Perhaps someone from the States can explain something to me, seeing as I'm a Brit. So some law gets passed by both Congress and Senate, and the President then signs it into law, effectively creating a government body (I'm thinking the new Consumer Financial Protection Bureau, but I'm sure there are plenty of other examples). This has been considered to be a Good Thing (TM), and then Congress then de-funds it, effectively shutting it down.
Why don't they just repeal the original law that created it in the first place?
In the US, it is much, much easier to prevent something from happening in politics than it is to get something to happen. The don't just repeal it because they don't have the votes, so the most effective way to attack it is to attack the funding for it. To pass almost anything, it requires a simple majority in the House, a 60% majority in the Senate (to pass the inevitable filibuster), and the president to sign it. The other alternative is 2/3 majority in House, 2/3 majority in Senate to override presendential veto. The Consumer Financial Protection Bureau was passed when Democrats controlled the House, Senate, and White House. Currently, the Democrats still control the Senate (with less than 60% majority) and the White House, but Republicans control the House. Republicans want to repeal the law, but Democrats want to keep it. If Republicans tried to repeal it, it would be blocked by the Democratic Senate and White House. If they try to defund it, that will still be blocked by the Democratic Senate and White House, but since that is tied into the federal government's budget which funds everything, eventually something has to give. It ends up becoming a giant game of chicken usually and at the last minute, a compromise that no one likes very much is approved.
Winning 2 divisions in 1 league out of 5 is NOT a sweep. A sweep would be winning all the leagues. 1 out of 5 isn't even close. Congratulations on the wins in the humanoid league.
For those that are curious, the other 4 leagues are simulation (focusing on team play and low barrier to entry), small size robot (hardware/software combined, wheeled robots), middle size robot (hardware/software combined, wheeled robots), and standard platform league (software only using real humanoid robots).
Survival is a terrible metric of intelligence. By that standard, lions and tigers and bears are the most intelligent species on the planet.
Forget lions and tigers or even insects and bacteria. By this metric, the water in the ocean, the sun we orbit around, and the vast expanses of space are far more intelligent than anything that has ever lived or any system we have put in place.
Let's suppose that the piece of junk has an injury radius of 50m, or 8000 sq m. There's 510 million sq m of surface area on the earth, so the chance that it will fall within 50m of me is 8,000::510,000,000 or 1 in 63,750. You're roughly 10x as likely to die in a car accident in any give year. That actually is much closer to significant than I thought it might be before I did the calculation. We do take steps to alter our risks of vehicular death: we use seat belts,we drive cautiously etc. But we don't take further steps, like installing a racing seat with a five point harness. Arguably, then, further marginal reductions in our death risk are not that meaningful to us, and we would very likely do nothing to reduce a risk that is already 1/10 as great as the point of diminishing returns for car risk.
Actually, it is much closer to significant than you thought because you made an error in the calculation. There's 510 million sq km of surface area on the earth, so the chance that it will fall within 50m of me is 8,000::510,000,000,000,000 or 1 in 63,750,000,000. You're roughly 10,000,000x as likely to die in a car accident in any give year.
Your calculations further down seem to have done the calculation correctly
The marketing date is a huge issue because 90% of the time, the people making the game have no buy-in regarding it. They're working towards being done when it's done, and then when they get told that they have six months when they need a year, things get implemented too fast and half-assed.
I think that's a poor excuse. The date needs to be agreed on between marketing and development well ahead of time. A rough date should probably be part of the initial game design where the game is being defined.
Even if marketing solely decides the date and doesn't give the date until 6 months out, development shouldn't just ignore that they need 1 year and stick their heads in the sand. At that point development needs to go back to marketing and try to work things out. If the date is fixed, then the game needs to be rescoped until it fits in the time available, even if it means removing features. It's better to have a slightly smaller, playable game then a larger, bug-ridden beast.
If both schedule and feature list are fixed, then it is managements fault for not manageing properly.
The game companies are killing their own market. I haven't bought many games since the games starting getting buggy and unbalanced on release. At $50 a game, I expect pretty near perfection. Lower the price and I'll be more lenient. Lately, I've been playing online games instead since once you find a good one you can play it for a long time and don't have to worry much about bugs anymore.
