If Mr. Jebara thinks that "within the next five to 10 years the system will be better than the world's best player," he needs to review his high-school physics.
It isn't just the angles at which the balls collide -- it's the how far off center the balls are when they collide, what kind of spin is on the cue ball when it begins the chain reaction, the mass of each of the balls, the felt table's coefficient of friction,.... To compensate for all of these variables will require many more (and varied) sensors than just a camera.
That's the what makes pool different from chess -- in chess, the friction of the pieces against the board doesn't matter. In many respects, pool is the ultimate game of applied physics. But all of the variables combine to make the game as much art as it is science.
Still, it is a pretty cool thing, and will be great for beginners to learn the basics.
For many of these systems, a VATS chip isn't even required -- installing the proper resistor where needed will do the trick. I installed one of these a couple of years ago, and it included directions on how to determine which resistor was correct and how to install it. The "key" here, as mentioned above, is that the car must receive the correct change in current, not some magic encryption key.
During my last tour in the Marine Corps, I worked with some Marines from the Marine Corps Modeling and Simulation Management Office (this was about two years ago). They were evaluating Close Combat III, and were talking to the developers about partnering to create a more realistic version. The Marines would privide advice about tactics, actual weapon performance under various conditions, etc. The developers would provide a commercial-quality training aid to the Corps.
I tried out the version of Close Combat they were starting from. As someone who spent time not only learning the basics of infantry tactics, but actually teaching those tactics to young officers, I can honjestly say that properly used, these games can be invaluable. It is not the shooting that needs to be taught, it is the ability to plan an attack and make decisions in the absence of perfect information about the enemy's location that is difficult to teach. Normally, this is taught with sand tables exercises, and then progresses to field exercises. For a major field exercise, the actual troops are often just training aids for the staff. It costs time and money to organize these exercises. But with a good simulation, I can do a week's worth of field training in an afternoon.
Even more important is the ability to play these games in multiplayer mode. When a more senior officer plays the part of the enemy, you now have something infinetely better than the finest AI system -- the human mind. What makes tactics difficult is this clash of independent wills. No matter how many times two people play against each other, it is unlikely either will ever be able to predict the other's actions.
Games like Doom, Soldier of Fortune, etc. are a lot of fun, and I play them regularly. But they bear no resemblance to reality. On the other hand, I found Close Combat III relatively realistic and challenging, while at the same time playable. I think the direction the Army and Marine Corps are going has a lot to commend it. It provides better training to our troops (which equates to fewer friendly casualties during war, as well as less collateral damage). And it saves a lot of money for the taxpayer. For the cost of a single, one-week, comany-level field exercise involving a single scenario, I could proably equip a computer lab with enough machines and software to train the company staff on a dozen different scenarios during that same week. And those resources can be re-used for every company in the battalion (and more).
The taxpayers save money, the troops get good training, and the gaming community gets more realistic games. Everybody wins!
Think about the time scales involved here. The Earth has existed for at least 5 billion years. There has been animal life for at least 100 million years. Ice Ages come and go in cycles of around 20,000 years. For around 5000 or so years, we humans have been able to record history. For around one hundred years, we have been able to take accurate (by modern standards) temperature readings.
So, we are basing global climate change forecasts on a sample that represents 2% of recorded history, and less than 2/3 of 1% of the periodicity of Ice Ages. Am I the only one here who thinks this sample is just a bit too small to make accurate predictions?
Not only do private pilots use them, but they are a standard part of military flight training. They are not conventional slide rules, but rather circular slide rules. The principle is the same, only the shape is different. Despite advances in electronics, I find mine to be faster and easier than a normal calculator for fuel computations.
I am currently a Marine Reservist working in the IT field. I left active duty in July. Since then, I have met several reservists that work in IT. To a person, their employers do not have a problem with it. There are a lot of potential benefits to the employer:
There are plenty of jobs in the Corps that are directly IT related. Talk to the recruiter about getting a guaranteed MOS (Military Occupational Specialty) assignment before enlisting. If he says that can't be done, tell him to go pound sand. It can be done, and he needs your body to fill his quota. See if you can get assigned to MOS 4066 - the title is something like Computer Systems Technician. Don't let the title fool you -- you will do everything from installing word-processing software to splicing fiber-optic cables. The Corps rarely lets a formal job description get in the way of assigning duties. If you can do the work, the work will be assigned to you. If not, you will be taught. You will gain valuable experience on real-world systems.
