Google’s new quantum computer is '100 million times faster than your PC'

How is a quantum computer going to be practical in the real world?Considering science is yet to unify the quantum world with the macro world.Correct me if I'm wrong.

a reply to: jonnywhite
a reply to: CallYourBluff

Yes, the internals are mind boggling, and it is better to stick to the 10k ft. view when attempting to explain how they work and how they are so different. The programmers here are really Mathematicians first. They can describe to these systems how to model problems in differential and integral calculus and matrices, which rely heavily on iterative convergence.

You won't see a quantum computer running complex video games, but imagine a quantum DSP in a video card....

In terms of our telecom infrastructure, the long lost heroes...we barely have enough on average for 10G/bit transport in most of our geographic space of the US. It will locally process data fast and determine a good set of "big data" but give it a few for it to actually deliver real network speeds. I agree, we don't know what these are used for in caged colos...could be anything.

a reply to: neoholographic

How much faster will it download porn?

originally posted by: DBCowboy
a reply to: neoholographic

How much faster will it download porn?

There's always one.

originally posted by: Shiloh7
Agree with you completely on that, in fact I am wondering how many jobs these monstrosities are going to kill off?

Right, so because the advancement of technology could mean redundancies, it must therefore not be a good thing?

By extension, should we halt technological discoveries and advances in order to maintain people doing the same job they did 30 years ago?

originally posted by: DBCowboy
a reply to: neoholographic

How much faster will it download porn?

It can download porn instantly, the question is, how long will it take you to decide what to watch?

a reply to: neoholographic

Dwave is not too far from my office in Vancouver.
About a 15 minute drive and am quite familiar
With their work on an ALL STATES AT ONCE
computing system. They just need to fully
Get past the "qbit decoherence" problem.

My suggestion is TO NOT try and to get qbits
To lock down to a final state (i.e. Quantum Annealing or Wave Function Collapse) but
Rather use the constant entanglement and
sudden "de-entanglement" of supercooled
atoms trapped in quantum wells to form
An ultrafast semi-classical computer.

Dwave is trying and somewhat succeeding
In building a multibit quantum supercomputer
That on a technical basis is really only good
For factoring huge numbers to break
Modern data encryption schema.
I.e Great for the NSA but unuseable
for the rest of us!

What's missing is all the if-then-else and true
Multibranching allowed in modern CPU chips
so that harder computer science problems
such as data mining can be tackled.

So my suggestion is to use the sudden de-entanglement/decoherence which
happens at predictable rates over a
predictable series of qbits AS THE
Computing engine itself.

Using an analogy, use the constant breaking
Of connections as BOTH a memory storage mechanism (i.e. X number of qbit decoherences
Over Y number of attoseconds can equal a solid
Large integer or real number) and as a branching mechanism. The key is use the fact you cannot actually MEASURE a Quantum State and/or a Quantum Position with any measure of certainty
Without causing "wave function collapse" BUT
You CAN measure the Quantum "breakage"
Left behind on nearby real world subatomic structures which are influenced by the Quantum Decoherence.

This means that Dwave can build chips that let
The qbits influence nearby atomic structure
At unimaginable sample rates likely in the attosecond-level of switching rates. As qbits
try to entangle and then self destruct (decohere)
at predictable rates per second you only measure
the atomic wreckage left behind on the areas
Immediately near or outside each quantum well.
That measurement of the direction and shape
Or atomic state of the left behind wreckage forms
The classical 1 or 0 binary bits (i.e. ON or OFF
bits) that can then form longer bitstreams,
integers and real numbers.

With attosecond switching rates, theoretically,
It would be possible to process Exabytes per second of quantum influenced decoherence wreckage that forms the classical bits and
bytes of real world data so that more difficult
Problems such as molecular simulations and
Modern text searching/data mining can be
Realized using the ULTRA-FAST switching rates
supported by a Quantum Computer.

I am saying that for a few tens of millions that DWAVE could ADD semi-classical computing
To their quantum computers that would be
Equivalent in CPU Horsepower to 20 or 30
IBM Blue Gene/Q supers in a package barely
The size of a single home refrigerator!

Talk about a money and electrical power savings!

a reply to: StargateSG7

As a suggestion, use trapped Xenon atoms
To influence nearby streams of Hydrogen,
lithium, calcium, sodium, potassium or other
More unstable elements which will act as the
Quantum Influenced medium which can be
examined and measured for Quantum Wreckage.

As the unstable medium trapped in
microchannels passes nearby each
single qbit quantum well, it gains
evidence of left behind quantum
decoherence wreckage, which can
Include charge disturbances or other
Measurable influences from a decohered
(or truly entangled!) qbit.

Quantum Wreckage of a specific shape, size, amount or type that is present (or not!) at a specific sample time is what becomes an agreed
Upon ON or OFF (i.e. 1 or 0) binary state which
Are then organized further into integers, real
numbers and text characters or pixels for
classical computation purposes.

