limited, and with Win2K's very stable nature, the stability arguement isn't that strong either. Until the desktop environment developers stop adding so many useless features and focusing on usability, Linux will really not be an upgrade for Windows NT users.
Which would be correct. Linux is a kernel; a kernel's job is to run programs.
.mid file doing a good impression of a tb-303 to convince me otherwise. Why else did Unreal Tournament use S3M, XM, and IT tracked music instead of MIDI, and most newer games like q3a use streaming MP3 or ogg anyway?
>>>>>
But those words carry a far larger meaning than the sentence implies. DirectX does only do sound, graphics, and input, but those words understate its power.
Just a tad, and DirectX Graphics's superiority is in areas that require hardware that most consumers (early adopters excluded) do not yet own. Even so, I hope Microsoft is enjoying its fleeting precious moments [gocollect.com] of superiority because they will end very soon.
>>>>>>>>
Why? The link you provide doesn't have anything to do with DirectX or OpenGL, so I surmise that you have no proof that OpenGL will suddenly stage a comeback. Even if OGL 2.0 gets rolling right now (it won't) it will quite awhile before the glacial-paced ARB will produce something useful.
Or "GLUT is infinitely weaker than DirectInput." Could you elaborate?
>>>>
Read the GLUT API, then read the DirectInput API. Juding from the GLUT 3.7 API reference GLUT supports three button mice, keyboards, and spaceballs. DirectInput can be programmed to handle anything, from mice with dozens of buttons buttons to 6 axis force-feedback joysticks to full-body cybersex suits. Also, GLUT's callback-based mechanism doesn't exactly scream "performance" in a game setting.
The sound quality of the General MIDI support on most consumer-grade sound cards sucks, and General MIDI has never been good for many genres of electronic music.
>>>>>>>
That's why DirectX doesn't use standard MIDI support anymore. On cards that have poor MIDI support, MS uses its (pretty good) software synth.
Point me to a
>>>>>
The music requirements of Q3 or UT aren't exactly terribly stressing. MIDI is uniquely suited to all sorts of things, such as dynamic adaptation to the game environment, that are hard to do with pre-recorded music. With DirectMusic, you can program scores that follow a general patterns, but fluctuate depending on conditions or at random times. The power is there, it is only a matter of time (DirectMusic only became mature fairly recently) until some clever developer decides to use it.
The sockets API is also protocol independent and can support any protocol for which your sockets implementation (such as BSDsock or Winsock2) has a backend.
>>>>>
True, but DirectPlay is a *much* more integrated solution. It does lobbies, transport, all sorts of things.
Why can't you use the game engine (built on d3d+dsound or opengl+openal) to do cut scenes, as Metal Gear Solid and Zelda 64 do? Or do you have some other reason for wanting to play movies?
>>>>>>>>
DirectX isn't necessarily just for games. You can use DirectX to get hardware acceleration for DVD playing and other multimedia uses. Also, many games use video quite well. The Final Fantasy series, for example, uses full motion video to a great effect. Also, game engines still aren't up to the quality of good FMV yet.
All three libraries have DirectX backends, but they also have backends for operating systems not controlled by Single Point of Failure Corp. [google.com]
>>>>>
Yes, but their functionality is a fairly limited subset of DirectX's features. Also, it is not fair to criticize DirectX just because MS sucks. DirectX is a quality technology, Microsoft or not.
What really bothers me about Gentoo (which is great, don't get me wrong) is the lack of good binary packages. The official policy seems to be compile-based. However, my poor 300MHz machine really can't handle compiling KDE every time a new release comes out, and sometimes you just want to type rpm -i *.rpm instead of waiting for a 15 minute compile to complete. If Gentoo had RPM integration (or maybe dpkg or something) then it could really be a force in the distro world.
In Win2K you'll get a segfault with that code. Or are you referring to Win9x, whose design goal wasn't stability but compatibility? Qualify correctly what you say.
