Why do Slashdotters have such a hard time with the theoretical? If I go outside throw a ball in the air, I can give you a general description of what it'll do, just from theory. Do I really have to go outside and throw the damn ball to prove my point?
You do realize that your comment is like asking me: "would you like to cite the papers that describe this 'conservation of energy' that you speak of?" That monopolies result in an economic loss, relative to competitive markets, is an idea in economics akin to "F=ma". Everybody uses it, you can pick up an elementary-level text and read about it.
Yea right. Most professors are leftist because they fall into an easy fallacy. That expertise in one area is transferrable. Being a first rate physicist does not mean you know jack about philosophy, politics or economics. But you would be hard pressed to find one who understood that.
You'd be hard-pressed to prove that baseless assertion. I'm talking about taking economics advice from economists, not taking economics advice from a physicist. However, even the physicist is more qualified to speak on economics than your average shmoe. A format training in mathematics and the scientific method is something that is widely applicable.
And the reason most fall for socialism is easy enough to understand if one understands human nature.
Red-herring. Who is talking about socialism here anyway? You're going to have a hard time finding a lot of socialist economists...
Socialism tells the chrome domes that they are the enlightened few, and that they should be running the world.
What the hell are you talking about? Socialism is by its very nature anti-classist and populist. A socialist utopia would be a country run by factory workers, something I assure you most academics would not find appealing. The idea that an enlightened few should run the world is not associated with socialism. Socialists have often held these ideas, but it is not a part of the ideology itself. Rather, these ideas have been associated with intellectuals in general, and indeed played a very prominent role in the thinking of our own Founding Fathers.
Especially since few have had much exposure to philosophy or political science, and of the few who have it is mostly of the marxist rewritten variety popular on college campuses.
Given your average professor and your average layman, which one do you really think would have lesser exposure to philosophy and political science? And I'm intrigued by this marxsist literature supposedly floating around college campuses. I assume you can provide references to back up your claims?
But they shouldn't be running the world, centralized planning dooesn't work, millions of mass graves and starving masses in lands that used to export food are proof for any with eyes to see.
Who exactly suggested centralized planning? There is a wide berth between the populist (socialist!) nature of America's current political climate and the absolutist tendencies of centrally-planned societies. In my opinion, a nice middle-ground is the original American Republic, where the power of government depends upon the people, but at the same time, the exact course of its actions is well-insulated from the masses.
No expert is as smart as the free market. Millions of average intellects interracting in the free market exhibits greater wisdom than any single, or small cabal of, geniuses
Most economists won't argue that point, at least in the abstract. What they will point out, however, is that the free market does have limitations in specific cases, stemming from its very nature. In those corner cases, a single entity (usually government, hopefully guided by a cabal of geniuses) can exhibit greater wisdom than the whole mass of people.
As for ecomists, they fall into two catagories: Marxist crackpots who should be dismissed from their tenured perches and the real ones who deal with reality.
In other words they can be seperated into ones you disagree with and ones you agree with? And I assume you are basing your categorization on your own extensive body of work, given you the ability to discern which economists are right and which are wrong?
$10bn is hardly an audacious claim given Microsoft's annual revenue. Their excess profit is enormous compared to those of companies in more competitive markets.
Losses, perhaps, over some ideal market -- a perfectly competitive market.
No, losses over a more competitive market. The more competitive the market, the lower the losses. In a perfectly competitive market, the losses are zero.
No markets are, in reality, perfect competition.
Doesn't matter. A lot of markets are close enough that the losses due to lack of perfect competition are very small. Even in moderately competitive markets like soft drinks, the economic losses are still much smaller than in a major monopoly like Microsoft.
There are huge barriers to entry in the OS world, simply because an OS is not just an OS but also a suite of programs, drivers, etc. That alone prevents the market from being perfectly competitive.