Why is anyone going to care about a weapon system everyone knows is a dud anyway?
The system has never once demonstrated that it works, every single test has either failed outright or been rigged. The only reason the program exists at all is to hand out taxpayer money to campaign contributors.
Even if the system doesn't work, it's still bad to have unsecured access to it. It's certainly a concern that some terrorist might be able to infiltrate the system and use it to shoot down passenger planes or cause some other kind of damage. The system is designed to destroy things, incoming missles in this case, how hard would it be to redirect that destructive force to cause problems?
Contract companies love to say "that's not in the contract, but we'd be happy to renegotiate and do it for $x."
And for good reason. Same reason that when you order a Dodge Neon they don't ship you a Dodge Viper. The contract is what is specifying what the government is buying. Change what the government is buying to enhance it and it's not really surprising that they want more money to produce it. Taking the least expensive option is usually the right option for the company even if it isn't in the system. It also doesn't surprise me in the least. The government really should have some boilerplate in their contracts saying system much follow established DoD security procedures as specified in DoD Standard Security Policies v10.43 or something along those lines.
With what they have so far, I don't see any reason not to just use a tracked vehicle. The sense of balance is impressive, but if you don't have all your weight mounted so high it becomes less of an issue.
Besides, most laptops come with LCD screens that look like crap if you try to scale an image to fit instead of just rendering it the right size to begin with.
Every time a patent story comes up, people always point out that the developer needs to have his invention protected. However, a patent does more than just prevent people from copying the inventor's invention. A patent prevents anyone else from independently developing the invention. This is why the invention needs to be non-obvious, which sadly seems to have effectively dropped from the patent approval process. Who protects the independent inventor who happens to be working on the same idea and developing a similar device independently? It is at least as important to protect this independent inventor as it is to protect the patent filer.
The whole patent system seems pretty ill-conceived to me. I don't particularly have a better solution, but strong enforcement of the prior art and non-obvious criterion would be a good start. The problem in my mind is that the whole patent system is based on the premise that inventions are developed from scratch. This just isn't the way most progress is done. Almost all inventions are combinations or refinements of existing technology. Sometimes those combinations/refinements are a large leap, sometimes they are rather obvious next steps.
In the case of hybrid cars, it's a combination of existing internal combustion engine and electric engine technology. The ground breaking part isn't the combination so much as it's the design to make a practical system. It's obvious that the technologies could be combined, it's not that much of a leap to see that it could possibly improve gas mileage based on existing understanding of effeciencies, but it's very unclear how to put the whole thing together such that you actually end up with better gas mileage.
A patent is only as good as your ability to enforce it. I think you'd be hard pressed to perform any patent litigation for under $10000. Since no one who can't afford the fee can effectively enforce the patent, this seems like a moot argument.
I was extremely impressed in the ext's; I simply had no idea how consistently well performing they were.
I'm not so sure about consistent performance. I've run in to the extreme slow down caused by many files in the same directory many times and it can be pretty painful. I'm not sure if this has been addressed at all, but from my understanding it is a limitation of the way the file system data is stored on disk with extra levels of indirection being required. With Reiser, I can have directories with 10,000s files with no problems.
ArrayList l = new ArrayList();
l.append("A String");
String s = (String)l.item(0);
Is an upcast. and I dare you to find a list implementation in any type-strong language that doesn't require an upcast in this situation. You need it to be able to store objects of an anonymous type on a list.
Whenever I want a container of objects, I always want them to be of the same type or have the same base type. Otherwise, it usually doesn't make any sense to put them in the same container. At least 95% of the time, they will all have the same type.
C++ allows me to express this:
std::vector l;
l.push_back("A String");
std::string s = l[0];
No upcast required because I can tell the compiler that I'm only going to store strings in my vector rather than always having to store an Object. This also makes Java harder to read because when looking at a container, it is impossible to tell from the variable declaration what type of object will be stored in the container.
I tried one that a friend had. It has to be the *MOST* irritating keyboard I have ever tried. I have big fingers and found it almost impossible to type more than a word without hitting 2 keys instead of 1 on some letter or other.
