Where is the future of graphics card miners?

Key points:

1. Among all PoW currencies, Ethereum’s anti-ASIC algorithm makes it the largest concentration of graphics card mining machines.

2. With the popularity of DDR5 particles, the Ethash algorithm will no longer be ASIC-resistant. Vitalik therefore proposed switching Ethereum to PoS mining. It is expected that by the end of 2020 at the latest, 80% of ETH will be produced by PoS and 20% by PoW.

3. Judging from the computing power on the Ethereum network, there are about 800,000 to 1 million graphics card mining machines running. Once Ethereum switches to PoS, these graphics card computing power will inevitably be freed up. Where will they go in the future?

4. The value of old ASIC-resistant PoW currencies such as ETC, ZEC, and XMR cannot accommodate so much computing power. Specifically, the daily electricity bill of 800,000 mining machines is far greater than the tokens generated by these networks every day. Therefore, the economic ecology of graphics card mining may face a large-scale reshuffle.

5. The worst case scenario is that a large number of GPU mining machines are shut down, resulting in a large amount of stray computing power in the ecosystem. These computing powers can be turned on at any time to carry out a 51% attack on any ASIC-resistant currency.

6. CZZ is a new PoW + 0 pre-mining project, with no financing and no pre-mining. CZZ is committed to becoming a public infrastructure and welcomes Ethereum miners to actively participate.

7. Since CZZ has just started, the total token stock is small. At this time, the intervention of computing power can affect the average cost of the coin, thereby affecting the future value of CZZ. However, for ETC, ZEC, XMR and other currencies with huge stock, the intervention of Ethereum miners can only increase the future currency cost. However, in the face of huge stock, the future coin acquisition cost has little impact on the average coin cost.

In today's digital assets, if calculated by market value, PoW still has an absolute dominance. Graphics card mining has always been a force that cannot be ignored in the PoW camp. The biggest advantage of graphics card mining machines over ASIC mining machines is that graphics card mining machines can switch programs at any time to mine different currencies. Since ASIC mining machines are single-function hardware, if the currency being mined does not perform well, there will be no other choice. Only when graphics card miners can develop robustly can ASIC-resistant mining algorithms be supported. How to effectively resist ASIC is still an important issue for the decentralized ideal of blockchain.

Fast forward to 2020, and we find that this era is not friendly to GPU miners. Ethereum, the largest currency that accommodates GPU miners, is about to switch to PoS. In 2020, Ethereum will complete the Constantinople upgrade, and the mining reward will be changed to 80% PoS and 20% PoW. What consequences will this have on the economic benefits of Ethereum miners?

The Ethereum miner group is roughly divided into two components, the GPU group and the ASIC group. Ethereum's mining algorithm relies on memory hard to resist ASIC. Due to the continuous update and iteration of technology, the cost performance of DDR chips has been greatly improved in recent years, and now DDR chips can be directly embedded in ASIC design. Now Ethereum's ASIC mining machines have just reached the critical point from the perspective of mining economy, and cannot create an overwhelming advantage for GPU mining machines. Most GPU mining machines are old machines, and the fixed investment costs have long been recovered, but they are more power-consuming. The fixed costs of newer ASIC mining machines may not be recovered in time, but they are more power-saving than GPUs to achieve the same computing power, so the operating costs are lower than GPU mining machines. Once Ethereum's PoW production is reduced, the graphics card mining machines that consume more power will definitely be eliminated.

So how many GPU mining machines are there on Ethereum? About 800,000 to 1 million. The biggest question for the GPU mining community now is, once Ethereum PoS, where will these computing powers go in the future? The value of old ASIC-resistant PoW currencies such as ETC, ZEC, and XMR cannot accommodate so much computing power. Specifically, the daily electricity bill of 800,000 mining machines is far greater than the tokens generated by these networks every day.

If most of these hashrates eventually choose to shut down, it will threaten the security of all ASIC resistant PoW currencies. Because the price of a graphics card mining machine in a shut-down state will basically return to zero, people with ulterior motives can hoard waste mining machines at extremely low prices, and there will be no operating costs as long as they are not turned on. When a 51% attack is needed on a network, it can be turned on instantly, and then shut down continuously after the attack is completed. This will undoubtedly cause a devastating blow to the entire ASIC resistant ecosystem.

Therefore, the urgent task now is the emergence of new ASIC-resistant PoW projects, or old coins like ETC need to appreciate by dozens of times before ETH 2.0 goes online. Of the above two possibilities, I think the former is easier to achieve and fairer to the miners. This is where we have to introduce the CZZ project.

