26 May 14
Originally posted by DeepThoughtIt was a long time ago my friend, when the Pentium 2 was state of the art 😲
If you are a computer scientist then you should not have any problems with the maths associated with introductory quantum theory. Quantum field theory is a little tougher and by the time it gets to string theory you need to know your way around the maths. To get the basic ideas you need to know what a partial derivative is and to understand what linear ...[text shortened]... to be able to calculate things like spherical harmonics, provided you understand what they are.
26 May 14
Originally posted by twhiteheadFeynman probably said that, but, from hazy recollection, I think that before him Niels Bohr had said something along the lines of "Anyone who thinks they have understood quantum mechanics has not understood it.". I had a look on Wikipedia but can't find the reference.
I think it was Richard Feynman - correct me if I am wrong, who claimed that nobody understood quantum theory. I think what he said, was that although we can do the math, we will never fully understand it intuitively - because a lot of it is not intuitive.
26 May 14
1 edit
Originally posted by twhiteheadDoing the exercises isn't just pen and paper.
I have given it some thought, and I disagree more the more I think it over.
I believe that the vast majority of physics concepts can be understood without any mathematics whatsoever. Although 'doing the math' may help to cement the concepts, - largely by forcing the student to think about it more, or seeing practical examples, I dispute that this is the ...[text shortened]... eneficial for my understanding to see examples of this in action than to do math problems on it.
Again let me say that it depends on your goal. If you want to know why "that when a wire passes through a magnetic field, an electric current is generated" then you need maths and there is no way around this. After you use the math and understand it you can explain it like you wish:verbally, graphically, etc.
I'd argue that if you only know why this phenomenon is so because someone explained it to you and didn't use maths to do it then you know it in a more rudimentary way. And there is nothing intrinsically wrong with that.
Edit: To put it more bluntly: you certainly don't need to know Quantum Mechanics at the same level that I use on my blog to use your microwaves. The thing is that if no one understood quantum mechanics at that level you wouldn't have a microwave to start with.
That is to say that what today may seem too abstract or theoretical will get an application tomorrow. And the engineers that are responsible for its implementation don't need to know the finer details the same way a theoretical physicist knows. And evidently the people that use it don't need to know it at the level that the engineer knows it, nor do they need to know it at the theoretical physicist level...
26 May 14
Originally posted by DeepThoughtFeynman: "I think I can safely say that nobody understands quantum mechanics."
Feynman probably said that, but, from hazy recollection, I think that before him Niels Bohr had said something along the lines of "Anyone who thinks they have understood quantum mechanics has not understood it.". I had a look on Wikipedia but can't find the reference.
Bohr: "Anyone who is not shocked by quantum theory has not understood it."
The two sentences are in a way complementary (see what I did there?).
Originally posted by adam warlockI have to disagree.
Again let me say that it depends on your goal. If you want to know why "that when a wire passes through a magnetic field, an electric current is generated" then you need maths and there is no way around this. After you use the math and understand it you can explain it like you wish:verbally, graphically, etc.
Firstly, 99% of the people who can do the maths ie work out what current will be generated given a particular magnetic field, haven't the foggiest idea why it happens.
Secondly, even when you know the next level of physics (the next step in 'why'😉 you may not need math to understand that next step.
I'd argue that if you only know why this phenomenon is so because someone explained it to you and didn't use maths to do it then you know it in a more rudimentary way.
And I disagree. I do not believe the math necessarily gives more insight. In fact it is my claim that the math only serves as a learning tool, or for practical use, it is not central to the concepts themselves.
Let me pose this:
Can evolution only be understood by doing the math? If not, why not?
The thing is that if no one understood quantum mechanics at that level you wouldn't have a microwave to start with.
I think you may be confusing the basic principles, with calculation of specific values in specific instances.
That is to say that what today may seem too abstract or theoretical will get an application tomorrow. And the engineers that are responsible for its implementation don't need to know the finer details the same way a theoretical physicist knows.
Yet engineers, do just as much math, only applied math rather than theoretical math. And I would argue that quite often they are brilliant at doing the math required for their job without fully understanding the physics behind it.
26 May 14
Originally posted by twhiteheadThe question of why a changing magnetic field induces a current in a wire is somewhat deep. I'd have said that engineers don't understand it. They are taught electro-magnetism up to Maxwell's equations and understand the how very well. They can do the associated calculations and know how to apply it all in practical situations. But you really need an understanding of things called principal bundles to understand why there should be an interaction between currents and fields in the first place. The why is down to how vectors defined on principal bundles behave. Principal bundles are mathematical objects. Since the electro-magnetic field is a projection of an object known as a connection it is quite hard to avoid mathematical discourse when discussing theory at the deepest level.
