[HN Gopher] Quantum computing's reality check
___________________________________________________________________
Quantum computing's reality check
Author : mathgenius
Score : 105 points
Date : 2023-12-23 17:23 UTC (5 hours ago)
(HTM) web link (spectrum.ieee.org)
(TXT) w3m dump (spectrum.ieee.org)
| coffeebeqn wrote:
| I was recently learning about Shors Algorithm which was kind of
| mind blowing until I found out it can't actually be executed on
| anything right now. We're still in the super early days. Like
| where microcomputers were 40-50 years ago
| ghaff wrote:
| Last I looked expert consensus on factoring RSA-2048 was about
| 15-20 years out but with a wide divergence of opinion. You
| might find a less optimistic take today,
| bawolff wrote:
| I was under the impression the 20 year prediction was for
| classical methods not quantum
| ghaff wrote:
| I can't speak to classical methods but, yes, that was the
| quantum prediction in report a couple years back.
| tippytippytango wrote:
| I'd say we need to go back further, quantum is in the era of
| the Babbage difference engine.
| Strilanc wrote:
| Related: "Quantum computing worst case scenario: we are
| Lovelace and Babbage" [1]
|
| For scale: Babbage's planned analytical engine had a word
| size of 50 digits, a clock rate of 7Hz, and a physical size
| of roughly a locomotive [2]. Contrast [3] where it's
| estimated that 600-digit superposed additions would run at
| 27Hz (by dedicating millions of qubits to magic state
| distillation of the underlying AND gates). Given current
| plans, a quantum computer capable of doing arithmetic
| operations as wide and as fast as the analytical engine would
| probably be larger than the analytical engine.
|
| We can see how to do reliable quantum computation in
| principle. The overhead of error correction makes it daunting
| in scale. It sure would be nice if someone came along and
| invented the quantum computing equivalent of a vacuum tube or
| a transistor.
|
| [1]: https://csferrie.medium.com/quantum-computing-worst-
| case-sce...
|
| [2]: https://medium.com/tech-is-a-tool/building-the-modern-
| comput...
|
| [3] "How to factor 2048 bit RSA integers in 8 hours using 20
| million noisy qubits" https://quantum-
| journal.org/papers/q-2021-04-15-433/
| oneshtein wrote:
| First micro-processor MP944 (1970) had 5360 transistors. (73k
| transistors with memory).
|
| Quantum computer with just 1123 qubits released this month.
| However, it doesn't work properly yet, because large systems
| doesn't work as quantum systems, so error rate is high. Large
| quantum computer is like Eniac at earthquake.
| delusional wrote:
| > We're still in the super early days. Like where
| microcomputers were 40-50 years ago
|
| There's a trap in that sort of thinking. It assumes that it's
| inevitable that we will at some point arrive at current day for
| quantum computers. That technology is a straight path in
| whatever direction we set out. I don't think that is a given.
| There a plenty of things out there that physics just can't do.
| Things that, no matter how much effort and thought and research
| we put into it, we just can't make the world do in a scalable
| affordable fashion. I don't have the knowledge to make a
| believable claim that quantum computers are one of those, but
| you have to consider that it might be.
|
| It's possible that quantum computers aren't at the stage of the
| 8086, but rather at the stage of the flying cars in jetsons.
| Doomed to forever be an unrealistic dream.
| bena wrote:
| You're right, it is not a given.
|
| Progress only looks like a straight shot in hindsight. Yes,
| we can take any invention today and trace every step back to
| the previous to make a single unbroken chain to the invention
| of the wheel.
|
| But for every airplane, there's a Bored Ape Yacht Club NFT.
| qmaybe wrote:
| The roadmap of IONQ has them producing 64 logical qubits with
| error correction within 2025. Whether or not they succeed, I
| think time will tell as theyve stopped publishing and a cofounder
| stepped down to return to teaching and research.
|
| Quantinuum, the Honeywell merger, expects billions of revenue in
| 2026. Honeywell gave up on transmons and joined up with their
| $260m investment into Cambridge.
|
| Speaking of transmons, Rigetti is facing delisting
|
| IBM is still going hard on transmons and selling mainly to
| universities.
