/sci/ - Science & Math

it's not a meme, but it's not just a fast computer. The only people that would ever use it would be researchers and governments.

you can run on a simulated one from IBM and I think Microsoft has one too.
The IBM one you could by cpu time for a real machine, it has like 4 qubits or something.
go give it a shot.

>>11349777
Because it is grant farming and nothing but a money sink.

>>11349777
Quantum computing was an idea a theoretical mathematician came up with in the 1960s, with its existance relying on technology to accurately analyse theoretical particles like photons and analyse their "spin state" or generate twin particles and use quantum entanglement tondo faster than light communication etc.

The problem is, we can barely observe an electron without changing its properties in some way, because the only way to do that is hitting it with another electron or taking some of its energy or changing its trajectory.

It's called "Heisenberg's uncertainty principal" which is the principal that you cannot analyse something that small with any certainty, you can only calculate what it's probably doing. Hence the electron probability cloud.

So if we could make a twin electron pair at the Super collider, and put it in a magnetic bottle, and put the other entangled electron at another lab and do FTL communication as the change in one electron is instantaneous with the other, you could stimulate one electron to a higher state, the twin emulates it, analyse the second electron and transmit binary at FTL,

The problem is we cannot accurately even observe the electron though, so the transmission particle, and reciever particle communication cannot be interpreted because the data is heavily corrupted,
Now with a computer you cannot send randomised binary messages with lot of errors, that's why quantum computers are referred to as "noisy" it's not noise lol, it's a random number generator.

So mentally ill scientists are trying to come up with a way to have a "quantum" decryption computer at the reciever end to filter out the noise using quantum magic.

So long story short.
We need to build a working computer, that uses another working computer to work, and both computers need to be working, to get the other computer working, and they both have to work before they work ;)

It's called mental illness.

>>11350066
Just for real though.

A quantum computer is a computer that runs an algorithm where you can send it any string of random binary code right.
Say 100011111101010101010010101

And that computer can translate that into any file you want of any size, so it could convert the binary I just typed into the video you accidently deleted, that doesn't exist anymore, because this is quantum right, and half the computer is in another dimension, so it plucks the string of binary from the multiverse where you didn't delete it, or goes back in time and gets it before you deleted it.

I'm so sorry, but this is madness.
That's literally what a quantum decrypter is supposed to do.

>>11349777
>>11349872
https://www.ibm.com/quantum-computing/technology/experience/

>>11350066
>>11350080
No that isn't what a quantum computer does, and you're mistaking entanglement (and you fucked that up too) for quantum computing

Its essentially just exploiting additive and subtractive qualities of waves to reinforce the correct answers to a given question and using destructive interference to weed out incorrect ones.
Its only efficient for a limited set of problems like factoring numbers and solving some graph based problems. The hype comes from current encryption methods being reliant on the difficulty of factoring large numbers.

Entanglement can be used for encryption though, but not transmission of data efficiently, and definitely not FTL. While the quantum state does "update" instantly, that change is indistinguishable from random noise by the receiver without the eigenvalues from the sender, which have to be sent classically. Thats where use as an encryption method comes in, those eigenvalues are useless without the quantum state and vice versa. Tampering with the state is impossible as well, as it'll decohere and become useless to the receiver at that point.

>>11349777
anyone read this book? it was just on my mind desu

>>11349777
This board is full of undergrads and retards who feel clever when criticizing common and well-researched ideas.
Quantum Computing is perfectly possible and performed today. Do it yourself via IBM's web interface if you like. D-Wave and Rigetti opened up as well.
We had two quantum computer in our basement when I was still at my Alma Mater.

>>11350402
How does the transistor work?
What's it made from, how does the bios work?
How is the data entered?
What can we use this for?
Where can I look at the programming language?
What is an example of a quantum algorithm ? Like in the programming language you enter into the quantum computer?