Undoubtedly. But the OS market doesn't need to be perfectly competitive. That'd be nice, but it's not essential. The losses could be minimized by making it more competitive. That's the whole idea behind anti-trust legislation. There are certain natural monopolies that are inevitable (power companies are a classical example). However, in order to maintain some semblence of efficiency in the system, monopolies must be regulated, and prevented from doing things to make the situation worse. In the case of Microsoft, there is a lack of controls that should be there.
THe OJ and Nicole example, or the Michael Jackson example involves things that may or may not have happened, and there is no way to know what really occurred. In contrast, whether Microsoft is a monopoly involves no such uncertainty. The courts get to decide who is and is not a monopoly. According to the courts, Microsoft is a monopoly. Now, you could say that this judgement is wrong, and that Microsoft's should not be labeled a monopoly, but that does not change the fact that, currently, as a result of previous court decisions, it is labeled a monopoly.
I'm surprised so many Slashdotters come to the defense of Microsoft in response to a story that merely says the obvious. Of course Microsoft's monopoly creates losses! If it didn't, it'd be the first monopoly in history not to! The fact that it is a monopoly, and that it uses business practices that are illegal (for good reason) isn't even under debate. They've been convicted of the charges already!
The economics behind the two situations are significantly different. The economic losses created by the existance of a monopoly in a particular market is a subject well-studied in economics. The fact that there are economic losses in a monopoly market, relative to an equivalent competitive market, is widely accepted. On the other hand, there has been very little study of losses created by the large-scale copying of cheaply replicable goods. In terms of large-scale economic impacts, it's such a new phenomenon that there just isn't a lot of knowledge on exactly what happens, or how to calculate such losses.
Your #4 is incorrect, and has been for about 50 years now. The "unseen hand" is a useful analogy, but its not entirely consistent with modern economic theory. Read up on anything written in the last 50 years regarding monopolies, and you'll learn that they can exist naturally, and indeed the government, by virtue of its coercive ability, is often the only way to break them. The remainder of your post, derived from this flawed premise, should thus be disregarded.
Oh, and FYI, professors are leftist because they actually study the world. I find it incredible that most people wouldn't ride in an airplane built by a layman, but are perfectly willing to listen to economic theory espoused by people unqualified to do so.
It'sn ot FIPS, it's FLOPS. A "FLOP" is pretty much accepted these days as an IEEE754 double-precision operation. The nature of these operations don't differ to much between modern processors.
Part of the reason is that not much significant has happened in computer science since the late 1980's. Artificial intelligence was going to be the next big thing, but that kinda petered out by the early 1990s. The industry has been rehashing the 1980's work ever since then.
Salaries are normalized with respect to the official workweek. You don't get paid X per hour for working 40 hours per week, but you get paid $xx,xxx per year, with the expectation that you'll put in a standard work week. Yes, its not a legal constraint, but it is convention. That's why offices that have salaried employees still have standard business hours, usually 7:30 to 4:30 or 8:00 to 5:00. These hours are calculated figuring the standard 40 hour workweek, plus the standard lunch hour.
If we're going to count the amount of time I spend thinking about Slashdot at work, how about we count the amount of time I sit thinking about work on my personal time? When I'm working on a project, code is going through my head all the time. I'll be in the shower thinking about an algorithm, or eating dinner trying to figure out where a bug came from. When it gets bad, I'll be trying to get to sleep, but I'll be distracted by code. When I finally do get to sleep, I'll dream about code. I'd like to see a study done about this...
I'd like to see a calculation of the amount of money lost by salaried employees who work more than the mythical 40 hour work week without anything resembling overtime. Let's do some math.
Let's consider just engineers. There are 2 million engineers in the US, nearly all of whom are salaried employees, nearly all of whom work over 40 hours a week. The average engineer makes 70-90K/yr. Let's take the average at 80k/yr. Now, assuming a 40-hour work-week and the standard 3-weeks vacation, that works out to about $41 an hour. Now, I'd say your average engineer would believe they work, on average, 50 hours per week. That's $40bn in lost wages for engineers alone, using conservative estimates. Now, consider the number of other overworked, salaried employees. The lost wages could easily run to 10x that!
