After watching this video someone protested that the Deutsch oracle WOULD outperform a classical computer if the black box had a very large number of inputs, such as one million. I suspect they were thinking that a single threaded computer simulation would struggle to simulate such a huge amount of operations that would all be occurring at the same time in the quantum scenario. But the point is that we could build the whole thing with classical components instead of quantum ones in which we use 3 bits to represent one of the 8 possible states that each qubit would be in. So the whole thing could be reproduced using classical components because everything in the Deutsh oracle is 100% deterministic. And the classical scenario with 3 million bits would work first time whereas the scenario with 1 million qubits would need to be re-run several times to gain a high probability that the result was not a false result due to errors. We should not pretend that quantum beats classical just because it would take a long time to simulate lots of qubits. It would be like claiming that 3 million switches acting in parallel will be faster than a single threaded computer simulation of them. We need to compare like with like, and when we do, classical always wins.
Yes, bold. It's sad that the truth is considered to be bold, when I consider Bell's and Spooky Action to be the boldest of them all, yet they somehow became mainstream physics.
you should make more videos that highlight the failed milestones of quantum computing and some of the internal contradictions inside quantum computing - especially regarding the format of the quantum system each camp uses as computing elements and the fact that they have become de-correlated in results. . Even some commentary on some of the most popular QM videos would be excellent! A much needed alternative , rational view. Cheers for all the great work
This is a very valuable topic to discuss and I tip my hat to you for bringing this up. I am of a slightly different opinion on the matter of quantum computing and I really think the forms and function of a quantum computer is really just that of a network that has exchange rules mediated using a power-law based coupling with a randomly varying oscillatory term added to it and that the behavior that is affiliated with the so-called quantum computation such as error correction, fast optimization solving and so forth is really just a consequence of the mean field of the network and that what is seen as the "quantum" behavior of these systems can really be explained by classical analogues.
Well done. Can you recommend other "simple maths" QM computing references, particularly any that might address deterministic vs non-deterministic models/theory?
Hi Jim, it's great to hear from you. As you will have discovered, in this video I have focused on the Deutsch-Jozsa algorithm (a.k.a. Deutsch oracle) which I said was 100% deterministic. And so when we perform the measurements to get actual binary outputs, there are no cases where the qubit is in state where the probability of the reading being a 1 or a 0 is anything other than 100% or 0%. I've not tackled the much more complicated cases such as Shor's algorithm for prime factorisation. The advocates of QM believe that via superposition and entanglement a huge amount of different paths can be tested at the same time, exploring all options and finding a solution in an exceptionally fast time. They claim that the main problem is how to extract the best solution as we are hampered due to the 'collapse' to binary values when we take the readings. This results in a complicated approach to eking out a final answer - all of which I believe is very complicated. I'm not sure if prime factorisation is the best example for a comparison of deterministic/classical computing versus quantum superposition & entanglement. Ideally we want something much simpler that can be explained in very simple terms. Sadly I don't know of any such simple example. If there was any convincing evidence that quantum computing could outperform classical computers for optimisation problems or prime factorisation then I would have more interest in trying to understand some of the algorithms used. But since there is no evidence I can't create any interest in trying to analyse them. I made this video quite a long time ago and I can't remember reading anything that was particularly good for explaining non-deterministic algorithms. But if you manage to find something then please let me know.
After watching this video someone protested that the Deutsch oracle WOULD outperform a classical computer if the black box had a very large number of inputs, such as one million. I suspect they were thinking that a single threaded computer simulation would struggle to simulate such a huge amount of operations that would all be occurring at the same time in the quantum scenario.
But the point is that we could build the whole thing with classical components instead of quantum ones in which we use 3 bits to represent one of the 8 possible states that each qubit would be in. So the whole thing could be reproduced using classical components because everything in the Deutsh oracle is 100% deterministic.
And the classical scenario with 3 million bits would work first time whereas the scenario with 1 million qubits would need to be re-run several times to gain a high probability that the result was not a false result due to errors.
We should not pretend that quantum beats classical just because it would take a long time to simulate lots of qubits. It would be like claiming that 3 million switches acting in parallel will be faster than a single threaded computer simulation of them. We need to compare like with like, and when we do, classical always wins.
If we go yet another decade without clear Quantum Supremacy, you may be right. You're bold as hell, I'll give you that.
Yes, bold. It's sad that the truth is considered to be bold, when I consider Bell's and Spooky Action to be the boldest of them all, yet they somehow became mainstream physics.
you should make more videos that highlight the failed milestones of quantum computing and some of the internal contradictions inside quantum computing - especially regarding the format of the quantum system each camp uses as computing elements and the fact that they have become de-correlated in results. .
Even some commentary on some of the most popular QM videos would be excellent! A much needed alternative , rational view.
Cheers for all the great work
This is a very valuable topic to discuss and I tip my hat to you for bringing this up. I am of a slightly different opinion on the matter of quantum computing and I really think the forms and function of a quantum computer is really just that of a network that has exchange rules mediated using a power-law based coupling with a randomly varying oscillatory term added to it and that the behavior that is affiliated with the so-called quantum computation such as error correction, fast optimization solving and so forth is really just a consequence of the mean field of the network and that what is seen as the "quantum" behavior of these systems can really be explained by classical analogues.
Well done. Can you recommend other "simple maths" QM computing references, particularly any that might address deterministic vs non-deterministic models/theory?
Hi Jim, it's great to hear from you. As you will have discovered, in this video I have focused on the Deutsch-Jozsa algorithm (a.k.a. Deutsch oracle) which I said was 100% deterministic. And so when we perform the measurements to get actual binary outputs, there are no cases where the qubit is in state where the probability of the reading being a 1 or a 0 is anything other than 100% or 0%.
I've not tackled the much more complicated cases such as Shor's algorithm for prime factorisation. The advocates of QM believe that via superposition and entanglement a huge amount of different paths can be tested at the same time, exploring all options and finding a solution in an exceptionally fast time. They claim that the main problem is how to extract the best solution as we are hampered due to the 'collapse' to binary values when we take the readings. This results in a complicated approach to eking out a final answer - all of which I believe is very complicated.
I'm not sure if prime factorisation is the best example for a comparison of deterministic/classical computing versus quantum superposition & entanglement. Ideally we want something much simpler that can be explained in very simple terms. Sadly I don't know of any such simple example.
If there was any convincing evidence that quantum computing could outperform classical computers for optimisation problems or prime factorisation then I would have more interest in trying to understand some of the algorithms used. But since there is no evidence I can't create any interest in trying to analyse them.
I made this video quite a long time ago and I can't remember reading anything that was particularly good for explaining non-deterministic algorithms. But if you manage to find something then please let me know.
Your doing the lords work here sir. I'm sure your position will stand the test of time.
Exactly! wtf is the scientific community playing at? I suppose its all about jobs as usual.
p.s David Deutsch believes in many worlds theory - lol.