Author Topic: Move over Watson, IBM unveils its most powerful computers ever  (Read 1140 times)

rrr314159

  • Member
  • *****
  • Posts: 1383
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #15 on: July 13, 2017, 12:00:29 PM »
IBM recently implemented "ballistic" transmission, mentioned above by jj2007. Normally electrons transmitting on a wire, or bus, as in the computer you're using, bounce a lot on the way, off of copper atoms. This not only heats them up and slows them down. More important for quantum computing (QC), it causes loss of coherence. That is, the wave function it started out with is destroyed by these random collisions. In "ballistic" transmission this effect is much reduced, to the point where they only bounce off the walls in a controlled fashion. This is much more energy-efficient, rather like superconducting, and it's coherent: the wave function is preserved. At least that's what they're aiming at, with partial success at this time. The word "ballistic" means it's as though the electrons are being "thrown" cleanly like a ball through the air, not bouncing around randomly. This is all very well and good, and will help implement QC, plus it has applications outside of QC also. So I don't mean to belittle it, it's good work. But it's nowhere near making QC useful. That may never happen. At best, it's some decades away. I, jj, anunitu, and all the other old farts around here, won't live to see it. But look on the bright side, you're probably not missing much.

As for the "God particle", the Higgs, I guess it's the real deal. They still haven't nailed down the spin AFAIK, and decay rates may not be exactly right, but it's close enough to validate electroweak theory. Maybe take a few more years to be really sure. Assuming it's Ok, that electroweak theory is a major home run IMHO, comparable to GR. Purely theoretical considerations led them to an amazing prediction, which turned out right. Very impressive. The problem is, they get a few of these successes, it goes to their heads. For every such success there are a dozen failures, like, probably, Dark Matter. Theory should never be trusted until backed up by experiment, because 12 times out of 13 (or so) it turns out wrong. Experiment without theory, OTOH, is always right.
« Last Edit: July 14, 2017, 03:49:50 AM by rrr314159 »
I am NaN ;)

felipe

  • Member
  • ***
  • Posts: 374
  • assemble the unassembled.
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #16 on: July 13, 2017, 03:50:34 PM »
One quantum physics problem that i found interesting is the fact that the observer seems to change the result of some experiments just by looking and doing nothing more. Even if the observer is a camera.   :dazzled:
Felipe.

rrr314159

  • Member
  • *****
  • Posts: 1383
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #17 on: July 14, 2017, 03:55:52 AM »
Observer effect is a long story. Conscious observer (apparently) causes wave function to collapse. If it's only a camera, that causes "decoherence" but not, arguably, collapse. Many physicists say the whole thing is a non-issue, but they're wrong. About the best ref I can give for a modern physicist is Lubos Motl. Google him with "collapse" and "observer". He's an oddball, BTW, but he does understand this issue well.
I am NaN ;)

K_F

  • Member
  • *****
  • Posts: 1294
  • Anybody out there?
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #18 on: July 14, 2017, 04:44:49 AM »
Quantum is nothing more than a probabilistic average... so it's probably the incorrect terminology to use for 'small' stuff.
 ;)
'Sire, Sire!... the peasants are Revolting !!!'
'Yes, they are.. aren't they....'

rrr314159

  • Member
  • *****
  • Posts: 1383
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #19 on: July 14, 2017, 04:51:08 PM »
Well, no, K_F, QM is more than just "a probabilistic average". Presumably you're talking about the so-called Statistical Ensemble Interpretation (or one of its variants). It correctly points out that we never actually know what an individual wave function is. Rather, we can only infer it by repeated trials on identically-prepared quantum systems. Therefore, (by some hand-waving) there's no such thing as collapse. Or at least, we can ignore it. Sorry, it doesn't work. Yes, I know there are some "great" physicists (i.e., physicists who are great at public relations) who push it on the gullible public. Surely you're too smart to fall for that? If not ... I have an infallible system for picking winning lottery tickets! Since you're such a good guy, I'll sell it to you for only $10,000 US! You can immediately use it to win millions! PM for details!!!  :biggrin:
I am NaN ;)

anunitu

  • Member
  • ****
  • Posts: 919
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #20 on: July 14, 2017, 06:19:58 PM »
Is that you Nigerian  Prince?  ::)

nidud

  • Member
  • *****
  • Posts: 1410
    • https://github.com/nidud/asmc
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #21 on: July 15, 2017, 12:40:28 AM »
Bit skeptical about this but I'm willing to invest in a modified version where I win thousands as appose to millions assuming the cost is reduced accordingly.

hutch--

  • Administrator
  • Member
  • ******
  • Posts: 4933
  • Mnemonic Driven API Grinder
    • The MASM32 SDK
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #22 on: July 15, 2017, 12:53:46 AM »
But what about those generous banana republic dictators who desperately need a loan of you bank account so they can park millions of whatever currency they say they are using for safety in your bank account so they don't get ripped off as they abscond from their country.   :P
hutch at movsd dot com
http://www.masm32.com    :biggrin:  :biggrin:

HSE

  • Member
  • ****
  • Posts: 553
  • <AMD>< 7-32>
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #23 on: July 15, 2017, 01:15:02 AM »
Or donations of sister Ruth Hamsom from Costa de Marfil (yes, she write spanish, perhaps also have a degree in QM! ).

