Do you have a bit of a complex? You do realize that Ford owns a fairly significant chunk of Mazda, right? I seriously doubt you getting cut off had anything to do with not driving an American car...
I don't have a complex, but you seem to. And seemingly you've never visited Detroit.
I'm fully aware that Mazda is owned mostly by Ford. And that my Mazda Tribute was made in the same American plant as the Ford Escape (great vehicles btw). But obviously people in Detroit don't seem to make that distinction. (Nor does the Mazda dealer as I had to pay import duty that I wouldn't of had to pay if I bought an Escape.)
I'm just recounting what happened to me when I was there. And I wasn't just cut off - the guy tried to push me into the shoulder when we were the only two cars in the vicinity on a three lane highway. I was warned about the consequences of renting a foreign car there before I went by several people.
Even if my experience was a coincidence, my friend in her Toyota got egged while waiting at a red light - that was deliberate.
That's funny, because even Australians know that spades and shovels are two different tools. A spade has a flat blade. A shovel is more like a scoop.
And. here I thought a spade was a suit in a deck of cards, and a shovel is what you beat your opponent over the head with when you discover him cheating.
With a car I can pull out the engine and sell it. Then I can insert a new different engine, make some modifications then sell the entire new car.
If I truly owned the software outright, I should be able to do the same thing: rip out the graphics engine, sell it, then insert my own engine and resell the game.
AFAIK it's not legal to build an exact copy of your car, either (copyrights, patents get in the way
No, its perfectly legal for me to buy all of the pieces of a car, assemble it then resell it. I just have to disclose that is what I did (I can't pretend there's a warranty for example).
Japanese consumers in many markets prefer Japanese brands.
Just as many Americans prefer to buy American brands. It's natural to be a bit nationalistic, although some people take it too far. Try driving a foreign car in Detroit. I've been deliberately cut off on the highway in a Mazda ( they tried to run me into a ditch) and a friend of mine had eggs thrown at her while she was in her Toyota.
Unless the x360 changed that, console games are NOT licensed, they are bought. There is no EULA in console games and the only contract applying to them is the sales contract.
Really? If that was true then I would be free to reverse-engineer and modify the game as I see fit. Much like a car that I buy then can modify and resell all or portions thereof.
Our rent is outrageous -- almost $18-$25 per square foot. I fully blame this on the Fed's easy credit and a massive amount of "new" money entering the property market. There is no way my main store's rent should be $5000 per month -- but it is! This is in a town of 4000 people, mind you.
That's insane (and hard to believe). Here in Toronto, population 5 million, retail rates are on the order of a dollar or two per square foot. Check here: http://toronto.craigslist.org/off/
You're arguing a different point. Did you miss the part where he said solenoid? No motor, no gears. High speed, high force. The downside? High energy cost.
Could be electric solenoid, but again they are not high force, and difficult to control (could explain the jerky movements some people saw in the video). Hydraulic would give you the needed force (and more) but I see no bulky connections for hydraulics.
I agree. However if you are depending on a high contrast glove then that is the same effect as a marker. You are containing the environment and nothing else in the scene can be the same colour as the glove. This is no longer a general scene.
Also if you wanted to know accurate depth, then you have to know the exact size of the glove to do a model match.
think it would be more like 10-20 fps
In a constrained scene as described probably, not in a general scene.
Really. I can easily think of several sites (primarily focused to developers) that used email to notify of new content. Saved going to their site and searching for the new stuff - just read the email when you had time to do so.
imagine a completely wireless world filled with all sorts of small mobile devices
You've just described sheer hell. Can't even watch a friggin movie now without somebodys wireless device going off. Even more of these things would be a nightmare.
Cool. So it is using a known size high contrast target (black and white), essentially a highly constrained environment. This approach does not easily extend to a general scene with unknown targets such as would be required for boxing. That situation is much harder to deal with and slower (like 1-5 fps on decent hardware) unless as I say you use some kind of optical target on the gloves.
I have some expirience in 3d tracking and I don't agree. My Nokia 6600 phone with 15 frame/sec camera and 109mhz CPU (without FPU) can detect 3d position of rectangular marker relativly to camera and make some 3d rendering with 8 frame/sec speed
So how exactly are you getting 3D information from a single camera? Are you assuming a known size of the marker or are you comparing successive frames? what is the accuracy of your depth resolution?
