ATI Rage Mobility was the defacto standard 3D chipset on many platforms. It had nothing to do with TV-out, as most machines using the ATI chipset didn't even offer that feature. That is what kept ATI in business, not selling a few consumer-level cards with TV-out. And SGI has been a has-been in 3D for YEARS. They decided to restructure around high-performance computing. Maybe you missed the part where they bought Cray and made that huge announcement. Cray, BTW has subsequently split off again as a separate company.
I have heard this same argument over and over again. There is pop culture, and there is the cutting edge, where innovation happens. This is true for all forms of art, not just games. The thing that suprises me most is this:
Where do the new ideas go if we can't have games like Viewtiful Joe, Shenmue, and Jet Grind Radio?
Is it just me, or did you just name 3 examples that directly contradict the statement you are trying to make? There are still plenty of innovative games out there, but like most innovative art, they are on the fringes, and are incorporated into pop culture in big gulps as society tries to reach beyond its borders. Just like Nirvana helped spur the grunge phenomenon in music (and like the New Wave movement 10 years earlier), we have the DOOM spurring on the FPS movement, the Command & Conquer inspiring the RTS explosion, and GTA3 inspiring an as-yet-to-be-named phenomenon.
OEMs make decisions based on cost, feature set and possibly name recognition, but only if it adds value to their product. They care about the business relationships they have with the vendors, and whether they can get price breaks, and whether the vendor's product integrates easily with their own.
Benchmarks that differ by a couple of percent depending on which test is run are not going to make a big difference in the overall decision process. If they made decisions based on benchmarks then ATI would have closed its doors many years ago, since until very recently they have been consistently outclassed by their competitors performance-wise for several years. However, ATI has done VERY well in the OEM market during this time not due to better performance, but due the the factors I listed.
JAD is a godsend. I wrote a very complex optimization method that was extremely effective in a couple of circumtstances. A couple of years later, those circumstances turn up again only in a different language. I can't find the source code anywhere, just the class file that had my great method in it. So, JAD comes to the rescue; it gave me a bunch code that used d1,d2,d3,... as my variables, but I already had a basic understanding of what each variable's role was, so it wasn't a problem for me to reverse-engineer my own code and finally port it to another language. I also made several back-ups of the source code this time.:-)
While I agree this is basically reverse discrimination, what I find MORE disturbing is the fact that they wasted so much time (and let's not forget thousands of dollars of taxpayer money) considering this law when there are about a million other much more important things they could have been working on.
that the reason is because they haven't had java running satisfactorily on sun hardware in a couple of years. It's a bit sadly ironic, but it runs *way* better on equivalent x86 servers than sun's sparc servers.
What he has proven is that under some conditions it might be possible to tweak some extra performance out of your memory AS LONG AS the only thing you do with your computer is run Synthetic Sandra.
The first major problem I have is that he didn't test enough data points to prove that the differences in memory performance were actually due to the tweaks he did, and not thanks to some other environmental condition like sun spots, power surges, or what he ate for breakfast. His ENTIRE test suite only eaked out a couple of percent difference; he would have to run each point several times just to establish that those couple of percent weren't a fluke.
Second, and more important, even if we assume his test data is 100% accurate, it doesn't simulate real-world conditions. Benchmarks are great for establishing ball-park performance numbers (within 1-2 sigma, *maybe*), but they are definitely not representative of how people use a computer. If I'm reading a web page, for instance, I might only access memory intermittently as I scroll the page up and down. But, then again maybe I'm listening to mp3s while I read that web page, so I am intermittently accessing one part of memory, while periodically (as in, at a regular interval) accessing another part of memory. And, let's not forget that I'll randomly startg up new applications, close old ones, etc. which means that there's going to be extra time where my solid-state memory is actually waiting on data from either virtual memory (usually equivalent to the HDD), my CD-ROM drive, or perhaps from the network. None of this is really being factored into the benchmark, but it's something I do all the time which has a much larger impact on my perceived memory performance than the difference between his worst test settings, and his best test settings.
Is this really patentable? The concept of secure communications between two networked devices have been around for years.
