And arguably, the G4 is a faster CPU (though I'd really like to see real-world numbers for this; the G4 is a beast when it comes to number crunching, but the Pentium-M is a beast when it comes to clock-effecieny).
Maybe because every user doesn't want to fight with 8 configuration files dealing with graphics, two to get a network card running, and be responsible for ensuring every piece of hardware works correctly right after being installed.
Face it, Linux is difficult. It's getting better, but OS X is already where Linux needs to be (though artificially; you control the hardware, you control the software). And it's worth it to us to pay the premium to get a machine that works.
Try that now, buy a machine, and in a year buy a new processor and try to stick it in the machine. You won't be able to do it. Then try it again in another year. Once again, you won't be able to do it.
I honestly don't see why the rest of the industry isn't crying foul; with AMD integrating all of the components, they're essentially taking jobs away from chipset manufacturers. Of course, chipsets can still be used for things like USB and such, but how long until AMD realizes the industry dream of Computer-on-chip?
But, this is a distant thing, and as long as Intel still exists, I'm not worried about AMD becoming a monopoly and completely locking us into our system's components. But I'm sure if Intel made these steps, everyone would be crying foul.
.9~ and 1 distinctly aren't the same number, it's a flaw in the number system, but it's nothing to huff about, it's easy to catch and remedy.
The reason for this, of course, stems back to why we have a need for.0~ in the first place; what's 1 / 0? Undefined, and there's nothing we can do about it; no amount of mathematics can change 1/0 into anything else. But, there is a way around this.
If we say a number is "arbitrarily close to zero", aka.0~, we can say that 1/.0~ is infinity, thus solving the problem we had. Anyone who's been through differential calculus (and some precal courses) can tell you that much.
But then of course, you are the one who fails to get it in this case; there is a way to get 1 from.0~; add.9~ to it. I've had people try to argue to me that it isn't possible, but algebraically it makes perfect sense; there is no reason.9(9-bar) should repeat any faster than.0(0-bar)1, and as long as.1 +.9 is 1.0, we get back to one.
How could anyone get conclusive results without operating inside of google, and inside of yahoo? It seems to me like the *best* you can do is have a bunch of websites log what URL's they're coming from, but that's inconclusive; some users will use one search engine repeatedly just to find an article, while other's will search for the same thing across a lot of search engines.
I just don't see how anyone could come to a result that's completely objective.
I honestly can't tell if you're being sarcastic or being truthful, and since both Multics and PL/I predate me (but I do know that Multics became UNIX), I can't really render an opinion. I do know that UNIX's success came from being built on a toolchain that was easily portable, which came from the not-so-feature-rich language C.
From the looks of the PL/I language, it seems like it must have taken decades to compile on the hardware of the time; unlike Java, it had to compile to machine code on every machine it ran on.
You, on the other hand, missed the point completely. Looks like the guy's looking to upgrade (the computer's ANCIENT). You, on the other hand, would be downgrading.
I think that it probably would be faster, but only marginally.
Or how about "Apple+Shift+Q"-> Log off. Or set the screen saver to lock the screen. Don't like screensavers? Use the blank one, set the time to low, and be done.
The problem with this idea is that the complexity of an operating system cannot be removed; there are a number of things an operating system must be able to do, and the only way to do that is by complex methods.
That being said, using a higher level language only moves the complexity from the language, and adds it to the compiler, ups the memory requirements, et cetera.
Changing the toolset doesn't change what you are building. The goal is the same no matter what the tool you use, and it just so happens that assembly and C are very good tools for operating system construction.
And this point I totally and completely agree with. Anyone looking to be a code slave needs to be pointed towards IT and business training, where the scientists should be grabbed up in any way possible.
To put it nicely, I just didn't agree with the way you dismissed the need for CS students; theoretical code still needs to be effecient, and we need it now more than ever. We're moving to a whole new bit-level, one that allows us to address thousands of times more memory. We're moving to a new computing paradigm; no longer can we serialize processes and hope that computers will get faster, thus making our code faster.
I'm only so defensive because I truely love computers and everything about them, from silicon to software, and I've been involved in them since birth, and I can tell you, I still have a long way to go, and thirty years from now, nothing's going to have changed about that. But the computers we have now; the entire machine-type, is going the way of the Dodo, and scientists as well as the layman, needs to be aware of this. If we give up the science, how can we ever be expected to use the technology granted to us by it?
But that's just it; computers are NOT Turing Machines; turing machines are so much better than current computers. Turing machines know no such thing as time or memory limits. Turing machines don't have to deal with the heat death of the universe.
