Quantum computing will be used as a tool for improving the worst time complexity of some classical algorithms.I doubt that we will use qubits , we will simply encode classical bits as qubits and use the properties of quantum mechanics to speed up some processes.

 

Quantum computing will be used as a tool for improving the worst time complexity of some classical algorithms.I doubt that we will use qubits , we will simply encode classical bits as qubits and use the properties of quantum mechanics to speed up some processes.

The advantage of quantum computing is not being faster processor, but rather, it's about designing algorithms that are better/faster when using quantum mechanics. Those algorithms can also provide insights on how to make classical machines run those classes of problems faster.

https://www.simonsfoundation.org/20...uter-beat-a-quantum-computer-at-its-own-game/

Short answer: Nobody knows yet if QC is actually better.

 

The advantage of quantum computing is not being faster processor, but rather, it's about designing algorithms that are better/faster when using quantum mechanics. Those algorithms can also provide insights on how to make classical machines run those classes of problems faster.

https://www.simonsfoundation.org/20...uter-beat-a-quantum-computer-at-its-own-game/

Short answer: Nobody knows yet if QC is actually better.

I dont think the best classical algorithm can outperform the best quantum algorithm for some task ever.

 

I dont think the best classical algorithm can outperform the best quantum algorithm for some task ever.

Possibly not. The juries out. The amount of performance improvement will depend on the particular class of problem. Some problems, such as the discrete log and integer factorization problems, the speed up is so pronounced that the basis for most commercially-used public key cryptosystems will be broken entirely. Modern classical computers would take billions of years to factor a 4096-bit RSA key, while Shor's algorithm will let a quantum computer do it in a few hours to days. But, just as with classical computers, we will not have the processing power needed to achieve that the day after we figure out how to get around some of the problems we currently have. A quantum computer will need about 10,000 clean qubits to accomplish it and currently we have something like 1000 noisy qubits. With the qubits we have today, it would take something like 100 million qubits. So there's a ways to go. But there's also no reason to believe that, at some point, the number of qubits won't grow exponentially for some period of time, either, and from 1000 to 100 million is only about 16 doublings.

Many classical algorithms can be sped up using Grover's algorithm can be sped up by the root of the problem size, which is a huge improvement, but far from an exponential speed up.

But many problems are likely to only see speedups of a couple orders of magnitude, at least unless someone comes up with a clever quantum algorithm (like Shor's) to tackle that particular problem.

 

Possibly full quantum computing will tend to be a bit closer than practical cold fusion,which has been "about 20 years down the road"for a very long time, a couple decades,I think. But Q C is only five years down the road. Or was that two years down the road ??
We will need to wait and see how fast it really is. THAT will be VERY INTERESTING INDEED!!!

 

Possibly full quantum computing will tend to be a bit closer than practical cold fusion,which has been "about 20 years down the road"for a very long time, a couple decades,I think

It was 10 years away when I got my physics degree 52 years ago.

 

It was 10 years away when I got my physics degree 52 years ago.

All of these things -- flying cars, commercial fusion, the cure for diabetes, etc., etc. -- are always just ten or twenty years away and have been for decades.

 

Flying cars are just around the corner, I'd say ... and by that, I mean that in 20 years or less autopiloted flying taxis will be a common sight in big cities ...

 

Flying cars are just around the corner, I'd say ... and by that, I mean that in 20 years or less autopiloted flying taxis will be a common sight in big cities ...

There could be flying taxis but I doubt that they will be "cars". Rather some variety of VTOL aircraft. THEN, Just Imagine MEXICO CITY with flying cabs!! AWESOME INDEED!!! ( You really need to have seen all the cabs there to understand this comment.)

 

( You really need to have seen all the cabs there to understand this comment.)

And you really need to know who you're talking to to understand that I DO understand ...

Either way, you're right ... I can't possibly imagine how flying taxis would be handled in that hellish city ...

