Uncle blocks in ethereum have dropped considerably since January 2018, down from just above 2,000 blocks a day to now only 250.
Improvements in the underlying protocol to increase propagation efficiency has been one contributing factor for this fall.
The drop in transaction numbers may be another. At the all-time high for uncles, ethereum was handling about 1.4 million transactions a day. Now it’s down to 400,000 transaction.
That’s a 3x drop, while uncle rates have fallen by close to 10x. Meaning that miners could now increase the gas limit to allow for greater capacity.
State Rent and Ethereum 1x
Despite the fall in uncle rates due to less blocks found at or near the same time, which would require for one of the blocks to sort of be discarded, miners have not increased the daily gas limit that puts a cap on transaction capacity.
That might be because an increase of the limit would increase the resources to run a node, including storage requirements as well as bandwidth because each block would contain more data.
Alexey Akhunov, an eth dev, is thus working on storage rent. That’s a simple proposal in one way in that you just delete non active contracts with a method to restore them. On the other hand, the proposal can be quite complex:
No, but I just realised that State fees proposal v2, if implemented up to change G, will probably allow restoring of the parity multisig (because creator of that library can reset nonce on his EOA that created): https://t.co/VqvlNDWfhJ
— Alexey Akhunov (@realLedgerwatch) February 9, 2019
Akunov has begun publishing a series of very technical articles on rent storage, with the general idea being an upgrade of the Proof of Work (PoW) ethereum chain to add more capacity while everyone waits for ethereum 2.0.
Initially they suggested such upgrade might go through as soon as this June, but then they said it would not be one big upgrade. There are many phases and June is just an estimate.
To make matters more complex, ethereum developers spent pretty much an entire dev call on a proposed PoW change to get rid of asics.
They concluded they were going to wait for an audit which might come back by the end of next month or in April. Then, if they do decide to go ahead, there might be a ProgPoW fork in June.
In addition, they have a fork planned for autumn called Metropolis Istanbul. Where ethereum 1x fits in this schedule is not very clear.
Prioritization and Coordination Still Lacking
The eth 2.0 schedule in many ways is irrelevant as far as eth 1x is concerned. The Beacon chain, for example, which has a testnet launch planned for next month, is basically a new blockchain that connects to PoW through a proof of burn mechanism initially.
Eth 2.0, however, won’t be ready in a usable form for ordinary ethereans until 2021 at best, with just staking on offer until then.
In the meantime, an ethereum 2.0 like blockchain might launch by the end of the year, with Gavin Wood of Polkadot stating this chain would effectively have all the main capabilities at launch.
Thus ethereum will have to provide something to keep dapp devs and users satisfied, but there doesn’t seem to be much coordination or prioritization between the Metropolis teams and eth 1x.
The teams do overlap, but eth 1x is a very new proposal. Making it unclear at this stage whether they have a revised plan on how all of it will fit within a schedule and within an estimated timeline.
The impression so far has been more of a wing it as you go, with communication and coordination between teams as well as prioritization still seemingly remaining a problem where the PoW teams are concerned.
There are indications that this might be beginning to change, but a revised roadmap that incorporates eth 1x is still lacking. Making it unclear when we might expect greater capacity in ethereum despite this significant fall in uncles.
Improvements in the underlying protocol to increase propagation […]