>>11350543
They're not transistors at all.
Mix of Silicon, Copper, Tantalum, and Silicon Oxide iirc, its far more complex than that but those are the basics.
BIOS isn't really applicable here, this isn't a home computer. Think of it more as a DSP.
A seperate conventional computer sends some signals to it. Its not like theres a keyboard n screen hooked up to one.
Unironically most of the work is done in Python right now. Here's an example if you wanna look: https://cirq.readthedocs.io/en/stable/
Check the tutorial, you can see a few examples.
This comes with a pretty complex simulator but can also interface with an actual quantum computer.

>>11350066
>man who claims QM is a Vatican Conspiracy also claims QIS is a mental illness
Well, at least you're consistent. If anyone in this thread actually wants to ask someone that works on building quantum computers a real question, fire away.

>>11350402
>responding to it
Come on anon. Just say no.

>>11349777
Nice trips. Quantum computing isn't a meme yet. There are legitimate reasons one might believe a proper quantum computer (fully fault-tolerant) could never be constructed. Unfortunately, very few on this board understand the field enough to actually know those reasons, so people here just cite "it's a conspiracy for grant money" and related shitposts. These people have no interest in actually learning any basic physics, let alone the subdiscipline of quantum info.

Personally, I am skeptically optimistic. I see no *fundamental* reason why we can't have an error-corrected device in the future, but this is far, far down the road. The field has made some great progress, but it's baby steps compared to the overhyped promises that people tend to talk about.

I would love to see one some day, not just because of applications, but for the actual feat of physics and engineering that comes with it. A proper quantum computer is essentially a controllable, macroscopic, topologically nontrivial phase of matter. That's fucking fascinating to think about.

>>11350409
If the previews from https://pragprog.com/book/nmquantum/quantum-computing are any indication, I would strongly suggest you stay away from this book. It looks like one of those crappy "learn how to program using this software" books (in this case, seems like IBM's Qiskit), but 1) you don't need a textbook for that, and 2) it's a waste of time at this stage anyway. If you want to get into the field, you need to learn the nuts and bolts of what quantum actually is, not how to implement Grover's search in Q# on a space of 4 elements with noisy, low fidelity qubits.

Pick up a standard textbook, like Nielsen and Chuang. Or check out John Preskill's notes, they're really fucking good. I also used "An Introduction to Quantum Computing" by Kaye, Laflamme, and Mosca in undergrad, it's decent if you don't want to start with N&C.

>>11350809
Oh hey! Someone that sounds like they actually think about QIS.

I think one of the most interesting things at the moment (especially with universal digital ec'd ft quantum processors such a ways off yet) are quantum simulators. Even if it's still an open question as to whether or not you can do anything meaningful with them sans error correction that you can't do with a classical computer, it's an interesting frontier to explore.

>>11349777
>I had this conversation talking with a pal at … a nice restaurant near one of America’s great centers of learning. Our waiter was amazed and shared with us the fact that he had done a Ph.D. thesis on the subject of quantum computing. My pal was convinced by this that my skepticism is justified; in fact he accused me of arranging this. I didn’t, but am motivated to write to prevent future Ivy League Ph.D. level talent having to make a living by bringing a couple of finance nerds their steaks.
https://scottlocklin.wordpress.com/2019/01/15/quantum-computing-as-a-field-is-obvious-bullshit/
But more seriously, the error correction needed is probably going to always outweigh any potential benefits that class of approaches to computation would otherwise have. And 99.9999% of anyone excited about quantum computing is realistically just a suggestible faggot who thinks "quantum" is a cool and intelligent sounding label to prefix shit with.

>>11349777
They are a meme the same way fusion is. Something that has the potential to revolutionise the world, has no meaningful results (fusion consumes more energy than it produces, quantum computers have no real world use yet) and pop-sci loves them.

Someday they may become something other than a meme, but today's not it.

>>11351664
>fusion... has no meaningful results
Hydrogen bombs are pretty meaningful.

>>11351664
>no real world use
You have no idea what you're talking about.

>>11351692
What practical use does today's quantum computers have? I'm super curious now.

>>11350066
keks

>>11350653

By when do you and your peers estimate to have a 'working' quantum computer? What will this computer be most useful for, in terms of applications? Will conventional encryption be obsolete against quantum computing power?