If I recall correctly from Win95 (this was a awhile ago), if you opened a folder that already had a window open, it'd just open another window for the folder (so you'd have two). Of course, that was a long time ago. Your point is well-taken, though, Win95 did have many elements of a spatial file browser.
I don't think the reactions are to the army using weapons on bad guys. I think the reactions are in the context of the army using the weapon as stated in the article --- on rioters. There is a big difference between the two, and a big difference between rubber bullets and the stuff they use on rioters here and a ray gun...
Bah. The problems properly arranging the period table just shows why they shouldn't bother to teach high-school level chemistry in schools. The damn subject never made any sense to me until I took quantum mechanics. It was then I realized that chemistry wasn't actually full of inconsistencies and special cases, and justified only by anthropomorphizing atoms ("he wants to have 8 electrons in his outer shell!"). It was just explained that way because teachers didn't think you were smart enough to tell you what was actually happening...
1% overall in 10 different places can still be 10%, but that's beside the argument here. I didn't say that you shouldn't do the other things, you should definitely optimize that big case too.
But we're not talking about 10 different 1%'s, we're talking about this particular 1%. The cost of changing it is high, and the gain is only 1%. If they other 9 1%'s don't involve moving large amounts of code into the kernel, well, you can have those, but leave this particular 1% alone.
Sorry, but that isn't true. X runs at an elevated privilege to access IO ports etc. It can definitely crash your system. Guaranteed.
Good point. X can crash your system by issuing improper DMA requests to the graphics card. But my point still stands. There is a very limited part of X (the graphics driver) that can crash the kernel. However, in a kernel-mode graphics system, there is a much larger part (even something like reading configuration files) that can crash the system.
I haven't written written one, but i have modified some, since they *crashed* my system. And your number naming for the rest is kind of pointless, I didn't tell you to put a whole X in the kernel. That would indeed be kernel-bloat.
I didn't say to put the whole X in the kernel. I was just talking about the parts you were talking about --- managing video memory, managing windows, etc. If you want to eliminate context switches, those things need to go into the kernel. That's still tens of thousands of lines of code.
That's basically what CC is! Is it only shorter to write "All rights reserved except reproduction for non-commerial use" than "licensed under Creative Commons"? It's simply a way of standardizing the legalease, so people can immediately know what rights they have to a particular work.
I stopped reading his crap a decade ago. It's interesting to see that he's still one of the stupidest people in tech journalism, and that he apparently hasn't shut up since then.
This is a bullshit argument, i have quite a quick radeon card. You're not telling me that card can't draw a browser window fast enough. Though your argumentation sounds allright for the rest, this can't be correct.
This statement has nothing to do with what I said. I didn't say that the bottleneck in drawing your browser window is the graphics card, I said the bottleneck in the overall X -> kernel -> graphics card communication process is the AGP/PCI-E bus. The reason your browser window doesn't draw fast enough is the reasons I mentioned earlier. I'll reiterate them, since you obviously didn't read my post very carefully:
1) Synchronization --- appearing fast is better than being fast and appearing slow. OS X is actually quite slow. But people think it has a very nice "feel" because everything is smooth, synchronized, and double-buffered. With your emphasis on saving a few irrelevent clock cycles here and there, you're completely missing the real things that make a GUI feel fast. Synchronized resize, double buffering --- all of those cost lots of cycles. But they're all absolutely necessary in making a UI feel fast.
2) Text compositing. The speed of your Radeon card doesn't really even come into play, because its barely even getting used! A major bottleneck in screen drawing is alpha-blending thousands of glyphs onto the screen for every character of text. On most cards, that alpha blending is not accelerated by the blending engine on your card. Instead, the CPU has to do a painfully slow read/modify/write for each pixel in the image. That's wicked slow.