|
| Google had sycamore but were still waiting to see if they've
| solved the cascading errors inherent to scaling
|
| Microsoft invested heavily in majorna fermions for compute --
| which might not even exist. They are not an authority on the
| physics
| klyrs wrote:
| What does "expects billions of revenue" mean, precisely?
| qmaybe wrote:
| It means that they are willing to claim to the public that
| there is a business case for their technology providing
| billions in value in that year. They do not have the
| bookings.
| klyrs wrote:
| That's just execs tossing out big numbers about a market
| they don't understand. They do that all the time. Like
| babies, they cannot help but fill diapers. That doesn't
| mean we need to discuss it at the water cooler.
|
| Do they have hardware? Is it available? How many nines? Do
| they have a useful software stack? Does it give developers
| meaningful leverage to solve problems? The most incredible
| thing about quantum hype is that ordinarily detail-focused
| engineer types are getting snowed and buying it because the
| punchline is too good to be true. But sometimes, these
| companies make testable claims: if we'd only talk about the
| externally verifiable ones, and leave the vapor out of it,
| I think it would make for a much more interesting
| conversation.
| qmaybe wrote:
| That's one way to put it.
|
| But more specifically they have customers with use cases
| studied in detail, where a modest quantum advantage
| becomes highly profitable, so all they need to do now is
| ship and the bookings will come in.
| klyrs wrote:
| Sorry, but that's exactly the kool-aid swilling nonsense
| I'm talking about. Did you notice that your "all they
| need" is literally a quantum-computer shaped hole in the
| plan? You're curiously bullish on exactly one brand of
| vaporware. _Why?_
| qmaybe wrote:
| I'm glad you brought this up.
|
| So there's different levels of bullshit right ?
|
| Quantum has a level of bullshit where an enterprise could
| tell a customer their systems will solve P=NP. Clearly we
| don't have information theory that supports that: we can
| only solve BQP better with them.
|
| The industry has customers with use cases proven for BQP
| with restricted quantum computers ready to go. The linked
| article leans to saying that's not going to be true, but
| the companies out there building contradict this and say
| their customers will be ready to pull the trigger on
| billions in quantum compute in the next 2-3 years at
| most.
|
| It could be another Tesla with self driving perpetually
| 2-4 years out. But look at waymo, clearly tech changes
| happen
|
| With what people are accomplishing today certainly seems
| closer to reality. But nobody has really shown us just
| yet
|
| As for "one brand of vaporwave" I'm cynical on
| quantinuum. If their tech was strong they would have
| raised without Honeywell.
| klyrs wrote:
| > I'm glad you brought this up.
|
| Look, I've been on this rock long enough to know when
| smoke is being blown up my ass and this phrase is at the
| top of the list.
| DebtDeflation wrote:
| >their customers will be ready to pull the trigger on
| billions in quantum compute in the next 2-3 years at most
|
| Right. I'm sure these customers would be willing to spend
| billions of dollars on a working quantum computer that
| solves a business problem for them in the next 2-3 years.
| What GP is pointing out, however, is that statement
| presupposes that said quantum computer will actually
| exist.
|
| I would bet money that no such machine will be available
| for purchase in 2-3 years. What will exist in 2-3 years
| are more press releases about new QCs with even larger
| numbers of noisy physical qubits that still don't amount
| to a single fully error corrected logical qubit that can
| factor 35 without resorting to tricks like
| precompilation. Along with more press releases
| proclaiming loudly that commercial QCs are a mere 2-3
| years away.
| the8472 wrote:
| IBM's quantum roadmap[0] also looks... curious. There's some
| modest scaling until 2028. Then there sudden orders-of-
| magnitude jumps in the following years. Perhaps that's backed
| by technology they have in the labs and plan to have ready by
| then, but without details it looks too good to be true.
|
| [0] https://newsroom.ibm.com/2023-12-04-IBM-Debuts-Next-
| Generati...
| dr_dshiv wrote:
| One of the big challenges is being able to frame real world
| concrete problems in terms that are solvable by quantum
| algorithms. There are a shocking lack of examples out there.
|
| Even when the hardware is there, it isn't clear how quantum
| computing can do useful things. But I suspect this is a solvable
| problem of good communications. But right now, experts in quantum
| don't seem to be able to provide these examples. Or they don't
| exist.
