it wont affect anything for a while. it will take a few years before devs can basically assume people have them. then on top of that they have to scale up their stuff so it takes advantage of the greater power.

^ Devs would basically have to learn programming all over because of the qubits.

Lots actually...its just one of those interwoven areas where you learn a crapload from a lot of sections of science and everything builds from there.  ...but I do admit I am a bit of a freak when it comes to how I spend my free time. I absolutely love reading up on the latest nerdy innovations =P

Lucky for me there actually are some girls out there who enjoy it too and when you have someone who is just as interested in this crap as you are it just works~

@omgwtfbbq,

I was kind of giving the dumbed down version since most people don't understand the basics of path logic well enough to follow it.  Just trying to explain as little new stuff as possible etc...

I honestly haven't heard much about the genetic algorithms yet sadly, but I do understand the basics of it, particularly the biology roots. 

Well, it will break the Microsoft and Intel/AMD domination of the industry, so it won't be a PC (as in IBM PC, the PC in PC vs. Mac). I'm voting Linux, BSD or OpenSolaris on a future IBM quantum architecture.

People thought that multi-core was too hard to program for, but programming tools have arisen which make it easier, and Intel claims to be developing a compiler for its upcoming Larabee GPGPU that  makes it entirely transparent.

There will be some overhead cost to abstract the complexity of quantum computers to the programmer, but if you're dealing with a 1000x speed-up you're still going to have unimagainable improvement over any future x86 designs. I predict we'll see an free GNU Compiler Collection 'Quantum architercture' compiler for C and C++ in a few years after the first mainstream processor is released.

Quantum computing is a different paradigm, you can't put a label on it such as "million times faster". For certain classes of problems, it gives an exponential speedup, which amounts to taking the logarithm of the running time (an operation which is not expressible by "x times faster" in the general case, since x gets bigger and bigger as the problem size grows). In other classes of problems (such as search), it gives a quadratic speedup, which amounts to taking the square root of the running time (again, not expressible as "x times faster" in general).

It's a totally different class of computing, which relies on very unintuitive physics, which (depending on taste) are best understood if you assume that there are parallel universes running different paths of the computation.

As Origin said, quantum computing is not going to replace traditional computing, since it's not applicable to all classes of problems. It can be used as a co-processor in a normal PC, if the PC needs to perform certain kinds of problems which benefit from the QC.

As for human brain emulation, it would be useful if the human brain relies on quantum effects, which is not known to be the case. However, by most estimates I've seen, traditional computing will be able to emulate a human brain in 20 years at most (at least in terms of hardware power, software is an unknown yet).

If you want to know how fucked up the concept of Quantum Computing is, here's an example:

Imagine you have an unsorted list of N things. If you want to search this list for a specific value, the best you can achieve with traditional computing is to check each value of the list, stopping when you find the one you're looking for. This method runs in linear time (the expected number of checks is 0.5 times N, which is proportional to N, hence linear time).

Now with a Quantum Computer, using Grover's algorithm you can actually search the list in sqrt(N) time! If you don't yet see how fucked up this is, consider that you're actually doing less operations than the number of items in the list, i.e., for a big enough problem, the traditional computer (or any human looking at the list) didn't even have time to check each element yet, no matter how fast it is!

Is there no way to abstract the general languages we use, such as C, to put all of the code in operations the qauntum computer is highly efficient at? Surely if the computer functions as a Turing machine it can do the exact same classes of computation as for example x86?

i'm gonna guess that since current languages are compiled to do bit operations, not really. you can probably develop some sort of virtual machine, but for that i won't even venture to guess how efficient it'll be.

my intuition would be that things are so completely different in a quantum computer than the current logic we use in classical machines would be far from being efficiently translatable.

that's probably completely BS though. am i in the right ballpark though?

come to think of it, probably with a quantum computer the following would be a simple thing to do: only run inefficient code blocks on the quantum computer by passing over the data. this would be a lot like writing optimized code in C while the main program is in python or something.

that actually sounds pretty straightforward!

Not all conceivable quantum computers are Turing complete, but some are. It all depends on the logic gates which are implemented:

http://en.wikipedia.org/wiki/Quantum_gate