FORTRANS

  • Member
  • ****
  • Posts: 946
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #24 on: July 18, 2017, 10:33:51 PM »
Hi,

So I don't mean to belittle it, it's good work. But it's nowhere near making QC useful. That may never happen. At best, it's some decades away. I, jj, anunitu, and all the other old farts around here, won't live to see it. But look on the bright side, you're probably not missing much.

   I subscribe to "Science News", the current issue is "July 8 & July 22,
2017".   A quote from the article "Quantum Computers Get Real" that
that may interest you:

"IBM has made its five-qubit computer accessible online for free, and
the company is planning a 50-qubit computer for commercial use."

Elsewhere:

"(IBM's 16-qubit computer is in beta testing, so Quantum
Experience users are just beginning to get their hands on it.)"

   A related article is "Birth of the Qubit", with this fun quote:

"Much more recently, the math describing quantum error correcting
has turned up in a completely unexpected context - efforts to
understand the nature of spacetime by uniting gravity with quantum
mechanics."

Cheers,

Steve N.

rrr314159

  • Member
  • *****
  • Posts: 1383
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #25 on: July 20, 2017, 05:06:21 AM »
Since there's such interest in my Winning Lottery Numbers scheme, I'll give it away FREE! As soon as you send me a charitable donation. Amount negotiable. I'm looking for $10,000 but hey, ten bucks ought to do it ... PM for details!!!!

@FORTRANS, that IBM online has been around a while, maybe a year?, but I couldn't access it a while ago. You'll find the capabilities are extremely limited, suitable only for playing with a real QC, not for solving a real problem. For instance any problem takes a number of runs, then you accept an answer when it passes some statistical test. Typically it might give the same answer (usually right) 2 out of 3 times, so you need 30 runs (or whatever, this is off top of my head) to reach statistically acceptable level. A researcher can thus study behavior of different algo approaches. It's useless for a solution, which a calculator (or even, mental arithmetic) will give you more reliably with a lot less trouble. Admittedly haven't read about it in a while but am confident this assessment is not far off.

AFAIK all the math involved, including error correcting, is valid and interesting, not surprised that cosmologists can get some mileage out of it. Well, all the math except some of the fantasies coming from David Deutsch and his group.
I am NaN ;)

rrr314159

  • Member
  • *****
  • Posts: 1383
Re: Move over Watson, IBM unveils its most powerful computers ever
« Reply #26 on: July 20, 2017, 05:27:33 AM »
Here's an interesting ANALOG QC being developed by Harvard and Russian researchers. Analog will probably achieve something useful someday, even if digital never does. But all the results they get are about the quantum system itself, not math problems or anything like that. As you can see from the abstract. I've bolded the key (weasel) words in it, for those not experienced in reading this sort of thing. When you see phrases like this, you know they haven't actually got anything concrete yet.

https://arxiv.org/abs/1707.04344

Probing many-body dynamics on a 51-atom quantum simulator

Controllable, coherent many-body systems provide unique insights into fundamental properties of quantum matter, allow for the realization of novel quantum phases, and may ultimately lead to computational systems that are exponentially superior to existing classical approaches. Here, we demonstrate a novel platform for the creation of controlled many-body quantum matter. Our approach makes use of deterministically prepared, reconfigurable arrays of individually controlled, cold atoms. Strong, coherent interactions are enabled by coupling to atomic Rydberg states. We realize a programmable Ising-type quantum spin model with tunable interactions and system sizes of up to 51 qubits. Within this model we observe transitions into ordered states (Rydberg crystals) that break various discrete symmetries, verify high-fidelity preparation of ordered states, and investigate dynamics across the phase transition in large arrays of atoms. In particular, we observe a novel type of robust many-body dynamics corresponding to persistent oscillations of crystalline order after a sudden quantum quench. These observations enable new approaches for exploring many-body phenomena and open the door for realizations of novel quantum algorithms.

BTW Google keeps announcing their competitor to IBM will be ready "any day now". With all their money it may be better, give IBM something to shoot at. This sort of competition might produce a viable digital QC in a few years, if indeed that's possible at all. Don't hold your breath!
I am NaN ;)