Copyright doesn't protect hardware, unless you're talking about masks,and there are special provisions for those.
So there is a difference in implementing in hardware and implementing in software-- software is already afforded protections under copyright (yes, they are different protections).
It isn't about penalizing software implementations-- It is about only providing enough incentive to innovate-- If you excessively compensate for a patent, the patent (law) is much more likely to be abused.
Excessive compensation implies some arbitrary large amount of profit over the cost of producing the invention. Unfortunately, this is a judgement call, but when profit goes to one individual or entity at the expense of others innovating, it seems clear that the balance is flawed.
No, what needs to be stopped is the American PAtent Office's (implicit) policy of 'if it hasn't been patented, it must be priginal/patentable.
Simply because someone hasn't found it in a patent search does NOT mean that a competent person in the field couldn't invent it.
'Obvious' appears to have become meaningless in the world of patents, and that is one of the large reasons that software patents are so insidious-- Much of what is patented is ludicrisly obvious.
Furthermore, the term of a patent (especially for software, where there is little investment in developing a new algorithm) is too long.
Even if you have someone working on something for a few years, the investment made for a software algorithm is orders of magnitude less than that for many other inventions.
Remember, the -original- purpose of patents was to encourage people to invent! All that requires is an adequate return on your investement-- if the return is 'too good' (or too long), then it ends up harming innovation.
Oh, and we have a 3d perspective. We use multiple cameras to track the ball, else, and I agree, it would likely be innacurate judging distance from the camera.
This is to head off those who would state otherwise.
Since I don't recall the details of QuesTec's system (and honestly, don't want to know the details, for obvious reasons), I can't tell you if they do the same or not, but since WE use multiple cameras, we have stereo vision, which implies that we can compute a depth-map from image disparities.
(ok ok, for those others who've studied computer vision, there is more to it yes, but you'll agree with the basic premise, put a couple of cameras down, know their orientation, position, and lens charactaristics, and given two images of not too disparate views, you'll be able to compute a pretty reasonable depth-map given sufficient texture.)
I work for Sportvision, and we do K-Zone, on ESPN. K-Zone does something quite similar to QuesTec, except that we put the strikezone in the video, for real.
I.e. we don't render a seperate 3d virtual world, our virtual world IS the real world (or at least, it verrrrry closely corresponds, since we work very hard to ensure that it is so).
Since I work for Sportvision, I can't talk about their product, but I can tell you that our product is quite good at tracking the pitches (optically), and we're also quite good at putting the video up where it is supposed to be (not to mention sizing the strikezone the way the rules state it should be sized, as opposed to having a fixed strikezone size).
Of course, I believe that the above poster is accurate, at least with regards to Sportvision's system-- Wouldn't you be unhappy (and searching for means to get rid of such a device) if there was a device that showed how you made mistakes in front of a national audience?? I know I sure as heck would!!
We also do other things, like tracking cars using differential GPS in Nascar Winston Cup events (soon to be the Nextel Cup.. I'm afraid of the fact that I know such things now..!!), and putting down virtual (yellow) first down lines for football games (on many different broadcasts), and puttind flags 'under' the ice for Olympic speedskating.
Anyway, that was a "ME Too!!" =) If you want to check out our propaganda, we have a website at the obvious address of: www.sportvision.com
No I don't speak for the company, I'm just a grunt, but I'm proud of my work.
Oh, and did I mention? We use Linux to make those all of our Nascar real-time special effects (things that point at cars, things that show how fast a car is relative to the pole car at any point on the track, etc, etc)
Do we use Windows too? Yup, but I like my linux boxes better =)
The simplex algorithm is an algorithm for finding an optimal solution to a set of inequalities. (i.e. a system of constraints)
I say 'an optimal solution' because there can be more than one.
If viewed in space, the solution set for a simplex problem is a convex, generally closed region of space. In isn't neccessarily closed, but it is necessarily convex if it is closed, and it is never concave.
The simplex algorithm is an algorithm whereby some of the points/vertices of that solution space are visited in a search, until an optimal vertex (i.e. solution) is discovered.
It isn't completely fool-proof, you can get into states where the algorithm bounces between two vertices forever (looping), but for most well-stated linearly constrained problems (in any number of dimensions, really), the simplex is a good way to find an optimal solution.
