I don't think that is right. I think gaussian has the wrong model for the human visual system for scaling and tends to result in images that are too blury. Gaussian might be optimal if you had no other information, but in this case we have a lot of special information about what is expected. For a start, pixels are square (particularly on LCDs), but gaussians are circular. bessel is just as much work as gaussian, but seems to be prefered, e.g.
No idea why you got a funny, but gaussian isn't really the best choice. There are standard libraries that can do the right thing anyway (libmagick, libgdkpixbuf, etc) so perhaps you should just grab one of them and use that.
You assume too much. My parents paid for my training. And I would never have got an advertising career! I think you are just jealous. There are scholarships for academic types too, and probably a comparable number, or more (for example, I got some scholarships for my various degrees by being good at stuff). The entrance requirements for a scholarship to the AIS are much more harsh than getting through year 12. And most people going through never make it big time.
Anyway, HECS is a very reasonable scheme. I also have a $13k hecs debt, but it's a very low interest loan (I think it is just indexed) and when I get a 'real job' I'll be able to pay it off in no time. Much better than systems used elsewhere - at least our system makes the student pay, not the parents! (Not that the students seem to understand that they're paying, considering the few that turned up to lectures)
I'm not sure whether you think the AIS is good or bad from your post. Yes, we have a goverment sponsored research institution into sport. But I don't see this as very different to say the US model with universities having a big emphasis on sports, or having corporate sponsorship of good athletes in certain games. The Australian model is clearly better bang for buck, and so if I'm going to pay for sports research I'd rather do it as efficiently as possible.
The Americans have a state subsidized dept. of defense which has no other goal but to "make our guys win".
I do think Australians are a little parochial about sport, and I do wish that more australians would play than watch on TV. Australians are on average, quite unhealthy. I also wish more money would be spent on other research, but perhaps not at the expense of sports research. (Less money on defence instead?) I think the drive for being good at sport is perhaps a little bit of arrogance, we like to think we're better than average, but in fact we're pretty much on target for an economy of our size.
Disclosure: I trained at the AIS, and my nephew is currently training for the australian swim team.
diamond has a band gap of 6.4eV, which, if my physics is correct, gives an LED with an emission frequency of E = hf, You have: c / (6.4eV/h ) You want: nm
* 193.72529 so assuming we can solve the materials engineering, perhaps UV sterilisation LEDs aren't impractical (I suspect that we'd have to make the case out of quartz to survive this kind of radiation!).
I'm glad you wrote this, that was my first reaction too. When I first saw UV LEDs, sterilization was my first thought, as the UV-C lamps are quite expensive. But I looked up the curve and went, bah. The big problem with the UV-C lamps is that they are expensive and fragile, perhaps someone will work out a way to make them cheaper and more robust (diamond film perhaps?). I doubt LEDs will ever be as efficient as gas discharge though.
The claim about LEDs being more efficient that light bulbs is also rather silly, as CFL are much more efficient, and are so cheap/lumen these days that there is little reason not to use them.
you will see that most of the papers come from US universities.
What I found surprising is that the although more papers to ACM or IEEE (I'm in CS) top tier conferences are from the US, the papers that I felt had original or good research were predominantly from elsewhere. It may be that the language barrier increases the required quality to get accepted, or that US organisations are more willing to throw papers at top tier publications.
A car that could only go 20 miles between recharges would not be a hit, not even if the recharge was done in a minute.
With electrical charging there is no need to have to stop especially to charge the car: I propose a car with a 20 mile travel distance and inductive loop charging at intersections, carparks and over km stretches of freeway. billing would be done by symmetric cryptography before the current was turned on.
It is straightforward to use a servo to lock the wheels (could probably be added in software to existing controllers!). As kinetic energy is the square of the velocity there is very little energy to absorb at these speeds anyway.
