Quantum computers could smash speed and security bottlenecks in electronics and completely change the world as we know it.
Or at least that's the promise of the technology, which always seems to be right around the corner. Now, however, a team led by researchers at Microsoft have said a working, useful quantum computing system finally seems to be within reach.
The team just published a paper about their scheme on the preprint server arXiv, and it predicts a working quantum computer in the next 10 years.
"Recent improvements in control of quantum systems make it seem feasible to finally build a quantum computer," the researchers write in their study.
This is a contentious announcement, as a story at The Register points out, because a company called D-Wave claims to have built the first quantum computer (pictured above). However, research published last year demonstrated that D-Wave's prototype was no faster than some regular computers. D-Wave disagrees, and said the researchers did not fairly test the computer.
The new study suggests a "hybrid" quantum computer - one embedded in a traditional, silicon-based computer - could process simulations that would be useful to materials scientists, for example. And their hybrid model, the researchers write, could also help solve a lot of the problems that have plagued efforts to build effective quantum computers.
How quantum computers work
The computers we're all familiar with use "bits" to represent information in "0"s and "1"s. Strings of bits can represent specific numerals and letters, for example, as well as more complex data like image and audio files.
Computer processors have improved over the years to handle more bit-based data, but engineers think a physical bottleneck is on the horizon: There's only so much you can do to shrink circuits to process more and more bits. Once that limit is reached, to improve performance computers may start to take up more space and use more energy.
Quantum computers might leap over this hurdle by not only increasing the number of bits, but changing their very nature. Instead of using bits that only have a "0" or "1" setting, quantum computers use a weird-but-true law of physics - where a particle can exist in multiple places at once - to make "qubits." Qubits have an extra setting, so they can exist as a "1," or a "0," or both at the same time.
This flexibility enables the devices to process much, much more information at once than regular computers. For example, a quantum computer could theoretically perform a complex calculation in a few minutes that might take a regular computer years to complete.
Developers dream of harnessing such speed to create exceedingly accurate weather models, revolutionize the cybersecurity and cryptography fields, and instantaneously sort huge piles of data (no wonder Microsoft is so interested).
How to build one in 10 years
Quantum computers and their qubits have the potential to shake up entire industries, including those trying to discover brand-new molecules, materials, and pharmaceuticals. The materials industry in particular, write the Microsoft researchers, "will profit enormously from the availability of quantum computers as special-purpose accelerators."
The problem is that quantum computers are very tricky to build.
Prototypes do exist, but they only work at temperatures near absolute zero - colder than the darkest depths of space. Also, the smallest disturbance can jostle the qubit particles and corrupt their data. The devices are also very expensive, since cooling isn't cheap and most designs call for pricey materials like diamond or cesium.
The hybrid model proposed by Microsoft researchers would partly address these hurdles by integrating a small quantum computer into a regular silicon-based computer. According to the group's study, the regular part could take most of the basic computing load, freeing up the quantum part to tackle really difficult problems. The whole device may be less expensive to build and operate than other quantum computers, too, since it'd require less cooling and use fewer expensive materials.
Qubit stability, however, is still a big snag for the device and one of the biggest roadblocks to the age of superfast computing. But Google's quantum research team has already made headway on that problem - they figured out how to stabilize a small chunk of qubits earlier this year.