Only 440,000 transactions were made in the past 24 hours, down from the usual 600,000 to 700,000 before as seen above.
Gas usage has fallen too, down to 33 billion at some point, now at 40 billion from the usual circa 50 billion.
Some of this fall in activity is probably because much is paused a few hours before and after the upgrade, but things should be normal by now.
They largely are. Uncle rates are falling in line with block time increases and block rewards as well as other stats are where you’d expect them.
When it comes to transactions, however, there’s been a slight change which could explain the decrease:
Ethereum Improvement Proposal (EIP) number 1884 reprices certain opcodes “to obtain a good balance between gas expenditure and resource consumption.”
The stated reason is that “an imbalance between the price of an operation and the resource consumption (CPU time, memory etc) has several drawbacks.”
The effects are: “Certain calls will become more expensive… Contracts that assume a certain fixed gas cost for calls (or internal sections) may cease to function.”
So they’ve broken some smart contracts, 680 of them just on the Aragon platform. In addition as some operations are now more expensive, it might no longer be cost effective to perform them. So explaining this drop in transactions.
There was considerable discussion around this EIP before it was decided to include it, but it’s not clear whether its effects were fully communicated as you’d expect Gemini to have been prepared otherwise and not run out of gas.
This break-ability of functions, and at times critical functions due to customized software or for whatever reason, is one justification for the usage of backwards compatible soft-forks in bitcoin.
They take the view protocol devs shouldn’t dictate to node operators what code they have to keep up with to be part of the network on a principled basis in regards to decentralization and also on a practical level in regards to giving them sufficient time to upgrade on their own terms.
In ethereum however it’s not clear whether softforks can be done. There were suggestions of potential attack vectors if eth does softforks in 2016, but how much of that was maybe a bit political and how much was technical it’s not too clear, especially as where it concerns maintenance upgrades, like the one on Sunday, you’d think there isn’t much reason to not do it with a softfork if it can be done.