I wouldn't be surprised if cars are 5% efficient. Remember, they sit idling at the lights, often run on the choke, carry around random extra baggage, are poorly tuned and run out of the optimal power band for most of the time. Try turning an isolated automatic transmision by hand sometime... Of these problems, electrical solves all but the extra baggage problem (and that would solve itself if energy cost more). Electric vehicles also get a little from regenerative braking.
What disturbs me is that pressing on the cell apparently doubled the voltage. TFA claims that this is due to the electrode making a better contact. But they are measuring the voltage with a digital multimeter, with an input impediance of at least 10M. This means that the fuel cell has an output impedance of the order of 10M too - you'd probably get more power out of a calculator solar panel on a stary night!
(alternatively, the finger is warming it up or something)
Actually, I have. They taste like (you'll never guess this!) Apples. Traditional aboriginal food (which generally means 'yuck, don't eat', but in this case they are very nice).
Yep, I've never watched a comedy twice. The BBC doesn't put all their comedy series on DVD first. The movie industry doesn't bother with comedys because they are not profitable. I've never reread a PTerry. etc. It's a nice idea, but it's quite clearly not the whole story.
Yep, but mp3 uses MDCT (http://en.wikipedia.org/wiki/MDCT) which has some additional properties which may make the original algorithm not very helpful (for example, it may not be able to use all of the symmetries, leading to it having to do a full FFT over the larger set, requiring 4 times as much space (and thus losing its original advantage).
Though it may not be directly useful (I haven't seen which FFT variants the library provides), the MDCT used by mp3 is a fairly similar algorithm and could probably be sped up the same way.
Slashdot Mirror
Once you have a break it would be hard to find it
I think TDR would solve this.
Under parental supervision and safe operating conditions children with guns can even be more safe then many adults.
:) Lord of the Flies with guns.
Considering the way kids treat each other without guns, I find this very amusing
I don't think that is right. I think gaussian has the wrong model for the human visual system for scaling and tends to result in images that are too blury. Gaussian might be optimal if you had no other information, but in this case we have a lot of special information about what is expected. For a start, pixels are square (particularly on LCDs), but gaussians are circular. bessel is just as much work as gaussian, but seems to be prefered, e.g.
No idea why you got a funny, but gaussian isn't really the best choice. There are standard libraries that can do the right thing anyway (libmagick, libgdkpixbuf, etc) so perhaps you should just grab one of them and use that.
You assume too much. My parents paid for my training. And I would never have got an advertising career! I think you are just jealous. There are scholarships for academic types too, and probably a comparable number, or more (for example, I got some scholarships for my various degrees by being good at stuff). The entrance requirements for a scholarship to the AIS are much more harsh than getting through year 12. And most people going through never make it big time.
Anyway, HECS is a very reasonable scheme. I also have a $13k hecs debt, but it's a very low interest loan (I think it is just indexed) and when I get a 'real job' I'll be able to pay it off in no time. Much better than systems used elsewhere - at least our system makes the student pay, not the parents! (Not that the students seem to understand that they're paying, considering the few that turned up to lectures)
I'm not sure whether you think the AIS is good or bad from your post. Yes, we have a goverment sponsored research institution into sport. But I don't see this as very different to say the US model with universities having a big emphasis on sports, or having corporate sponsorship of good athletes in certain games. The Australian model is clearly better bang for buck, and so if I'm going to pay for sports research I'd rather do it as efficiently as possible.
The Americans have a state subsidized dept. of defense which has no other goal but to "make our guys win".
I do think Australians are a little parochial about sport, and I do wish that more australians would play than watch on TV. Australians are on average, quite unhealthy. I also wish more money would be spent on other research, but perhaps not at the expense of sports research. (Less money on defence instead?) I think the drive for being good at sport is perhaps a little bit of arrogance, we like to think we're better than average, but in fact we're pretty much on target for an economy of our size.
Disclosure: I trained at the AIS, and my nephew is currently training for the australian swim team.
The technology that replaces the Gasoline Engine has to be of devistating(sic) power.
Why? Would you care to give your reasons for this?
