No. First, on these you are mostly limited by the thermal noise of the sensor which is miles above the photon noise for this application. Then you are still thinking that a pixel receives the same flux (power per surface area) as a traditional camera. This is not correct as each of the pixels collect flux from a much larger angular portion of the scene (due to the lack of optical focusing).
I do, why am I wrong? Typical usage : I log onto distant machine, start working in command line (vim, python, matlab -nodesktop), then at some point I will need to display a couple of graphs or images. That's a relatively small graphical payload for which I *do not* want to use VNC. With ssh -X I get the windows to be displayed locally just as if I was doing the work on my light-weight terminal.
So I guess they will go with dynamic content insertion to protect from ad-blocking (via JS, no?). But at the same time they will need indexing... So shouldn't we just move on and spoof our user-agents as crawl-bots?
you need four things: the largest aperture possible, the best-quality optical systems and cameras/CCDs, the least interference from the atmosphere, and the analytical techniques and power to make the most of every photon
Nope, you only need the best acquisition method possible to get out the most information per photon. This is not equivalent to having the best CCD camera or the best optical system. It actually can be the opposite now that we have a lot of processing capabilities (of both light beams and digital imaging). We can, for instance, make high resolution images from low resolution sensors while being less sensitive to noise; or even introducing known aberrations in the system to correct for the shortcomings of traditional imaging techniques.
They have released this demo : https://www.youtube.com/watch?v=kw0-JRa9n94 which looks pretty decent. You can find some artifacts (mainly the occlusion of the little robot which could be better). The depth of field looks pretty cool in the second part and the resolution seems decent (at least for the 1080p camera and for the few frame it is actually in focus, might not be perfect for the eye though). I have no idea on the volume/weight of the device though.
... presented by Inigo Quilez and Pol Jeremias, last year at GDC. Shadertoy had this for more than a year (if you want to warm up your GPU : https://www.shadertoy.com).
No it isn't. Rule of thumb : 1 high end GPU = 12x 4 cores in single precision, 8x 4 cores in double. People getting more than two orders of magnitude differences are comparing (highly-)unoptimized code. Optimizing code on CPU or GPU is hard.
I cannot believe you on this. How was this done? Using cudaMemset, clEnqueueFillBuffer or something equivalent? Or your own code (coalesced memory access...)?
It seems that the CNES took out an article today saying that the instrument was repaired. See the cached content here.
It is saying that the core of the instrument is a titanium sphere keeping inner sensors in a 0.25 mbar vacuum (at most). This is about 40000x less than atmospheric pressure on earth. What they saw is that the pressure inside the sphere went up from 0 to 30 mbar in a month. They found a production defect on one of the sensors connectors going through the shell but it seems that they were able to seal it correctly with some resin. They also found a second leakage problem around the pump creating the vacuum, which could be fixed as well.
They are currently running tests to determine if everything is ok. According to the article, the launch could be re-scheduled to March 18th.
One of the main problem of this on the desktop is the bandwidth. Having true heterogeneous memory arch between CPU and GPU but with slower than current in-device bandwidth (around 250GB/s) would be a waste for a vastly larger applications pool than for those which would benefit from it.
These stories get more and more attention of the media and every time they will emphasize that this is considered as a federal crime for which penalty is severe fines and possibly jail time. But this does not seem to be at all effective with the population. The question is, are people doing this out of a really bad intention or are just not intelligent enough to understand the risks and the sentences they are facing for, literally, no personal gain?
Exactly. Spitzer is capable of detecting it as a very faint signal, knowing that the actual source is a star. If they are using EM communication, the prower is probably orders of magnitude smaller than the star emission itself. Good luck, catching that with sufficient SNR in the radio domain...
Take 2 polarized filters, and measure the amount of light that gets through as a function of the angle between them. With a classical model of polarization, you'd expect it to fall directly with the angle, but instead it falls of as cos^2 of the angle.
The classical E.M. theory perfectly predicts the cos^2 term. See Malus law.
What's really weird, though, is that of you take 2 polarizing filters at right angles, such that no light gets through, then stick a third between them at a 45 degree angle, then it's as bright as one filter alone.