Sorry, for the rather rambling explanation.
If you're truely interested, do a web search for "simplex method"
I expect that if they are cheating, they should be caught at it.
It is premature to conclude that they are cheating, though it seems likely. In any case were they cheating, either they are doing less of it, or they are less effective at it.
That being said, if they are, I hope their noses are rubbed in it as well. I'm not a big fan of liars, and that is exactly how I percieve this kind of cheating.
YES... *(&^#*$(&^ It is a cheat. Look, to rehash an analogy, what the driver is doing in this benchmark is equivalent to your car disregarding the steering input and driving straight to your workplace, regardless of buildings, etc.
This not what you the user (or developer) wants. You want the dang thing to do what you tell it to do.
It is a CHEAT.
Here is an example of an optimization: You realize that you'll be driving in snow and ice a lot, so you let some air out of your tires to provide more contact between the tires and the surface. This results in greater traction across snow and ice, but less speed on dry pavement.
It is one thing to optimize a specific render path in a driver for a game. Frankly, its a good thing, but cheating i.e. shortcutting and not executing the commands I send to the card is pure and utter crap and their noses should be rubbed in it.
Conceeded: they have different capabilities and features. The point of a benchmark is to take the common featureset among those disparate features and capabilities, and compare them, thus, apples-to-apples.
It may be that an oranges-to-bananas comparison is more interesting, or more applicable, but the purpose of a BENCHMARK is apples-to-apples comparison, else the information it provides is useless.
Example: Take a WW-II era B-29 and an Airbus-A300, and you use this to compare the American and European aerospace industries? I'd call you a moron.
Unlike art, graphics hardware, software, and silicon can be compared QUANTITATIVELY. I conceed that paintings cannot be compared on an apples-to-apples basis w.r.t subjective measures. Anything where "subjective measures" is done in low quantities is not an apples-to-apples comparison, however, quantify this subjective measure.. i.e. sampling the subjective measurement of many, gives it more of an apples-to-apples feel (since it becomes a quantitative measurement.. i.e. you can say that 67% of people prefer Van Gogh to Dali)
You are implying that since paintings cannot be compared on an apples-to-apples basis, that nothing can.
I vehemently disagree.
When the set of behaviours for a system has an intersection with the set of behaviours for another system, then the performance of the systems can be qualitatively compared on the basis of that intersection.
Now, onto why per-game optimizations have nothing to do with this cheating in the benchmark--
When I send a vertex program (etc) to the graphics card, or tell it to clear the back-buffer, I expect the damn hardware to to exactly what I ask of it.
It is one thing to cut corners on precision, or optimize a certain render path, and another completely to disregard instructions...
Analogy: How would you feel if you were headed to work and your car decided to disregard the steering input and headed in a straight line towards work because it was a 'short cut'??
On the other hand, knowing that you would be travelling in snow and icy conditions, you let a little air our of your tires. This is an optimization for winter weather, since more of the tire surface is in contact with the ground. The car still does what you ask it to, but it performs better in snow and ice, and worse on dry pavement.
This is more the case when optimizating a specific render path.
I expect application-level optimizations to occur on an application-level. Anything else is a breach of specification and I want them to get their grubby hands OFF god dangit.
Allright?
Furthermore, a failure to provide an objective, apples-to-apples comparison does not imply that was not the purpose.
I've read the article, and it implies that ATI was (perhaps) cheating on one test (Game 4), while they found NVidia cheating across the board.
Also, ATI's changes (cheats or not) accounted for a less than 2% change overall, where NVidia's changes accounted for 20+% That is a big difference.
It seems unlikely that ATI was cheating as much. I'd think they'd try to cheat in nearly the same ways give the efficacy of the cheats were they trying to cheat.
I do live television special effects for sports, and while I care a -GREAT DEAL- about the performance of the graphics cards (If I screw up, millions of people see it.. right away), we have small enough volume (you only need one system to make graphics for millions of people), that ATI and NVidia don't just hand out their cards to us.
Would I prefer it that way? (Who doesn't like free goodies??) Heck yes! I'd like to get the latest card and evaluate its robustness (Very important to television...) right away so that I can qualify it for use in our systems only a few weeks after it comes out instead of months.