And then I did some real research and found thatr ide gives down to 250nm
/h )
http://en.wikipedia.org/wiki/Aluminum_gallium_nit
http://en.wikipedia.org/wiki/Aluminum_nitride gives down to 210nm
diamond has a band gap of 6.4eV, which, if my physics is correct, gives an LED with an emission frequency of E = hf,
You have: c / (6.4eV
You want: nm
* 193.72529
so assuming we can solve the materials engineering, perhaps UV sterilisation LEDs aren't impractical (I suspect that we'd have to make the case out of quartz to survive this kind of radiation!).
I'm glad you wrote this, that was my first reaction too. When I first saw UV LEDs, sterilization was my first thought, as the UV-C lamps are quite expensive. But I looked up the curve and went, bah. The big problem with the UV-C lamps is that they are expensive and fragile, perhaps someone will work out a way to make them cheaper and more robust (diamond film perhaps?). I doubt LEDs will ever be as efficient as gas discharge though.
The claim about LEDs being more efficient that light bulbs is also rather silly, as CFL are much more efficient, and are so cheap/lumen these days that there is little reason not to use them.
you will see that most of the papers come from US universities.
What I found surprising is that the although more papers to ACM or IEEE (I'm in CS) top tier conferences are from the US, the papers that I felt had original or good research were predominantly from elsewhere. It may be that the language barrier increases the required quality to get accepted, or that US organisations are more willing to throw papers at top tier publications.
Absolutely.
A car that could only go 20 miles between recharges would not be a hit, not even if the recharge was done in a minute.
With electrical charging there is no need to have to stop especially to charge the car: I propose a car with a 20 mile travel distance and inductive loop charging at intersections, carparks and over km stretches of freeway. billing would be done by symmetric cryptography before the current was turned on.
Near 0 rpm, the motor won't stop you.
It is straightforward to use a servo to lock the wheels (could probably be added in software to existing controllers!). As kinetic energy is the square of the velocity there is very little energy to absorb at these speeds anyway.
I wouldn't be surprised if cars are 5% efficient. Remember, they sit idling at the lights, often run on the choke, carry around random extra baggage, are poorly tuned and run out of the optimal power band for most of the time. Try turning an isolated automatic transmision by hand sometime... Of these problems, electrical solves all but the extra baggage problem (and that would solve itself if energy cost more). Electric vehicles also get a little from regenerative braking.
The original maker should some experiments!
What disturbs me is that pressing on the cell apparently doubled the voltage. TFA claims that this is due to the electrode making a better contact. But they are measuring the voltage with a digital multimeter, with an input impediance of at least 10M. This means that the fuel cell has an output impedance of the order of 10M too - you'd probably get more power out of a calculator solar panel on a stary night!
(alternatively, the finger is warming it up or something)
I don't think you can prove axioms?
Actually, I have. They taste like (you'll never guess this!) Apples. Traditional aboriginal food (which generally means 'yuck, don't eat', but in this case they are very nice).
I think Goedel proved this in 1939.
Actually, Appleberries exist already - I've got one growing on my front fence.
Yep, I've never watched a comedy twice. The BBC doesn't put all their comedy series on DVD first. The movie industry doesn't bother with comedys because they are not profitable. I've never reread a PTerry. etc. It's a nice idea, but it's quite clearly not the whole story.
Yep, but mp3 uses MDCT (http://en.wikipedia.org/wiki/MDCT) which has some additional properties which may make the original algorithm not very helpful (for example, it may not be able to use all of the symmetries, leading to it having to do a full FFT over the larger set, requiring 4 times as much space (and thus losing its original advantage).
This is unlikely though.
Though it may not be directly useful (I haven't seen which FFT variants the library provides), the MDCT used by mp3 is a fairly similar algorithm and could probably be sped up the same way.
I think your speed of light is about 10^7 times to slow.
$ units
2438 units, 71 prefixes, 32 nonlinear units
You have: c
You want: 45cm/fortnight
* 8.0584213e+14
(thus insuring publication)
That's a clever idea: Take insurance out against the paper being rejected.