Intel Delivers 17-qubit Superconducting Chip with Advanced Packaging to QuTech

[ty_ger]

Maybe I'm living under a rock or something, but I've yet to see any of these quantum computers actually "doing" anything. Anybody have a link?

Nope, you are not living under a rock. Various people have various excuses for why we haven't made much real progress; such as blaming computer scientists for using the wrong approaches, or not yet having an effective quantum programming language, or quantum computer manufacturer's lying about their chips' capabilities.

I think that this is quite enlightening and relevant without all the finger pointing:

Skip to 5:20, if you wish.

 

[Prima.Vera]

Makes one wonder how these things would become a household item with that extreme cooling needed.

They won't. Why do you need a quantum computer in your house?? This will be so strictly regulated by the Governments that even a Private corporation won't be allowed to own those...

 

[ty_ger]

They won't. Why do you need a quantum computer in your house?? This will be so strictly regulated by the Governments that even a Private corporation won't be allowed to own those...

Agreed. Having the the theoretical power to easily break any public key encryption system would certainly need to be heavily regulated.

 

[FordGT90Concept]

They won't. Why do you need a quantum computer in your house?? This will be so strictly regulated by the Governments that even a Private corporation won't be allowed to own those...

Private companies have more D-Wave computers than the government does at present. I think Google has several by itself while ORNL (US government) only just got one.

 

[ty_ger]

Private companies have more D-Wave computers than the government does at present. I think Google has several by itself while ORNL (US government) only just got one.

But none of the Quantum Computers work yet. So............

If/Once they do work, there would suddenly become a serious threat to the security of the infrastructure which our modern society is built on, and heavy regulation would be forced, wanted, and even welcomed by the majority.

 

[FordGT90Concept]

D-Wave does work. The latest model is guaranteed to have 2000+ functioning qubits.

D-Wave 2000Q quantum computers are available this quarter for shipment, with systems also accessible to subscribers remotely over the internet.

D-Wave is the leader in the development and delivery of quantum computing systems and software, and the world’s only commercial supplier of quantum computers.

 

[Rehmanpa]

But the biggest question that no-one has asked yet is this, will it run minecraft??

 

[ty_ger]

D-Wave does work. The latest model is guaranteed to have 2000+ functioning qubits.

Garbage. It is not a real general purpose quantum computer. It can only find solutions to a small group of types of problems; and with very questionable performance.

D-Wave's "Quantum computers" are NOT general purpose quantum computers. They can only do quantum annealing, which allows a small subset of problems to be solved. They can't run Shor's or Grover's algorithms, as these aren't quantum annealing problems. It's also still an open question whether D-Wave's machines even provide any speedup over classical simulated annealing systems.

 

[FordGT90Concept]

Because qubit processors are naturally slow at processing binary data. It doesn't make sense to buy a multi-million dollar specialized processor to solve 1+1. Qubit processors excel in areas where deep branching is necessary. If I understand it correctly, the 2000 qubit model could solve 2000 if...then statements nested in each other in a few cycles. A traditional binary processor would take thousands of times longer with significant memory overhead to solve the same problem.

 

[Bytales]

Agreed. Having the the theoretical power to easily break any public key encryption system would certainly need to be heavily regulated.

What will be the Point of "encrypting" files, if goverments will be able to use their regulated Quantum Computer Clusters to break the encryption whenever they please.
Might as well not encrypt at all.

 

[Frick]

A qubit is very well defined, like a bit. A bit is true or false while a qubit is true, false, or both. They may take different approaches to creating a quantum processor but a qubit is a qubit. Intel could have a vastly higher cycle rate than D-Wave does but, at this point, I doubt it.

Sure. But as with bits, there are a myriad of ways to get that qubit to do work, and because of quantum magic it's really hard to verify their operation. In a decade or two when we know more we might be able to evaluate performance and different architectures, but until then it's apples and oranges.

 

[Basard]

I seem to remember, in the late 90's, seeing something on the discovery channel about these.... Of course my memory has been clouded over the last fifteen years.... But progress remains the same. Qubits "could" do this or "may be able to" do that... yadda yadda... We never had this much trouble with transistors.

I say that if you're going to go through all the trouble to cool these things down to near absolute zero, then you may as well just throw a binary chip in there and clock it to a billion terahertz.

 

[bug]

I seem to remember, in the late 90's, seeing something on the discovery channel about these.... Of course my memory has been clouded over the last fifteen years.... But progress remains the same. Qubits "could" do this or "may be able to" do that... yadda yadda... We never had this much trouble with transistors.

I say that if you're going to go through all the trouble to cool these things down to near absolute zero, then you may as well just throw a binary chip in there and clock it to a billion terahertz.