So, as computing machines approach the power of turing machines, we should see more and more improvements with what we can do with our machines. And Computer Science is the *key* science behind these improvements. How will we manage all of the memory a 64-bit computer allows us to address? How are we going to write software that's more sympathetic to having the ability to run 64 concurrent threads once we've hit the megahertz wall? (I am uniformally convinced we have; when I saw the Pentium 4 trying its best to squeeze out a few more drops of performance, I knew we'd found the theoretical wall. Not saying it isn't possible to make a chip run faster, just saying that we now know how close we are to physics no longer being a friendly ally). These aren't problems for engineers; these are problems for Computer Scientists; the guys who design the compilers so that engineers can implement them. This is the kind of work we need someone sitting in a lab playing with different ways to optimize algorithms by breaking them into pieces and running them inside of seperate processes, not the guy who's going to use the technology to make sure our next generation video games look as real as humanly possible.
We must not stop Computer Science. We must not stop science, period, as putting a foot down to stop science is putting an end to that entire branch of thinking. The problem is, too many Computer Scientists get into the profession wanting to be programmers; people who should be signing up for IT professions. Too many CS students get by without having to deal with all of the mathematics and algorithm design. Too many get by without even knowing how a CPU addresses the memory inside of their computer; "If I just learn Java and C#, I can join the work force and write me a video game!!!111oneonewon".
If anything, this recent pullback in CS is simply out of this reasoning. We're using the name CS to stand for Code Slave.
That's like saying "We don't need to teach the kids art in kindergarden anymore, because we already have had plenty of great artists in all of the different art forms, and now we need applied artists, like archetects". (excuse the downcast).
Computer and Software Engineers THRIVE off the sciences created by Computer Scientists. Too many people think CS is all about writing source code, but really, it's just like any other science; it's research, research, research.
Breakthroughs are still left to be found in all the fields, and new fields are just now being created (bioinformatics anyone?), and if we just give up on the science now, we won't have engineers implementing it later.
Don't believe I don't see your part about needing other sciences to migrate to using computers, but who do you think design the algorithms for integrating other sciences into Computer Science, the Engineers who build solutions, or the scientists in other fields who haven't had the training in mathematics or the algorithms to make things more efficient. And before you give me that crap about "computers being fast enough and having enough memory these days to deal with shitty programming", think about this; the *simplist* of protein folding implementations requires hundreds and hundreds of CPU hours, even cutting a dozen off of it means massive cost cutting for the organizations using it.
Face it; telling people to stop moving to CS is like telling people to stop moving towards Physics, or any other discrete science; it's stupid, short sighted, and just plain wrong.
Well, there may be no reason to spell it wrong, but I'd be damned if I've heard anyone pronounce it correctly. It's got the opposite problem most Russian words are endowed with (too many vowels).
What language does "Touareg" come from anyway and what the smegma does it mean?
A script in the story posting system would be more effiecent to both copy and pasting and a Firefox extension. Besides, it'd really only be a small hack; simply load all links into an array, and go through the array finding ".com|.net|.org" etc, and inserting in the nessicary "nyud.net:8090" or whatever it is. Not that difficult and would instantly leave us with the ability to choose.
Honestly, who said they were "stopping" with music videos?
You can't just jump into a market, you've got to have research, you've got to have money, and you've got to have a product. Right now, the only people who really have a product for feature length movies is Sony with the PSP, and that's a huge experiment; nobody knows what the long term is, and whether or not the PSP UMD format will go the way of the Beta or not.
Apple will go as they can into the media market. I'd really like to see the Tablet Mac before I'd see distributed movies, simply because a feature length really needs a larger screen, whereas a music video doesn't really need the size or resolution.
I, along with many others, really like music videos; they're a form of expression just as music is, and generally they tell a short story of why the song is the song (well, except for rap videos, but I don't wanna talk about that).
You're pricing rationale on the other hand makes entirely no sense. If it did, song's would also be priced against how long they are, but they aren't. Music videos cost more per minute because they cost more to make than a feature length movie does, per minute. A feature length movie may cost 10 Million dollars, but it's at least an hour long, most averaging out about a hundred minutes, meaning that per minute, they've only spent 100k. Music videos on the other hand, can cost ten thousand to five hundred thousand dollars, leading to as few as a few dollars per minute, to 150k per minute. Generally, though, production costs are kept down because the artist isn't paid to shoot the video; they do it out of promotion of their music.
Lastly, this content war's never going to end. There will always be someone selling entertainment, and what exactly they're selling will change as much as the direction of the wind. If you care about the medium, you'll continue to pay, and if you don't, you won't. It's that simple.