On that note, I'm also constrained by lack of support for Linux on the latest cards (at times). For example, the 9800 doesn't yet have an accelerated linux driver. Dangit! Now, I love the 9700 pro, but I'd love to have that 256 meg on-card.. It is amazing how quickly you can eat up texture memory when you're doing things the card manufacturers didn't think of (like chroma-keying, video mapping, interlaced frame rendering, blah blah)
At least for/in the US. In other countries (i.e. under different treaties and copyright laws), in some cases copyright is EXPLICIT... i.e. if you don't put the notice on there, it doesn't count.
I forget the name of the convention/treaty, but in certain countries, for instance, if you don't put the "(C)" in there, it isn't copyright..
Re:Download AND Pay?
on
The Law and P2P
·
· Score: 3, Informative
No offence, but your phrasing STINKS.
Downloading music isn't illegal. Downloading movies isn't illegal.
Doing either of the above in breach of copyright law IS illegal.
Blanket statements such as: "Downloading music is illegal" are misleading and inaccurate.
On to your other points.
Fair use is and has been eroding. The fact that you can't buy your Inu Yasha DVDs straight from Japan and play them on your DVD player is a perfect example of this. Modify your dvd player to play discs from other regions? Face years in Jail!
I'm all for fair play-- That is why it is my intention to screw those who screw me in equal measure.
How do you feel about price fixing? I don't believe -for a moment- that I'll be getting back my fair share of money from their activities.
IMHO, what they are suffering from now is what happens when you act the part of the big bully- People don't invite you to the party, and work around you.
There is a HUGE difference between illegal and wrong. You seem to be making the assertion that illegal==wrong. *BUZZZZ* AAhh, you're a loser at the game of morality!
Since when does the law==justice==right/wrong? The law is an approximation of a justice system. It isn't even arguably the best we can do!! Do you condone slavery? You DO know that people were allowed to -own- people not too many years ago, right?.. And a matter of fact, it protected under law...
Keep in mind that the RIAA isn't really in the business of producing music. Music has never been a way to make money (for the majority of people who play/produce it throughout history)... The RIAA is in the business of making STARS. Celebrities.. i.e. trademarks, -are- a good way of making money. The music is incidental (And if you're going to argue that it isn't, what is the purpose of a for-profit corporation? )
Actually, had this been done, oh say 5 years ago, that would have fallen into fair-use, so long as the clips were relatively short (i.e. a smal part of the work) IANAL
Actually, the compiler support in.net (2002, or the upcoming 2003) is much MUCH better than it was in MSVC++6.0
2002+ the Intel C++ compiler gets you a nearly ISO standards compliant (insofar as ANY development environment is ISO c++ compliant), and 2003 should be there without the addon.
Since I use 'advanced c++' features pretty often, it isn't even possible for me to use 6.0.
In short, MSVC++.net (2002|2003) is a vast improvement in the compiler area.
Keep in mind that these manufacturers are not yet experiencing economies-of-scale... that is to say that their unit costs are much higher since they are selling perhaps thousands as opposed to (for example) Logitech's millions.
I'm not saying its inexpensive.... I'm not saying they may be making more profit off of each device than Logitech.. I'm saying that if you sell thousands of devices, you will have to charge more to keep your employees payed.
Breakdown of additional costs (above the $1 more plastic) 1) Tooling (at the factory. This is expensive) 2) People (This is REALLY expensive) 3) Marketing 4) Support
Keep that in mind. Producing innovative designs is often more expensive than commodity designs-- Few places have done the tooling for 'em, which means higher costs per-unit until they become a commodity.
If someone who has a patent sues someone indicating "patent infringement," then, as far as I know, the burden of proof of innocense rests on the shoulders of the accused.
Slashdot Mirror
Hahah. no, that is the revision of the ATI driver, not the kernel you need to use.
do a web search for:
ati linux 2.9.12
-or-
ati linux 2.9.13
You'll find prerelease drivers that work fine with 4.3
I need to proofread more often. I can count more grammatical mistakes in the above comment than dollars in my wallet.
Copyright doesn't protect hardware, unless you're talking about masks,and there are special provisions for those.
So there is a difference in implementing in hardware and implementing in software-- software is already afforded protections under copyright (yes, they are different protections).
It isn't about penalizing software implementations-- It is about only providing enough incentive to innovate-- If you excessively compensate for a patent, the patent (law) is much more likely to be abused.