No, you would have less power than a single polarizer. This also very well explained by Jones calculus.
No, with entanglement it would seem that you cannot force one particle to a state so that the other is switching as well. Thus, you cannot effectively use this as a communication channel.
A very crude picture can be stated as the following : you send two letters containing the same unique number (0...9) to both Alice and Bob (who also know this rule). When Alice opens her letter and reads the number, she knows that Bob has the same number but she cannot use that to communicate a particular number to Bob directly. WARNING : This explanation is using the hidden variable model which is wrong, as the EPR paradox/Bell's inequality and these researchers are proving. But this is the only simple explanation I know (with my limited knowledge of QM) to convey the fact that entanglement can not be used as a communication channel.
This quantity of fuel is treated as a potential risk in France and in the EU mostly due to industrial catastrophes such as Seveso (Italy) in 76 or AZF in 01. The zoning laws are now pretty strict for sites having large stocks of flammable/explosive materials. It also appears that this data-center has several aerial container.
For the noise, my guess is that the court ruling is only temporary and will require the owner of the facility to shield the sources.
Better yet, why not require two-factor authentication for large and online purchases where the card isn't swiped? If the purchase is large or the card isn't swiped, simply send a verification code to the customer's phone for that transaction that they have to enter.
This is already implemented in Europe for online purchases (some banks at least). It took more than 25 years to get the chips in the US, I guess we'll have to wait another 15 years or so...
You can use Stellarium (http://www.stellarium.org/, or with the help of your package manager) to get a preview of tonight's sky at your terrestrial location (not accounting for cloud cover though). This includes a simulation of the actual eclipse.
If we can't measure it, it does not mean that they do not exist. They could be weaker to some extent, yes, but it could be also that the "sources" are so evenly distributed around us that the superposition of their waves at our location almost cancel itself.
Slashdot Mirror
http://www.kshitijmarwah.com/index.php?/research/compressive-light-field-photography/ (From MIT Media Lab)
No. First, on these you are mostly limited by the thermal noise of the sensor which is miles above the photon noise for this application. Then you are still thinking that a pixel receives the same flux (power per surface area) as a traditional camera. This is not correct as each of the pixels collect flux from a much larger angular portion of the scene (due to the lack of optical focusing).
I do, why am I wrong?
Typical usage : I log onto distant machine, start working in command line (vim, python, matlab -nodesktop), then at some point I will need to display a couple of graphs or images. That's a relatively small graphical payload for which I *do not* want to use VNC. With ssh -X I get the windows to be displayed locally just as if I was doing the work on my light-weight terminal.
So I guess they will go with dynamic content insertion to protect from ad-blocking (via JS, no?). But at the same time they will need indexing... So shouldn't we just move on and spoof our user-agents as crawl-bots?
you need four things: the largest aperture possible, the best-quality optical systems and cameras/CCDs, the least interference from the atmosphere, and the analytical techniques and power to make the most of every photon
Nope, you only need the best acquisition method possible to get out the most information per photon. This is not equivalent to having the best CCD camera or the best optical system. It actually can be the opposite now that we have a lot of processing capabilities (of both light beams and digital imaging).
We can, for instance, make high resolution images from low resolution sensors while being less sensitive to noise; or even introducing known aberrations in the system to correct for the shortcomings of traditional imaging techniques.
They have released this demo : https://www.youtube.com/watch?v=kw0-JRa9n94 which looks pretty decent. You can find some artifacts (mainly the occlusion of the little robot which could be better). The depth of field looks pretty cool in the second part and the resolution seems decent (at least for the 1080p camera and for the few frame it is actually in focus, might not be perfect for the eye though).
I have no idea on the volume/weight of the device though.
... presented by Inigo Quilez and Pol Jeremias, last year at GDC. Shadertoy had this for more than a year (if you want to warm up your GPU : https://www.shadertoy.com).
(*) Drivers not included.
No it isn't. Rule of thumb : 1 high end GPU = 12x 4 cores in single precision, 8x 4 cores in double.
People getting more than two orders of magnitude differences are comparing (highly-)unoptimized code.
Optimizing code on CPU or GPU is hard.