>>11351802
>'working' quantum computer
If you mean a universal digital quantum computer with all the bells and whistles? I have a running bet with a few of my peers, and the year I put my money on was 2039. Anyone that tries to tell you anything less than a decade out right now is deliberately hyping you or has little working knowledge in the field.

>most useful for, in terms of applications
I'm actually not focused on looking to the applications for that full-on device - it's still a ways into the future. We have devices today that cannot be exactly modeled by a classical machine with sufficient controllability to shed light on chemical and condensed matter phenomena that we have literally no other way of directly studying. I happen to think the most transformative potential for quantum processors in, say, the next three decades are in these areas of quantum simulation and quantum chemistry. And if any of the problems that it looks like we may be able to shed some light on - like calculating reaction rates in the Haber process - come to pass, they will change the way materials and chemical engineering work. And we have devices in labs today that might already be used for some of these problems.

> conventional encryption
If we get the full-on quantum computer, yes. Yet, we already know how to make networks that quantum computers can't break. We've started to build them, and given how much time we have until we are likely to see any device that actually could break encryption be made I would wager anything worth protecting will already be out of reach by the time those secrets are earnestly threatened by a quantum computer.

It's hyped because it's potentially "scary" and grabs peoples attention, but this is kind of the most boring / trival part of quantum computer applications. The algorithm is simple and provable, which is nice, but it's close to the least inspiring part of quantum information science in my book.

>>11349777
Jokes on you, I am working on such a project myself right now. The idea is to create a lattice structure of superconductive walls and topological insulator on the bottom to trap a Majorana Fermion which will be the base for quantum computer.
Whether this is a meme or not I can tell you after I finish my work :^ )

>>11352060
>Majorana Fermion-based quantum computing
>anything but a meme

>>11351993

I have nothing intelligent to contribute to this thread myself just wanna say thanks for posting!

>>11352093
Hey, if it's suddenly a success I get the credit
If it's a meme, I still get my thesis done :^)

>>11352126
Hey man, I'm not saying it's not worth trying stuff that seems ridiculous or impossible - we do lots of cool stuff that way - but using that as the physical layer for a quantum computer is definitely a meme. Hopefully you learn some cool and useful stuff along the way - there are plenty of other platforms that are similarly bad candidates but we've learned a ton from trying anyway.

>>11351664
Buddy, fusion research is insanely exciting right now, esp with things like SFS Z-pinch and FRC projects moving towards net energy prototypes in the next 3 to 5 years.

>>11351618
I'm biased since I basically work with Hamiltonian simulation and related ideas, but yeah, I agree. It will be interesting to see what sorts of models we might be able to either simulate time evolution explicitly or calculate some useful quantities. I just recently saw a talk where they showed (theoretically and experimentally) that certain observables can be estimated to much higher precision than the actual state fidelity might otherwise indicate. There's a certain robustness here that might make pre-FT devices, if not useful, at least interesting.

>>11351993
Pretty based, I hope we'll be able to simulate FeMoco within 2 decades, but again I lean towards skeptical optimism. Currently I just don't see a practical way of studying a ~100 qubit system on a NISQ device, and if we expect to have some level of fault tolerance by then, there best approach I know of requires something like 3000 logical qubits IIRC (I think they uaed qubitization to avoid some of the insane costs of magic state distillation).

>>11349777

The only possible progress in quantum computing is information. Syntax error. There is no real world application for it besides holding slaves. What we have today is as useful as it will ever be 1000 years from now.

too stochastic and unstable.

>>11349875
Third post best post.
Ask the physicists in your uni department, all those not working on quantum computers will tell you it's a meme.
Biggest one since climate change.

>>11349779
>when /sci/ thinks it's /g/

You're wrong. I want a quantum computer so I can break your encryption.

>>11350066
The mathematician was theoretical?

>>11352611
is the talk online somewhere?

>>11353204
No, it was just a weekly talk within my group. I don't even think the preprint is up yet.

>>11353158
Once QC breaks all the forms of encryption it can break we could all just swap to any other kind that it can't and carry on as normal

>>11354062
This. Or double key length.