 

Flying cars are just around the corner, I'd say ... and by that, I mean that in 20 years or less autopiloted flying taxis will be a common sight in big cities ...

Yep. And that has been the claim for, quite literally, more than a century.

In fact, the first patents for them date back to 1841. The first actual flying car to 1917.

 

Yep. And that has been the claim for, quite literally, more than a century.

In fact, the first patents for them date back to 1841. The first actual flying car to 1917.

And the first (imagined or proposed) flight to the moon since the middle ages .. with technical details later added by Edgar Allan Poe ...

This time, though, there are actual working proof of principle (or prototypes, if you will ) being tested out there ... the Market will decide their fate, as always

 

And the first (imagined or proposed) flight to the moon since the middle ages .. with technical details later added by Edgar Allan Poe ...

This time, though, there are actual working proof of principle (or prototypes, if you will ) being tested out there ... the Market will decide their fate, as always

And there have been actual, drivable and flyable flying cars out there for seventy years.

Building a workable flying car is not the hard part.

Managing the traffic will be the big, possibly insurmountable, problem. For instance, the Denver metropolitan area has something like 100 to 150 aircraft in the sky at one time at the peak. Imagine what it would take to handle ten times that. Now figure that there can be about half a million cars on the road at the same time over that same area that's more than three orders of magnitude more. DIA, which is the fourth busiest airport in the world in terms of takeoffs and landing, averages less than eighty and hour. Yet that is so busy that private pilots are strongly discouraged from operating there, require pre-coordinated clearances and have high landing fees (something like $200). You have to have special equipment to operate within 30 nautical miles of DIA. The Class B, at the surface, extends out 10 NM from the airport (so it covers over 300 sq NM) and you must obtain specific permission from ATC to enter that zone -- and they are not shy about not granting it.

People have this naive notion that they will be able to treat their flying car like their normal car. Get it in whenever they want to go somewhere, go there and land (and just where are all these takeoff and landing areas going to be), do whatever they want, and then head back when it suits them. Nothing will be further from the truth, except in the most remote areas. Proponents, when pressed on this, always wave it away with, "Oh, it will all be automated." As if saying it makes it so.

 

Inside the sub-zero lair of the world's most powerful computer
View attachment 361754
​

Hi,

So what was the task?
It added 1+1=2 a zillion times

 

And there have been actual, drivable and flyable flying cars out there for seventy years.

Building a workable flying car is not the hard part.

Managing the traffic will be the big, possibly insurmountable, problem. For instance, the Denver metropolitan area has something like 100 to 150 aircraft in the sky at one time at the peak. Imagine what it would take to handle ten times that. Now figure that there can be about half a million cars on the road at the same time over that same area that's more than three orders of magnitude more. DIA, which is the fourth busiest airport in the world in terms of takeoffs and landing, averages less than eighty and hour. Yet that is so busy that private pilots are strongly discouraged from operating there, require pre-coordinated clearances and have high landing fees (something like $200). You have to have special equipment to operate within 30 nautical miles of DIA. The Class B, at the surface, extends out 10 NM from the airport (so it covers over 300 sq NM) and you must obtain specific permission from ATC to enter that zone -- and they are not shy about not granting it.

People have this naive notion that they will be able to treat their flying car like their normal car. Get it in whenever they want to go somewhere, go there and land (and just where are all these takeoff and landing areas going to be), do whatever they want, and then head back when it suits them. Nothing will be further from the truth, except in the most remote areas. Proponents, when pressed on this, always wave it away with, "Oh, it will all be automated." As if saying it makes it so.

The way I envision it, is that most individuals would get into their aircraft (cars that can fly will be useless, I think) and then select a destination, after which the vehicle will connect to a centralized navigation system. Said system will coordinate all aircraft and guide each to its destination. My point being that the exception will be manually operated aircraft.

Unless I'm mistaken, I believe it's much easier to control and coordinate aircraft than automobiles. I'm under the impression that from a computer's perspective, flying has a lot less factors to deal with.