| reikonomusha wrote:
| This is a good reference: https://quantumalgorithmzoo.org/
| fxj wrote:
| The big unsolved problem in QC is the lifetime of the QC itself
| until decoherence. Imagine a computer where you can execute a
| maximum of 1800 commands (IBM Heron) and then it is broken. That
| is the status of QC at the moment. A QC might store TB of data
| and do searches with O(1), but there is no way (at the moment) to
| upload a TB database to the QC. What we need is a quantum
| processor that lives for hours or days in a coherent state, but
| what we have is milli seconds.
|
| Just my 2 ct.
|
| edit: IBMs Roadmap shows a QC with 1 Billion commands (gates)
| after 2033. With that machine they could (in principle) upload a
| 100MB database and do searches.
|
| source: https://www.tomshardware.com/tech-industry/quantum-
| computing...
| reikonomusha wrote:
| It's not a big mystery how this will get accomplished however,
| which is why participants in the field seem hopeful (if modest
| and tempered). The theory of quantum error correction is rich
| and pretty well developed. It's just that engineering a system
| which has enough runway to be error corrected requires a lot of
| development and innovations in a lot of directions: qubit
| design and fabrication, quantum compilers, rapid experimental
| procedures, etc.
|
| Edit to edit: Thinking of gate depths as permitting such-and-
| such megabyte databases as being "uploaded" isn't really a good
| or accurate metric in my opinion.
| fxj wrote:
| What puzzles me with IBMs roadmap is their scaling:
|
| 2025 156 qubits, 5000 gates
|
| 2028 156 qubits, 15K gates
|
| 2029 200 qubits, 100M gates
|
| 2033 2000 qubits, 1B gates
|
| The jump from 15K to 100M gates looks fishy to me. Maybe I am
| wrong but I doubt that will work that way.
| mettamage wrote:
| The key is to give an optimistic roadmap and then work on
| delivering it as fast as possible.
|
| Maybe I am getting a bit cynical but it seems that is what
| all companies and workers are doing.
| pclmulqdq wrote:
| Lying enough to get people excited to fund you, but not
| quite enough to catch a fraud charge.
| echelon wrote:
| There's a huge difference between saying "we can do" and
| "we think we will be able to do".
| layer8 wrote:
| It reads like wishful thinking, or an upper bound of what
| could happen in the best-case scenario.
| sampo wrote:
| When will they (or anyone) be able to run Shor's algorithm
| to factor the number 35? There was a failed attempt in
| 2019.
|
| https://en.wikipedia.org/wiki/Shor%27s_algorithm#Physical_i
| m...
| sesm wrote:
| Error correction can fix bit flips, but how can it help if
| qubits loose entanglement?
| sebzim4500 wrote:
| It's an unintuitive theorem in quantum computation that as
| long as you can protect against two types of errors (bit
| flips and sign flips) you are protected against almost all
| errors.
| packetlost wrote:
| I'm a bit biased, but non-superconducting modalities have
| considerably better coherence time properties. Neutral-atom,
| for example, has on the order of _seconds_. It has other
| constraining issues, but in theory the coherence times are
| better
| qmaybe wrote:
| Neutral atoms don't fare much better for gate fidelity, error
| correction algorithms are the only way to compute useful
| problems then probably
|
| Edit: quera is having its moment with the darpa announcement
| and the 48-logical qubit accomplishment in lab
| red75prime wrote:
| > do searches with O(1)
|
| Nope. Grover's algorithm allows only O(sqrt(n)) search.
| zozbot234 wrote:
| The bigger question is how many qubits exactly you can keep
| arbitrarily entangled on your preferred time scale. Because a
| linear increase in the number of qubits gives you an
| exponential increase in capability (compared to a classical
| computer) for problems that are amenable to the quantum
| approach. So if adding qubits turns out to be exponentially
| difficult, then QC will not amount to much since you could've
| done the same thing in a classical simulation. If it can be
| done more or less arbitrarily with a non-exponential cost, it's
| a true asymptotic change for the kinds of problems QC can
| address.
| panarky wrote:
| _> lives for hours or days in a coherent state_
|
| Or a few seconds, but with better error correction.