As some additional information, here is a JPEG image
indicating a POSSIBLE educated guess for the workings
of a classically-oriented Quantum Computing Systems:



For the microlaser part, I suggest this paper
to get you get started on modifying a microlaser
configuration to make a super-fast and tiny
chippable laser interferometer:

Optical memory based on ultrafast wavelength
switching in a bistable microlaser
by Sergei V. Zhukovsky and Dmitry N. Chigrin

Theoretical Nano-Photonics Group,
Institute of High-Frequency and Communication Technology,
Faculty of Electrical, Information and Media Engineering,
University of Wuppertal,

Rainer-Gruenter-Str.
21, D-42119 Wuppertal, Germany.



2009 Optical Society of America
OCIS codes: 210.4680, 130.4815, 190.1450, 140.3948, 230.5298

We propose an optical memory cell based on
ultrafast wavelength switching in coupled-cavity
microlasers, featuring bistability between modes
separated by several nanometers. A numerical
implementation is demonstrated by simulating
a two-dimensional photonic crystal microlaser.
Switching times of less than 10 ps, switching
energy around 15–30 fJ and on-off contrast
of more than 40 dB are achieved.

Theoretical guidelines for optimizing
the performance of the memory cell in
terms of switching time and energy are drawn.

---

Even though in the report the maximum switching rate
is "only" in the PicoSecond range, switching rates that
still give many petabytes a second computation rates
with only 512 Qbits is possible today! Just add more
Qbits (i.e. 4096 Qbits should about do it!) to get into
the Exabytes per second of data processing rates!

The key is to use the consistently measured (exact time-domain
measurement aka Sample Rate per second) and highly-specific
CHANGE STATES of the quantum event influencing medium
AS both the storage system AND as the branching system.

i.e. Value X state change = an if-then-else or
Branching command and Value Y state change equals
a binary ON or OFF storage bit (8 state changes = one byte)
32 state changes = 32-bit integer or real number and
64 state changes = 64-bit integer or real number.

Can I ask something about quantum computing?

How does it figure out the quickest path? Traditional computers have to compute the paths to figure out which is quicker. We're teh same way, generally. We compute the numbers. Computers are really popular for determine routes for shipping product across land and sea, for this reason. Because it involves numbers and comparing results.

So do quantum computers just know? They don't need to compute it? You ask and they answer immediately. Seems odd to me.

Magic. That's what comes to mind. It doesn't compute because it already knows. It's everywhere. It's kind of like God. But because we're not everywhere at once, we have trouble understand.

a reply to: jonnywhite

I think this is an absurdly simplified explanation, but the best way of saying is that it computes via all routes possible, at the same time. It's like this, you have a car come to a fork in the road. You know both paths lead to the destination, and one of them is faster, but you don't know which is which. What happens is that the car goes down both routes simultaneously, and whichever one it emerges from first is the one it went down.

Quantum is weird. REALLY weird. I'm not even sure I explained it properly. Someone else will likely be able to.

originally posted by: Eilasvaleleyn
a reply to: jonnywhite

I think this is an absurdly simplified explanation, but the best way of saying is that it computes via all routes possible, at the same time. It's like this, you have a car come to a fork in the road. You know both paths lead to the destination, and one of them is faster, but you don't know which is which. What happens is that the car goes down both routes simultaneously, and whichever one it emerges from first is the one it went down.

Quantum is weird. REALLY weird. I'm not even sure I explained it properly. Someone else will likely be able to.

How does Quantum physics know the quickest path? How is it even able to know the distances of each path (to compare)? Traditional computers have to compute the distance. So quantum physics doesn't?

For computing the best path, this is like having a time travelling computer. Its computing happens at "other times" and produces an instant answer for us, giving the impression it already had the answer.

In my mind I cannot fathom something which doesn't need to compute, you understand? So either it's computing in our time or it's computing in other times. I cannot comprehend something which doesn't compute at all; exists outside time. That something is like God or magic.

www.wired.com - Everywhere God in a Flash: The Quantum Physics of Photosynthesis...

Almost no energy is lost in between. That’s because it exists in multiple places at once, and always finds the shortest path.

“The analogy I like is if you have three ways of driving home through rush hour traffic. On any given day, you take only one. You don’t know if the other routes would be quicker or slower. But in quantum mechanics, you can take all three of these routes simultaneously. You don’t specify where you are until you arrive, so you always choose the quickest route,” said Greg Scholes, a University of Toronto biophysicist.

By analyzing changes to a laser beam sent through the protein immediately afterwards, the researchers were able to extrapolate what was happening inside — an ultra-high-tech version of shadows on a screen. They found that energy patterns in distant molecules fluctuated in ways that betrayed a connection to each other, something only possible through quantum coherence.