It could automatically insert a second link with the text "[1]" with links to Coral caches next to the original link and "[2]" to mirrordot. That way the original link is untouched, but the first and second are automatically available if the server is down.
Same here. I _love_ the Clash, and always thought of them as awesome musicians.. this just furthers my opinion of them.
I admit to filesharing, but of the good bands I've always returned to buying their CDs simply out of support and out of the extra materials you get with the CD (lyrics without the noisy ads, tons of meta information about the music's production, album art, etc). You just can't beat CDs for some things.
The P4 is superior in the ranks of servers because it has a really great data delivery system (800MHz FSB, quadpumped of course), a great set of very fast encoders, but I'm afraid that its pipeline was simply so long that after it's used its bus to fill itself up with data, it often has to flush it all out because it's a branch prediciting chip, and it's predictions fail often.
The P4 is a superior in the ranks of a server chip because it is fast. Even with a low IPC, it can get a lot of work done, simply because it does run so many cycles.
So yes, the pipeline was constructed to ramp the speed up, but it also facilitiates a crapload of in flight instructions, a crapload of code to be simultaniously executed (if you've got Hyperthreading enabled, you'll see this number continue to rise). But as SOON as you hit a branch, prepare to shutdown and lose 30 cycles of work.
The kind of server I seeing a Pentium 4 being dedicated to would be streaming media encoding or rendering. Since it doesn't look like Intel's going to give up on the Netburst arch, they're probably working their asses off trying to refactor the chip, and make a Pentium M of Netburst. Just because a chip was designed to be fast as hell, and not get much work done, doesn't mean that the archetecture was bunk, and that's what I'm trying to argue. But in the end, it really is just an experiment, the third coming of x86 implementations, super RISCy underneat.
Slashdot Mirror
After knowing about how the Game Industry is a sweatshop for video games, I figured I'd make games for my own good, if I made any at all.
Thank goodness everyone knows that people who own Xboxes only play Halo... I'd hate to see that one start popping up in dorm rooms.
Are you smoking? Hell's still trying to restart the fires from all of the news we had a couple weeks back with Apple and the like.
I've heard reports that it may take until Bush is out of office for the hell-fires to restart.
And arguably, the G4 is a faster CPU (though I'd really like to see real-world numbers for this; the G4 is a beast when it comes to number crunching, but the Pentium-M is a beast when it comes to clock-effecieny).
Maybe because every user doesn't want to fight with 8 configuration files dealing with graphics, two to get a network card running, and be responsible for ensuring every piece of hardware works correctly right after being installed.
;).
Face it, Linux is difficult. It's getting better, but OS X is already where Linux needs to be (though artificially; you control the hardware, you control the software). And it's worth it to us to pay the premium to get a machine that works.
Oh, and the eye candy's definitely better
Try that now, buy a machine, and in a year buy a new processor and try to stick it in the machine. You won't be able to do it. Then try it again in another year. Once again, you won't be able to do it.
I honestly don't see why the rest of the industry isn't crying foul; with AMD integrating all of the components, they're essentially taking jobs away from chipset manufacturers. Of course, chipsets can still be used for things like USB and such, but how long until AMD realizes the industry dream of Computer-on-chip?
But, this is a distant thing, and as long as Intel still exists, I'm not worried about AMD becoming a monopoly and completely locking us into our system's components. But I'm sure if Intel made these steps, everyone would be crying foul.
Isn't this whole concept a dupe of BuddyZoo.com, which was on Slashdot two years ago?
Sounds like an ad-ridden dupe to me.
.9~ and 1 distinctly aren't the same number, it's a flaw in the number system, but it's nothing to huff about, it's easy to catch and remedy.
.0~ in the first place; what's 1 / 0? Undefined, and there's nothing we can do about it; no amount of mathematics can change 1/0 into anything else. But, there is a way around this.
.0~, we can say that 1/.0~ is infinity, thus solving the problem we had. Anyone who's been through differential calculus (and some precal courses) can tell you that much.
.0~; add .9~ to it. I've had people try to argue to me that it isn't possible, but algebraically it makes perfect sense; there is no reason .9(9-bar) should repeat any faster than .0(0-bar)1, and as long as .1 + .9 is 1.0, we get back to one.
The reason for this, of course, stems back to why we have a need for
If we say a number is "arbitrarily close to zero", aka
But then of course, you are the one who fails to get it in this case; there is a way to get 1 from
How could anyone get conclusive results without operating inside of google, and inside of yahoo? It seems to me like the *best* you can do is have a bunch of websites log what URL's they're coming from, but that's inconclusive; some users will use one search engine repeatedly just to find an article, while other's will search for the same thing across a lot of search engines.