3) Layout. Layout is the single biggest bottleneck in the whole chain. Every time you resize your window and you're left with a blank gray area where your content should be, do you think that area exists because X can't keep up? Hell no! It exists because the layout engine can't keep up!
In your last post that was still 6%. Check your numbers.
Learn to read. The 6% is how much time you'd save in the buffer flush process if the context swithch overhead was zero. The 1% is how much CPU time you'd save overall.
And you keep compare those 1000 cycles to some bigger number, probably the total number of cycles you CPU does each second. Stop that, compare it to the version without the context switch please!
What do I care about the version without context switches? I'm a programmer, all I care about is the bottom line. I don't care if task A is a thousand times faster if it only takes 1% of my CPU time to begin with. The key to designing a fast system is concentrating on the things that are slow. If text layout takes 30% of your time, and context switches eat up 1%, you'd be a moron to go to a lot of trouble optimizing the latter. Even if you could make it 1000 times faster, you'd gain nothing!
If such a change can optimize a call to draw some lines (a buffer full) from several thousand to several hunderd, that's a *huge* improvement. Several such improvements can add up quickly and will mage the whole of X feel less sluggish.
That just makes no sense mathematically. Why bother trying to make several minor things 10 times faster, just to gain 1% here and 1% there, when making one major thing even 2 times faster will gain you much more?
BTW, what difference would it make anyways? XFree86 runs with elevated privileges already, and can crash your system anytime it wants to.
X cannot crash your system if it goes down. It can delete all your files, but like any userspace process, it cannot crash the kernel.
What i'd suggest is to put a simple graphics driver into the kernel, videomemory management and maybe some synchronisation between apps in there. In the worst case that are 5000 lines, plus some 10000 for each graphics driver.
Hah! Have you ever written a graphics driver? Go read the DRI or Xgl mailing lists sometime. Mode setting alone is hundreds of kilobytes of code! Proper, generic video memo
Slashdot Mirror
http://en.wikipedia.org/wiki/Deadweight_loss
Why do Slashdotters have such a hard time with the theoretical? If I go outside throw a ball in the air, I can give you a general description of what it'll do, just from theory. Do I really have to go outside and throw the damn ball to prove my point?
You do realize that your comment is like asking me: "would you like to cite the papers that describe this 'conservation of energy' that you speak of?" That monopolies result in an economic loss, relative to competitive markets, is an idea in economics akin to "F=ma". Everybody uses it, you can pick up an elementary-level text and read about it.
Yea right. Most professors are leftist because they fall into an easy fallacy. That expertise in one area is transferrable. Being a first rate physicist does not mean you know jack about philosophy, politics or economics. But you would be hard pressed to find one who understood that.
You'd be hard-pressed to prove that baseless assertion. I'm talking about taking economics advice from economists, not taking economics advice from a physicist. However, even the physicist is more qualified to speak on economics than your average shmoe. A format training in mathematics and the scientific method is something that is widely applicable.
And the reason most fall for socialism is easy enough to understand if one understands human nature.
Red-herring. Who is talking about socialism here anyway? You're going to have a hard time finding a lot of socialist economists...
Socialism tells the chrome domes that they are the enlightened few, and that they should be running the world.
What the hell are you talking about? Socialism is by its very nature anti-classist and populist. A socialist utopia would be a country run by factory workers, something I assure you most academics would not find appealing. The idea that an enlightened few should run the world is not associated with socialism. Socialists have often held these ideas, but it is not a part of the ideology itself. Rather, these ideas have been associated with intellectuals in general, and indeed played a very prominent role in the thinking of our own Founding Fathers.
Especially since few have had much exposure to philosophy or political science, and of the few who have it is mostly of the marxist rewritten variety popular on college campuses.
Given your average professor and your average layman, which one do you really think would have lesser exposure to philosophy and political science? And I'm intrigued by this marxsist literature supposedly floating around college campuses. I assume you can provide references to back up your claims?