I have to disagree.
Firstly, 99% of the people who can do the maths ie work out what current will be generated given a particular magnetic field, haven't the foggiest idea why it happens.
Secondly, even when you know the next level of physics (the next step in 'why'😉 you may not need math to understand that next step.
[b]I'd argue that if you only k ...[text shortened]... iant at doing the math required for their job without fully understanding the physics behind it.
27 May 14
1 edit
Originally posted by twhitehead"Firstly, 99% of the people who can do the maths ie work out what current will be generated given a particular magnetic field, haven't the foggiest idea why it happens."
I have to disagree.
Firstly, 99% of the people who can do the maths ie work out what current will be generated given a particular magnetic field, haven't the foggiest idea why it happens.
Secondly, even when you know the next level of physics (the next step in 'why'😉 you may not need math to understand that next step.
[b]I'd argue that if you only k ...[text shortened]... iant at doing the math required for their job without fully understanding the physics behind it.
Yes you're right but what I've said is the converse of what you state. I said that if you want to know why you have to do the maths. Do you see the difference between your statement and mine?
"Secondly, even when you know the next level of physics (the next step in 'why' you may not need math to understand that next step"
Forgive me for not taking your word for it. Here's what I propose. I'll ask you to explain an advanced physical concept that you know nothing about and you have to tell me the why it is so without using math.
"And I disagree. I do not believe the math necessarily gives more insight. In fact it is my claim that the math only serves as a learning tool, or for practical use, it is not central to the concepts themselves. "
I'll repeat here my last answer.
""Can evolution only be understood by doing the math? If not, why not?
I don't consider myself to know enough about evolution to answer this question. My main points are of course related to Physics.
"I think you may be confusing the basic principles, with calculation of specific values in specific instances."
No I'm telling you that some people have to get their hands really dirty in mathematics at the first stages of the development, then come the analogies, then come the people the repeat the analogies without really understanding what's going on.
"Yet engineers, do just as much math, only applied math rather than theoretical math."
Can you help in quantifying the amount of math one does so that I can see if I agree or disagree with your statement?
"And I would argue that quite often they are brilliant at doing the math required for their job without fully understanding the physics behind it"
And I'd support you in your argument.
28 May 14
4 edits
Originally posted by menace71that is interesting although I detect the article has definitely got a grain of extremely badly flawed reasoning with its last comment which was:
http://www.extremetech.com/extreme/177316-d-wave-disentangled-google-explains-the-present-and-future-of-quantum-computing
????
Manny
"...In fact, given the hundreds of billions of dollars poured in to the development of modern computers, it would be astonishing if scientists invented a new computing solution capable of beating conventional equipment in all respects in just a handful of years...."
-err no, it wouldn't be "astonishing" because that is a flawed inference because, if only they took full account of how much a quantum computer performance should be potentially enhances with each extra qubit added to it, they would see that how much money is currently poured into improving the performance of classical computers would be virtually irrelevant!
Lets say, for the sake of argument that, thanks to huge money that is currently poured into classical computers, classical computers processing power would probably increases by 100 fold in the next 10 years. Now, assuming maximum connectivity between qubits in any new quantum computer, regardless of whether that improvement in classical computers actually takes place, for quantum computers to become competitive with classical computers, each will have to use 'many' qubits (at least 50 of them but 200 would make them EXTREMELY competitive! ). But, assuming maximum connectivity between qubits, you would only have to add an extra 8 qubits to get more than 100 fold increase processing power out of the quantum computer!
I assume that if you have already managed to make an effective quantum computer that uses, say, 100 qubits, it should be relatively easy to make one that uses just an extra 8 qubits simply because 8 is much less than 100. Thus it wouldn't make much difference to the probability of whether, say, ten years from now, quantum computers would be able to compete with classical computers even if the processing power of classical computers were increased by, say, 100 fold.
28 May 14
Originally posted by humyI have to point out that quantum computers only compete with classical computers on certain classes of problem. For sequential computing, they offer no advantage and would require similar clock speeds to compete.
Thus it wouldn't make much difference to the probability of whether, say, ten years from now, quantum computers would be able to compete with classical computers even if the processing power of classical computers were increased by, say, 100 fold.