On the side of economics, other than the cost of fuel (or energy), a per-flight fare will be applied, along with an annual ownership tax of sorts.

But I don't see that scenario happening any sooner than the next 40 to 50 years.

 

The way I envision it, is that most individuals would get into their aircraft (cars that can fly will be useless, I think) and then select a destination, after which the vehicle will connect to a centralized navigation system. Said system will coordinate all aircraft and guide each to its destination. My point being that the exception will be manually operated aircraft.

Unless I'm mistaken, I believe it's much easier to control and coordinate aircraft than automobiles. I'm under the impression that from a computer's perspective, flying has a lot less factors to deal with.

On the side of economics, other than the cost of fuel (or energy), a per-flight fare will be applied, along with an annual ownership tax of sorts.

But I don't see that scenario happening any sooner than the next 40 to 50 years.

Aviation has a lot more factors. Consider that cars driven by everyday people, many with a few hours of experience, routinely travel in dense packs down a highway at 75 mph separated by a couple hundred feet front and back and four to six feet side to side and think nothing of it. Aircraft separations in IFR (instrument flight rules) is three nautical miles laterally and one thousand feet vertically. This is the minimum separation to operate legally.

Card drivers don't worry about the weather unless it gets extreme. Pilots worry about it constantly because turbulence can and does happen frequently and unexpectedly. Updrafts and downdrafts or even just high temperatures can make it impossible to maintain course and/or altitude (been there, done that). Car drivers seldom have issued with icing (icy roads, yes, icing of the car itself, not so much). Icing in an aircraft is a real issue, even at above zero temperatures because of the temperature drops associated with air pressure changes associated with just generating lift. And that's ignoring icing concerns with the power plant.

So, each vehicle needs maneuvering space to deal with these. These are things that your centralized navigation system can't plan for other than ensuring that each vehicle has a clear buffer around it so that it can make split-second adjustments to these perturbations. So let's say that each aircraft is given a bubble that is 500 ft in diameter and that your nav system routes each vehicle so that no two bubbles intersect, ensuring a one thousand foot separation between vehicles. How many vehicles could be in the air at one time?

Let's use Denver as an example. The land area of the entire Denver Metropolitan Area is about 8300 sq mi (that a LOT more than just Denver, it lays across ten counties). How high up with these vehicles operate? They are going to want to go no higher than they have to. Climbing is what is going to use up a very disproportionate amount of energy, plus put extra wear and tear on the drive components. Plus, it take time and distance to climb and descend, so intracity flights won't have the time to ascend very high. Air taxis currently operate between 300 ft and 1000 ft AGL typically, higher in congested areas since they have to be 1000 ft above the highest manmade obstacle within a 2000 ft radius. But many waivers are granted by the FAA to allow operations below this, underscoring how much operators want to avoid climbing up. So let's use a 3000 ft thick operating volume for our example. That gives us an operating volume of 694e12 cubic feet. Each vehicle's bubble, at 500 ft, has a volume of 524e6 cubic ft. Now imagine that each bubble is a marble and they are all packed together as tight as possible, like marbles in a jar. The packing efficiency is about 75%, so we could fit 982 thousand of them into that volume. But, they couldn't move! Imagine removing half of the marbles in the jar. Could the rest move relatively freely from one position to another without ever bumping into another one? Not even close! Even at 10% packing, they could constantly have to be dodging each other, which would significantly increase the cost for operators, increase travel times, and aggravate frustrations. But let's say that we believe that our centralized nav system can handle this -- including all of the real time adjustments needed as each vehicle deviates slightly is speed, course or altitude due to a myriad of variables. At 10% packing, you are new down to fewer than ten thousand vehicles over the entire Denver metropolitan area -- and a highly disproportionate number of these vehicles are going to want to operating over a small fraction of this space (the central business district and a few other areas), but they are restricted to similar volume densities as elsewhere. So, pushing things to pretty extreme limits that are likely not to be acceptable to anyone, passengers, operators, or regulators, and assuming that traffic density will be every bit as heavy at the edges of the area and over open space regions as over the tech centers and downtown, you can support about about 1% to 2% of the number of air vehicles as they already have ground vehicles. In practice, because of low demand over much of the area, the peak numbers that the system can expect to see and handle are probably going to be range of a thousand or two. Another point to consider, ground vehicles are much more able to operating in overly-dense traffic conditions at peak times than air vehicles. They have the ability to sit in bumper to bumper traffic that moves at a crawl, but still gets people to their destinations. That is not an option for air vehicles. They can't hover a couple feet apart indefinitely.

