Why is the North American Bitcoin Mining Machine Chip Company doomed to fail?

Colin Wu

According to Forbes on August 24, as part of the round B financing, Blockstream, a bitcoin infrastructure company, raised $210 million at a valuation of $3.2 billion.

The new capital is accompanied by the acquisition of intellectual property rights of Spondoolies, an Israeli bitcoin mining hardware manufacturer specializing in mining machine ASIC chips. The acquisition amount is not disclosed.

The Spondoolies team will also join Blockstream to establish the company's ASIC department. By raising funds, the company hopes to expand its mining business and develop new products“We actually expect Blockstream branded mining machines on the market in the third quarter of next year, "said Adam back, CEO of Blockstream.

Coindesk introduced that Blockstream was founded by CEO Adam back, bitcoin core developer Gregory Maxwell and 9 others (including Pieter wuille, Erik Svenson, Jonathan Wilkins, Austin hill and Jorge Timon).Famous people who have worked in the company include Andrew poelstra, Samson mow, Lawrence Nahum, Christopher Allen and Matt Corallo. It provides a source of funding for bitcoin developers.

Blockstream's mining facilities have a total capacity of 300 MW and has also purchased a large number of bitcoin mining machines from Whatsminer. On January 28, 2021, blockstream announced that it had purchased a Whatsminer mining machine worth US $25 million from MicroBT.

Spondoolies is an Israeli mining machinery company. According to people familiar with the matter, it had a certain strength before 2016, but with the shipment of new models such as Antminer S9, which swept overseas miner manufacturers, Spindoolies had to exit sadly.

According to CCN, as of May 4, 2016, Spondoolies tech had stopped operation, and its seven employees asked to dissolve and liquidate the company "because of competition with Chinese bitcoin hardware developers". The third generation chip design of Spondoolies still stays at 28nm, while S9 has reached the position of 16nm chip. Shortly after leaving, guy core, the founder of Spondoolies, turned to privacy coin and founded beam.

Due to the 521 renovation, almost all large mines in China were closed, and a large number of mines were transferred to the United States. The US mining pool, led by DCG foundry, is also providing loan services to the mine through its powerful financial group, so as to introduce its hashrate.

On August 15, Barry, founder of DCG, tweeted that the foundry mine pool from the United States has ranked first in the world. However, this should only be a short period of 24 hours to unearth the total number of blocks, and the current ranking is stable at the seventh position.

But in the short term, Chinese mining machine companies will not face too much risk of domestic policy. Their policy risks actually come from the United States, high process chips were severely inspected by the United States and may block high process technology from Chinese enterprises. In addition, if Taiwain has a war in the future, Bitmain, which depends on TSMC, will be under strong pressure. At that time, the capacity side may also be an opportunity for North American chip companies.

However, on the design side, it is very difficult for North American companies to catch up. Different from mining farms and pools, bitcoin chips have huge design barriers. Except that Whatsminer is relatively smooth all the way (inheriting the previous experience to a certain extent), Bitmain and Canaan have invested a lot of money and kept trying and making mistakes to reach the competitive level of mining machine chips today.

Although the logic of bitcoin miner design is simple, the technical difficulty is not small: it is necessary to design a hash core as efficient as possible on the low voltage close to the physical process limit, and calculate with the limit voltage. It is necessary to be familiar with the chip back-end design and the process nodes of the wafer factory, and use a large number of unconventional designs to design competitive miner chips. Take the bitcoin chip, the difficulty of the front end of the chip is very low, and the difficulty lies in the back-end optimization.

This Israeli company did 28nm in 2015, under the current high-end process competition in the industry, which is mainly around 7nm, the patents of Spondoolies in that year did not play a great role. On the other hand, chips are more competition for design talents. The lack of chip talents in North America is not conducive to the development of local mining machines. On the contrary, China is pushing chip industry, with a steady stream of chip talents.

Finally, chip companies in mainland China also have geographic advantages and have close cooperation with TSMC, Samsung, and SMIC.  Blockstream is very difficult to build cooperation with wafer factories and obtain production capacity. The financing amount of US $210 million may be enough to build a team and carry out, but it is not enough to make several trials and errors. From this perspective, the mining machinery field is still a relatively solid situation in China for the time being. Blockstream may be able to build mining machines, but it is difficult to be competitive. More will give up halfway or cannot be mass produced for their own mining.

In contrast, the bigger competitors of the three mining machinery manufacturers may come from China in the future. As Ethereum will completely turn to POS next year, all mining chip talents and companies will flow into the field of BTC mining machine chip design. This may be the reason why the chairman of Bitmain, Micree Zhan, made an "death threat" to his resigned employees.