It always amuses me to see "oh, by blind, senile grandmother coulda done this" comments. The point is, they didn't.
If all these things are as obvious as people like to claim, why don't they patent them? Is it that maybe they were only obvious after someone stated them?
Perhaps because they didn't think they were worthy of a patent?
On second thought, maybe I agree with you. I been meaning to send in my patent application for a method for temporarily storing water in an accessible form using a cylindrical device with one end capped. I'm thinking of calling it a "cup". You interested in licensing it?
I recommend going to www.vote-smart.org for information about politicians. It's run by a bunch of volunteers and they refuse all money from anyone except private individuals. The have voting records, statements made, and the opinions of all of the position groups for every politician and candidate.
UNSW (Uni of New South Wales, Aust). Always is. They have won every game they have competed in, with an average lead of 30 goals (no joke).
Actually, UNSW lost games to both Carnegie Mellon University and LRP (Paris) their first year. Also, they have never won a game by more than 14 goals and the games were all close their first year. I'd say the difference is that UNSW spends a lot of time tuning there systems for the task at hand. Many of the other universities spend more time doing basic research, which won't make you play better soccer this year, but might a few years down the road. If you measure success in terms of published papers, then UNSW is behind many of the other universities. That being said, their system is impressive. Hopefully, this year we (I'm from Carnegie Mellon University) can give them a bit more of a challenge. The gap certainly closed between 2000 and 2001. It should be exciting to see what happens this year.
OK, are these robots REAL robots (ie. pre-programmed automatons) or the lame "robot wars" robots (ie. glorified RC cars)?
These are real robots. The robots are completely controlled by computers in all leagues. Human intervention is limited to starting/stoping the robots and enforcing rules. All decisions during play are made by the robot players themselves.
I'm on the University of Virginia team, and we're one of two teams to qualify representing the US for the simulation league (which doesn't use real robots, and is thus a lot more fast paced). The other team being AT&T Research Labs.
Just a small clarification. The parent is referring to the simulator league. There are also several US teams entering in other leagues. For example, Carnegie Mellon University is entering in the small-size league and the Sony legged league.
I was at the RoboCup98 in paris, it will be interesting to see how things have changed. The match that I saw involved a few brick looking robots moving around in a random manner. It was the first time that I saw the Sony Dogs... Do the robots now wear boots?
The robots now look much slicker. They also perform a lot better. The teams no longer look random at all. They are certainly capable of scoring. Last year saw many scores of around 10-1. The robots still go barefoot.
How does one compete in this tournament? It looks like all the teams have identical equipment: do the companies (like Sony) sponsor them?
There are several different leagues that one can enter. There is a simulator league, small-size league, middle-size league, Sony legged league, a humanoid league (new for this year), and a couple rescue leagues.
The simulator league has no hardware. The software is downloadable from sourceforge. This is the easiest league to get started in since there isn't a lot of money involved and you can even enter remotely. Keep in mind that a lot of researchers are working hard on this problem, so most of the teams are really good. The Sony legged league is run by Sony. Sony provides a lot of support. You have to make a proposal to Sony to get into this league. A committee of researchers and Sony personel decide which teams to allow in. Everyone uses the Sony hardware in this league. The small-size and middle-size leagues have hardware built by the teams. It usually costs at least $10,000 for all the hardware needed for a small-size team. Middle-size teams run from $3,000-$40,000 per robot (4 on 4 competition). See www.robocup.org for more details. Registration is already over for this year, but there will be another competition next year (in Italy, I believe).
Has been able to win against the best chess player. So if I understand correctly, the logic involved to win a game isn't really the problem, it is more of a mechanical problem, isn't it? Recent video games are already able to beat really good video game player, and there are not even super-killer-incredibly-intelligent applications.
There are plenty of software problems. It turns out that chess is actually easy compared to the stuff we do everyday like walking. It's just that we have a lot of specialized wet-ware for the normal tasks, so we don't notice how difficult they are. Control and perception are major problems. Planning and coordination are also major problems. Chess is easier because things don't move while you are thinking and you don't have to pick up the pieces if you do it like Deep Blue did. The video games are usually easier for the computer than the real world because they have perfect information about what is going on in a convenient format for them. They almost always have a huge interface advantage as well. The humans attention is usually divided and forced to operate through a relatively clunky interface where the computer doesn't mind keeping track of 80 things at one time. The fact that the human ever wins at the hardest levels shows how stupid the computers are.