A camera is probably not the best way to do this unless you put some retro reflective matters on the gloves to make the image processing easier. I agree with the other poster, some sort of proximity sensor or light curtain would be easier to implement.
Canada has some of the worst anti-freedom laws: these laws are completely manipulated by large groups (unions, corporations, military, etc).
What a load of crap. Canada has much less restrictive personal freedom laws than the U.S. Possession of marijuana for personal use is legal here, as is copying music / video / software for personal use, nude beaches, women topless in public, even swinging is now legal. As a Canadian citizen I have many more freedoms than someone in the U.S (whether you agree with these freedoms or not is irrelevant). My government doesn't make illegal wiretaps of me like the U.S. government was recently caught doing of its citizens. For Gods sake blowjobs are even against the law in some U.S. states. I challenge you to name one significant freedom that Americans enjoy that is illegal here in Canada.
Actually, making a profit is the ONLY way you know you are helping people properly. Profit means you are offering someone a product they want at a price they want, so both parties are gaining something.
Another complete load of crap! Take an economics course at some point and educate yourself how the world really works.
Less regulation = safer, cheaper products. More regulation = monopolistic corporations that can take advantage.
What a naive blanket statement. So deregulation of for example the nuclear industry so they can provide cheaper electricity by any means possible would be a good thing? Get a grip.
However lots of people still work in the city and it ends up being a tradeoff of bigger properties and long commutes vs. smaller properties or condos and a short commute.
Here in Toronto, if you work downtown and live in a suburb, you are probably dealing with a 1.5 hour average commute each way if you try and drive in. I also know people who live and work downtown their entire lives and have never bothered to get a drivers license because they've never needed one.
I'm actually in the opposite situation, and part of a growing trend (in Canada at least). I live downtown (in a detached house, ~3000 sqft) so I can take advantage of the amenities the city has to offer, but I commute to work in the suburbs. The commute for me is a lot shorter than someone trying to get into the city because the traffic is a lot lighter going out.
Computers can calculate faster, no doubt, but we far excel computers as dealing with a changing environment. Especially when that environment is uncontrolled.
There were very few camera based vision systems in DARPA (and the ones that were there performed badly) simply because it is extremely difficult to pick useful information out of an image. It is hard enough trying to extract where the road is and if there are any obstacles on it, let alone detect and deal with unexpected happenings. Laser systems perform better because it simplifies the information into two quantities; an obstacle, or not an obstacle.
Even a robot catching a ball is on the forefront of robotic research. This is simple to any of us over about 5 years of age but to a robot / vision system this is highly complex. Success has only been achieved in highly controlled and restricted scenarios - basically no unknowns.
So yes, computers have fast response times but it is practically irrelevant in this application for dealing with unexpected situations.
If the engine is off you are running off of just the battery (belts run the alternator, alternator only runs when engine runs). So the engine would HAVE to be on.
Huh? Why can't the game run off of battery alone?
I am not too sure how good it is on your brakes to continualy pump them
Car is entirely drive by wire. Pumping the brake pedal while the engine is off doesn't do anything to the physical brakes.
I suspect the biggest obstacle to comfortably using exoskeletons is responsiveness. If you want to move your hand, you just think about it and it takes a few milliseconds to move. With an exoskeleton, you have to hit the sensors (perhaps past their critical point), and the hydraulics/whatever has to kick in and move it. How long does that take?
You're talking about the response time here of the system, and yes on a big system it gets to be an issue. Rule of thumb is to keep the response time to under 100ms worst case which is usually doable.
Another issue is not only removing the apparent weight of what you are holding but also the inertia. Removing the weight in a control loop just requires good velocity sensors which are commonplace, but you still feel the weight when you change velocity. Removing the inertia means you have how measure the acceleration very accurately so you either get the acceleration directly from the sensor (high precision = expensive) or take the derivative of the velocity sensor which introduces delay and noise. Stable inertial compensation is not trivial.
Slashdot Mirror
I don't have a complex, but you seem to. And seemingly you've never visited Detroit.
I'm fully aware that Mazda is owned mostly by Ford. And that my Mazda Tribute was made in the same American plant as the Ford Escape (great vehicles btw). But obviously people in Detroit don't seem to make that distinction. (Nor does the Mazda dealer as I had to pay import duty that I wouldn't of had to pay if I bought an Escape.)
I'm just recounting what happened to me when I was there. And I wasn't just cut off - the guy tried to push me into the shoulder when we were the only two cars in the vicinity on a three lane highway. I was warned about the consequences of renting a foreign car there before I went by several people.