Since Satoshi Nakamoto released the Bitcoin project, digital currencies have achieved remarkable achievements in terms of market value. However, since there has been no super application, Bitcoin or other digital currencies have not yet participated in large-scale daily transactions. We must reflect on what other obstacles are ahead? As we said before, the most important obstacles are roughly the following three: poor usability, unfairness, and lack of economic model. This article mainly analyzes these three issues.

As we all know, the speed of Bitcoin transfer has always been criticized. Less than 10 transactions are processed per second, and the transaction confirmation time is often as long as an hour. Although the main network is decentralized, users often need to use third-party centralized services to use it normally. This is also the weakest link of Bitcoin. Since EOS and TRON, the "third-generation public chain" projects have made great improvements in tps and other issues compared to BTC and ETH networks, but what they have lost is decentralization.

Decentralization is the soul of any digital currency. Blockchain technology has two incarnations, distributed ledger and trust generation machine. The former is actually a gradual improvement of the database structure, while the latter is a revolutionary advancement of blockchain technology. You can imagine that whether you buy a token of a centralized project, or a stock or fund, the personal reputation of the founder will play a decisive role. But do people know who Satoshi Nakamoto is? As far as he is concerned, his reputation can be said to be zero. So why do people still believe in Bitcoin? Because everyone is assured of the competition mechanism of Bitcoin's accounting rights, and everyone trusts the Bitcoin mainnet. As for who Satoshi Nakamoto is, it is no longer important. This is the essence of a decentralized network.

In summary, only two types of blockchain projects can achieve good results. One is a centralized project, but it requires the centralized operator to have strong social credibility, such as the central bank digital currency planned by a sovereign state, or Libra jointly initiated by multinational companies such as Facebook. The other is a completely decentralized project, such as Bitcoin. In a completely decentralized state, everyone's attention will return to the essence of trust brought by cryptography: in cryptography we trust.

CZZ does not have any super node system, any privilege system, no financing, no pre-mining, and all participants are treated equally. Only in this way can CZZ become a public infrastructure without obvious beneficiaries. The earliest Bitcoin issued tokens with the fairest "zero pre-mining + PoW" mechanism, which is a tradition worth adhering to. For example, many projects issued later often pre-mined 50%-99% of the tokens with private placements, and these tokens were almost 0 cost. The more projects with nearly 0 cost tokens, the more unfair it is to other participants, the lower the desire of subsequent participants, and the easier it is to be "dumped" by big dealers. On the contrary, if you want a token to become truly valuable, you must ensure that there are no nearly 0 cost tokens in the entire network.

The "zero pre-mining + PoW" mechanism is a relatively effective mechanism for controlling the cost of obtaining coins for the entire network, but it has a shortcoming that must be solved. When the computing power of the entire network is insufficient in the early stage, it is also easy to generate a large number of nearly 0 cost tokens. Specifically, the tokens generated by a machine are equal to the proportion of the computing power of the machine in the entire network, multiplied by the number of coins issued by the PoW chain in a period of time. For example, when a chain produces 10,000 tokens a day, the computing power of a certain machine is 100 hashes per second, and the computing power of the entire network is only 1,000 hashes per second. The computing power of the machine accounts for 10% of the computing power of the entire network, and 1,000 tokens can be expected to be generated in a day. Therefore, when the computing power of the entire network is particularly low, a laptop may be able to mine a large number of tokens, and the cost of obtaining tokens at this time is very low. If the early low computing power stage is too long, it is easy for the entire network to generate a large number of nearly 0 cost tokens, which is extremely unfavorable to the development of the entire project.

In response to this problem, CZZ's solution is to set a minimum difficulty for the entire network. On a normal PoW network, there is generally a predetermined block interval, such as 10 minutes for Bitcoin. When the actual block speed is faster than 10 minutes, the mining difficulty will gradually increase. When the actual block speed is slower than 10 minutes, the mining difficulty will gradually decrease. This is done to ensure that the block interval does not change drastically due to computing power. CZZ sets a minimum difficulty on this basis. Before the computing power of the entire network cannot reach the minimum difficulty, the block time will be very long. For example, when the computing power of the entire network is 100 h/s, the minimum block difficulty is 1m, and it takes about 3 hours to produce a block.

Although the block speed will be very slow in the early stage, this sacrifice is acceptable considering that the transfer frequency will not be very high in the early stage. Its advantage is that even in the low computing power stage of the entire network, it can effectively ensure that the mining cost is above a certain level, fundamentally eliminating the possibility of all tokens with nearly 0 cost. If some people think that the value of CZZ tokens is higher than the mining cost paid, these people will most likely not sell their results obtained by proof of work at a low price. If another part of the people do not recognize the value of CZZ at this stage, the cost of obtaining CZZ coins will serve as a threshold to block them from the outside. Therefore, the minimum mining difficulty limit itself is also an automatic filter that puts early CZZ tokens in the hands of those who recognize it most.