Author: Mark Friedenbach 2017-10-20 18:55:55
Published on: 2017-10-20T18:55:55+00:00
The current block time of ten minutes for Bitcoin was chosen by Satoshi as a tradeoff between confirmation time and the amount of work wasted due to chain splits. However, with advances in technology and lack of a rigorous formula to determine the optimal rate, there may be room for optimization in this number. While 10-second block times would result in a lot of chain splits due to Bitcoins 12-13 second propagation time (to 95% of nodes), it is possible to decrease the block time without much issue. However, reducing the block interval comes with a large number of centralizing downsides that are documented elsewhere and getting down to 1 second or less on a global network is simply impossible due to the speed of light. Blockchains with fast confirmation times are believed to suffer from reduced security due to high stale rates. As blocks take time to propagate through the network, if miner A mines a block and miner B happens to mine another block before miner A's block propagates to B, miner B's block will end up wasted and not contribute to network security. Moreover, there is the centralization issue where a mining pool with more hash power has a higher risk of producing a stale block. With these two effects combined, blockchains which produce blocks quickly are very likely to lead to one mining pool having a large enough percentage of the network hash power to have de facto control over the mining process.Furthermore, if the average block time is reduced, the block size should be reduced accordingly. In a hypothetical 5-minute block size Bitcoin blockchain, there would be twice the block space available for miners to include transactions. This could lead to the blockchain growing up to twice the rate, which is known to be bad for decentralization, and transaction fees might go down, making it cheaper for spammers to bloat the UTXO sets. There have been numerous proposals that tried to overcome the downsides of faster blocks, the most noteworthy probably being the "Greedy Heaviest Observed Subtree" (GHOST) protocol. If you want point of sale support, it is suggested to look into the excellent work the Lightning teams have done.
Updated on: 2023-06-12T21:45:15.044718+00:00