But they shouldn't be running the world, centralized planning dooesn't work, millions of mass graves and starving masses in lands that used to export food are proof for any with eyes to see.
Who exactly suggested centralized planning? There is a wide berth between the populist (socialist!) nature of America's current political climate and the absolutist tendencies of centrally-planned societies. In my opinion, a nice middle-ground is the original American Republic, where the power of government depends upon the people, but at the same time, the exact course of its actions is well-insulated from the masses.
No expert is as smart as the free market. Millions of average intellects interracting in the free market exhibits greater wisdom than any single, or small cabal of, geniuses
Most economists won't argue that point, at least in the abstract. What they will point out, however, is that the free market does have limitations in specific cases, stemming from its very nature. In those corner cases, a single entity (usually government, hopefully guided by a cabal of geniuses) can exhibit greater wisdom than the whole mass of people.
As for ecomists, they fall into two catagories: Marxist crackpots who should be dismissed from their tenured perches and the real ones who deal with reality.
In other words they can be seperated into ones you disagree with and ones you agree with? And I assume you are basing your categorization on your own extensive body of work, given you the ability to discern which economists are right and which are wrong?
$10bn is hardly an audacious claim given Microsoft's annual revenue. Their excess profit is enormous compared to those of companies in more competitive markets.
Losses, perhaps, over some ideal market -- a perfectly competitive market.
No, losses over a more competitive market. The more competitive the market, the lower the losses. In a perfectly competitive market, the losses are zero.
No markets are, in reality, perfect competition.
Doesn't matter. A lot of markets are close enough that the losses due to lack of perfect competition are very small. Even in moderately competitive markets like soft drinks, the economic losses are still much smaller than in a major monopoly like Microsoft.
There are huge barriers to entry in the OS world, simply because an OS is not just an OS but also a suite of programs, drivers, etc. That alone prevents the market from being perfectly competitive.
Undoubtedly. But the OS market doesn't need to be perfectly competitive. That'd be nice, but it's not essential. The losses could be minimized by making it more competitive. That's the whole idea behind anti-trust legislation. There are certain natural monopolies that are inevitable (power companies are a classical example). However, in order to maintain some semblence of efficiency in the system, monopolies must be regulated, and prevented from doing things to make the situation worse. In the case of Microsoft, there is a lack of controls that should be there.
THe OJ and Nicole example, or the Michael Jackson example involves things that may or may not have happened, and there is no way to know what really occurred. In contrast, whether Microsoft is a monopoly involves no such uncertainty. The courts get to decide who is and is not a monopoly. According to the courts, Microsoft is a monopoly. Now, you could say that this judgement is wrong, and that Microsoft's should not be labeled a monopoly, but that does not change the fact that, currently, as a result of previous court decisions, it is labeled a monopoly.
I'm surprised so many Slashdotters come to the defense of Microsoft in response to a story that merely says the obvious. Of course Microsoft's monopoly creates losses! If it didn't, it'd be the first monopoly in history not to! The fact that it is a monopoly, and that it uses business practices that are illegal (for good reason) isn't even under debate. They've been convicted of the charges already!
The economics behind the two situations are significantly different. The economic losses created by the existance of a monopoly in a particular market is a subject well-studied in economics. The fact that there are economic losses in a monopoly market, relative to an equivalent competitive market, is widely accepted. On the other hand, there has been very little study of losses created by the large-scale copying of cheaply replicable goods. In terms of large-scale economic impacts, it's such a new phenomenon that there just isn't a lot of knowledge on exactly what happens, or how to calculate such losses.
Your #4 is incorrect, and has been for about 50 years now. The "unseen hand" is a useful analogy, but its not entirely consistent with modern economic theory. Read up on anything written in the last 50 years regarding monopolies, and you'll learn that they can exist naturally, and indeed the government, by virtue of its coercive ability, is often the only way to break them. The remainder of your post, derived from this flawed premise, should thus be disregarded.
