This is one of the most illuminating and inspiraitonal videos on quantum computing. Mr. Sank seems so passionate about the subject and excited to share the knowledge with people who would like to know the fundamentals of quantum computing. Thank you~
I’m an electronics technician not an engineer, so this is definitely the most easy to understand video I’ve seen on quantum computers. I’m assuming one day , Q computer technicians will be needed so I’m trying to get a basic grasp of it. My question is - how many computations does the computer perform before an accurate answer can be obtained using probability?
I'm still confused how logical operations are performed and the results observed. A good amount of time of this video was spent focusing on explaining the protection the qubit circuitry from any interference, even so much as a photon, but then it was said that logic operations are sent to the qubits for processing, and no explanation was given. (1) How are the logic operations sent to the qubits for processing without changing the qubits into an unknown state? (2) And how are the results observed without altering the answer? Observation involves interaction with the qubit, even if it is so much as a photon or a very sensitive electromagnetic field sensor that responds by sending out a signal (which itself generates a new electromagnetic field), and if the qubits are supposed to be protected from such things, how are answers observed?
Had to keep it to around ten minutes and couldn't cover everything. We'll get to control and readout at some point. But in short: single qubit controls are microwave frequency pulses for bit flips and lower frequency pulses for frequency shifts. Readout is a bit more complicated and is described in publicly published literature, but for a pedagogical discussion you could try my PhD thesis: web.physics.ucsb.edu/~martinisgroup/theses/Sank2014.pdf
For (1): There are microwave generators that send microwave pulses down control lines connected to the qubits. Different pulses have different effects, and within the set of possible effects there are quantum gates such as the Hadamard gate. Quantum computation is basically just the application of a series of known gates in a specified order, so in principle these pulses are sufficient to perform any computation. In practice there is sufficient noise that the computations are still limited to short times; for very long running computations error correcting codes such as the surface code will be needed. For (2): In the simplest case, measurement always comes at the end of the computation. So it doesn't matter that you've caused damage that would prevent the computation from being continued further. In more complicated cases, you need to be careful to measure only the bare minimum of what you need and leave everything else unaffected.
This video was excellent. I can say that I have a very basic understanding of quantum computers. How can we observe superposition without changing it? Due to the cooling?
Amazing video! thank you! well explained. I think its great to learn more about how errors get produced! We should have a think tank for the subject. Qubit fab is the future. I can't wait to hear more from the team regarding its progress. Also what classes / degrees are required to get into this type of work?
How noise can be reduced by Quatumcomputing, perhaps through using ledsignals by transmition of qubits. Just a thought Thanks for explanation and kind regards.
Thanks for the video Daniel, If I got it well, you are proposing a new chip (8 times colder than space) simulating the operations of Cryostat. The superposition is ensured by the Superconductivity at steady rate, but we could have some errors. My questions: Does this chip also manipulate true electrons inside for the superposition states (as the Cryostat), or States are hardcoded in the simulation? Is it possible to use this chip to create small quantum computers (like PCs) and make them available for the public? Correct me if I made a mistake, please.
2:57 Shouldn't the red arrow from (000,101) leads to (101,000) not (100,000), just as blue arrow (000, 100) leads to (100,000)? Sorry for my lack of location notation, but i hope you understand my question.
Cool video, thanks! Would be great to know more about the current challenges you guys are facing, maybe in a follow-up? Also would be really interesting to see a lab tour type of video. You mentioned improving the superconducting material purity - is that a part of the research you do? So it's not just software / hardware development, it's also material science and physics? Sounds very interesting.
Yes, materials science, fabrication methods, and the associated physics is critical to our field. Regarding follow-up videos going into more depth on the challenges we're working on: that's exactly what I had in mind when scripting this video. I wanted to bridge from the computer science concepts to the hardware so that in future videos we can talk about how logic gates work and what kinds of things go wrong on the physical level. I recently got permission to release a slide presentation that may be helpful: docs.google.com/presentation/d/1UsXVoqzu9rnpwiC3nN7YVG2YA9mvHgM84KH_JMXwFJE/edit?usp=sharing
I was hoping for something that goes into the physical representation of qubits in a bit more detail, more youtube vids here I go. Unlike suggested by the diagram at ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-uPw9nkJAwDY.html the energy levels between 0 and 1 and 1 and 2 are not the same, and that is important because we don't want it to go to level 2 when we add energy, we only want 0 and 1, see e.g. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-t5nxusm_Umk.html
Micro-fab begin folding those metals atleast 22 times like a samurai sword as to remove imperfections, or ions (slag). If their ions are removed the q-bit won't have opportunity to flip. If I understand correctly you just want clean electricity to flow through no hurdles & aren't their water cooled computers.?. Tie that in some how. LoL I know nothing just really curious how it all works. I do electrical work (shade tree electrician) but point I guess like from transmission lines to service lines into the services of a house to an AC adaptor to a circuit board to services of what ever using. Accept dirty electricity exist & that's what I feel your saying you have to remove, and I also think your guys cryostats are cool. Like really really cool.
Why can't we make a quantum cube which doesn't interact with the other particles? Instead of a flat chip, we can make a closed cube inside which qubits are built.
HERE IS THE BIGGEST LIE: 8:13 "RELATIVELY STANDARD COMMERCIAL PRODUCTS", hahaha. THEY ARE SO STANDARD THAT, SO FAR, NOBODY HAS COME UP WITH ANYTHING USEFUL.