I think there's a huge misconception on slashdot about what a patent is, exactly. You cannot patent a general concept. What you patent is a specific method of doing something. For example, there are several hundred patents issued for different types of can-openers. Obviously, the concept of opening a can has been around since cans were invented. But, not surprisingly the methods used to get at the contents of a can have been quite varied. In fact the thing we now recognize today as the canonical can-opener wasn't invented until 1925. Before then there were all kinds of crazy methods of opening cans that tended to spill stuff all over the place. For further reading on this subject I highly recommend The Evolution of Useful Things by Henry Petroski.
Does this mean Atari or Sony or Nintendo can't use encryption in their games?
No. They will either have to find a different method than that patented (which shouldn't be too hard, the MS patent is fairly specific in its claims), or license the technology from MS.
Graphics processor vs. general-purpose CPU
on
Future of 3d Graphics
·
· Score: 4, Insightful
If these cards are getting so powerful at computations then why do we need a Intel/AMD processor at all? Just make a graphics card with more transistors and drop the traditional processor..
Because GPUs are specialized processors. They are only good at a couple of things: moving data in a particular format around quickly, and linear algebra. It is possible to do general-purpose calculations on a GPU, but that's not what it is good at, so you'd be wasting your time.
This is akin to asking why you shouldn't go see a veterinarian when you get sick. Because veterinarians specialize in animals. Sure, they might be able to treat you, but since their training is with animals you might find their treatments don't help as much as going to see a regular doctor.
If SCO were really concerned about losing IP, they could have discreetly contacted the parties in question, demonstrated their case, and maybe worked out some kind of licensing agreement.
Instead the first thing they do (before even contacting the companies in question) is to file a lawsuit. This is like taking my neighbor to court because his dog did his business on my front lawn. If I ask my neighbor politely to fix the problem, he probably will. If he doesn't, THEN I might take more serious action, but not before.
The first step in any dispute is to try to reach some kind of resolution outside of a courtroom setting. That SCO did not take this step indicates to me that they are up to no good.
So I should put my job on the line to save the world? What if I had kids to feed? Should they suffer to save the innocense of (already) molested kids?
This is poor reasoning. By actively ignoring this kind of activity, you are putting YOUR OWN CHILDREN at risk. What if instead of giving authorities the opportunity to trace the origins of the pornography on the machine, you DON'T report it, and the same people who made the child pornography on the guy's computer later molest YOUR kids for profit?
I don't understand your sentiment here, anyway. Sure whistle-blowers might get the shaft sometimes, but I'd like to believe that doing the right thing gets rewarded more often than it gets punished. If I'm ever working at a company and get the impression I could lose my job by doing the right thing, I don't want to continue working for that company anyway. I'd rather risk being fired than teach my children to live in constant fear.
Thanks, I read that article when it came out. In fact I did a lot of my own additional analysis on the data (I think the author credited me when he updated his paper) and found that you can get 5 rentals/month on a basic account with no trouble, and only minor problems occur with 6. Also, the queue lag vs. rentals/month is determined by how much you are paying. That is, if you want to rent more movies per month, then upgrade your subscription to a higher level, and you will continue to have high availability in your queue.
On the empirical evidence side, I've had no troubles at all getting any movie I want in a timely fashion from Netflix.
Thanks, I already understand that perfectly, and you'd know that if YOU had bothered to read my other responses in this thread. It is not my practice to respond to essentially the exact same post 500 times (as you put it), but I thought the irony here was particularly funny.
Take for example Tokyo and New York City. The actual amount of land used in the center city is quite small, small enough that walking or using a mass-transit system becomes quite viable.
You definitely cannot do that in Los Angeles, that's to be sure--it's so spread out that you'll need exorbitant amounts of money to build a mass transit system the cover the whole Los Angeles Basin.
It's not that it has to be small per se, but high-density. To take your Tokyo example, there are folks who commute to and from Tokyo from other cities in Japan every day. The rail lines in Japan are VERY extensive; it's hard to go asomewhere that doesn't have a train station within a reasonable distance (say, a bike ride). If you also factor in trolleys and buses, there are very few places in Japan you can't get to using public tranportation. The thing that makes it all work is that the entire country is fairly densely populated. You couldn't afford infrastructure to build a rail line to transport only 500-1,000 people/day at a reasonable price, but 20,000-50,000 people is another story.