Actually, bonobo's are a species in the family Pongidae, which also includes chimps, gorillas and orangutans. While they were regarded as a subspecies of chimpanzee until 1933, they are actually a seperate species entirely. As for being new, I never said Bonobos were new. I said that the knowledge that Bonobos are hypersexual gives a new meaning to the Bonobo component framework.
You're the one who wanted to get technical...
DirectX does only three things: graphics, sound, and input.
>>>>>
That's like saying that all Linux does is run programs!
OpenGL handles graphics
>>>>>
Well, but Direct3D is a tad ahead at the moment.
GLUT handles input,
>>>>>
But if infinately weaker than DirectInput
and OpenAL [openal.org] can do sound.
>>>>>
But does it do MIDI like DirectMusic?
Then what about protocol-independant multiplayer (DirectPlay), and multimedia (DirectShow)?
Other multimedia programming libraries include Allegro [sourceforge.net], SDL [libsdl.org], and ClanLib [clanlib.org], all of which can coexist peacefully with OpenGL graphics.
>>>>>
They can do the same with DirectX, but because of all of DirectX's features, you don't need extra libs.
Are you kidding? GLUT is so lame compared to DirectInput. And you know why? IT WASN'T DESIGNED AS A PRODUCTION API!
Kinda like how Glide won? Or MeTaL? Or PowerSGL? Direct3D is not extendible (game developers stopped MS from making it extendible) and it owns the majority of the consumer 3D market. All the graphics cards makers are actively in support of it, while most offer less support for OpenGL. The fact that a standard interface like DirectX allows manufacturers to reach *everyone* with a 3D card, not just those with particular 3D cards outweighs any advantages of proprietory APIs. The game developers like it because it simplifies development for them. The hardware manufacturers like it because it broadens the use of 3D hardware. Consumers like it because they aren't stuck buying unsupported crap (generally...)
I don't know how much many of you know this, but bonobo's are actually a species of monkey with hyper-sexual behavior. Gives a whole new meaning to the idea of interfacing bonobo components with each other...
That's a stupid comment. If someone has been using the 2.4.10-preX series and can volunteer some information, isn't it easier if he just talks about it rather than making everyone go through the processes of trying it out?
Not necessarily. Say you mmap a 2GB database file. Each process that maps the file uses about 2MB of page tables to maintain the mapping (even more in FreeBSD). If only some parts of the file are accessed often (which could very well be possible in a case like this) then the page tables mapping the less-used regions would be sitting in memory doing nothing. Sure you could have all sorts of complex logic in the program that tries to only map parts of files at a time, etc, but it is cleaner just to mmap the whole thing and let the OS deal with it. Page tables should be treated no differently from regular pages. The swap-out daemon should make no distinction between types of pages (other than locked pages of course!) It should choose which pages to page out based on how often they are touched. If a page table doesn't get touched often, it should be paged out with the anonymous memory that doesn't get touched often.
If someone came out with an easy way to make universal installers for linux programs, every developer out there would snap it up in a heartbeat.
>>>>>>>>>>>
I think its a reluctance of developers to standardize on anything. Some people are vehmently against the LSB, even though it tries to solve exactly the problems you mention.
I know this horse has been beat down to the ground, but it still bugs me. How can Linux get anywhere on the desktop without good performance? Win2K runs great on my relatively old 300MHz machine, but Linux/KDE seems much less responsive. It just seems to me that the Linux desktop is too much of a compromise. You can have cool features with KDE2, but you abandon performance. You can have a super-fast desktop with E17, but you have to put up with a tiny software base that actually uses E17's features natively. Win2K manages to get lots of features with good performance, so there is no reason to have to choose the lesser of several evils on the Linux desktop!
Linux's memory footprint is a lot larger because of several things. First, Linux can't page out kernel memory. On my Win2K machine, of the 36MB of kernel memory, only 8MB is unpagable. Since most of the OS's features aren't being used all the time, this can lower the memory footprint of a running machine. Also, Linux can't page out the page tables, while Windows 2000 can. This becomes a larger problem in something like XCOM (used in Mozilla) where lots of sizable files are mapped in at the same time, but not necessarily used all the time.