Quantum Supremacy

21 views
Skip to first unread message

John Clark

unread,
Oct 23, 2019, 6:41:09 AM10/23/19
to everyth...@googlegroups.com
Today Google officially published in the journal Nature it's article claiming their 53 Qubit Quantum Computer (it was actually 54 Qubit but one Qubit didn't work) could solve a problem in 3 minutes and 20 seconds that it would take Summit, the world's largest supercomputer, 10,000 years to solve: 


Very recently IBM, another company working on Quantum Computers, has disputed Google's claim:


IBM says they have found a classical algorithm that if running on the largest classical supercomputer in the world could do what Google's Machine did in 2.55 days, and if all 54 Qubits had been working duplicated it in 5.80 days; that's still a lot longer than 3 minutes and 20 seconds but it's not 10,000 years. It should be noted that unlike Google IBM has not actually performed the calculation to prove what they say is true.

 John K Clark

spudb...@aol.com

unread,
Oct 23, 2019, 1:11:02 PM10/23/19
to everyth...@googlegroups.com
You once stated in this mailing list, that you thought the human world will change, if and when we hit 100 successful qubit operations. Does your conjecture on this still hold, or have you modified or discarded this? Solving huge problems as you highlighted below?


--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv300RcJsVqJAeH9509UBV%2BuSKTZ8wNMeATeON5s--2cuA%40mail.gmail.com.

John Clark

unread,
Oct 23, 2019, 2:52:51 PM10/23/19
to everyth...@googlegroups.com
On Wed, Oct 23, 2019 at 1:11 PM spudboy100 via Everything List <everyth...@googlegroups.com> wrote:

> You once stated in this mailing list, that you thought the human world will change, if and when we hit 100 successful qubit operations. Does your conjecture on this still hold, or have you modified or discarded this? 
 
Google was working with 53 imperfect Qubits and still did something very impressive, but the number of states the computer can be in doubles with each added qubit so if they had 100 perfect Qubits that would indeed change the world beyond recognition. Very recently Dario Gil, the head of IBM’s research lab in Yorktown Heights New York said "Imagine you had 100 perfect qubits. You would need to devote every atom of planet Earth to store bits to describe that state of that quantum computer. By the time you had 280 perfect qubits, you would need every atom in the universe to store all the zeros and ones."

If your Qubits aren't perfect then you're going to need quantum error correction which will use up a lot of your Qubits, so a few pretty good Qubits is better than a lot of crappy ones. Gil has developed a new way to measure the power of a Quantum Computer he calls "quantum volume" which takes in account both the number of Qubits and their quality. That's what's so exciting about Microsoft's attempt to make a Topological Quantum Computer, if they're successful their Qubits will be of extremely high quality requiring little error correction. 

John K Clark

spudb...@aol.com

unread,
Oct 25, 2019, 6:45:51 PM10/25/19
to everyth...@googlegroups.com
Very good JC, thanks for the explanation. I am siding on the notion that we can achieve 80-100 perfect qubits soon, because let's face it, the species needs a whole lot of changing, beyond the vaunted, encryption/decryption pursuit, journalists seem to dream of. Beyond recognition seems like a pretty good deal right now. Here's hoping!


-----Original Message-----
From: John Clark <johnk...@gmail.com>
To: everything-list <everyth...@googlegroups.com>
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit
Reply all
Reply to author
Forward
0 new messages