Excessive compensation implies some arbitrary large amount of profit over the cost of producing the invention. Unfortunately, this is a judgement call, but when profit goes to one individual or entity at the expense of others innovating, it seems clear that the balance is flawed.
No, what needs to be stopped is the American PAtent Office's (implicit) policy of 'if it hasn't been patented, it must be priginal/patentable.
Simply because someone hasn't found it in a patent search does NOT mean that a competent person in the field couldn't invent it.
'Obvious' appears to have become meaningless in the world of patents, and that is one of the large reasons that software patents are so insidious-- Much of what is patented is ludicrisly obvious.
Furthermore, the term of a patent (especially for software, where there is little investment in developing a new algorithm) is too long.
Even if you have someone working on something for a few years, the investment made for a software algorithm is orders of magnitude less than that for many other inventions.
Remember, the -original- purpose of patents was to encourage people to invent! All that requires is an adequate return on your investement-- if the return is 'too good' (or too long), then it ends up harming innovation.
Oh, and we have a 3d perspective. We use multiple cameras to track the ball, else, and I agree, it would likely be innacurate judging distance from the camera.
This is to head off those who would state otherwise.
Since I don't recall the details of QuesTec's system (and honestly, don't want to know the details, for obvious reasons), I can't tell you if they do the same or not, but since WE use multiple cameras, we have stereo vision, which implies that we can compute a depth-map from image disparities.
(ok ok, for those others who've studied computer vision, there is more to it yes, but you'll agree with the basic premise, put a couple of cameras down, know their orientation, position, and lens charactaristics, and given two images of not too disparate views, you'll be able to compute a pretty reasonable depth-map given sufficient texture.)
We do this too.
On TV even!
I work for Sportvision, and we do K-Zone, on ESPN.
K-Zone does something quite similar to QuesTec, except that we put the strikezone in the video, for real.
I.e. we don't render a seperate 3d virtual world, our virtual world IS the real world (or at least, it verrrrry closely corresponds, since we work very hard to ensure that it is so).
Since I work for Sportvision, I can't talk about their product, but I can tell you that our product is quite good at tracking the pitches (optically), and we're also quite good at putting the video up where it is supposed to be (not to mention sizing the strikezone the way the rules state it should be sized, as opposed to having a fixed strikezone size).
Of course, I believe that the above poster is accurate, at least with regards to Sportvision's system-- Wouldn't you be unhappy (and searching for means to get rid of such a device) if there was a device that showed how you made mistakes in front of a national audience?? I know I sure as heck would!!
We also do other things, like tracking cars using differential GPS in Nascar Winston Cup events (soon to be the Nextel Cup.. I'm afraid of the fact that I know such things now..!!), and putting down virtual (yellow) first down lines for football games (on many different broadcasts), and puttind flags 'under' the ice for Olympic speedskating.
Anyway, that was a "ME Too!!" =)
If you want to check out our propaganda, we have a website at the obvious address of: www.sportvision.com
No I don't speak for the company, I'm just a grunt, but I'm proud of my work.
Oh, and did I mention? We use Linux to make those all of our Nascar real-time special effects (things that point at cars, things that show how fast a car is relative to the pole car at any point on the track, etc, etc)
Do we use Windows too? Yup, but I like my linux boxes better =)
The simplex algorithm is an algorithm for finding an optimal solution to a set of inequalities. (i.e. a system of constraints)
I say 'an optimal solution' because there can be more than one.
If viewed in space, the solution set for a simplex problem is a convex, generally closed region of space. In isn't neccessarily closed, but it is necessarily convex if it is closed, and it is never concave.
The simplex algorithm is an algorithm whereby some of the points/vertices of that solution space are visited in a search, until an optimal vertex (i.e. solution) is discovered.
It isn't completely fool-proof, you can get into states where the algorithm bounces between two vertices forever (looping), but for most well-stated linearly constrained problems (in any number of dimensions, really), the simplex is a good way to find an optimal solution.
Sorry, for the rather rambling explanation.
If you're truely interested, do a web search for "simplex method"
I expect that if they are cheating, they should be caught at it.
It is premature to conclude that they are cheating, though it seems likely. In any case were they cheating, either they are doing less of it, or they are less effective at it.