The only reason they are staging this competition is that combat seems to be the only way that men express themselves or advance their knowledge. Isn't it enough that a robot can track, intercept and guide an object without having to turn it into a "battle" complete with winners and losers?
As a graduate student who has been working on robots in RoboCup for 4 years, I think you are missing the point. AI made many discoveries and developed many algorithms during the '70s by focussing on the challenge of making computers play chess. At the time, it was believed that if the problem of playing chess was solved we would be able to do almost anything we wanted with robots. Over time, it became very clear to roboticists that being able to play chess was not helping us build better robots. The RoboCup initiative was started in order to provide a new grand challenge for roboticists to work on and provide a common framework for comparisons. Soccer was chosen because it exemplifies many of the things that robots are traditionally bad at. It is an environment with multiple robots cooperating (and yes competing), lots of dynamic activity, difficult sensing problems, and challenging motor control problems. Basically, everything robots are bad at. Soccer was chosen over other alternatives simply because it is more popular around the world which in turn makes it easier to get funding. All of the research from soccer applies to many other domains that people really care about. The RoboCup organization has also recognized the need to do other things than soccer and also run a search and rescue domain at the competition each year which is equal in stature to all of the other leagues. The search and rescue problem is to find people trapped after natural or other disasters. This was started a couple of years ago.
And they use the term "PMPO" - "Peak Music Power Output". Fine, putting aside the fact that this term has no accepted definition in electrical engineering - let's say that those little Taiwanese-made speakers contain an amplifier with a big bank of capacitors to dump out enough current to achieve 250 watts peak. If the power supply to them is only 9V, the capacitors would never get above 9V. If the speakers themselves have a standard nominal impedance of 8 ohms, then we can calculate
Not to say they aren't lying, they are, but capacitors can be charged to a higher voltage than the source using a voltage doubler circuit or a flyback voltage multiplier. Doesn't give you any more power but does trade current for voltage. See this page for some example circuits.
Slashdot Mirror
Perhaps someone from the States can explain something to me, seeing as I'm a Brit. So some law gets passed by both Congress and Senate, and the President then signs it into law, effectively creating a government body (I'm thinking the new Consumer Financial Protection Bureau, but I'm sure there are plenty of other examples). This has been considered to be a Good Thing (TM), and then Congress then de-funds it, effectively shutting it down.
Why don't they just repeal the original law that created it in the first place?
In the US, it is much, much easier to prevent something from happening in politics than it is to get something to happen. The don't just repeal it because they don't have the votes, so the most effective way to attack it is to attack the funding for it. To pass almost anything, it requires a simple majority in the House, a 60% majority in the Senate (to pass the inevitable filibuster), and the president to sign it. The other alternative is 2/3 majority in House, 2/3 majority in Senate to override presendential veto. The Consumer Financial Protection Bureau was passed when Democrats controlled the House, Senate, and White House. Currently, the Democrats still control the Senate (with less than 60% majority) and the White House, but Republicans control the House. Republicans want to repeal the law, but Democrats want to keep it. If Republicans tried to repeal it, it would be blocked by the Democratic Senate and White House. If they try to defund it, that will still be blocked by the Democratic Senate and White House, but since that is tied into the federal government's budget which funds everything, eventually something has to give. It ends up becoming a giant game of chicken usually and at the last minute, a compromise that no one likes very much is approved.
Winning 2 divisions in 1 league out of 5 is NOT a sweep. A sweep would be winning all the leagues. 1 out of 5 isn't even close. Congratulations on the wins in the humanoid league.
For those that are curious, the other 4 leagues are simulation (focusing on team play and low barrier to entry), small size robot (hardware/software combined, wheeled robots), middle size robot (hardware/software combined, wheeled robots), and standard platform league (software only using real humanoid robots).
Survival is a terrible metric of intelligence. By that standard, lions and tigers and bears are the most intelligent species on the planet.
