Schrödinger’s Cat vs. Moore’s Limit

Nov 05, 2018
Blog

Today the neural networks are all the rage. They are capable of writing poems, driving a car or helping you to choose the best vacation photos. What is more, they even learn themselves! Jolly good and rightly so. But in the meantime there’s a technology under development, that will change usual calculation principles. Let’s sort out what a quantum computer is, what this all is about, and what a cat has to do with it.

What are the quantum calculations?

Familiar to us computers work on transistors and use bits as the minimum unit of information. A bit can be one or zero, true or false. Using bit combinations, one can construct expressions of formal logic and depict algorithms. Taking a fair amount of bits, you can digitize almost anything.

In quantum computing, the minimum unit is called a qubit. And the point here is that it can be zero and one at the same time. By the principle of superposition, the quantum bit is in all possible states at once. An example of superposition (i.e. state mixing) is the famous Schrödinger’s cat. Until we open the box, the cat can be equally alive (one) or not (zero). It is not so easy to comprehend as quantum mechanics is by definition one of the most difficult sections of physics to perceive. It’s not for nothing that one of its principles is called quantum entanglement. But the general idea is that by moving from bits to qubits, it becomes possible to solve certain tasks millionfold faster through processing all possible states at the same time, rather than going through them one after another. Some 50 qubits will be enough to solve everything that we can offer such computers.

Let us say that someone needs to break the cryptographic protection of a bank. To do it, one needs to decompose a 500-digit number into prime factors. Such a task will take a modern supercomputer only 5 billion years. But a quantum will find a solution in a matter of seconds – and you can withdraw money from the accounts. But so far everything is not that simple.

Why is it important?

Traditional computing is completely tied to silicon chips. With the development of technology, we can place more transistors smaller in size on the same microprocessor area. This very indicator affects the increase in computing power in new processors. The Founder of Intel, Gordon Moore, formulated a law named after himself – according to that law the number of transistors doubles approximately every two years. But now technologies are finally approaching Moore’s limit — the moment when it is physically impossible to make a transistor smaller. In the end, the size of a silicon atom is 0.2 nanometers, and the latest development by the Taiwanese chip maker TSMC (they work with AMD and Apple, for example) is a 7-nanometer processor. It is also not an option to make processors bigger to accommodate more transistors, as the speed of light imposes its limitations, the data on the chip cannot be transmitted faster than it. For a while, we will continue to enjoy more efficient algorithms and parallel calculations. But a quantum computer is for the time the only possibility for an increase in computing power not to stop one day.

What does it take?

Actually, the main difficulty is the creation of the qubit itself and providing it with guaranteed quantum uncertainty, that is, the absolute absence of the influence of external factors. After all, as soon as at least something has affected the quantum state, the superposition disappears, and a steep qubit becomes a boring bit. So far, it takes a lot of experiments to find the perfect qubit and conditions for it, for example, with the angular momentum of an electron of a particular substance at temperatures close to absolute zero, or with ions in vacuum traps, or with the laboratory obtaining new materials, which are both a superconductor and a ferromagnet (a substance that destroys the state of a superconductor at a certain temperature) … In brief, the work is in progress, and the possibility of its success has already been proven. So the result is only a matter of time. Most likely, a full-fledged quantum computer will appear within seven years.

Where is the catch?

In fact, many companies have already announced prototyping but there is a number of problems:

1. Are they really quantum? The paradox of that very ‘cat’ is that it will cease to be in superposition as soon as we open the box to see how it is doing. So it is impossible to check the status of qubits.

2. How to check them? Indeed, traditional methods cannot be used to determine whether the result of quantum calculations is correct. But there is some progress. You can find a problem that can be solved only by quantum computing or teach quantum computers to double-check each other, and also to select for checks the tasks, the final result of which we know. For example, digitization of the structure of our DNA cells.

3. How to use a quantum computer? Unfortunately, there are not too many tasks that require quantum computation. However, it is already obvious that not only mathematics and physics but also chemistry, biology, astronomy, and sociology have such tasks. So, have a computer, will work.

If you want to figure out how this works, pay attention to the Quantum Development Kit package developed by Microsoft. It includes a quantum computing emulator, a specially developed language for them, Q #, and many reference materials. See what the future might be.

Previous articleNext article