The production of Elbrus is transferred from Taiwan to Zelenograd, where lithographic systems are also being developed

The comment of Konstantin Trushkin speaks in favor of the first option,
MCST Deputy General Director for Marketing: “We see that domestic factories can create worthy processors with sovereign Russian technologies for critical information infrastructure (it includes communication networks and information systems of government agencies, energy, financial, telecommunications and a number of other companies. – RBC), areas information security and other markets. To achieve sufficient practical performance, we see a large reserve in reworking the source code and optimizing the software for the Elbrus architecture.

More about the possibilities of the domestic manufacturer

With a high degree of probability, it can be argued that if it is not possible to agree with foreign factories, then modern processors will have to be abandoned. At the moment, Mikron can produce chips using 65 nm technology, but in the case of MCST, only 90 nm process technology will be available. Unfortunately, in this case, the range of tasks that domestic chips can perform will be quite narrow. These are communication systems, thin clients, cash registers, etc. The operation of such chips in more or less modern PCs, laptops and other systems is out of the question.

Experts believe that there will be few computers with such chips, and software developers will have to adapt software to domestic chips. But, since MCST is the only company in the Russian Federation that can produce processors with its own architecture, then it, likely to support.

If you are reading our blog, you may be interested in these texts:

How to optimize spending on infrastructure rental
Ready-made container registry – who needs it and how to use it
How we launched a remote cloud region in Uzbekistan

And this despite the fact that Russian buyers of processors and equipment where they are installed were not very pleased with the results. So, at the end of last year, Sberbank tested two types of servers on Elbrus, after which it was concluded that it would not be possible to use such systems in the organization’s infrastructure.

In addition, domestic systems and state-owned companies were criticized, which were dissatisfied with the high price, not very high performance and low energy efficiency of chips and equipment. Then representatives of state corporations said that they were not against switching to domestic equipment, but only if it could be used in regular work scenarios, and all this would cost no more than in the case of working with foreign systems.

There is also an opinion that domestic chips will find their buyer for those tasks where particularly high performance is not needed. Plus, software for domestic systems can still be configured to get maximum performance from the chips. But, as mentioned above, the range of tasks that will be solved in this case will not be very extensive.

The problem is that as long as the chips are produced in the thousands, not the millions, their cost will remain high. But there will obviously not be millions, since the domestic market of Russia is not as large as, say, in China. Now the volume of production of silicon wafers for chips with a topology of 180-90 nm is something like 3000 per month. Mikron will be able to double its wafer output by 2025, but these are still not figures that will significantly cut the cost of the chip.

There is hope?

It is possible that the problem with the outdated technical process will be solved and it will be possible – but this is clearly not a matter of the next year or even two. But still, there is a construction site in Zelenograd, where work on the construction of the factory has started. She, according to the plan, will produce processors using 28-nm technology, which will be used to assemble domestic electronic devices of various types. The construction is planned to be completed at the end of 2024, but, of course, this does not mean that the factory will immediately start producing chips.

As we already wrote, the factory is being built as part of a large-scale plan for the development of the domestic electronics industry, the plan was approved two years ago, in 2020. The national project is designed until 2030. Its author is the Ministry of Industry and Trade of Russia, 22 working groups participated in the preparation of the document. According to the plan, the state plans to invest 3.19 trillion rubles in the project. The funds will be distributed in four main areas – for the development of infrastructure, domestic electronic products, increasing demand for it and the personnel component.

Construction is being carried out on the site of two demolished old Angstrem buildings. The total construction area is 50 thousand square meters. The buildings will house the research and educational infrastructure of the Moscow Institute of Electronic Technology.

But if the plans to develop domestic lithographic systems for the production of chips using the 28 nm process technology are implemented, then this will happen no earlier than 2030, that is, in 8 years. And all this time, the chip production plant will require support of hundreds of millions of rubles a year – and it doesn’t matter if something is produced or not. Production of this type requires huge amounts of investment annually.

In general, it is too early to say anything definite about success in this direction. Probably, it will be possible to determine more or less accurately in a couple of years, when something will already be built and details about the lithographic systems being developed will become known. In the meantime, we will have to be content with chips created using 90 nm topology and older technologies.

Similar Posts

Leave a Reply