Forget lions and tigers or even insects and bacteria. By this metric, the water in the ocean, the sun we orbit around, and the vast expanses of space are far more intelligent than anything that has ever lived or any system we have put in place.
Let's suppose that the piece of junk has an injury radius of 50m, or 8000 sq m. There's 510 million sq m of surface area on the earth, so the chance that it will fall within 50m of me is 8,000::510,000,000 or 1 in 63,750. You're roughly 10x as likely to die in a car accident in any give year. That actually is much closer to significant than I thought it might be before I did the calculation. We do take steps to alter our risks of vehicular death: we use seat belts,we drive cautiously etc. But we don't take further steps, like installing a racing seat with a five point harness. Arguably, then, further marginal reductions in our death risk are not that meaningful to us, and we would very likely do nothing to reduce a risk that is already 1/10 as great as the point of diminishing returns for car risk.
Actually, it is much closer to significant than you thought because you made an error in the calculation. There's 510 million sq km of surface area on the earth, so the chance that it will fall within 50m of me is 8,000::510,000,000,000,000 or 1 in 63,750,000,000. You're roughly 10,000,000x as likely to die in a car accident in any give year.
Your calculations further down seem to have done the calculation correctly
The marketing date is a huge issue because 90% of the time, the people making the game have no buy-in regarding it. They're working towards being done when it's done, and then when they get told that they have six months when they need a year, things get implemented too fast and half-assed.
I think that's a poor excuse. The date needs to be agreed on between marketing and development well ahead of time. A rough date should probably be part of the initial game design where the game is being defined.
Even if marketing solely decides the date and doesn't give the date until 6 months out, development shouldn't just ignore that they need 1 year and stick their heads in the sand. At that point development needs to go back to marketing and try to work things out. If the date is fixed, then the game needs to be rescoped until it fits in the time available, even if it means removing features. It's better to have a slightly smaller, playable game then a larger, bug-ridden beast.
If both schedule and feature list are fixed, then it is managements fault for not manageing properly.
The game companies are killing their own market. I haven't bought many games since the games starting getting buggy and unbalanced on release. At $50 a game, I expect pretty near perfection. Lower the price and I'll be more lenient. Lately, I've been playing online games instead since once you find a good one you can play it for a long time and don't have to worry much about bugs anymore.
Why is anyone going to care about a weapon system everyone knows is a dud anyway? The system has never once demonstrated that it works, every single test has either failed outright or been rigged. The only reason the program exists at all is to hand out taxpayer money to campaign contributors.
Even if the system doesn't work, it's still bad to have unsecured access to it. It's certainly a concern that some terrorist might be able to infiltrate the system and use it to shoot down passenger planes or cause some other kind of damage. The system is designed to destroy things, incoming missles in this case, how hard would it be to redirect that destructive force to cause problems?
Contract companies love to say "that's not in the contract, but we'd be happy to renegotiate and do it for $x."
And for good reason. Same reason that when you order a Dodge Neon they don't ship you a Dodge Viper. The contract is what is specifying what the government is buying. Change what the government is buying to enhance it and it's not really surprising that they want more money to produce it. Taking the least expensive option is usually the right option for the company even if it isn't in the system. It also doesn't surprise me in the least. The government really should have some boilerplate in their contracts saying system much follow established DoD security procedures as specified in DoD Standard Security Policies v10.43 or something along those lines.
America's Army is more of a recruiting tool then propaganda. I consider it more like an ad.
Cool technology, but doesn't seem very practical.
With what they have so far, I don't see any reason not to just use a tracked vehicle. The sense of balance is impressive, but if you don't have all your weight mounted so high it becomes less of an issue.
640x480? What year is this again?
Besides, most laptops come with LCD screens that look like crap if you try to scale an image to fit instead of just rendering it the right size to begin with.
Every time a patent story comes up, people always point out that the developer needs to have his invention protected. However, a patent does more than just prevent people from copying the inventor's invention. A patent prevents anyone else from independently developing the invention. This is why the invention needs to be non-obvious, which sadly seems to have effectively dropped from the patent approval process. Who protects the independent inventor who happens to be working on the same idea and developing a similar device independently? It is at least as important to protect this independent inventor as it is to protect the patent filer.