Oh, and FYI, professors are leftist because they actually study the world. I find it incredible that most people wouldn't ride in an airplane built by a layman, but are perfectly willing to listen to economic theory espoused by people unqualified to do so.
It'sn ot FIPS, it's FLOPS. A "FLOP" is pretty much accepted these days as an IEEE754 double-precision operation. The nature of these operations don't differ to much between modern processors.
Part of the reason is that not much significant has happened in computer science since the late 1980's. Artificial intelligence was going to be the next big thing, but that kinda petered out by the early 1990s. The industry has been rehashing the 1980's work ever since then.
Salaries are normalized with respect to the official workweek. You don't get paid X per hour for working 40 hours per week, but you get paid $xx,xxx per year, with the expectation that you'll put in a standard work week. Yes, its not a legal constraint, but it is convention. That's why offices that have salaried employees still have standard business hours, usually 7:30 to 4:30 or 8:00 to 5:00. These hours are calculated figuring the standard 40 hour workweek, plus the standard lunch hour.
If we're going to count the amount of time I spend thinking about Slashdot at work, how about we count the amount of time I sit thinking about work on my personal time? When I'm working on a project, code is going through my head all the time. I'll be in the shower thinking about an algorithm, or eating dinner trying to figure out where a bug came from. When it gets bad, I'll be trying to get to sleep, but I'll be distracted by code. When I finally do get to sleep, I'll dream about code. I'd like to see a study done about this...
I'd like to see a calculation of the amount of money lost by salaried employees who work more than the mythical 40 hour work week without anything resembling overtime. Let's do some math.
Let's consider just engineers. There are 2 million engineers in the US, nearly all of whom are salaried employees, nearly all of whom work over 40 hours a week. The average engineer makes 70-90K/yr. Let's take the average at 80k/yr. Now, assuming a 40-hour work-week and the standard 3-weeks vacation, that works out to about $41 an hour. Now, I'd say your average engineer would believe they work, on average, 50 hours per week. That's $40bn in lost wages for engineers alone, using conservative estimates. Now, consider the number of other overworked, salaried employees. The lost wages could easily run to 10x that!
If I recall correctly from Win95 (this was a awhile ago), if you opened a folder that already had a window open, it'd just open another window for the folder (so you'd have two). Of course, that was a long time ago. Your point is well-taken, though, Win95 did have many elements of a spatial file browser.
I don't think the reactions are to the army using weapons on bad guys. I think the reactions are in the context of the army using the weapon as stated in the article --- on rioters. There is a big difference between the two, and a big difference between rubber bullets and the stuff they use on rioters here and a ray gun...
Bah. The problems properly arranging the period table just shows why they shouldn't bother to teach high-school level chemistry in schools. The damn subject never made any sense to me until I took quantum mechanics. It was then I realized that chemistry wasn't actually full of inconsistencies and special cases, and justified only by anthropomorphizing atoms ("he wants to have 8 electrons in his outer shell!"). It was just explained that way because teachers didn't think you were smart enough to tell you what was actually happening...
The new table was invented in Great Britain...
Windows never had a spatial finder. "Large icons" have nothing to do with spatial file browsing.
GNOME defaults to spatial browsing.
1% overall in 10 different places can still be 10%, but that's beside the argument here. I didn't say that you shouldn't do the other things, you should definitely optimize that big case too.
But we're not talking about 10 different 1%'s, we're talking about this particular 1%. The cost of changing it is high, and the gain is only 1%. If they other 9 1%'s don't involve moving large amounts of code into the kernel, well, you can have those, but leave this particular 1% alone.
Sorry, but that isn't true. X runs at an elevated privilege to access IO ports etc. It can definitely crash your system. Guaranteed.
Good point. X can crash your system by issuing improper DMA requests to the graphics card. But my point still stands. There is a very limited part of X (the graphics driver) that can crash the kernel. However, in a kernel-mode graphics system, there is a much larger part (even something like reading configuration files) that can crash the system.