I cannot believe you on this. How was this done? Using cudaMemset, clEnqueueFillBuffer or something equivalent? Or your own code (coalesced memory access...)?
It seems that the CNES took out an article today saying that the instrument was repaired. See the cached content here.
It is saying that the core of the instrument is a titanium sphere keeping inner sensors in a 0.25 mbar vacuum (at most). This is about 40000x less than atmospheric pressure on earth. What they saw is that the pressure inside the sphere went up from 0 to 30 mbar in a month. They found a production defect on one of the sensors connectors going through the shell but it seems that they were able to seal it correctly with some resin. They also found a second leakage problem around the pump creating the vacuum, which could be fixed as well.
They are currently running tests to determine if everything is ok. According to the article, the launch could be re-scheduled to March 18th.
No, you are supposed to look at the diff log...
"Real governments use version control..."
Provided by the objective function is parallelization yes. Because SA is not.
One of the main problem of this on the desktop is the bandwidth. Having true heterogeneous memory arch between CPU and GPU but with slower than current in-device bandwidth (around 250GB/s) would be a waste for a vastly larger applications pool than for those which would benefit from it.
These stories get more and more attention of the media and every time they will emphasize that this is considered as a federal crime for which penalty is severe fines and possibly jail time. But this does not seem to be at all effective with the population.
The question is, are people doing this out of a really bad intention or are just not intelligent enough to understand the risks and the sentences they are facing for, literally, no personal gain?
Exactly. Spitzer is capable of detecting it as a very faint signal, knowing that the actual source is a star. If they are using EM communication, the prower is probably orders of magnitude smaller than the star emission itself. Good luck, catching that with sufficient SNR in the radio domain...
Yeah, but having a lot of RAM means that you also have plenty of space for cache and that's very comfortable.
Take 2 polarized filters, and measure the amount of light that gets through as a function of the angle between them. With a classical model of polarization, you'd expect it to fall directly with the angle, but instead it falls of as cos^2 of the angle.
The classical E.M. theory perfectly predicts the cos^2 term. See Malus law.
What's really weird, though, is that of you take 2 polarizing filters at right angles, such that no light gets through, then stick a third between them at a 45 degree angle, then it's as bright as one filter alone.
No, you would have less power than a single polarizer. This also very well explained by Jones calculus.
No, with entanglement it would seem that you cannot force one particle to a state so that the other is switching as well. Thus, you cannot effectively use this as a communication channel.
A very crude picture can be stated as the following : you send two letters containing the same unique number (0...9) to both Alice and Bob (who also know this rule). When Alice opens her letter and reads the number, she knows that Bob has the same number but she cannot use that to communicate a particular number to Bob directly.
WARNING : This explanation is using the hidden variable model which is wrong, as the EPR paradox/Bell's inequality and these researchers are proving. But this is the only simple explanation I know (with my limited knowledge of QM) to convey the fact that entanglement can not be used as a communication channel.
"and time runs backward as well as forward."
It had to be published today, right Doc?
This quantity of fuel is treated as a potential risk in France and in the EU mostly due to industrial catastrophes such as Seveso (Italy) in 76 or AZF in 01. The zoning laws are now pretty strict for sites having large stocks of flammable/explosive materials. It also appears that this data-center has several aerial container.
For the noise, my guess is that the court ruling is only temporary and will require the owner of the facility to shield the sources.
Better yet, why not require two-factor authentication for large and online purchases where the card isn't swiped? If the purchase is large or the card isn't swiped, simply send a verification code to the customer's phone for that transaction that they have to enter.
This is already implemented in Europe for online purchases (some banks at least). It took more than 25 years to get the chips in the US, I guess we'll have to wait another 15 years or so...
You can use Stellarium (http://www.stellarium.org/, or with the help of your package manager) to get a preview of tonight's sky at your terrestrial location (not accounting for cloud cover though). This includes a simulation of the actual eclipse.
Meanwhile, bandwidth will grow by 20%...
If we can't measure it, it does not mean that they do not exist. They could be weaker to some extent, yes, but it could be also that the "sources" are so evenly distributed around us that the superposition of their waves at our location almost cancel itself.