I just don't see how anyone could come to a result that's completely objective.
I honestly can't tell if you're being sarcastic or being truthful, and since both Multics and PL/I predate me (but I do know that Multics became UNIX), I can't really render an opinion. I do know that UNIX's success came from being built on a toolchain that was easily portable, which came from the not-so-feature-rich language C.
From the looks of the PL/I language, it seems like it must have taken decades to compile on the hardware of the time; unlike Java, it had to compile to machine code on every machine it ran on.
You, on the other hand, missed the point completely. Looks like the guy's looking to upgrade (the computer's ANCIENT). You, on the other hand, would be downgrading.
I think that it probably would be faster, but only marginally.
Or how about "Apple+Shift+Q"-> Log off. Or set the screen saver to lock the screen. Don't like screensavers? Use the blank one, set the time to low, and be done.
The problem with this idea is that the complexity of an operating system cannot be removed; there are a number of things an operating system must be able to do, and the only way to do that is by complex methods.
That being said, using a higher level language only moves the complexity from the language, and adds it to the compiler, ups the memory requirements, et cetera.
Changing the toolset doesn't change what you are building. The goal is the same no matter what the tool you use, and it just so happens that assembly and C are very good tools for operating system construction.
And this point I totally and completely agree with. Anyone looking to be a code slave needs to be pointed towards IT and business training, where the scientists should be grabbed up in any way possible.
To put it nicely, I just didn't agree with the way you dismissed the need for CS students; theoretical code still needs to be effecient, and we need it now more than ever. We're moving to a whole new bit-level, one that allows us to address thousands of times more memory. We're moving to a new computing paradigm; no longer can we serialize processes and hope that computers will get faster, thus making our code faster.
I'm only so defensive because I truely love computers and everything about them, from silicon to software, and I've been involved in them since birth, and I can tell you, I still have a long way to go, and thirty years from now, nothing's going to have changed about that. But the computers we have now; the entire machine-type, is going the way of the Dodo, and scientists as well as the layman, needs to be aware of this. If we give up the science, how can we ever be expected to use the technology granted to us by it?
But that's just it; computers are NOT Turing Machines; turing machines are so much better than current computers. Turing machines know no such thing as time or memory limits. Turing machines don't have to deal with the heat death of the universe.
So, as computing machines approach the power of turing machines, we should see more and more improvements with what we can do with our machines. And Computer Science is the *key* science behind these improvements. How will we manage all of the memory a 64-bit computer allows us to address? How are we going to write software that's more sympathetic to having the ability to run 64 concurrent threads once we've hit the megahertz wall? (I am uniformally convinced we have; when I saw the Pentium 4 trying its best to squeeze out a few more drops of performance, I knew we'd found the theoretical wall. Not saying it isn't possible to make a chip run faster, just saying that we now know how close we are to physics no longer being a friendly ally). These aren't problems for engineers; these are problems for Computer Scientists; the guys who design the compilers so that engineers can implement them. This is the kind of work we need someone sitting in a lab playing with different ways to optimize algorithms by breaking them into pieces and running them inside of seperate processes, not the guy who's going to use the technology to make sure our next generation video games look as real as humanly possible.
We must not stop Computer Science. We must not stop science, period, as putting a foot down to stop science is putting an end to that entire branch of thinking. The problem is, too many Computer Scientists get into the profession wanting to be programmers; people who should be signing up for IT professions. Too many CS students get by without having to deal with all of the mathematics and algorithm design. Too many get by without even knowing how a CPU addresses the memory inside of their computer; "If I just learn Java and C#, I can join the work force and write me a video game!!!111oneonewon".
If anything, this recent pullback in CS is simply out of this reasoning. We're using the name CS to stand for Code Slave.
I feel this post being greatly disturbing.
That's like saying "We don't need to teach the kids art in kindergarden anymore, because we already have had plenty of great artists in all of the different art forms, and now we need applied artists, like archetects". (excuse the downcast).
Computer and Software Engineers THRIVE off the sciences created by Computer Scientists. Too many people think CS is all about writing source code, but really, it's just like any other science; it's research, research, research.
Breakthroughs are still left to be found in all the fields, and new fields are just now being created (bioinformatics anyone?), and if we just give up on the science now, we won't have engineers implementing it later.
