A good tube amplifier should probably sound the same as a good amplifier using field-effect transistors.
You know that the electrodes in tubes get contaminated and evaporate, right? And no two tubes respond exactly the same. It is possible that the nonlinearities sound good to your ears, or you have a whole lot of unintended low pass filtering built into your tube amp, plus some unintended resonances, and you kind of like the modified product. I on the other hand, prefer to hear sound reproduced exactly as it was recorded.
You forgot how crappy sound systems were in the old days. Today even cheap digital systems can often outperform the best equipment in an old time studio. Now when you get to analog components you might have a point, although there have been huge advances there too.
there are still ways to tell the difference, especially on tracks that have any waveform that are close to a fundamental harmonic of the sampling rate. You can actually hear the sampling beat frequencies injected into the music and distortion
So you have not heard of Nyquist frequency or sinc filtering. If you can actually hear beat frequencies than your hardware (software?) is misdesigned. It can be mathematically proven that you can't hear such beat frequencies in a properly engineered system. Of course I realize I am telling this to someone who *believes* in vinyl, so it probably fell on deaf ears.
Anyway, I can guarantee that you are wrong about the other guy being wrong. This is science.
OK, today I learned that the signal on a neural axon may be richer that a single scalar frequency value, in particular, inter-pulse intervals are known to carry information for some neurons. So I agree that the current conception of neural networks is too simplistic, or too unlike the real thing to be able to imitate it accurately. And now I got what I wanted, clear evidence that the total bandwidth of the brain is higher than has been fashionable to suppose. Not stupidly magical like the parent article proposes, but orders of magnitude beyond the 100 billion simple processing elements we get just by counting neurons.
Thanks for that. A quick survey showed me that neural coding on the axon is now known to be richer than a simple firing rate, in particular the inter-spike interval is significant in some cases, including Purkinje cells. Who knows what else is significant, maybe the timing of a single action potential in reference to some clock, in some cases. But it's clear that the task of reverse engineering C elegans is much bigger than just determining connectivity and synapse weights. The null result is important here: "we wired up a neural net exactly like the one we found in the worm, and it didn't work like the worm." So, short of advances in decoding neuroanatomy to find all the missing pieces of the puzzle, it would be useful to try and determine how the C elegans circuit diagram needs to be modified to replicate the behavior of the worm more accurately. This would hint at the kind of biological processes we can expect to discover as tools improve.
This also suggests that the current computational neural net dogma is too simplistic.
Older than hipsters, the top hits look like retired baby boomer stuff. But not all of it, some of it looks like grandkids of baby boomers. Who probably get disks and players as presents from their grandparents.
Renaissance fairs are big in Europe. Everybody dresses like The Gap is 400 years in the future and there is no such thing as medical science. I understand they have a lot of fun.
I'm familiar with the neurology of the retina and visual cortex thank you. You would seem to have your own private definition of Turing machine. Please, at least say state machine, and you're still wrong.
It's nice to see AMD get a foothold in the tablet market. But the 3000 series APUs, also announced today, are more important imho. For example, the 15 watt 3700U is ideal for a midrange ultrabook. With it's 10 Vega GPU cores it can handle 1080p gaming just fine, in additional to that Zen+ Ryzen goodness. I'm thinking you're going to see AMD ultrabooks at the top of bestseller lists for the first time ever.
There is no new secret quantum CPU technology that is going to push Moore's law along.
Not entirely true. Advances in understanding of quantum mechanics are also key to shrinking classical computers. For example, overcoming tunnelling issues.
The synaptic weights have not been mapped, instead machine learning algorithms were used. Not dissing the project at all, I think there is a huge amount to learn from it, including helping to determine whether our model of how neurons work is essentially complete or not. I'm betting on not, and that a whole lot more processing goes on in the neuron than what happens at the synapses alone.
it may mean that we may see an end to how dense things can be packed
EUV will take us through another factor of 8 or so density increase more or less smoothly without relying on as yet unknown breakthroughs beyond what is required to get past the current 7nm hump.
Then don't discount the possibility of breakthroughs. For example, somebody might figure out a way to mass produce nanotube transistors, maybe good for a further factor of 8.
by limiting myself to only one vehicle, I'm being a whole lot more green than the twits who keep a stable of them.
Is this a demonstration of the intellectual capacity of a pickup truck owner? It's how much and what you drive, not how many you own that affects the environment. I'm pretty sure your pickup puts out more smog than a sedan, other things being equal.
I am not sure you know what a neural net is, or a Turing machine. There is such a thing but this is a combination of neural networks and state machines. And "physical 3D network of neurons"... WTH are you talking about?
Please at least read the Wikipedia article on neural nets.
Argument by analogy, as useless as most argument by analogy. And you worked a false premise in there too: nobody I know understands the brain as a kind of computer. But we have analyzed the anatomy of the brain sufficiently to be sure that it is, ahem, a kind of neural network. How does that work for your steam engine analogy?
Slashdot Mirror
If you're a heavy metal fan you probably can't hear much of anything anyway, so no worries.
A good tube amplifier should probably sound the same as a good amplifier using field-effect transistors.
You know that the electrodes in tubes get contaminated and evaporate, right? And no two tubes respond exactly the same. It is possible that the nonlinearities sound good to your ears, or you have a whole lot of unintended low pass filtering built into your tube amp, plus some unintended resonances, and you kind of like the modified product. I on the other hand, prefer to hear sound reproduced exactly as it was recorded.