What I just read above disturbs me. I was led to believe quantum entanglement or coherence have no effect on the macro scale world. And yet photosynthesis is very much a macro scale thing, enabling plants to efficiently use the sun. In principle we can too. We're curretnly struggling to efficiently use hte sun. Superposition might help us. And if it does, I should be disturbed even moreso. The cat is not in the box anymore. It's living among us now and it's both dead and alive. Scary times.

originally posted by: jonnywhite
So do quantum computers just know? They don't need to compute it? You ask and they answer immediately. Seems odd to me.

We have algorithms that can answer certain problems in polynomial time. For example, what your GPS uses is either Dijkstra's or A* (pronounced A-Star) which have a run time of O(|E|+|V|log|V|) and O(B^D) respectively. Dijkstra's is usually faster except on small problems but A* is easier to implement and is all that most people really know how to do. So what you'll typically find is that people break a problem up into many small steps each solved by A* rather than one large problem with Dijkstra's.

Determining optimal shipping really isn't all that dissimilar from a traveling salesman problem, which we currently have no way of efficiently solving. In fact, just about all shipping uses non optimal routes for this reason, it's just that we don't know the actual optimal solutions and we have no way to find them within the necessary time constraints. If you could solve this problem you would become the richest person on earth overnight, as every single nation and transport heavy industry would pay you 100's of millions of dollars for the solution.

The way quantum computers work is that each qubit is in all possible states until measured, this means that the solution in a quantum computer contains all possible solutions. Then some algorithms are used (I don't know much about what they use) in order to determine which machine state is the most likely to contain the correct answer. This state is then measured locking the qubits into a specific configuration and then compared to a known answer by a regular computer. Then the process is repeated on the next most likely state. Comparisions are a very fast operation for computers so this process allows for millions of comparisons to be made in the time an ordinary computer could compute just a handful of results.

a reply to: 711117

You're assuming, of course, that he even does think...

Call me paranoid, but I'm not a big fan of this sort of thing.

originally posted by: jonnywhite
How does Quantum physics know the quickest path? How is it even able to know the distances of each path (to compare)? Traditional computers have to compute the distance. So quantum physics doesn't?

For computing the best path, this is like having a time travelling computer. Its computing happens at "other times" and produces an instant answer for us, giving the impression it already had the answer.

In my mind I cannot fathom something which doesn't need to compute, you understand? So either it's computing in our time or it's computing in other times. I cannot comprehend something which doesn't compute at all; exists outside time. That something is like God or magic.

Technology isn't magic. Let me explain using a knapsack problem example, you may be familiar with the problem.

You're in a restaurant and your group has $55 between them. You decide collectively that you want to leave a $9 tip which leaves you $46 for food. Your group is hungry so you want the most food you can possibly get, meaning your goal is to spend exactly $46. The menu prices are as follows
Chips - $1
Drink - $3
Quesadilla - $4
Soup - $4.50
Sandwich - $6
Cheese Sticks - $7
Burger - $8

Additionally these food items have the following "fullness value" (you'll feel more full with a higher score)
Chips - 1
Drink - 4
Quesadilla - 6
Soup - 9
Sandwich - 9
Cheese Sticks - 13
Burger - 12

Determine the optimal configuration of food that should be purchased.

A normal computer will find this problem very challenging and take a long time to solve (dynamic programming makes it faster but it's still slow... and very memory intensive). There are literally billions and perhaps trillions of possible solutions to calculate, store, and compare, and over 99.99% of those configurations aren't even valid possible answers. Once you do have the solution of all possible answers you need to check them all.

For a quantum computer the problem works a little different. Because the machine exists simultaneously in all possible states it only needs to look at those states where the answer sums to exactly $46. When over 99.9999% of all possible answers are wrong, this is a 1,000,000 fold decrease in the time it takes to arrive at a valid solution. Instead of having to figure out all of the $46 solutions, you're already provided with them. From that point you merely need to iterate through the list and record what configurations result in the highest fullness value.

So, it's not that a quantum computer is computing in some magical dimension, it's that they don't need to determine the answer at all. The machine already possesses all possible answers. It only has to select from those possible answers those that meet the criteria to be a possible correct answer, and part of that correct answer includes the work taken to arrive at that solution. From there we can just compare answers and see if it's the most correct one.

All the PRON!

originally posted by: BatheInTheFountain
All the PRON!

A quantum computer wouldn't actually be all that good for porn. A quantum computer is very good at reducing problems with infinite or uncountable answers into a finite sequence. However, video rendering is about turning a static string of binary into colored pixels on a screen, even 3d rendering such as virtual reality or one day holograms isn't that much different. They would be very inefficient at such a task because the solution that you're looking for is already known.

It also wouldn't download any faster than any other machine with a very good internet connection.