Don't believe I don't see your part about needing other sciences to migrate to using computers, but who do you think design the algorithms for integrating other sciences into Computer Science, the Engineers who build solutions, or the scientists in other fields who haven't had the training in mathematics or the algorithms to make things more efficient. And before you give me that crap about "computers being fast enough and having enough memory these days to deal with shitty programming", think about this; the *simplist* of protein folding implementations requires hundreds and hundreds of CPU hours, even cutting a dozen off of it means massive cost cutting for the organizations using it.
Face it; telling people to stop moving to CS is like telling people to stop moving towards Physics, or any other discrete science; it's stupid, short sighted, and just plain wrong.
Well, there may be no reason to spell it wrong, but I'd be damned if I've heard anyone pronounce it correctly. It's got the opposite problem most Russian words are endowed with (too many vowels).
What language does "Touareg" come from anyway and what the smegma does it mean?
A script in the story posting system would be more effiecent to both copy and pasting and a Firefox extension. Besides, it'd really only be a small hack; simply load all links into an array, and go through the array finding ".com|.net|.org" etc, and inserting in the nessicary "nyud.net:8090" or whatever it is. Not that difficult and would instantly leave us with the ability to choose.
Honestly, who said they were "stopping" with music videos?
You can't just jump into a market, you've got to have research, you've got to have money, and you've got to have a product. Right now, the only people who really have a product for feature length movies is Sony with the PSP, and that's a huge experiment; nobody knows what the long term is, and whether or not the PSP UMD format will go the way of the Beta or not.
Apple will go as they can into the media market. I'd really like to see the Tablet Mac before I'd see distributed movies, simply because a feature length really needs a larger screen, whereas a music video doesn't really need the size or resolution.
I, along with many others, really like music videos; they're a form of expression just as music is, and generally they tell a short story of why the song is the song (well, except for rap videos, but I don't wanna talk about that).
You're pricing rationale on the other hand makes entirely no sense. If it did, song's would also be priced against how long they are, but they aren't. Music videos cost more per minute because they cost more to make than a feature length movie does, per minute. A feature length movie may cost 10 Million dollars, but it's at least an hour long, most averaging out about a hundred minutes, meaning that per minute, they've only spent 100k. Music videos on the other hand, can cost ten thousand to five hundred thousand dollars, leading to as few as a few dollars per minute, to 150k per minute. Generally, though, production costs are kept down because the artist isn't paid to shoot the video; they do it out of promotion of their music.
Lastly, this content war's never going to end. There will always be someone selling entertainment, and what exactly they're selling will change as much as the direction of the wind. If you care about the medium, you'll continue to pay, and if you don't, you won't. It's that simple.
It could automatically insert a second link with the text "[1]" with links to Coral caches next to the original link and "[2]" to mirrordot. That way the original link is untouched, but the first and second are automatically available if the server is down.
Same here. I _love_ the Clash, and always thought of them as awesome musicians.. this just furthers my opinion of them.
I admit to filesharing, but of the good bands I've always returned to buying their CDs simply out of support and out of the extra materials you get with the CD (lyrics without the noisy ads, tons of meta information about the music's production, album art, etc). You just can't beat CDs for some things.
I'm not looking at "Amount of free space: " I'm looking at "################_______". Oh, it looks like I've used about two thirds of my hard drive.
And this is easily accessible; not buried behind anything, so really, they didn't steal this. Give credit where credit is due.
The P4 is superior in the ranks of servers because it has a really great data delivery system (800MHz FSB, quadpumped of course), a great set of very fast encoders, but I'm afraid that its pipeline was simply so long that after it's used its bus to fill itself up with data, it often has to flush it all out because it's a branch prediciting chip, and it's predictions fail often.
The P4 is a superior in the ranks of a server chip because it is fast. Even with a low IPC, it can get a lot of work done, simply because it does run so many cycles.
So yes, the pipeline was constructed to ramp the speed up, but it also facilitiates a crapload of in flight instructions, a crapload of code to be simultaniously executed (if you've got Hyperthreading enabled, you'll see this number continue to rise). But as SOON as you hit a branch, prepare to shutdown and lose 30 cycles of work.
The kind of server I seeing a Pentium 4 being dedicated to would be streaming media encoding or rendering. Since it doesn't look like Intel's going to give up on the Netburst arch, they're probably working their asses off trying to refactor the chip, and make a Pentium M of Netburst. Just because a chip was designed to be fast as hell, and not get much work done, doesn't mean that the archetecture was bunk, and that's what I'm trying to argue. But in the end, it really is just an experiment, the third coming of x86 implementations, super RISCy underneat.