The Corps will send you to school. You can get formal training above and beyond the minimum required -- and it won't cost your employer a dime.
There are a lot of Reserve billets that allow a much more flexible schedule than the standard one-weekend-a-month, two-weeks-a-year. My slot allows me to break that up and rearrange it to better mesh my civilian schedule with my military schedule. It all depends on what is available and what you want to do. If you want to be an infantryman, you will have to drill with the rest of your unit. I work for an office-bound unit (11 years of field service, with the infantry and various helicopter squadrons -- I earned some time indoors), so I can tailor my schedule better.
There are plenty of opportunities, many of which can directly pay off for your employer. What is available depends on what you want to do and where you live. It is at least worth looking into.
If you would like some detailed info, especially in the Northern Virginia area, go ahead and drop me a line, I'd be happy to answer any questions (no, I am not a recruiter): mcdowall@crosslink.net
The problem is, the deep ocean presents more difficult problems than the vacuum of space. Among them:
Pressure. On the surface of the Earth, it is about 15 psi. Going into orbit requires a pressure vessel that doesn't need to maintain a differential any more than that (15psi vs 0 psi). In the water, pressure increases by 1 atmosphere for every 33 feet of depth. Since the average depth of the ocean is 4,000 meters, any sub-surface vehicle suited for deep-sea research must maintain a pressure differential on the order of 5500 psi at 4,000 meters. That is one heck of an engineering problem.
Coming home. For space exploration, we have largely solved the problem of re-entry into the atmosphere. But the same pressure differentials mentioned above present problems returning to the surface from the deep ocean. If subjected to more than 3 atmospheres of pressure, humans are limited in the amount of time they can spend in that environment before requiring decompression. In addition, submarines can only ascend or descend so fast before the rate of change of pressure begins having adverse effects on the vehicle. In submarines, pressure vessel failures tend to be catastrophic and fatal (ever hear of USS Thresher?).
There are other problems, but those are two of the biggest. Solving them requires a lot of money, time, and energy (undersea propulsion being another big challenge). It's not as easy as it first seems.
Before I comment, allow me to point out my qualifications for commenting. I spent 11 years in the Marine Corps, where I served tours as a helicopter pilot (including one tour as an instructor), a Forward Air Controller, and as the Deputy Director of the Office of Science and Innovation at the Marine Corps Combat Development Command (where I attended many briefings on UCAVs).
The technology is nowhere near ready enough to be used for large-scale air-to-air combat. Too much of a pilot's situational awareness comes from peripheral vision and and minute visual cues gleaned from another aircraft's movements. Limiting the controller to a view that is essentially what can be displayed on a single monitor will make said controller less than effective -- it will make the UCAV a sitting duck. And so far, there are no algorithms that can out-think a person directly controlling an aircraft in the dynamic environment of air-to-air combat.
The primary mission area being evaluated for possible UCAV use is air-to-ground combat -- Close Air Support (CAS) and Deep Air Support (DAS). The smart money favors deployment for DAS long before the technology is ready to support CAS. DAS is where there are no friendlies in the area, and there is time to pick and choose targets so as to avoid collateral damage and civilian casualties. But CAS is another story entirely. By definition, CAS is carried out where close coordination with nearby ground troops is necessary. And despite what may be on the drawing boards, the average rifleman is still not carrying a notebook computer connected to a frequency-hopping radio. CAS coordination is done by a Forward Air Controller (FAC) talking to the pilot on a radio. The FAC must mark either the target or his own position. This is generally done with either some form of smoke or a signal light. Sometimes, it is done with simple mirror flashes. Then, the FAC "talks the pilot on" to the target. The kind of coordination necessary to avoid bombing one's own troops accidentally requires (for now) a live pilot in the cockpit of the bomber. Even with a live pilot in the cockpit, fully briefed on the entire area, accidents happen. Even when pilots are very familiar with the training range, and have flown the target area dozens of times, there are slip-ups (ever hear of Vieques?).
All of which makes this a fine academic discussion, but leads me to believe we will still have pilots in the cockpit of most combat aircraft for the foreseeable future. The one area that seems wide open for UCAVs is DAS.