What they excel at is parallel computations that have a single answer.
It must be noted that adding extra cores to classical computers is similar in that it gives no benefit to sequential programs.
28 May 14
5 edits
Originally posted by twhiteheadYes, I already know all this. However, the inference made by that article is still false and exactly for the reasons I just explained. Explaining that in other words; the conclusion cannot be true that "...it would be astonishing if scientists invented a new computing solution capable of beating conventional equipment in all respects in just a handful of years...." BECAUSE of the premise "...given the hundreds of billions of dollars poured in to the development of modern computers ..." although that conclusion is still probably true for other reasons OTHER than that premise given. In other words, even though the conclusion of the inference is probably true, the inference itself is STILL false and thus it is the inference, NOT the conclusion, that I am criticizing here. I just don't like shoddy logic even if shoddy logic by pure coincidence just happens to lead to the correct conclusion as here, which is why I pointed this flaw out here.
I have to point out that quantum computers only compete with classical computers on certain classes of problem. For sequential computing, they offer no advantage and would require similar clock speeds to compete.
What they excel at is parallel computations that have a single answer.
It must be noted that adding extra cores to classical computers is similar in that it gives no benefit to sequential programs.
28 May 14
Originally posted by humyI am not sure what you are arguing. If I read it correctly you are misunderstanding what the article is saying. They are saying it would be astonishing for a new technology to suddenly take over from an older technology that has had a lot invested in it. I don't think their astonishment has anything to do with actual technical considerations.
Yes, I already know all this. However, the inference made by that article is still false and exactly for the reasons I just explained. Explaining that in other words; the conclusion cannot be true that "...it would be astonishing if scientists invented a new computing solution capable of beating conventional equipment in all respects in just a handful of years ...[text shortened]... st happens to lead to the correct conclusion as here, which is why I pointed this flaw out here.
28 May 14
Originally posted by twhitehead
I am not sure what you are arguing. If I read it correctly you are misunderstanding what the article is saying. They are saying it would be astonishing for a new technology to suddenly take over from an older technology that has had a lot invested in it. I don't think their astonishment has anything to do with actual technical considerations.
They are saying it would be astonishing for a new technology to suddenly take over from an older technology that has had a lot invested in it.
It is that inference that I believe is false when it is applied to quantum computers in this case although there are very good other reasons to think quantum computers will not take over classical computers any time soon, the main ones being we seem to be still quite far away from making one that connects enough qubits and with enough connectivity and long enough to enable it to do a calculation significantly faster than what any classical computer can do. Once these technical challenges are achieve, how much money has been spent on classical computers would become almost irrelevant.
29 May 14
Originally posted by humyI think the main problem is that they require very delicate technology that all works at liquid helium temperatures. They may start supplementing super-computers soon, but it will be a while before they replace the P.C..They are saying it would be astonishing for a new technology to suddenly take over from an older technology that has had a lot invested in it.
It is that inference that I believe is false when it is applied to quantum computers in this case although there are very good other reasons to think quantum computers will not take over classical c ...[text shortened]... re achieve, how much money has been spent on classical computers would become almost irrelevant.
For some problems, such as factorizing very large products of prime numbers, even a quantum machine with a clock speed of 1Hz would be faster than the most powerful super-computer currently running. For other problems you may as well use a conventional machine.
29 May 14
5 edits
Originally posted by DeepThoughtI have researched the possible applications of quantum computers and come to the conclusion that the only one that counts i.e. the only one that has any real usefulness or value to humanity is certainly not the usual applications stated (finding large prime numbers and decryption being two of them ) but rather is with making them do certain types computer simulations esp computer simulations of quantum systems. This is because such computer simulations of quantum systems could greatly speed up research into them and lead to faster advances in solar cell design, superconductors, LEDs, electronics, photochemistry, catalysts, understanding of how protein molecules fold and understanding of chemical reactions esp organic chemical reactions in life chemistry.
I think the main problem is that they require very delicate technology that all works at liquid helium temperatures. They may start supplementing super-computers soon, but it will be a while before they replace the P.C..
For some problems, such as factorizing very large products of prime numbers, even a quantum machine with a clock speed of 1Hz would b ...[text shortened]... uper-computer currently running. For other problems you may as well use a conventional machine.
However, I don't see how quantum computer would ever replace classical computers, only merely supplement them for a certain limited range of but important applications.