Of course we didn't have this much trouble with transistors. Transistors weren't doing anything new, they were simply taking over the duties of tubes. A qubit is, among other things, a new type of "electric" component. And we do need to figure out how that influences designs, logic and yes, what type of work they can help with.

 

[Boosnie]

On the subject of cooling there are already devices called cryocoolers that are the size of a small thermos and can reach cryogenic temperatures.

you forgot to check what wikipedia says for "cryogenic temperatures".

 

[Boosnie]

Flight was impossible for millennia

Pleas, for the sake of my sanity, stop with this nonsense.
Cooking was impossible for millennia
Sharpening a stick was impossible for millennia
Rowing a boat was impossible for millenia.

We are not talking about building a relatively simple device that will be readily mass produced and available in a not too distant future.
We are talking about pinpoint an exact basic particle out of a gazillion others and keeping it masked from another gazillion continuously zipping through and around, perturbing the em field ad destroying the measurament.
This requires a crapton of energy and you will be never, ever, ever, never be able to do it on your lap on your couch.
The energy bill will not change, ever. Maybe you can devise clever methods to do it, but energy is energy, what you put in is what you get out.

 

[Basard]

what you put in is what you get out.

Until you add the human factor.... Then you get bitcoins.

 

[FordGT90Concept]

Sure. But as with bits, there are a myriad of ways to get that qubit to do work, and because of quantum magic it's really hard to verify their operation. In a decade or two when we know more we might be able to evaluate performance and different architectures, but until then it's apples and oranges.

Not really. Any logic processor can be used to do the same work a quantum processor can do but there's clear performance advantages for specific types of problems. Quantum computers are very much ASIC. Like I said, for simple math problems CISC/RISC will be far more economical. Quantum computers for solving problems that CISC/RISC struggle with (like deep branching logic). In those tests, Qunatum computers are millions of times faster. For companies and governments that need that kind of compute power, the price is worth it. This one relatively small machine (about the size of an upright fridge) can do as much compute work as a massive cluster super computer without complex code to distribute the work.

Deep Wave requires 25 kw of continuous power. That's about as much as your average electric furnace. This is not something people will be using at home any time soon.

 

[ty_ger]

What will be the Point of "encrypting" files, if goverments will be able to use their regulated Quantum Computer Clusters to break the encryption whenever they please.
Might as well not encrypt at all.

Umm.... ... .. . .. . your password is sent from your computer to the server where it is verified.

Maybe the "point of Encrypting your connection" will be to keep 7 billion other humans on this earth from draining your bank account every second of your life. I said infrastructure, not files.

If a general purpose quantum computer is ever physically possible, either it will need to be an extremely rare and heavily regulated item, or some genius will need to find a brand new way of securing our $19 trillion dollar infrastructure to replace the current method we use which was found as a mathematical law back in the year 1640; and every website, server, and service everywhere on the planet will need to change the way that they send, receive, and verify password information both server-side and client-side.

 

[FordGT90Concept]

Quantum entanglement has a solution for that too:
https://www.livescience.com/60645-quantum-encrypted-message-sent-between-buildings.html
https://www.sciencenews.org/article/china-quantum-satellite-adds-two-new-tricks-repertoire

 

[ty_ger]

Because qubit processors are naturally slow at processing binary data. It doesn't make sense to buy a multi-million dollar specialized processor to solve 1+1. Qubit processors excel in areas where deep branching is necessary. If I understand it correctly, the 2000 qubit model could solve 2000 if...then statements nested in each other in a few cycles. A traditional binary processor would take thousands of times longer with significant memory overhead to solve the same problem.

You are massively over-simplifying things. No, it isn't simply that these d-wave processors are unable to process binary data. The problem is that these d-wave processors are unable to perform the majority of types of quantum calculations. Therefore they "don't work". The d-wave processors are only able to perform a small subset of types of quantum calculations. They are only able to perform optimization-type calculations where the best solution to a problem is the output. They are unable to perform the holy grail of quantum calculations where every possible outcome is revealed at the same time; unless they test every possible outcome individually at a speed which is just as fast or slower than any other standard general purpose computer.

 

[ty_ger]

Quantum entanglement has a solution for that too:
https://www.livescience.com/60645-quantum-encrypted-message-sent-between-buildings.html

Yes, in 2015, I have read about quantum encrypted message tests. Until a solution is available worldwide, until the solution is refined to the point that information can be reliably transmitted or received from anywhere in the world, and until the entire infrastructure is converted -- at the cost of a massive amount of money -- , only a very few will be allowed to own a general purpose quantum computer. That is of course assuming that a general purpose quantum computer will some day exist.

 

[Prima.Vera]

Just develop a standard binary supercomputer of at least 1 Exaflop performance, and make it design a new Quantum Supercomputer technology.

 

[Manoa]

I wanne this quantum, it the mhz
for quantum ray tracing video card
I wanne see what monsters realy looks like lel