You have hit the nail on the head. I just finished a tour doing technology search for the Marine Corps, and the military is looking VERY closely at fuel cells. The fact is, they are not yet efficient enough. When it comes to a power/weight ratio, you still can't beat an internal combustion engine.
Slashdot Mirror
If Mr. Jebara thinks that "within the next five to 10 years the system will be better than the world's best player," he needs to review his high-school physics.
It isn't just the angles at which the balls collide -- it's the how far off center the balls are when they collide, what kind of spin is on the cue ball when it begins the chain reaction, the mass of each of the balls, the felt table's coefficient of friction,.... To compensate for all of these variables will require many more (and varied) sensors than just a camera.
That's the what makes pool different from chess -- in chess, the friction of the pieces against the board doesn't matter. In many respects, pool is the ultimate game of applied physics. But all of the variables combine to make the game as much art as it is science.
Still, it is a pretty cool thing, and will be great for beginners to learn the basics.
For many of these systems, a VATS chip isn't even required -- installing the proper resistor where needed will do the trick. I installed one of these a couple of years ago, and it included directions on how to determine which resistor was correct and how to install it. The "key" here, as mentioned above, is that the car must receive the correct change in current, not some magic encryption key.
During my last tour in the Marine Corps, I worked with some Marines from the Marine Corps Modeling and Simulation Management Office (this was about two years ago). They were evaluating Close Combat III, and were talking to the developers about partnering to create a more realistic version. The Marines would privide advice about tactics, actual weapon performance under various conditions, etc. The developers would provide a commercial-quality training aid to the Corps.
I tried out the version of Close Combat they were starting from. As someone who spent time not only learning the basics of infantry tactics, but actually teaching those tactics to young officers, I can honjestly say that properly used, these games can be invaluable. It is not the shooting that needs to be taught, it is the ability to plan an attack and make decisions in the absence of perfect information about the enemy's location that is difficult to teach. Normally, this is taught with sand tables exercises, and then progresses to field exercises. For a major field exercise, the actual troops are often just training aids for the staff. It costs time and money to organize these exercises. But with a good simulation, I can do a week's worth of field training in an afternoon.
Even more important is the ability to play these games in multiplayer mode. When a more senior officer plays the part of the enemy, you now have something infinetely better than the finest AI system -- the human mind. What makes tactics difficult is this clash of independent wills. No matter how many times two people play against each other, it is unlikely either will ever be able to predict the other's actions.
Games like Doom, Soldier of Fortune, etc. are a lot of fun, and I play them regularly. But they bear no resemblance to reality. On the other hand, I found Close Combat III relatively realistic and challenging, while at the same time playable. I think the direction the Army and Marine Corps are going has a lot to commend it. It provides better training to our troops (which equates to fewer friendly casualties during war, as well as less collateral damage). And it saves a lot of money for the taxpayer. For the cost of a single, one-week, comany-level field exercise involving a single scenario, I could proably equip a computer lab with enough machines and software to train the company staff on a dozen different scenarios during that same week. And those resources can be re-used for every company in the battalion (and more).
The taxpayers save money, the troops get good training, and the gaming community gets more realistic games. Everybody wins!
Think about the time scales involved here. The Earth has existed for at least 5 billion years. There has been animal life for at least 100 million years. Ice Ages come and go in cycles of around 20,000 years. For around 5000 or so years, we humans have been able to record history. For around one hundred years, we have been able to take accurate (by modern standards) temperature readings.
So, we are basing global climate change forecasts on a sample that represents 2% of recorded history, and less than 2/3 of 1% of the periodicity of Ice Ages. Am I the only one here who thinks this sample is just a bit too small to make accurate predictions?
Not only do private pilots use them, but they are a standard part of military flight training. They are not conventional slide rules, but rather circular slide rules. The principle is the same, only the shape is different. Despite advances in electronics, I find mine to be faster and easier than a normal calculator for fuel computations.
I am currently a Marine Reservist working in the IT field. I left active duty in July. Since then, I have met several reservists that work in IT. To a person, their employers do not have a problem with it. There are a lot of potential benefits to the employer:
There are plenty of jobs in the Corps that are directly IT related. Talk to the recruiter about getting a guaranteed MOS (Military Occupational Specialty) assignment before enlisting. If he says that can't be done, tell him to go pound sand. It can be done, and he needs your body to fill his quota. See if you can get assigned to MOS 4066 - the title is something like Computer Systems Technician. Don't let the title fool you -- you will do everything from installing word-processing software to splicing fiber-optic cables. The Corps rarely lets a formal job description get in the way of assigning duties. If you can do the work, the work will be assigned to you. If not, you will be taught. You will gain valuable experience on real-world systems.