Even if my experience was a coincidence, my friend in her Toyota got egged while waiting at a red light - that was deliberate.
Dinosaurs are not in the original movie but as far as I recall they are in the book.
With a car I can pull out the engine and sell it. Then I can insert a new different engine, make some modifications then sell the entire new car.
If I truly owned the software outright, I should be able to do the same thing: rip out the graphics engine, sell it, then insert my own engine and resell the game.
No, its perfectly legal for me to buy all of the pieces of a car, assemble it then resell it. I just have to disclose that is what I did (I can't pretend there's a warranty for example).I doubt people will learn. We've had the automobile for about a hundred years now and most people still don't know how to drive them properly.
A single camera has no depth perception. A red glove held several feet in front of a red sheet will not be detected as a glove.
I agree. However if you are depending on a high contrast glove then that is the same effect as a marker. You are containing the environment and nothing else in the scene can be the same colour as the glove. This is no longer a general scene.
Also if you wanted to know accurate depth, then you have to know the exact size of the glove to do a model match.
In a constrained scene as described probably, not in a general scene.Really. I can easily think of several sites (primarily focused to developers) that used email to notify of new content. Saved going to their site and searching for the new stuff - just read the email when you had time to do so.
You've just described sheer hell. Can't even watch a friggin movie now without somebodys wireless device going off. Even more of these things would be a nightmare.
Cool. So it is using a known size high contrast target (black and white), essentially a highly constrained environment. This approach does not easily extend to a general scene with unknown targets such as would be required for boxing. That situation is much harder to deal with and slower (like 1-5 fps on decent hardware) unless as I say you use some kind of optical target on the gloves.
So how exactly are you getting 3D information from a single camera? Are you assuming a known size of the marker or are you comparing successive frames? what is the accuracy of your depth resolution?
A camera is probably not the best way to do this unless you put some retro reflective matters on the gloves to make the image processing easier. I agree with the other poster, some sort of proximity sensor or light curtain would be easier to implement.
For that configuration of robot, either it moves real slow, or if they've geared it to move quickly then it has no force behind it.
You simply can't have high force and high speed that matches what a human can do in that size of a robot with motors of today.
This article isn't a dupe; its a followup.
However lots of people still work in the city and it ends up being a tradeoff of bigger properties and long commutes vs. smaller properties or condos and a short commute.
Here in Toronto, if you work downtown and live in a suburb, you are probably dealing with a 1.5 hour average commute each way if you try and drive in. I also know people who live and work downtown their entire lives and have never bothered to get a drivers license because they've never needed one.
I'm actually in the opposite situation, and part of a growing trend (in Canada at least). I live downtown (in a detached house, ~3000 sqft) so I can take advantage of the amenities the city has to offer, but I commute to work in the suburbs. The commute for me is a lot shorter than someone trying to get into the city because the traffic is a lot lighter going out.
Sure, blame the only party that isn't driving the country to hell.
Computers can calculate faster, no doubt, but we far excel computers as dealing with a changing environment. Especially when that environment is uncontrolled.
There were very few camera based vision systems in DARPA (and the ones that were there performed badly) simply because it is extremely difficult to pick useful information out of an image. It is hard enough trying to extract where the road is and if there are any obstacles on it, let alone detect and deal with unexpected happenings. Laser systems perform better because it simplifies the information into two quantities; an obstacle, or not an obstacle.
Even a robot catching a ball is on the forefront of robotic research. This is simple to any of us over about 5 years of age but to a robot / vision system this is highly complex. Success has only been achieved in highly controlled and restricted scenarios - basically no unknowns.
So yes, computers have fast response times but it is practically irrelevant in this application for dealing with unexpected situations.
You previously stated that you rejected Darwinism, but now you state you accept evolution as a fact.
Stop screwing around and wasting everyones time.
You're talking about the response time here of the system, and yes on a big system it gets to be an issue. Rule of thumb is to keep the response time to under 100ms worst case which is usually doable.
Another issue is not only removing the apparent weight of what you are holding but also the inertia. Removing the weight in a control loop just requires good velocity sensors which are commonplace, but you still feel the weight when you change velocity. Removing the inertia means you have how measure the acceleration very accurately so you either get the acceleration directly from the sensor (high precision = expensive) or take the derivative of the velocity sensor which introduces delay and noise. Stable inertial compensation is not trivial.