|
| https://www.nature.com/articles/s41586-022-05434-1
| bawolff wrote:
| I don't think its just skeptics. Afaik most researchers (other
| than those who are trying to attract investments) have been very
| clear that we are still a ways off from practical quantum
| computing, and even then the applications of QC is limited to
| specific domains.
| reikonomusha wrote:
| I agree with this. Unimaginable hype was mostly from those with
| an obvious incentive to raise money, to put themselves on a
| shortlist for a Nobel, and all that. Most of the ground-floor
| researchers and experimentalists have been extraordinarily
| realistic, though sometimes passive or quiet since there's an
| incentive to not bite the hand that feeds them.
|
| It also doesn't help that the giant PR machines of IBM, Google,
| IonQ, et al. have been commanding the narrative. At this point
| you'd think transmons, ions, and atoms are the only
| commercially/at-scale interesting options. :)
| echelon wrote:
| > the applications of QC is limited to specific domains.
|
| Breaking crypto alone would make it worth its weight in gold.
| shwouchk wrote:
| Do you forsee any sort of positive impact from destroying the
| ability to transact online?
| meltyness wrote:
| Skepticism of digital computing was a pretty open-air phenomena
| too.
|
| Specialization and generalization of hardware is also cyclical
| same as centralized and decentralized services.
| YouWhy wrote:
| I'm yet to see a single business case of money entering the QC
| value chain outside of (a) other QC businesses and (b) moonshot
| investments.
|
| After ~15 years of literally the world's smartest people trying
| to come up with exactly that (some of which are dear friends),
| the only way conscionable way I can regard QC is as a vapor
| bubble.
| pclmulqdq wrote:
| Quantum computing may be the "cold fusion" of the 2000's.
|
| In 2000, it was a fantastic looking technology that looked like
| it was going to leapfrog classical computing in a whole lot of
| ways. Now, classical computing has gotten so fast that (for
| example) O(sqrt(n)) searching of an in-memory structure is just
| not that exciting - O(n) is totally fine with a 100 GB dataset
| for many cases, and loading your database into your quantum
| computer would be O(n) anyway. Ideas for quantum machine
| learning have been supplanted by LLMs, and Ising optimization
| machines have already failed in the free market compared to a
| lot of classical computers.
|
| The remaining problems of interest are quantum simulation and
| encryption cracking, both of which are relatively niche
| markets.
| daxfohl wrote:
| If the encryption cracking market gets off the ground, would
| a quantum encryption hardening market come into existence, or
| can classical encryption algorithms suffice?
| adrian_b wrote:
| There would be only minimal impact of quantum computers
| over the bulk encryption algorithms (i.e. the 128-bit keys
| used today when speed is prioritized over security would
| become deprecated).
|
| Where changes would be needed is in digital signatures and
| certificates and in the key exchange algorithms for the
| establishment of communication connections in the public
| Internet, where pre-shared secret keys are not used.
|
| Many algorithms have been studied, but they are
| significantly less efficient than those used today.
| nmadden wrote:
| > Many algorithms have been studied, but they are
| significantly less efficient than those used today.
|
| Actually, several post-quantum algorithms are
| considerably _faster_ than current algorithms. But they
| have much larger ciphertexts, signatures, and public
| keys.
| bjornsing wrote:
| Ever since I understood how quantum computers actually work, I've
| been a skeptic.
| NanoYohaneTSU wrote:
| It won't work. It's just another tech scam that will fail, but at
| least we got great research out of it.
| reikonomusha wrote:
| If you mean quantum computers can't fundamentally work: you can
| publish a peer-reviewed paper about why it wouldn't work and
| the field would be immensely interested.
| pclmulqdq wrote:
| Generally, fields of academic study are not interested in
| accepting publications on why their field is unnecessary,
| even when you come at them with evidence.
| klyrs wrote:
| Gil Kalai has some well-researched opinions along these
| lines, and he seems to have a mutually respectful
| relationship with Aaronson.
| seiferteric wrote:
| Moore's law was enabled by lithography which enabled a pretty
| clear path forward on how to shrink transistors over time. AFAIK
| there is no similar enabling technology for qubit. Maybe that
| will change, but until then, I don't really see much hope for
| quantum computers future.