As it happens, I used to live in Japan, and now live in Los Angeles. There are some areas that I suspect could benefit greatly by a train line, even if you couldn't cover all of L.A. county. At the very least, you could do like the suburbs of Chicago (and elsewhere, like Atlanta), and have huge parking decks with free/cheap parking next to the local train station. I used this *all the time* when I was living in those areas. I imagine a line that ran up the coast through Long Beach, the South Bay cities, and up into LAX, and a similar line from Santa Monica south to LAX would work very well, and probably help alleviate the ever-present traffic jam on 405 just North of LAX. One would hope at least that there are enough people sick of being stuck in that traffic jam to give public transportation a shot.
While I agree that in theory it's possible, I'm not so sure you can just incorporate outside sources so easily into a CMM L5 project, since all of the previously existing development isn't up to snuff. There would be no design documents or QA for the kernel until it was brought in-house, so you'd have to spend several man-years just doing QA on the existing code to "prove" that it was safe.
That said, what the heck would be the problem with releasing a blueprint of the critical systems?
I didn't say there's a specific problem with it. I said that under the current political environment there's no way it would happen. Public documents are being pulled off shelves faster than you can say "national security." Do you really expect the current administration would allow such sensitive information to be openly published?
However my first point was that up until this point, the linux kernel has not been controlled under a CMM level 5 process (far, far from it), and thus incorporating it into the space shuttle software would basically require rewriting it from scratch just to make sure it met CMM level 5 requirements. Since they would have to rewrite anything from scratch anyway, there is no point in using OSS that I can think of, unless an OSS project comes along that happens to be certified CMM level 5.
p.s. also sorry about the broken HTML, my / key seems to be a little stiff.
Second, suppose despite point #1, they decide to use the linux kernel on the space shuttle. Obviously, they'd have to adapt the kernel to suit their needs, since most of the hardware on the shuttle is custom designed and built for it. Under the GPL they would have to release any changes they make to the kernel back into the public domain. This would be equivalent to providing a very detailed blueprint of how all the critical systems on the space shuttle function. Especially given the current political environment, do you really think the administration is going to divulge this kind of information to the public?
[sound of door hitting me] Me: *groan*...oh, hi Bob, I was just picking up some paper clips! Bob: Do you realize you have 3 of them stuck to your face?
Slashdot Mirror
ATI Rage Mobility was the defacto standard 3D chipset on many platforms. It had nothing to do with TV-out, as most machines using the ATI chipset didn't even offer that feature. That is what kept ATI in business, not selling a few consumer-level cards with TV-out. And SGI has been a has-been in 3D for YEARS. They decided to restructure around high-performance computing. Maybe you missed the part where they bought Cray and made that huge announcement. Cray, BTW has subsequently split off again as a separate company.
Is it just me, or did you just name 3 examples that directly contradict the statement you are trying to make? There are still plenty of innovative games out there, but like most innovative art, they are on the fringes, and are incorporated into pop culture in big gulps as society tries to reach beyond its borders. Just like Nirvana helped spur the grunge phenomenon in music (and like the New Wave movement 10 years earlier), we have the DOOM spurring on the FPS movement, the Command & Conquer inspiring the RTS explosion, and GTA3 inspiring an as-yet-to-be-named phenomenon.
OEMs make decisions based on cost, feature set and possibly name recognition, but only if it adds value to their product. They care about the business relationships they have with the vendors, and whether they can get price breaks, and whether the vendor's product integrates easily with their own.
Benchmarks that differ by a couple of percent depending on which test is run are not going to make a big difference in the overall decision process. If they made decisions based on benchmarks then ATI would have closed its doors many years ago, since until very recently they have been consistently outclassed by their competitors performance-wise for several years. However, ATI has done VERY well in the OEM market during this time not due to better performance, but due the the factors I listed.
Ugh, of course I know that. I was making a joke, not writing a physics text.
People can be SO anal retentive online.
I agree. They should just define "1 kilogram" as the weight of a 1-kilogram bag of sugar. Sorta like the jargon file's definition of recursion.
JAD is a godsend. I wrote a very complex optimization method that was extremely effective in a couple of circumtstances. A couple of years later, those circumstances turn up again only in a different language. I can't find the source code anywhere, just the class file that had my great method in it. So, JAD comes to the rescue; it gave me a bunch code that used d1,d2,d3,... as my variables, but I already had a basic understanding of what each variable's role was, so it wasn't a problem for me to reverse-engineer my own code and finally port it to another language. I also made several back-ups of the source code this time. :-)
While I agree this is basically reverse discrimination, what I find MORE disturbing is the fact that they wasted so much time (and let's not forget thousands of dollars of taxpayer money) considering this law when there are about a million other much more important things they could have been working on.
