Google Unveils 72-Qubit Quantum Computer With Low Error Rates (tomshardware.com) 76
An anonymous reader quotes a report from Tom's Hardware: Google announced a 72-qubit universal quantum computer that promises the same low error rates the company saw in its first 9-qubit quantum computer. Google believes that this quantum computer, called Bristlecone, will be able to bring us to an age of quantum supremacy. In a recent announcement, Google said: "If a quantum processor can be operated with low enough error, it would be able to outperform a classical supercomputer on a well-defined computer science problem, an achievement known as quantum supremacy. These random circuits must be large in both number of qubits as well as computational length (depth). Although no one has achieved this goal yet, we calculate quantum supremacy can be comfortably demonstrated with 49 qubits, a circuit depth exceeding 40, and a two-qubit error below 0.5%. We believe the experimental demonstration of a quantum processor outperforming a supercomputer would be a watershed moment for our field, and remains one of our key objectives."
According to Google, a minimum error rate for quantum computers needs to be in the range of less than 1%, coupled with close to 100 qubits. Google seems to have achieved this so far with 72-qubit Bristlecone and its 1% error rate for readout, 0.1% for single-qubit gates, and 0.6% for two-qubit gates. Quantum computers will begin to become highly useful in solving real-world problems when we can achieve error rates of 0.1-1% coupled with hundreds of thousand to millions of qubits. According to Google, an ideal quantum computer would have at least hundreds of millions of qubits and an error rate lower than 0.01%. That may take several decades to achieve, even if we assume a "Moore's Law" of some kind for quantum computers (which so far seems to exist, seeing the progress of both Google and IBM in the past few years, as well as D-Wave).
According to Google, a minimum error rate for quantum computers needs to be in the range of less than 1%, coupled with close to 100 qubits. Google seems to have achieved this so far with 72-qubit Bristlecone and its 1% error rate for readout, 0.1% for single-qubit gates, and 0.6% for two-qubit gates. Quantum computers will begin to become highly useful in solving real-world problems when we can achieve error rates of 0.1-1% coupled with hundreds of thousand to millions of qubits. According to Google, an ideal quantum computer would have at least hundreds of millions of qubits and an error rate lower than 0.01%. That may take several decades to achieve, even if we assume a "Moore's Law" of some kind for quantum computers (which so far seems to exist, seeing the progress of both Google and IBM in the past few years, as well as D-Wave).
Several decades? (Score:1)
Re: (Score:2)
No, first we need the improved battery tech that allows for higher capacities and almost instant recharging. Last I checked, it was 3-5 years away from being on store shelves.
Re: (Score:2)
Is your comment replying to the topic of quantum computers? Or are you talking about something else like flying cars?
Why would quantum computers need batteries? Wouldn't they be plugged in into the grid?
Re: (Score:2)
Are they saying that is allowable?
Re: (Score:1)
Actually no. 4cents tips is an insult, it should be at least 40, and that should be rounded to 50 cents. :-P
Re:Several decades? (Score:5, Insightful)
2 + 2 = 4.04
Are they saying that is allowable?
Yes. There are plenty of problems that are extremely hard to solve, but very easy to verify. An obvious example from cryptanalysis is factoring a 256 bit composite number into two 128 bit primes. Who cares if it is wrong 1% of the time? It is trivial to detect and toss out those errors just by multiplying the factors.
Good comment. (Score:2)
Re: Several decades? (Score:2)
that isnâ(TM)t how it works afaik.
these processors are probabilistic, âoe1% errorâ means they have a 1% standard deviation around the correct answer. They never give you the correct answer, but generate enough answers and you can imply the correct one.
Depending on who you ask, you also have to multiply that error by the number of qubits being used to get the actual sd of your result.
They are basically really expensive hardware random number generators with programable bias.
Re: Several decades? (Score:2)
qubit errors are normally distributed afaik.
obviously the distribution of the result is how those distributions combine.
Still just expensive random number generators, which, imho is gibberish to even call âcomputersâ(TM), they are so far from turing complete they wouldnâ(TM)t even count as a co processor from the 90s.
very high bandwidth HWRNG is useful tho, just not for anything they are claiming.
