[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 43 KB, 496x502, Jet.jpg [View same] [iqdb] [saucenao] [google]
5292264 No.5292264 [Reply] [Original]

Will quantum computers be able to predict the turbulent flows?

>> No.5292271

>>5292264
prove that quantum computers can solve problems that non-quantum computers can't

>> No.5292286

>>5292271
They can solve quantum problems like Schrodinger's cat.

>> No.5292303

>>5292271
They can efficiently solve some (restrictive) classes of problems that regular computers cannot solve efficiently (hear: the asymptotic complexity goes from exponential in time to polynomial in time).

Turbulent flows, meteorological predictions etc are hard to solve because of the chaos theory statement that these systems have considerably different behaviors if you change even-so-slightly the initial conditions, which means that to predict the correct behavior of the system for T+1 periods instead of T, the precision your need is multiplied. Your required precision grows exponentially as you try to predict the behavior of the system for a linearly increasing period of time.

Quantum computers do NOT solve this better than regular computers.

>> No.5292557

Wii quantum computers ever work as normal processors or are they only good for some problems?

>> No.5292560

>>5292557

*will

>> No.5292570

I can't wait for Daniela to make a video on quantum computing. I really want to know more about that subject.

>> No.5292608

>>5292303
So in other words, you don't know of any instance of a problem that quantum computers can solve that regular computers cannot.

>> No.5292654

>>5292608
It certainly won't help you find a qt gf

>> No.5292862

>>5292654
Can quantum computing simulate me a qt? waifu

>> No.5292871
File: 484 KB, 250x188, 1352136205025.gif [View same] [iqdb] [saucenao] [google]
5292871

>>5292862
You can already simulate your own waifus.

>> No.5292884

>>5292271
Provably, there are none that can't be solved given unlimited time. However, if you want to factor a 50,000 digit product of two large primes within the lifespan of the universe....

>> No.5292944

All turbulent flows can be computed numerically if the grid spacing is equal to the Kolmogorov lengthscale and the time step is equal to the Kolmogorov timescale (http://en.wikipedia.org/wiki/Kolmogorov_microscales).). Unfortunately, this takes a very long time and because of this, simulations of real turbulence are limited to very small scales.

The estimated time for the simulation of the flow over a 787 at Mach 0.8 (80% of the speed of sound) using a real turbulent flow is on the order of 10^81 years. So, if a quantum computer was three times as fast as current computers that simulation time would be reduced to 3.33*10^80 years.

The best chance at developing computational models of turbulence would be to develop better approximate turbulence models (i.e. Large Eddy Simulation).

>> No.5292953

>>5292608

I think the purpose of quantum computers is not to solve new kinds of problems, but to solve the old ones more quickly.

>> No.5292985

>>5292944
It's not that a quantum computer is 3 times faster, or even some constant factor. It's that it is exponentially faster. The bigger your problem, the bigger your savings. Better models are great, too, but truly massive parallelization is king.

>>5292557
Yes, you can simulate a normal processor with a quantum one.