That being said, if they are, I hope their noses are rubbed in it as well. I'm not a big fan of liars, and that is exactly how I percieve this kind of cheating.
YES ... *(&^#*$(&^ It is a cheat.
Look, to rehash an analogy, what the driver is doing in this benchmark is equivalent to your car disregarding the steering input and driving straight to your workplace, regardless of buildings, etc.
This not what you the user (or developer) wants. You want the dang thing to do what you tell it to do.
It is a CHEAT.
Here is an example of an optimization:
You realize that you'll be driving in snow and ice a lot, so you let some air out of your tires to provide more contact between the tires and the surface. This results in greater traction across snow and ice, but less speed on dry pavement.
It is one thing to optimize a specific render path in a driver for a game. Frankly, its a good thing, but cheating i.e. shortcutting and not executing the commands I send to the card is pure and utter crap and their noses should be rubbed in it.
OK, I have to call B*llshit over this comment.
.. i.e. you can say that 67% of people prefer Van Gogh to Dali)
Conceeded: they have different capabilities and features. The point of a benchmark is to take the common featureset among those disparate features and capabilities, and compare them, thus, apples-to-apples.
It may be that an oranges-to-bananas comparison is more interesting, or more applicable, but the purpose of a BENCHMARK is apples-to-apples comparison, else the information it provides is useless.
Example: Take a WW-II era B-29 and an Airbus-A300, and you use this to compare the American and European aerospace industries? I'd call you a moron.
Unlike art, graphics hardware, software, and silicon can be compared QUANTITATIVELY. I conceed that paintings cannot be compared on an apples-to-apples basis w.r.t subjective measures. Anything where "subjective measures" is done in low quantities is not an apples-to-apples comparison, however, quantify this subjective measure.. i.e. sampling the subjective measurement of many, gives it more of an apples-to-apples feel (since it becomes a quantitative measurement
You are implying that since paintings cannot be compared on an apples-to-apples basis, that nothing can.
I vehemently disagree.
When the set of behaviours for a system has an intersection with the set of behaviours for another system, then the performance of the systems can be qualitatively compared on the basis of that intersection.
Now, onto why per-game optimizations have nothing to do with this cheating in the benchmark--
When I send a vertex program (etc) to the graphics card, or tell it to clear the back-buffer, I expect the damn hardware to to exactly what I ask of it.
It is one thing to cut corners on precision, or optimize a certain render path, and another completely to disregard instructions...
Analogy:
How would you feel if you were headed to work and your car decided to disregard the steering input and headed in a straight line towards work because it was a 'short cut'??
On the other hand, knowing that you would be travelling in snow and icy conditions, you let a little air our of your tires. This is an optimization for winter weather, since more of the tire surface is in contact with the ground. The car still does what you ask it to, but it performs better in snow and ice, and worse on dry pavement.
This is more the case when optimizating a specific render path.
I expect application-level optimizations to occur on an application-level. Anything else is a breach of specification and I want them to get their grubby hands OFF god dangit.
Allright?
Furthermore, a failure to provide an objective, apples-to-apples comparison does not imply that was not the purpose.
I've read the article, and it implies that ATI was (perhaps) cheating on one test (Game 4), while they found NVidia cheating across the board.
Also, ATI's changes (cheats or not) accounted for a less than 2% change overall, where NVidia's changes accounted for 20+% That is a big difference.
It seems unlikely that ATI was cheating as much. I'd think they'd try to cheat in nearly the same ways give the efficacy of the cheats were they trying to cheat.
I do live television special effects for sports, and while I care a -GREAT DEAL- about the performance of the graphics cards (If I screw up, millions of people see it.. right away), we have small enough volume (you only need one system to make graphics for millions of people), that ATI and NVidia don't just hand out their cards to us.
Would I prefer it that way? (Who doesn't like free goodies??) Heck yes! I'd like to get the latest card and evaluate its robustness (Very important to television...) right away so that I can qualify it for use in our systems only a few weeks after it comes out instead of months.
On that note, I'm also constrained by lack of support for Linux on the latest cards (at times). For example, the 9800 doesn't yet have an accelerated linux driver. Dangit! Now, I love the 9700 pro, but I'd love to have that 256 meg on-card.. It is amazing how quickly you can eat up texture memory when you're doing things the card manufacturers didn't think of (like chroma-keying, video mapping, interlaced frame rendering, blah blah)
I believe the magical form is:
.. And I agree, just putting the (C) was not sufficient.