Another question: Who is going to pay to develop, install, operate, and maintain this centralized navigation system when the practical scale of operations is going to be so limited?

So what will we see?

I would predict a modest increase in air taxi services. Perhaps tenfold. Cities that currently have no air taxi services will likely see some spring up. I don't expect that there will be much increase in all of privately owned vehicles and it will probably be the same people that already use air taxi services.

 

Returning to the original topic of quantum computing, and how it is going to be so incredibly much faster, and thus somehow more powerful,just picture this: Artificial intelligence gets started running on a quantum machine, and some of the present level "bad actors" start using it for assorted criminal activities. Do you suppose that law enforcement will be difficult?? Does anybody else see some unhappy situations developing because of that?? Such as loss of all personal freedoms? And a whole lot of other issues as well??

It is time that a lot of folks should start thinking about those secondary results of things that "seem like a good idea at the time."

Currently we have a lot of terribly toxic materials, and a lot of really dangerous weapons , which have the potential of ruining everything if a few folks goof up.
So the importance of deciding just how much more danger should be created as a secondary result must be considered by those able to do something about it. All I can do is suggest that folks stop and think carefully.

 

Returning to the original topic of quantum computing, and how it is going to be so incredibly much faster, and thus somehow more powerful,just picture this: Artificial intelligence gets started running on a quantum machine, and some of the present level "bad actors" start using it for assorted criminal activities. Do you suppose that law enforcement will be difficult?? Does anybody else see some unhappy situations developing because of that?? Such as loss of all personal freedoms? And a whole lot of other issues as well??

It is time that a lot of folks should start thinking about those secondary results of things that "seem like a good idea at the time."

Currently we have a lot of terribly toxic materials, and a lot of really dangerous weapons , which have the potential of ruining everything if a few folks goof up.
So the importance of deciding just how much more danger should be created as a secondary result must be considered by those able to do something about it. All I can do is suggest that folks stop and think carefully.

Quantum computing will be better or equal of classical computing because quantum gates can simulate classical gates while classical gates cannot simulate superposition which is how many algorithms are sped up.

 

Quantum computing will be better or equal of classical computing because quantum gates can simulate classical gates while classical gates cannot simulate superposition which is how many algorithms are sped up.

What happens when quantum computing is utilized to run artificial intelligence?? How likely is that??
AND THEN WHAT HAPPENS when "bad actors start using that combination??? IS that possibility probable??? How could it even be controlled??

THAT eventuality is my concern.

 

What happens when quantum computing is utilized to run artificial intelligence?? How likely is that??
AND THEN WHAT HAPPENS when "bad actors start using that combination??? IS that possibility probable??? How could it even be controlled??

THAT eventuality is my concern.

Superposition alone doesnt speed up a algorithm because you can only have access to one measurable at a time.What increases speed is phase kickback(
https://en.wikipedia.org/wiki/Phase_kickback
Think of it like this:

if you have a function f(x) and f(x) is 1 by 1 (for every different x you get a different f(x)) then classical and quantum computing are the same thing.However if you have a function that with different x's you get the same output of the function then classically to learn about the result of different x's you need to run the program as many times as the number of your inputs.Phase kickback uses superposition in a way that you dont need to check for every x to learn how the group of inputs with the same output would behave so you dont need to check a function over its entire domain.