Quantum computers

Are these or can these be real? Every time I try to read how they work, I run out of brain.

Is there a simple explanation somewhere, or is it a combination of wishful thinking and scam?

Cheers

--
Syd
Reply to
Syd Rumpo
Loading thread data ...

Neither. The basic idea is that an ensemble of N qubits can interrogate

2**N quantum states at once even if they're just 2-state qubits.

Thus in principle the calculation becomes exponentially fast as N increases.

Personally I wish a plague on all their houses, because the one important application of quantum computers is breaking strong crypto, which would be A Bad Thing.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

I can only help with my very personal opinion:

In the very old times there was the BIT: 1 or 0 (things are or NOT are) simply no doubt dot,

then about in the '80 born the FUZZYBIT, reality is not only BW but with shades of grey a temperature maybe freezing - cold - warm - hot and so-on

then in the 90 born the QBIT the quantum bit where the bit is simultaneously 1 and 0 the consecration of the caos (positive: the caos is not so bad (I think of my desk disorder)) so I'm simultaneously married and single, male and female, happy and sad (probably a better description of the new times...).

The tecnology made computer ever faster but there is not a totally new architecture concept, so search for novelty is good, imagine the ram in a computer accessed word by word moving an address register, the quantum address register "scan" every address in memory (QRAM?) simultaneously! A tremendous computing power multiplication.

Difficult will be extract from that "noise" the wanted answer, maybe the Quantum computer will give You every possible answer (good and bad) how to choose the right? this will be a job for the next generation of computer (see about the GaAs computer and the genial Seymour Cray) someone must think after bit , fuzzybit, qbit how to name it.

bye delo

"Syd Rumpo" ha scritto nel messaggio news:ms1sa7$tro$ snipped-for-privacy@dont-email.me...

Reply to
delo

OK, I sort of 'understand' that a photon can be in two different polarization states at once, for example, but how does that relate to doing a calculation? How does or can 'an ensemble of N qubits ... interrogate 2**N quantum states'? What is the nature of this interrogation? What does it mean? What does it look like?

This is my problem. I understand 'fully' how transistors make gates, how gates make logic, how logic makes memory and processors and how Gates makes billions. I also realise that a caveman wouldn't understand this. Am I a caveman witnessing magic?

If they could break strong crypto, the corollary would be no crypto of this type, because it would be pointless. Maybe long term that wouldn't be such a bad thing.

Cheers

--
Syd
Reply to
Syd Rumpo

If the interactions between qubits are small, a single quantum state of the whole system is the product of states of the individual qubits. (This is just separation of variables as in elementary PDEs.)

Thus a system of N 2-level qubits has 2**N states, all of which can be contributing at once.

How you make algorithms out of this, I don't know, but I briefly collaborated with a couple of the gurus of the business, so I'm confident that you can. (An analogue computer, I'd believe.) ;)

Well, if that happens, we'll either all go back to paper and pencil, or else send all our money directly to the Russian mob just to avoid the suspense.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

It represents every possible number that could exist in that number of bits simultaneously. You get to ask it a question and for the right choice of question you can factor composite primes very very fast.

The Wiki entry isn't too bad although you would need to read Deutsch's papers and Shor's factorisation algorithm to see why it could be earth shattering for codebreakers if it can be made to work.

It is hard to know if classified secret quantum computers with this capability already exist. It isn't beyond the bounds of possibility. The WWII Colossus computer was so good at what it did that only within the last decade did general purpose CPUs become fast enough to rival it.

The serious players would switch to quantum encryption where you can tell if someone has attempted to read your message in transit.

One wierd quirk of quantum computing is that if it is really possible to build a quantum computer with a non trivial number of bits then it becomes quite likely that we are living inside a simulation.

At a hand waving level quantum computing for simple comparisons has some interesting properties that will make sense to electronics engineers.

Binary logic in an if statement has two basic states 0 and 1. Quantum logic in an if statement has four basic states A,C,G & T (and so does DNA). Wavefunction collapses to a match in a single comparison.

When you nest binary logic statements with each successive level of nesting you get 2^N distinguishable patterns. Ie 2,4,8,16 ...

In quantum logic with three comparisons you get 21 (ish) outcomes which is coincidentally the same as the number of amino acids coded for by DNA triplets. It could all be coincidence but it is highly suggestive.

--
Regards, 
Martin Brown
Reply to
Martin Brown

Why? If "everyone" can "break" any (government) code, would seem to me that the government's insanity might be broken, permanently.

Reply to
Robert Baer

Well, take a problem, bit-by-bit...

Reply to
Robert Baer

It will be a very long time before quantum computers wind up in PCs or phones, if they ever do, so decryption capabilities will continue to be very asymmetric for the foreseeable future. (Probably nobody's cell phone will ever contain a helium cryostat, for instance.)

Thus only governments and large (mostly criminal) organizations will be able to afford crypto-breaking. And the tyranny of administrative law we now live under is likely to get even worse, so even if you could afford a quantum computer, if they found you ferreting around in the NSA's systems, you'd be begging for the death penalty before they were done with you.

IOW, our situation will be the same as it is now, except that we'll be as defenseless as Frenchmen. (Crypto is illegal in France.)