Copyright (C) ENTITY DATE
But, I'm not sure.
At least for/in the US. In other countries (i.e. under different treaties and copyright laws), in some cases copyright is EXPLICIT... i.e. if you don't put the notice on there, it doesn't count.
I forget the name of the convention/treaty, but in certain countries, for instance, if you don't put the "(C)" in there, it isn't copyright..
No offence, but your phrasing STINKS.
.. And a matter of fact, it protected under law...
Downloading music isn't illegal.
Downloading movies isn't illegal.
Doing either of the above in breach of copyright law IS illegal.
Blanket statements such as: "Downloading music is illegal" are misleading and inaccurate.
On to your other points.
Fair use is and has been eroding. The fact that you can't buy your Inu Yasha DVDs straight from Japan and play them on your DVD player is a perfect example of this. Modify your dvd player to play discs from other regions? Face years in Jail!
I'm all for fair play-- That is why it is my intention to screw those who screw me in equal measure.
How do you feel about price fixing? I don't believe -for a moment- that I'll be getting back my fair share of money from their activities.
IMHO, what they are suffering from now is what happens when you act the part of the big bully- People don't invite you to the party, and work around you.
There is a HUGE difference between illegal and wrong. You seem to be making the assertion that illegal==wrong. *BUZZZZ* AAhh, you're a loser at the game of morality!
Since when does the law==justice==right/wrong?
The law is an approximation of a justice system. It isn't even arguably the best we can do!!
Do you condone slavery? You DO know that people were allowed to -own- people not too many years ago, right?
Keep in mind that the RIAA isn't really in the business of producing music. Music has never been a way to make money (for the majority of people who play/produce it throughout history)... The RIAA is in the business of making STARS. Celebrities.. i.e. trademarks, -are- a good way of making money. The music is incidental (And if you're going to argue that it isn't, what is the purpose of a for-profit corporation? )
Actually, had this been done, oh say 5 years ago, that would have fallen into fair-use, so long as the clips were relatively short (i.e. a smal part of the work)
IANAL
Actually, the compiler support in .net (2002, or the upcoming 2003) is much MUCH better than it was in MSVC++6.0
.net (2002|2003) is a vast improvement in the compiler area.
2002+ the Intel C++ compiler gets you a nearly ISO standards compliant (insofar as ANY development environment is ISO c++ compliant), and 2003 should be there without the addon.
Since I use 'advanced c++' features pretty often, it isn't even possible for me to use 6.0.
In short, MSVC++
Ok, good to know.
Once it exists in the spec, then its up the application writers to actually implement it.
In this regard, thusfar, http is behind.. Hopefully it will catch up soon.
Does HTTP support file continuation now?
If it does, and I'm not aware of it, I'd like to be... but thusfar I'm unaware of that capability existing as part of the HTTP.
When transferring large files, continuation is invaluable.
Keep in mind that these manufacturers are not yet experiencing economies-of-scale... that is to say that their unit costs are much higher since they are selling perhaps thousands as opposed to (for example) Logitech's millions.
I'm not saying its inexpensive.... I'm not saying they may be making more profit off of each device than Logitech.. I'm saying that if you sell thousands of devices, you will have to charge more to keep your employees payed.
Breakdown of additional costs (above the $1 more plastic)
1) Tooling (at the factory. This is expensive)
2) People (This is REALLY expensive)
3) Marketing
4) Support
Keep that in mind. Producing innovative designs is often more expensive than commodity designs-- Few places have done the tooling for 'em, which means higher costs per-unit until they become a commodity.
I tend to think that it is a 'new' computer when I change its name.
THe rest doesn't matter one whit. If I put a new motherboard in the system, and still call the box 'foo', its 'foo', not 'new foo'.
Its an problem strife with ambiguity, and unfortunately, the law doesn't care what most people think...
Yup, but it can still ruin your life.
Well, I'll gve you an example then--
If someone who has a patent sues someone indicating "patent infringement," then, as far as I know, the burden of proof of innocense rests on the shoulders of the accused.
Now -that- I agree with.