You forgot how crappy sound systems were in the old days. Today even cheap digital systems can often outperform the best equipment in an old time studio. Now when you get to analog components you might have a point, although there have been huge advances there too.
there are still ways to tell the difference, especially on tracks that have any waveform that are close to a fundamental harmonic of the sampling rate. You can actually hear the sampling beat frequencies injected into the music and distortion
So you have not heard of Nyquist frequency or sinc filtering. If you can actually hear beat frequencies than your hardware (software?) is misdesigned. It can be mathematically proven that you can't hear such beat frequencies in a properly engineered system. Of course I realize I am telling this to someone who *believes* in vinyl, so it probably fell on deaf ears.
Anyway, I can guarantee that you are wrong about the other guy being wrong. This is science.
OK, today I learned that the signal on a neural axon may be richer that a single scalar frequency value, in particular, inter-pulse intervals are known to carry information for some neurons. So I agree that the current conception of neural networks is too simplistic, or too unlike the real thing to be able to imitate it accurately. And now I got what I wanted, clear evidence that the total bandwidth of the brain is higher than has been fashionable to suppose. Not stupidly magical like the parent article proposes, but orders of magnitude beyond the 100 billion simple processing elements we get just by counting neurons.
Thanks for that. A quick survey showed me that neural coding on the axon is now known to be richer than a simple firing rate, in particular the inter-spike interval is significant in some cases, including Purkinje cells. Who knows what else is significant, maybe the timing of a single action potential in reference to some clock, in some cases. But it's clear that the task of reverse engineering C elegans is much bigger than just determining connectivity and synapse weights. The null result is important here: "we wired up a neural net exactly like the one we found in the worm, and it didn't work like the worm." So, short of advances in decoding neuroanatomy to find all the missing pieces of the puzzle, it would be useful to try and determine how the C elegans circuit diagram needs to be modified to replicate the behavior of the worm more accurately. This would hint at the kind of biological processes we can expect to discover as tools improve.
This also suggests that the current computational neural net dogma is too simplistic.
Vinyl is the music equivalent of homeopathy.
Or the data processing equivalent of typewriters.
Older than hipsters, the top hits look like retired baby boomer stuff. But not all of it, some of it looks like grandkids of baby boomers. Who probably get disks and players as presents from their grandparents.
Renaissance fairs are big in Europe. Everybody dresses like The Gap is 400 years in the future and there is no such thing as medical science. I understand they have a lot of fun.
I'm familiar with the neurology of the retina and visual cortex thank you. You would seem to have your own private definition of Turing machine. Please, at least say state machine, and you're still wrong.
Hardly. Vehicles depreciate mainly by miles, not time. You will own the vehicles longer, or someone will.
It's nice to see AMD get a foothold in the tablet market. But the 3000 series APUs, also announced today, are more important imho. For example, the 15 watt 3700U is ideal for a midrange ultrabook. With it's 10 Vega GPU cores it can handle 1080p gaming just fine, in additional to that Zen+ Ryzen goodness. I'm thinking you're going to see AMD ultrabooks at the top of bestseller lists for the first time ever.
Is this people trying to be stupid for stupid's sake?
Trying to teach us what it is like to be a person who owns a pickup.
There is no new secret quantum CPU technology that is going to push Moore's law along.
Not entirely true. Advances in understanding of quantum mechanics are also key to shrinking classical computers. For example, overcoming tunnelling issues.
And even if you could, you don't know how the weights change over time.
The synaptic weights have not been mapped, instead machine learning algorithms were used. Not dissing the project at all, I think there is a huge amount to learn from it, including helping to determine whether our model of how neurons work is essentially complete or not. I'm betting on not, and that a whole lot more processing goes on in the neuron than what happens at the synapses alone.
it may mean that we may see an end to how dense things can be packed
EUV will take us through another factor of 8 or so density increase more or less smoothly without relying on as yet unknown breakthroughs beyond what is required to get past the current 7nm hump.
Then don't discount the possibility of breakthroughs. For example, somebody might figure out a way to mass produce nanotube transistors, maybe good for a further factor of 8.
by limiting myself to only one vehicle, I'm being a whole lot more green than the twits who keep a stable of them.
Is this a demonstration of the intellectual capacity of a pickup truck owner? It's how much and what you drive, not how many you own that affects the environment. I'm pretty sure your pickup puts out more smog than a sedan, other things being equal.
More to the point, what are they smoking, and what meds are they on?
As I understand it, Teslas are selling as fast as they can build them.
I am not sure you know what a neural net is, or a Turing machine. There is such a thing but this is a combination of neural networks and state machines. And "physical 3D network of neurons"... WTH are you talking about?
Please at least read the Wikipedia article on neural nets.
Mine was not an analogy, it was reductio ad absurdum.
Physics always was overly complex. When we get to be God then we will design the universe much more simply, won't we?
Oh come on mods, that was super funny
Argument by analogy, as useless as most argument by analogy. And you worked a false premise in there too: nobody I know understands the brain as a kind of computer. But we have analyzed the anatomy of the brain sufficiently to be sure that it is, ahem, a kind of neural network. How does that work for your steam engine analogy?