If we ever go to internet voting, we'll have handed our last freedom to the spooks. They can trivially penetrate any system the rest of the govt might design, so they'll have the power to elect whatever government they like, unless the Russians or the Chinese beat them to it.

As the old military maxim goes, "Base your tactics on the enemy's capabilities, not his intentions."

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

What form does this product take?

How do I get the output? Where do I get the output?

Sorry, Phil, this is getting me nowhere - I must have some sort of quantum mental block (qublock?). Let's say I wanted to make an adding machine. Just integers, say 0..9 plus 0..9 to give (with luck) 0..18 as an answer. How would I do this with a quantum computer?

Maybe that's trivial, I really just don't know.

Cheers

--
Syd
Reply to
Syd Rumpo

I think your phone or hand calculator is enough to "QUANTIFY" that problem. Hell, even I can "QUANTIFY" that in me noggin. Really, I think people have gottne soft in the brain, those terms only impress the unknowing, and from the looks of it, there are quite a few!

Jamie

Reply to
M Philbrook

Just like it sounds: you multiply them together.

Details vary depending on the scheme.

Dunno. Quantum algorithms aren't my long suit. They're completely different from classical ones, though.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

nice... I was a "fly on the wall" for a conversation, I assume has happened (at least) hundreds of times... to paraphrase "Well maybe", said some smart guys. "We have to know", said Darpa.

George H.

Reply to
George Herold

France ended its most problematic anti-encryption laws in 1999...

(I agree with the rest of your argument.)

Reply to
David Brown

It is a case of great theory, but the practice is wishful thinking. In theory, quantum computing can let you scale some algorithms because quantum bits (qubits) can hold multiple states at one - 2^n states for n binary qubits - and the quantum computer can operate on all those states simultaneously. Think SIMD taken to extremes.

In practice, however, things are rather different. The more qubits you have, the less stable the system and the more interference between them. The coherence breaks down, and you are unable to get data into or out of the system. So saying that a 6-qubit quantum computer can work with

64 states at a time is a bit like saying an analogue computer can do 64-bit addition with an op-amp - all you need is a 64-bit ADC!

The current record for quantum computers (ignoring for the moment D-Waves semi-quantum systems that work in a significantly different way) is IIRC to factorise 21. We don't need to worry about breaking encryption any time soon.

(There have been claims that these factorisations also factored a series of larger numbers as a sort of by-product. But that is a bit like saying your "Hello, world!" program also calculates the reverse of the string "!dlrow ,olleH" - true, but not particularly useful.)

While I expect quantum computing to improve, I do not expect there will ever come a point where a quantum computer can beat conventional computing on /any/ of cost, speed, size, power, accuracy or ease of use. Thus they will remain of academic interest, along with neural networks, DNA computers, and other fads.

The only related topic where "quantum" is actually used is for quantum encryption. This is sold on the basis that it is possible to detect if someone has hacked it. It turns out that this is not true - there are ways to gain information from the line without being detected. And it would be vastly easier, more reliable, faster and cheaper just to use bigger numbers in your RSA encryption and /know/ that no one has hacked it, rather than finding out /if/ someone has hacked it.

Reply to
David Brown

The solution wavefunction ends up in the register you read back.

Think of it as casting the problem in such a form that only the solution interferes constructively and all the non-answers cancel out.

You read it back - forcing the superposition of states onto a solution.

You wouldn't bother to do something so simple with a quantum computer.

The sort of things where it excels are searching and factorising.

Grovers algorithm and Shor's algorithm respectively.

formatting link
formatting link

There aren't any really accessible article on this topic. This is about the closest I can find:

formatting link

I recall seeing one ages ago that enumerated the quantum branching factors for small numbers 1 through 4 nested if statements but I can no longer find a copy of it online.

--
Regards, 
Martin Brown
Reply to
Martin Brown

You might find this interesting:

NIST recently changed its guidance regarding the (non-quantum) "Suite B" cryptographic algorithms.

"... will initiate a transition to quantum resistant algorithms in the not too distant future. Based on experience in deploying Suite B, we have determined to start planning and communicating early about the upcoming transition to quantum resistant algorithms. Our ultimate goal is to provide cost effective security against a potential quantum computer."

formatting link

Regards, Allan

Reply to
Allan Herriman

What I think is a much more interesting question: how do you program a quantum computer? How does it store its instructions, and how are they sequenced?

What I have read about them up to now more or less suggests the task it performs is hardwired, like with an analog computer. We have all become accustomed to "stored program" computers, which in the general case are much more useful.

Reply to
Rob

Could you do something like a PCB autorouter. The quantum computer tries

*every possible* combination of routes and chooses the actual optimum (according to set design rules).

Or more generally any circuit design problem I suppose.

Or...

--

John Devereux
Reply to
John Devereux

I think we're all getting mightily embezzled. Quantum effects only show in experiments that are repeated many, many, many, times. You either have to arrange for a very large number of parallel operations or for a similar number in succession to see a result emerging. Where is the miracle in that? It's just statistics.

Jeroen Belleman

Reply to
jeroen Belleman

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.