The Corps will send you to school. You can get formal training above and beyond the minimum required -- and it won't cost your employer a dime.
There are a lot of Reserve billets that allow a much more flexible schedule than the standard one-weekend-a-month, two-weeks-a-year. My slot allows me to break that up and rearrange it to better mesh my civilian schedule with my military schedule. It all depends on what is available and what you want to do. If you want to be an infantryman, you will have to drill with the rest of your unit. I work for an office-bound unit (11 years of field service, with the infantry and various helicopter squadrons -- I earned some time indoors), so I can tailor my schedule better.
There are plenty of opportunities, many of which can directly pay off for your employer. What is available depends on what you want to do and where you live. It is at least worth looking into.
If you would like some detailed info, especially in the Northern Virginia area, go ahead and drop me a line, I'd be happy to answer any questions (no, I am not a recruiter): mcdowall@crosslink.net
Pressure. On the surface of the Earth, it is about 15 psi. Going into orbit requires a pressure vessel that doesn't need to maintain a differential any more than that (15psi vs 0 psi). In the water, pressure increases by 1 atmosphere for every 33 feet of depth. Since the average depth of the ocean is 4,000 meters, any sub-surface vehicle suited for deep-sea research must maintain a pressure differential on the order of 5500 psi at 4,000 meters. That is one heck of an engineering problem.
Coming home. For space exploration, we have largely solved the problem of re-entry into the atmosphere. But the same pressure differentials mentioned above present problems returning to the surface from the deep ocean. If subjected to more than 3 atmospheres of pressure, humans are limited in the amount of time they can spend in that environment before requiring decompression. In addition, submarines can only ascend or descend so fast before the rate of change of pressure begins having adverse effects on the vehicle. In submarines, pressure vessel failures tend to be catastrophic and fatal (ever hear of USS Thresher?).
There are other problems, but those are two of the biggest. Solving them requires a lot of money, time, and energy (undersea propulsion being another big challenge). It's not as easy as it first seems.
The technology is nowhere near ready enough to be used for large-scale air-to-air combat. Too much of a pilot's situational awareness comes from peripheral vision and and minute visual cues gleaned from another aircraft's movements. Limiting the controller to a view that is essentially what can be displayed on a single monitor will make said controller less than effective -- it will make the UCAV a sitting duck. And so far, there are no algorithms that can out-think a person directly controlling an aircraft in the dynamic environment of air-to-air combat.
The primary mission area being evaluated for possible UCAV use is air-to-ground combat -- Close Air Support (CAS) and Deep Air Support (DAS). The smart money favors deployment for DAS long before the technology is ready to support CAS. DAS is where there are no friendlies in the area, and there is time to pick and choose targets so as to avoid collateral damage and civilian casualties. But CAS is another story entirely. By definition, CAS is carried out where close coordination with nearby ground troops is necessary. And despite what may be on the drawing boards, the average rifleman is still not carrying a notebook computer connected to a frequency-hopping radio. CAS coordination is done by a Forward Air Controller (FAC) talking to the pilot on a radio. The FAC must mark either the target or his own position. This is generally done with either some form of smoke or a signal light. Sometimes, it is done with simple mirror flashes. Then, the FAC "talks the pilot on" to the target. The kind of coordination necessary to avoid bombing one's own troops accidentally requires (for now) a live pilot in the cockpit of the bomber. Even with a live pilot in the cockpit, fully briefed on the entire area, accidents happen. Even when pilots are very familiar with the training range, and have flown the target area dozens of times, there are slip-ups (ever hear of Vieques?).
All of which makes this a fine academic discussion, but leads me to believe we will still have pilots in the cockpit of most combat aircraft for the foreseeable future. The one area that seems wide open for UCAVs is DAS.
You have hit the nail on the head. I just finished a tour doing technology search for the Marine Corps, and the military is looking VERY closely at fuel cells. The fact is, they are not yet efficient enough. When it comes to a power/weight ratio, you still can't beat an internal combustion engine.