The whole patent system seems pretty ill-conceived to me. I don't particularly have a better solution, but strong enforcement of the prior art and non-obvious criterion would be a good start. The problem in my mind is that the whole patent system is based on the premise that inventions are developed from scratch. This just isn't the way most progress is done. Almost all inventions are combinations or refinements of existing technology. Sometimes those combinations/refinements are a large leap, sometimes they are rather obvious next steps.
In the case of hybrid cars, it's a combination of existing internal combustion engine and electric engine technology. The ground breaking part isn't the combination so much as it's the design to make a practical system. It's obvious that the technologies could be combined, it's not that much of a leap to see that it could possibly improve gas mileage based on existing understanding of effeciencies, but it's very unclear how to put the whole thing together such that you actually end up with better gas mileage.
A patent is only as good as your ability to enforce it. I think you'd be hard pressed to perform any patent litigation for under $10000. Since no one who can't afford the fee can effectively enforce the patent, this seems like a moot argument.
I was extremely impressed in the ext's; I simply had no idea how consistently well performing they were. I'm not so sure about consistent performance. I've run in to the extreme slow down caused by many files in the same directory many times and it can be pretty painful. I'm not sure if this has been addressed at all, but from my understanding it is a limitation of the way the file system data is stored on disk with extra levels of indirection being required. With Reiser, I can have directories with 10,000s files with no problems.
C++ allows me to express this: No upcast required because I can tell the compiler that I'm only going to store strings in my vector rather than always having to store an Object. This also makes Java harder to read because when looking at a container, it is impossible to tell from the variable declaration what type of object will be stored in the container.
What are you talking about?
I tried one that a friend had. It has to be the *MOST* irritating keyboard I have ever tried. I have big fingers and found it almost impossible to type more than a word without hitting 2 keys instead of 1 on some letter or other.
It always amuses me to see "oh, by blind, senile grandmother coulda done this" comments. The point is, they didn't.
If all these things are as obvious as people like to claim, why don't they patent them? Is it that maybe they were only obvious after someone stated them?
Perhaps because they didn't think they were worthy of a patent?
On second thought, maybe I agree with you. I been meaning to send in my patent application for a method for temporarily storing water in an accessible form using a cylindrical device with one end capped. I'm thinking of calling it a "cup". You interested in licensing it?
I recommend going to www.vote-smart.org for information about politicians. It's run by a bunch of volunteers and they refuse all money from anyone except private individuals. The have voting records, statements made, and the opinions of all of the position groups for every politician and candidate.
UNSW (Uni of New South Wales, Aust). Always is. They have won every game they have competed in, with an average lead of 30 goals (no joke).
Actually, UNSW lost games to both Carnegie Mellon University and LRP (Paris) their first year. Also, they have never won a game by more than 14 goals and the games were all close their first year. I'd say the difference is that UNSW spends a lot of time tuning there systems for the task at hand. Many of the other universities spend more time doing basic research, which won't make you play better soccer this year, but might a few years down the road. If you measure success in terms of published papers, then UNSW is behind many of the other universities. That being said, their system is impressive. Hopefully, this year we (I'm from Carnegie Mellon University) can give them a bit more of a challenge. The gap certainly closed between 2000 and 2001. It should be exciting to see what happens this year.
OK, are these robots REAL robots (ie. pre-programmed automatons) or the lame "robot wars" robots (ie. glorified RC cars)?
These are real robots. The robots are completely controlled by computers in all leagues. Human intervention is limited to starting/stoping the robots and enforcing rules. All decisions during play are made by the robot players themselves.
I'm on the University of Virginia team, and we're one of two teams to qualify representing the US for the simulation league (which doesn't use real robots, and is thus a lot more fast paced). The other team being AT&T Research Labs.
Just a small clarification. The parent is referring to the simulator league. There are also several US teams entering in other leagues. For example, Carnegie Mellon University is entering in the small-size league and the Sony legged league.
I was at the RoboCup98 in paris, it will be interesting to see how things have changed. The match that I saw involved a few brick looking robots moving around in a random manner. It was the first time that I saw the Sony Dogs... Do the robots now wear boots?
