The world's fastest computer will work like a human brain.

A large group of scientists and researchers are developing the fastest computer known to mankind, which should work like a human brain. In October 2013, a conference was held in Switzerland dedicated to the ambitious Human Brain Project, which brought together scientific personnel from 135 scientific and state institutions to create a computer brain. The project cost is estimated at 1.6 billion dollars.

Recall that this is one of the two grand initiatives of the decade - along with graphene.

The human brain is the most complex machine out of all existing ones, so there is nothing surprising in the fact that technology tends to be similar to it. A computer being developed must be 1000 times faster than the fastest computer we use today.

Before you remember the 2001 Space Odyssey, we warn you: not every computer brain is HAL 9000.

The first phase of the project is expected to last about 10 years, and during this time scientists will try to better understand the functions of the brain. Then scientists hope to comprehend how we learn, think, see and hear.

Currently, computer stuffing reaches a speed of one quadrillion operations per second. But Henry Markram, director of the Human Brain Project, is not enough.

"Well-known manufacturers of supercomputers, such as IBM, Cray, Intel and Bull, are planning to create a machine that surpasses all essential things by about 2020," said the director in an interview with Fox News. - "Therefore, we are confident that we will have the cars that we need."

These incredibly fast computers need new forms of memory and the power of scientists who can develop new storage technologies.

The main goal of the project, as indicated on its website, is “to bring together our rapidly growing knowledge of the human brain”. Modeling the human brain will give an idea of ​​the inner workings of the brain and where our thoughts and emotions originate. The implications go beyond technology; such modeling will help us learn how to heal the human brain or improve its individual functions.

The project is still in the first planning stage. The brain is about the same complexity as the universe. In one kilogram of substance there are as many neurons as there are stars in the Milky Way. Powerful? Powerfully.

Neurons in the brain - how many stars are in the galaxy

Popularists of science like to say this phrase when people try to estimate the scale and complexity of their own brain. Recall: in an ordinary galaxy about 150-200 billion stars. That is why the media likes to measure the distance by football fields, the mass - by the number of fully loaded Boeing 747s, energy - by Hiroshima and so on.

Even if we cannot imagine either one or the other, an attempt to compare two numbers — stars and neurons — leads people into a state of awe and quiet horror. And as conscious beings, we love to look for equivalents and correspondences; the better, if the unit of measure is epic or important in itself - a star or a neuron.

For many years, neurologists have said that there are about 100 billion neurons in the human brain. Remarkably, none of them published a scientific piece of paper that would confirm the number. But in 2009, Azevedo and colleagues still found a more accurate number. Do you know how much?

How many neurons in the human brain ? Approximately 86 billion neurons.

Recent calculations of the stars of the Milky Way showed a number from 200 to 400 billion. Almost, but neurons are still smaller. Does not converge at all, to be precise.

But how did scientists find out that there are 86 billion neurons? How did they come to this number? The easiest way would be to estimate the number of neurons in one part of the brain, and then extrapolate to the rest. Remarkably, the same method is used to count the stars in the galaxy. But he has a few problems:

1. The density of neurons in the brain is not uniform. For example, in the cerebellum is almost half of all neurons of the central nervous system, but it is much less than half the brain in volume. By the way, if you want to admire the beautiful gallery of neurons in colors, you are welcome.

2. Even in one area of ​​the brain it is difficult to count the neurons, since they are very dense, transparent and intertwined with each other - it is even difficult to separate them. One of the ways to “illuminate” neurons is to use a staining technique that will make neurons so visible that they can be counted. The classic method is the “Golgi stain,” named after the Nobel Prize winner Camillo Golgi. This method allows you to calculate a small percentage of neurons, making them visible, and then extrapolating to the rest. But it is also inaccurate.

The new method, which gives the number of 86 billion - is pure and unique.

The method involves the dissolution of the cell membranes of brain cells and the creation of a homogeneous mixture - such a "puree". The nuclei of these cells are painted in different colors in order to distinguish neurons from glia, and this makes it possible to calculate the number of cell nuclei belonging to neurons, and then to scale and get the total number. The advantage of this method is to resolve the issue of heterogeneous density of neurons in one brain region with extrapolation to the whole brain.

Well, without extrapolating in any way. But maybe by the Human Brain Project we will finally know the exact number.

The article is based on materials https://hi-news.ru/computers/samyj-bystryj-kompyuter-v-mire-budet-rabotat-kak-mozg-cheloveka.html.