Layer 2 is a popular direction in the market this year. Since the airdrop of arbitrum, its ecology has experienced a rapid explosion, and the TVL on the chain has reached a new high. However, after the popularity of arbitrum, many users have begun to focus on zksync era and The two layer 2 starknet chains are on the chain. However, after experiencing the interactive activities of Argent and Braravos, and after users conducted a large number of on-chain transactions and experiences, everyone found that the experience of starkent was extremely poor. In addition to the high gas fee on the chain, the transaction confirmation was also difficult. It is very slow, and because of this, many gamers have complained about Starknet.
We can clearly see the difference between Starknet and other layer2 chains through the block generation speed. We randomly compare the block speed on the chain. Optimism basically produces 2-3 blocks in 1 minute, and the transactions in each block are close to 100. tx.
Arbitrum produced 18 blocks within 5 seconds, with an average of about 3 transactions in each block.
Zksync produces about 45 blocks in one minute, and there are about 10-20 transactions in each block.
From the data point of view, zksync and arbitrum are currently relatively fast in terms of layer 2 confirmation speed. Optimismâs block production speed is slower than that of the Ethereum main network, and its tps is relatively low. Of course, this is also due to the current popularity of optimization, but it is basically the same. able to satisfy the use. For Starknet, the block data we see is relatively worse.
It can be seen that Starknet produces one block per minute on average, which is relatively slow, so the time for users to wait for transaction confirmation is at least about 1 minute. However, we also see that the number of tx accommodated in each block has increased relative to optimization, so the overall In terms of performance, the current overall performance of Starknet is about half worse than the performance of Optimism, and it is currently the peak period of Starknet network data, while the Optimism network data is declining. Through actual experience, the transaction confirmation time of Starknet is generally between 3-10 minutes. This is mainly because many transactions are in a sorting state, and there are too many transactions that have not been confirmed by the block, thus causing the slow confirmation speed on the chain. , and in fact the TPS on Starknet is around 0.07, which is much smaller than the Ethereum mainnet.
Of course, regarding this situation, in fact, Starknet officials have publicly stated that the main task in 2023 is to improve network performance. The reduction in block generation speed makes the sequencer the focus of improvement, so as to ensure that blocks can be processed under the existing block generation speed. It can be launched stably, thereby reducing everyone's experience time (more than 5 minutes) to the theoretical 1-2 minutes, improving the experience, and then considering further increasing the block generation speed, and subsequently reducing transaction fees. Another reason why Starknet currently has difficulty reaching the performance of chains such as Arbitrum is that it does not directly adopt the EVM virtual machine and solidity language. Instead, it uses cairo VM. The new VM also means building from scratch, so relatively speaking, development progress and The optimization is also slower than other layer 2, so that there is a big gap between starknet and arbitum, zksync, etc. According to the progress of the project team, it is expected that EIP-4844 will make significant changes to the status quo in the process of improving performance. Starknet officially plans to exit v0.12.0 in the second quarter of this year, including the transition to Rust-based Sequencer and the new Rust-Cairo VM, which is expected to reduce block execution time, thereby improving throughput and reducing latency, in v0.13.0 It is planned to reduce the overall transaction cost, and the block production speed will be further improved in version 0.15.0. By then, the overall performance of Starknet may be further improved, and the user experience will also be improved accordingly. It is expected to be officially launched at the end of this year or next year.
EIP-4844 has been the focus of discussions on the recent Ethereum upgrade. The main core is to select blob type transactions not to be executed by EVM, which can reduce the gas fee of blob transactions. Blob transactions are specifically designed for layer2 to transmit data. The data types involved, so in this way, layer2 transactions can be specially processed, thereby reducing the EVM resources occupied by the overall layer2 transaction. The blob has a 30-day life cycle and will be deleted after 30 days, so it will not affect the overall performance. Worry. Finally, we can see from the technical route that the overall performance of Starknet has not been fully exerted at present, and the team still needs to overcome many difficulties. Starknet is still in the early stage of the main network, and it will still take a certain amount of time before the ecological explosion, and if we rush here Coin airdrops may not have a positive impact, and users are generally more inclined to exaggerate the disadvantages of products. Therefore, for Starknet, coin airdrops may be relatively far away, and users who want to bet on airdrops still have opportunities.