| cma wrote:
| You only need a linear increase in number of error-corrected
| qubits to get exponential gains on the limited subset of
| problems they can improve, whereas Moore's law has provided an
| exponential increase in the number of elements for classical
| computers.
|
| So, Moore's law isn't needed for something similar to Moore's
| law gains in quantum computing, a linear Moore's law will do if
| quantum error correction doesn't face scaling laws for the
| whole ensemble.
|
| That might mean they only keep pace with classical computing if
| Moore's law continues to hold there, but eventually that hits
| the Landauer limit without new physics (or possibly reversible
| computing).
| adgjlsfhk1 wrote:
| IMO this is very wrong. Most real applications of quantum
| computers will need thousands of error correct qbits (aka
| millions of real qbits). At the current growth rate in qbits
| of <100 qbits per year, if the growth was linear, it would be
| about 10000 years before we had quantum computers solving
| real problems.
| kevindamm wrote:
| I get the impression that cavity-QED is a similar path for QC
| that lithography was for classical compute. But I'm not at all
| an expert, just a curious and interested observer.
| citizenpaul wrote:
| I got curious about quantum computers so I ordered and read (most
| of) Quantum Computation and Quantum Information: 10th Anniversary
| Edition.
|
| It pretty much proved to me that quantum computing is at best
| 100+ years away or at worst a pipe dream fantasy. Until
| decoherence at scale is solved there is no way quantum computing
| will be useful beyond current computing abilities. There is not
| even a hint that this problem will be solved anytime soon.
| reikonomusha wrote:
| Except published research demonstrates continual improvements
| in coherence time and implementations of error correction
| protocols. So "not even a hint" is at best hyperbolic or at
| worst just wrong.
| echelon wrote:
| I felt this way about AI in college. And yet here we are.
| Vervious wrote:
| I don't see how you can read and understood Nielsen and Chuang
| in one sitting, unless you are already a quantum computation
| theorist. I also don't see how reading what is essentially an
| algorithms textbook can lead you to develop an informed opinion
| about the state of quantum computer engineering...
|
| it's like reading saying "I was curious about how computer
| software works so I ordered and read CLRS and I don't think
| faster computers are anywhere on the horizon in 100 years..."
| curuinor wrote:
| they didn't say or give the implicature that it was in one
| sitting. might be months
| odyssey7 wrote:
| Like Bletchley Park during the last world war, the best-case
| scenario for intelligence is that the government has quantum
| computers that break public-key encryption and nobody finds out.
| It's hard to know what machines exist behind the scenes.
| brcmthrowaway wrote:
| Prescient article:
| https://scottlocklin.wordpress.com/2019/01/15/quantum-comput...
| kromem wrote:
| General purpose QC isn't going to happen all that soon, and may
| always be faced with scaling issues.
|
| But we _will_ see QC in a very specific application before the
| decade is out.
|
| It's kind of a perfect marriage between ML and QC. If you want
| stochastic outputs and aren't going to know what's going on at
| each step in the network anyways, the issues of error correction
| around measurement aren't nearly as crippling as they are for
| general purpose QC.
|
| While initially the work will be converting from classical to
| optoelectronic hardware for matching current operations, such as
| MIT's work this year, once we've seen greater availability of
| optoelectronic hardware I suspect we'll see algorithms for ML
| developed that would only work in photonic networks and fully
| exploit the quantum properties therein.
| B1FF_PSUVM wrote:
| > If you want stochastic outputs and aren't going to know
| what's going on at each step in the network anyways
|
| You have a way with words, I'm already more at ease with the
| perspective of a schizophrenic AGI ...
| kromem wrote:
| It's not really any different from the current state of
| models, it's just that moving the same constraints to a
| different hardware foundation opens up significant
| performance gains.
| osigurdson wrote:
| The fact that a working quantum computer exists and works at all
| is mind blowing to me.
| geijoenr wrote:
| What is happening with QC is very sad.
|
| Research on this topic should be a long term scientific endeavor,
| like nuclear fusion has been for many years.
|
| Not the circus it has been so far.
___________________________________________________________________
(page generated 2023-12-23 23:00 UTC)