There were great expectations for ETH 2.0, but widespread misconceptions. There are two corrections to the statement "ETH 2.0 gas fee will be lower". First, the 2.0 upgrade is a long process, and the "Merge" expected to be completed within this year will not lower gas fee, but the "Shard" will. Second, even if the "Shard" is completed, it will only ensure a reduction in gas fee for L2, while L1 will probably remain a "noble chain" forever.

Gas fee is essentially the price of energy, so it is determined by supply and demand. The supply here is the space available for computation and storage, called scalability, as measured by TPS and block size.

Of course you can, as Elon Musk wishes, "speed up block time 10X, increase block size 10X & drop fee 100X", which is easy to do so by just changing two parameters. But the end result is to transform ETH into another blockchain like Polygon or BSC. In fact, the block size and block time of blockchains have almost reached theoretical limits in order to accommodate decentralization.

Therefore, the only way left is to improve TPS, but which cannot be improved by the merging of ETH 1.0 and 2.0. This explains the first misconception, that once the "Merge" is complete, ETH gas fee won't change at all (block time is slightly reduced, but it doesn't make a big difference).

Sharding is the technology that can improve TPS, but which is a process, not a moment like the "Merge", so it's hard to predict the final TPS. Based on the current information, we can confirm that there will be 64 shard chains in phase 1, but this does not mean that the capacity of ETH will increase by 64 times, because the capacity of each shard chain is not equal to the current capacity of ETH. A reasonable estimate is that the capacity of each chain will be 1/3 to 1/2 as large as it is today, thus increasing the overall size by about 21–32 times. At current rate of progress, this is not expected to be completed until late 2023.

However, even if all goes according to schedule, it does not mean that L1's gas fee will come down. Going back to the previous point, price is determined by supply and demand, and we have just calculated the change in supply, but ignored the demand. Even conservative estimates suggest that trading volume on ETH will increase at least 5 times in the future. If the price of the ETH also increases 5 times, then gas fee in USD will not change much.

Some people might think that 64-shards is just the first phase, and eventually it will grow to 1024. This is only in the theoretical stage, but in reality there is a "minimum user count" problem involved here. Sharding will lead to a dispersion of nodes, so a single shard chain is not as secure as ETH is today. To ensure that a shard chain is not easily attacked and has enough redundancy (including data availability sampling), the minimum number of nodes should be at least hundreds. Hence, it seems difficult to justify a sharded blockchain having more than a few hundred shards.

With the gradual increase of demand, the gas fee of L1 will be higher and higher (maybe go down in the short term but go up in the long term), and the so-called gas fee reduction actually refers to L2. L2 creates an off-chain transaction execution environment independent of L1 and uploads the results to any shard chain after processing the transaction. Essentially, this reflects the "Impossible Trinity" of blockchain, where decentralisation is the reason why ETH can't endlessly improve scalability. Rollups, on the other hand, does not need to pay attention to these and only needs to focus on improving scalability, with L1 taking care of security and decentralization. Therefore, Rollups does not have a "minimum user count", resulting low maintenance, hence there is no upper limit on the number of Rollups.

The current gas fee on the OP Rollup is 1/8–1/3 of that on ETH, and which on the ZK Rollup is expected to be 1/100–1/40. Therefore, after the "Shard", the gas fee on the ZK Rollup will be further reduced to 1/7000–1/3000 (ZK Rollup will gradually become dominant due to lower gas). Here it is necessary to explain why sharding can significantly reduce L2's gas fee but not necessarily L1's, for two reasons. First, there is a practical limit to the number of shard chains, while the number of L2s can keep increasing, thus meeting an infinite expansion in demand. Second, ETH is a good store of value, the price of which is likely to rise as the crypto market expands, but L2's governance token is just fuel that no one will want to consume if it's too expensive.

In addition, the more transactions on L2, the higher the total cost of buying L1 block space, so it is possible that L2 gas fee gradually decreases while L1 gas fee gradually increases in the future. Perhaps by then all the transaction activity on L1 will be data uploads from L2, and no more individual users will be found.

Reference：

1.https://vitalik.ca/general/2021/05/23/scaling.html

2.

Follow us

Twitter: https://twitter.com/WuBlockchain

Telegram: https://t.me/wublockchainenglish