I suspect it's actually some clever strategy to reduce the /. effect. We all know how lazy geeks are.
that the reason is because they haven't had java running satisfactorily on sun hardware in a couple of years. It's a bit sadly ironic, but it runs *way* better on equivalent x86 servers than sun's sparc servers.
What he has proven is that under some conditions it might be possible to tweak some extra performance out of your memory AS LONG AS the only thing you do with your computer is run Synthetic Sandra.
The first major problem I have is that he didn't test enough data points to prove that the differences in memory performance were actually due to the tweaks he did, and not thanks to some other environmental condition like sun spots, power surges, or what he ate for breakfast. His ENTIRE test suite only eaked out a couple of percent difference; he would have to run each point several times just to establish that those couple of percent weren't a fluke.
Second, and more important, even if we assume his test data is 100% accurate, it doesn't simulate real-world conditions. Benchmarks are great for establishing ball-park performance numbers (within 1-2 sigma, *maybe*), but they are definitely not representative of how people use a computer. If I'm reading a web page, for instance, I might only access memory intermittently as I scroll the page up and down. But, then again maybe I'm listening to mp3s while I read that web page, so I am intermittently accessing one part of memory, while periodically (as in, at a regular interval) accessing another part of memory. And, let's not forget that I'll randomly startg up new applications, close old ones, etc. which means that there's going to be extra time where my solid-state memory is actually waiting on data from either virtual memory (usually equivalent to the HDD), my CD-ROM drive, or perhaps from the network. None of this is really being factored into the benchmark, but it's something I do all the time which has a much larger impact on my perceived memory performance than the difference between his worst test settings, and his best test settings.
So what you're saying is you plan to pre-emptively patent their policy to prevent patent proliferation? Preposterous!
Am I the only one picturing a lawn gnome with a wire coming out of his butt?
I think there's a huge misconception on slashdot about what a patent is, exactly. You cannot patent a general concept. What you patent is a specific method of doing something. For example, there are several hundred patents issued for different types of can-openers. Obviously, the concept of opening a can has been around since cans were invented. But, not surprisingly the methods used to get at the contents of a can have been quite varied. In fact the thing we now recognize today as the canonical can-opener wasn't invented until 1925. Before then there were all kinds of crazy methods of opening cans that tended to spill stuff all over the place. For further reading on this subject I highly recommend The Evolution of Useful Things by Henry Petroski.
No. They will either have to find a different method than that patented (which shouldn't be too hard, the MS patent is fairly specific in its claims), or license the technology from MS.
Because GPUs are specialized processors. They are only good at a couple of things: moving data in a particular format around quickly, and linear algebra. It is possible to do general-purpose calculations on a GPU, but that's not what it is good at, so you'd be wasting your time.
This is akin to asking why you shouldn't go see a veterinarian when you get sick. Because veterinarians specialize in animals. Sure, they might be able to treat you, but since their training is with animals you might find their treatments don't help as much as going to see a regular doctor.
Why is there a court case in the first place?
If SCO were really concerned about losing IP, they could have discreetly contacted the parties in question, demonstrated their case, and maybe worked out some kind of licensing agreement.
Instead the first thing they do (before even contacting the companies in question) is to file a lawsuit. This is like taking my neighbor to court because his dog did his business on my front lawn. If I ask my neighbor politely to fix the problem, he probably will. If he doesn't, THEN I might take more serious action, but not before.
The first step in any dispute is to try to reach some kind of resolution outside of a courtroom setting. That SCO did not take this step indicates to me that they are up to no good.
This is poor reasoning. By actively ignoring this kind of activity, you are putting YOUR OWN CHILDREN at risk. What if instead of giving authorities the opportunity to trace the origins of the pornography on the machine, you DON'T report it, and the same people who made the child pornography on the guy's computer later molest YOUR kids for profit?