I haven't written written one, but i have modified some, since they *crashed* my system. And your number naming for the rest is kind of pointless, I didn't tell you to put a whole X in the kernel. That would indeed be kernel-bloat.
I didn't say to put the whole X in the kernel. I was just talking about the parts you were talking about --- managing video memory, managing windows, etc. If you want to eliminate context switches, those things need to go into the kernel. That's still tens of thousands of lines of code.
That's basically what CC is! Is it only shorter to write "All rights reserved except reproduction for non-commerial use" than "licensed under Creative Commons"? It's simply a way of standardizing the legalease, so people can immediately know what rights they have to a particular work.
I stopped reading his crap a decade ago. It's interesting to see that he's still one of the stupidest people in tech journalism, and that he apparently hasn't shut up since then.
This is a bullshit argument, i have quite a quick radeon card. You're not telling me that card can't draw a browser window fast enough. Though your argumentation sounds allright for the rest, this can't be correct.
This statement has nothing to do with what I said. I didn't say that the bottleneck in drawing your browser window is the graphics card, I said the bottleneck in the overall X -> kernel -> graphics card communication process is the AGP/PCI-E bus. The reason your browser window doesn't draw fast enough is the reasons I mentioned earlier. I'll reiterate them, since you obviously didn't read my post very carefully:
1) Synchronization --- appearing fast is better than being fast and appearing slow. OS X is actually quite slow. But people think it has a very nice "feel" because everything is smooth, synchronized, and double-buffered. With your emphasis on saving a few irrelevent clock cycles here and there, you're completely missing the real things that make a GUI feel fast. Synchronized resize, double buffering --- all of those cost lots of cycles. But they're all absolutely necessary in making a UI feel fast.
2) Text compositing. The speed of your Radeon card doesn't really even come into play, because its barely even getting used! A major bottleneck in screen drawing is alpha-blending thousands of glyphs onto the screen for every character of text. On most cards, that alpha blending is not accelerated by the blending engine on your card. Instead, the CPU has to do a painfully slow read/modify/write for each pixel in the image. That's wicked slow.
3) Layout. Layout is the single biggest bottleneck in the whole chain. Every time you resize your window and you're left with a blank gray area where your content should be, do you think that area exists because X can't keep up? Hell no! It exists because the layout engine can't keep up!
In your last post that was still 6%. Check your numbers.
Learn to read. The 6% is how much time you'd save in the buffer flush process if the context swithch overhead was zero. The 1% is how much CPU time you'd save overall.
And you keep compare those 1000 cycles to some bigger number, probably the total number of cycles you CPU does each second. Stop that, compare it to the version without the context switch please!
What do I care about the version without context switches? I'm a programmer, all I care about is the bottom line. I don't care if task A is a thousand times faster if it only takes 1% of my CPU time to begin with. The key to designing a fast system is concentrating on the things that are slow. If text layout takes 30% of your time, and context switches eat up 1%, you'd be a moron to go to a lot of trouble optimizing the latter. Even if you could make it 1000 times faster, you'd gain nothing!
If such a change can optimize a call to draw some lines (a buffer full) from several thousand to several hunderd, that's a *huge* improvement. Several such improvements can add up quickly and will mage the whole of X feel less sluggish.
That just makes no sense mathematically. Why bother trying to make several minor things 10 times faster, just to gain 1% here and 1% there, when making one major thing even 2 times faster will gain you much more?
BTW, what difference would it make anyways? XFree86 runs with elevated privileges already, and can crash your system anytime it wants to.
X cannot crash your system if it goes down. It can delete all your files, but like any userspace process, it cannot crash the kernel.
What i'd suggest is to put a simple graphics driver into the kernel, videomemory management and maybe some synchronisation between apps in there. In the worst case that are 5000 lines, plus some 10000 for each graphics driver.
Hah! Have you ever written a graphics driver? Go read the DRI or Xgl mailing lists sometime. Mode setting alone is hundreds of kilobytes of code! Proper, generic video memo