The robots now look much slicker. They also perform a lot better. The teams no longer look random at all. They are certainly capable of scoring. Last year saw many scores of around 10-1. The robots still go barefoot.
How does one compete in this tournament? It looks like all the teams have identical equipment: do the companies (like Sony) sponsor them?
There are several different leagues that one can enter. There is a simulator league, small-size league, middle-size league, Sony legged league, a humanoid league (new for this year), and a couple rescue leagues.
The simulator league has no hardware. The software is downloadable from sourceforge. This is the easiest league to get started in since there isn't a lot of money involved and you can even enter remotely. Keep in mind that a lot of researchers are working hard on this problem, so most of the teams are really good. The Sony legged league is run by Sony. Sony provides a lot of support. You have to make a proposal to Sony to get into this league. A committee of researchers and Sony personel decide which teams to allow in. Everyone uses the Sony hardware in this league. The small-size and middle-size leagues have hardware built by the teams. It usually costs at least $10,000 for all the hardware needed for a small-size team. Middle-size teams run from $3,000-$40,000 per robot (4 on 4 competition). See www.robocup.org for more details. Registration is already over for this year, but there will be another competition next year (in Italy, I believe).
Has been able to win against the best chess player. So if I understand correctly, the logic involved to win a game isn't really the problem, it is more of a mechanical problem, isn't it? Recent video games are already able to beat really good video game player, and there are not even super-killer-incredibly-intelligent applications.
There are plenty of software problems. It turns out that chess is actually easy compared to the stuff we do everyday like walking. It's just that we have a lot of specialized wet-ware for the normal tasks, so we don't notice how difficult they are. Control and perception are major problems. Planning and coordination are also major problems. Chess is easier because things don't move while you are thinking and you don't have to pick up the pieces if you do it like Deep Blue did. The video games are usually easier for the computer than the real world because they have perfect information about what is going on in a convenient format for them. They almost always have a huge interface advantage as well. The humans attention is usually divided and forced to operate through a relatively clunky interface where the computer doesn't mind keeping track of 80 things at one time. The fact that the human ever wins at the hardest levels shows how stupid the computers are.
The only reason they are staging this competition is that combat seems to be the only way that men express themselves or advance their knowledge. Isn't it enough that a robot can track, intercept and guide an object without having to turn it into a "battle" complete with winners and losers?
As a graduate student who has been working on robots in RoboCup for 4 years, I think you are missing the point. AI made many discoveries and developed many algorithms during the '70s by focussing on the challenge of making computers play chess. At the time, it was believed that if the problem of playing chess was solved we would be able to do almost anything we wanted with robots. Over time, it became very clear to roboticists that being able to play chess was not helping us build better robots. The RoboCup initiative was started in order to provide a new grand challenge for roboticists to work on and provide a common framework for comparisons. Soccer was chosen because it exemplifies many of the things that robots are traditionally bad at. It is an environment with multiple robots cooperating (and yes competing), lots of dynamic activity, difficult sensing problems, and challenging motor control problems. Basically, everything robots are bad at. Soccer was chosen over other alternatives simply because it is more popular around the world which in turn makes it easier to get funding. All of the research from soccer applies to many other domains that people really care about. The RoboCup organization has also recognized the need to do other things than soccer and also run a search and rescue domain at the competition each year which is equal in stature to all of the other leagues. The search and rescue problem is to find people trapped after natural or other disasters. This was started a couple of years ago.
And they use the term "PMPO" - "Peak Music Power Output". Fine, putting aside the fact that this term has no accepted definition in electrical engineering - let's say that those little Taiwanese-made speakers contain an amplifier with a big bank of capacitors to dump out enough current to achieve 250 watts peak. If the power supply to them is only 9V, the capacitors would never get above 9V. If the speakers themselves have a standard nominal impedance of 8 ohms, then we can calculate
Not to say they aren't lying, they are, but capacitors can be charged to a higher voltage than the source using a voltage doubler circuit or a flyback voltage multiplier. Doesn't give you any more power but does trade current for voltage. See this page for some example circuits.