I don't understand your sentiment here, anyway. Sure whistle-blowers might get the shaft sometimes, but I'd like to believe that doing the right thing gets rewarded more often than it gets punished. If I'm ever working at a company and get the impression I could lose my job by doing the right thing, I don't want to continue working for that company anyway. I'd rather risk being fired than teach my children to live in constant fear.
Thanks, I read that article when it came out. In fact I did a lot of my own additional analysis on the data (I think the author credited me when he updated his paper) and found that you can get 5 rentals/month on a basic account with no trouble, and only minor problems occur with 6. Also, the queue lag vs. rentals/month is determined by how much you are paying. That is, if you want to rent more movies per month, then upgrade your subscription to a higher level, and you will continue to have high availability in your queue.
On the empirical evidence side, I've had no troubles at all getting any movie I want in a timely fashion from Netflix.
Try Netflix. You can watch and return movies at your leisure, and they have a much, much better selection of movies than Blockbuster.
This is why I love slashdot.
Thanks, I already understand that perfectly, and you'd know that if YOU had bothered to read my other responses in this thread. It is not my practice to respond to essentially the exact same post 500 times (as you put it), but I thought the irony here was particularly funny.
It's not that it has to be small per se, but high-density. To take your Tokyo example, there are folks who commute to and from Tokyo from other cities in Japan every day. The rail lines in Japan are VERY extensive; it's hard to go asomewhere that doesn't have a train station within a reasonable distance (say, a bike ride). If you also factor in trolleys and buses, there are very few places in Japan you can't get to using public tranportation. The thing that makes it all work is that the entire country is fairly densely populated. You couldn't afford infrastructure to build a rail line to transport only 500-1,000 people/day at a reasonable price, but 20,000-50,000 people is another story.
As it happens, I used to live in Japan, and now live in Los Angeles. There are some areas that I suspect could benefit greatly by a train line, even if you couldn't cover all of L.A. county. At the very least, you could do like the suburbs of Chicago (and elsewhere, like Atlanta), and have huge parking decks with free/cheap parking next to the local train station. I used this *all the time* when I was living in those areas. I imagine a line that ran up the coast through Long Beach, the South Bay cities, and up into LAX, and a similar line from Santa Monica south to LAX would work very well, and probably help alleviate the ever-present traffic jam on 405 just North of LAX. One would hope at least that there are enough people sick of being stuck in that traffic jam to give public transportation a shot.
While I agree that in theory it's possible, I'm not so sure you can just incorporate outside sources so easily into a CMM L5 project, since all of the previously existing development isn't up to snuff. There would be no design documents or QA for the kernel until it was brought in-house, so you'd have to spend several man-years just doing QA on the existing code to "prove" that it was safe.
I didn't say there's a specific problem with it. I said that under the current political environment there's no way it would happen. Public documents are being pulled off shelves faster than you can say "national security." Do you really expect the current administration would allow such sensitive information to be openly published?
My mistake on misunderstanding the GPL.
However my first point was that up until this point, the linux kernel has not been controlled under a CMM level 5 process (far, far from it), and thus incorporating it into the space shuttle software would basically require rewriting it from scratch just to make sure it met CMM level 5 requirements. Since they would have to rewrite anything from scratch anyway, there is no point in using OSS that I can think of, unless an OSS project comes along that happens to be certified CMM level 5.
p.s. also sorry about the broken HTML, my / key seems to be a little stiff.
The submitter said "some" projects.
What on earth makes you think they'd use linux or other OSS to develop the space shuttle software? First of all, the development process for space shuttle software is quite possibly the most rigorous software development process in the world (it is , BTW). There isn't a chance in hell open-source software would be allowed into a level 5 process, because it's not controlled properly. They would essentially have to rewrite any OSS software they used from scratch, just to meet CMM level 5 requirements.
Second, suppose despite point #1, they decide to use the linux kernel on the space shuttle. Obviously, they'd have to adapt the kernel to suit their needs, since most of the hardware on the shuttle is custom designed and built for it. Under the GPL they would have to release any changes they make to the kernel back into the public domain. This would be equivalent to providing a very detailed blueprint of how all the critical systems on the space shuttle function. Especially given the current political environment, do you really think the administration is going to divulge this kind of information to the public?
[sound of door hitting me]
Me: *groan*...oh, hi Bob, I was just picking up some paper clips!
Bob: Do you realize you have 3 of them stuck to your face?