Mining nodes with full blockchain to end the centralization of ASIC mining

Important changes in the Bitcoin protocol implemented will end with some problems, such as the centralization of mining.

On January 3, 10 years of the generation of the Bitcoin genesis block by Satoshi Nakamoto are commemorated. We have had a particularly tough year, but 2019 looks promising with Nakamoto's return to the scene. Everyone by now knows about the 'nour' message, but I put a link so that you can read it. His return is not accidental and is based on the need to make important adjustments. Perhaps the most important is ending centralized Bitcoin mining.

Satoshi Nakamoto, father of Bitcoin, is Nick Szabo

The Roman betrayal of Bitmain

Currently Bitmain is the private company that generates the most ASICs in the world and controls large mining farms. If we look at the data and what we know, AntPool is a farm mostly controlled by Bitmain. On the other hand we have BTC.com, controlled by Bitmain since Jihan Wu and Roger Ver collaborate. We must remember that Wu and Ver teamed up to create Bitcoin Cash.

With just two Pools we have 30% of the power of the Bitcoin network. ViaBTC and SlushPool could join the conspiracy and 51% of the power of the network would be surpassed. This allows the generation of blocks controlled by them and therefore double spending or blocking transactions would be possible.

Another possibility could be a malicious attack on some of your competitors using DDoS. This would cause a decrease in the power of the network. Open secret is that Bitmain before selling the new ASICs puts them to mine. If we have some actors disappear, the outside of AntPool and BTC.com are combined and new malicious miners are injected ... 51% attack served.

Yes, they are complicated scenarios and involve a high cost, but it is the problem of the centralization of mining. A few actors can take control of the network. How do we solve it?

Full nodes to mine

One of the first options currently being considered is that to be a miner you have to be a full node. This means that you must have a Bitcoin client running on your system. The most important thing is that you will have a complete copy of the blockchain.

Does this eliminate ASICs? Completely. An ASIC has many chips for computing, but the amount of memory they have is minimal. It is interesting that an ASIC has a lot of computing power and its capacity is limited for the operating system and the mining software. The blockchain is stored by the Pool servers, to which the ASICs inject power. Basically it would be a central unit with data and many computers doing calculations for this server.

Currently the Bitcoin blockchain has not yet reached 200MB, but an ASIC does not have the capacity to store it. The other problem is that there can be very large changes in the size of the blocks. 1MB in 2018 (almost 2019) is a joke although by 1998 it was a huge amount. Sure, Bitcoin is the evolution of Nick Szabo's Bit Gold idea and back then 1MB was a world. Nowadays any reggaeton song already occupies 5MB in the best of cases.

Obviously it is not a complete and durable solution, but it is necessary. First to eliminate the centralization of mining. Second, because the size of the blocks will increase due to, among others, the new encryption and other improvements that seek to increase the number of transactions per second (not validations per second)

Incentivize full nodes

Ethereum Casper is a part of the transition phases from Proof-of-Work to Proof-of-Stake. Right now these two are operating on the network simultaneously. In Bitcoin, to favor the creation of nodes, something similar but stable over time would be sought. It seeks to continue with PoW but adding PoS in order to incentivize the creation of nodes and improve network saturation problems.

Work is being done on increasing the number of transactions per second that the network supports (not validations). This would mean going from 3.000-4.000tx / s to close to 14.000tx / s, a brutal jump. Logically tripling transactions is worthless if 7tx / s continue to be validated. That is why it is sought to add complete nodes and encourage this function.

This part is still in the development phase and we are looking at how to solve it correctly so that there are no problems. It is speculated that it could escalate to between 50-100 validations per second. I would like to be clearer, but it is still in the code optimization and testing phase.

Will the users go back to mining with GPU?

It is clearly a brutal adjustment of the Bitcoin infrastructure that may sound crazy, but it would be key. Users could go back to mining BTC, since we must remember that it was conceived as an extra income (in part). Going back to graphics cards will allow all users with a home computer to get bitcoins, something unthinkable.

This transition can be very beast and possibly force a hard reset of the difficulty. Currently, if all the ASICs were turned off, the network could stall and the injection of graphics to the network does not correct the problem. Drastic measures should be applied to lower the difficulty, not to 1 as during the early days of Bitcoin. It should go down to something GPU manageable.

Here comes an additional positive factor. NVIDIA does not have the capacity to supply 20 million NVIDIA GeForce RTXs for mining as it relies on TSMC. Maybe opting for the GeForce GTX and Radeon RX would be a solution, but we would have the mess with gamers again. Be that as it may, it seems that the graphs will once again be protagonists in Bitcoin.

Conclusion

If everything evolves as up to now, we will see how Bitcoin undergoes important changes at the hands of Szabo. Very important and necessary changes, since the value of Bitcoin (not the price) is stagnant. It takes a thorough review of the protocol and return to the original idea of ​​Szabo (and not these copies of medium hair that have come out like mushrooms). Do not miss the next installments where I will talk about block sizes, quantum processors or SHA-256.