Re: (Score:2)
Re: (Score:2)
Imagine something like this picture, but with ~83,000 of them:
http://www.dvinfo.net/forum/at... [dvinfo.net]
Given people are very frustrated by even a single stuck pixel, 83,000 would b
Re: (Score:2)
The real question is "What are the important use cases?".
If I thought the only use case was breaking net security, I'd say why bother, there are other, cheaper, ways. But it also seems to be good at modeling molecule interactions. What else?
Use cases (Score:2)
You mentioned molecular interactions, and that's no small category when you think biochemistry. Quantum computers may usher in a new age in finding drugs and vaccines, as we will be able to model the chemical processes involved and search for complex molecules that can cause a desired behavior.
Re: Use cases (Score:1)
Re: Use cases (Score:2)
Our current computers, including our quantum computers, work VERY far from the theoretical minimum computational energy.
Re: (Score:2)
It requires nuclear fusion reactors to operate and as we all know, they're 'only' a few decades away at most.
Also see: Optical disks that store 15TB, I'm pretty sure those have been a few years away for 25 years.
Re: (Score:3)
Apparently they're up to 360 terabytes on a 3.75 inch disk.
Ol Musky put one in the glovebox of his roadster!
https://techcrunch.com/2018/02... [techcrunch.com]
Re: (Score:2)
Re: (Score:2)
Nah. Fusion reactors are a decade away, and have been since the 1950s.
Re: (Score:2)
You've also got to understand what "quantum supremacy" means, in effect it means that a quantum computer will outperform a supercomputer emulating a quantum computer solving an artificial problem of no use to anyone that quantum computers are good at but normal computers aren't. Go, quantum computers!
A bit like saying I can beat Mike Tyson any day of the week, provided he's been heavily sedated, clocked with a lead pipe, handcuffed, and tied down. Yeah, look at what a great boxer I am, it's me supremacy n
Re: (Score:2)
Only if it works at all. In actual reality, Physics has always come up with inaccuracies and limits when you go to the extremes. The same will happen here.
Re: (Score:2)
Yup
Re: (Score:2)
Crypto! Most feared subject. (Score:2)
Re: (Score:2)
Again, such news should mention, when this quantum computer will crack typical asymmetric cryptos and all that long-term stored encrypted https dumps with embarrassing photos (yours too!), can be decrypted by Google or NSA.
Is it still the case with Forward Secrecy ciphers?
Re: (Score:2)
Is it still the case with Forward Secrecy ciphers?
Yes. https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
I will readily admit that this is not my wheelhouse, but I was under the impression that Shor's Algorithm would effectively halve the key size. And that it meant the brute force time dropped orders of magnitude, but if FS was used that the key for each message would still need to be brute forced independently. Is this correct?
If there is an underlying flaw in AES, only accessible with quantum computing, I have not heard of it. I would be interested to know if I am mistaken.
Supremacy (Score:2)
The supremacy remark is just supremacy over a classical computer. What a laugh.
We're not in Penrose territory yet.
Re: (Score:2)
Well, if you are laughing, you are in Penrose territory.
Re: (Score:2)
You can tell because d-wave now has 2000 "quibits" machines. In particular, quantum supremacy means you can run an actual algorithm with super-positioned program states (quantum logic, tiffoli gates), not just a fixed equation with superpositioned quibit registers.
What d-wave does is quantum annealing - it has one "hardcoded", specific algorithm it can run. [wikipedia.org]
Only certain linear matrix algebra benefits from fast annealing (of no
Re: (Score:2)
In actual reality, when the inputs scale, the largest thing you can tolerate is O(n log n) or the algorithm is basically irrelevant. Pretty much means that out-scaling Shor's algorithm is not a problem, the numbers just need to get a bit larger.
What can it factor? (Score:2)
Re: (Score:2)
Re: (Score:2)
72
That reminds me of Multics
Re: (Score:2)
That is not a general result. It just says there is one 5 digit number that can be factored this way.
Welcome our new overlords (Score:1)
So they basically can connect a few transistors (Score:2)
This is so far from a demonstration of actual usefulness as a computing device, it is pathetic. An no, there is no "Moore's Law" for QC. About 30 years ago, they were at 4 Qbits. Now they are at 72? Sounds more like a linear scaling or worse to me.
Re: (Score:2)
The madness and the experiments in practical entanglement started back then.
Random Circuits? (Score:2)
Man I am so sick of hearing people who don't know anything about QM talk about QM.