Quick writeup to explain what's happening about the gas limit raise:
Different from almost all other protocol matters, the gas limit is directly controlled by the validators. With every block, a validator votes whether to raise or lower the gas limit. So, very unlike the EIP process, there is no "standard process" to raise the gas limit.
Now, the gas limit has *never* been increased since the merge. However, before the merge, it was an established practice that miners would not use this power arbitrarily, but typically only after core devs signalled it was ok to change it.
Recently, some of us have started signalling that we think it is time for an increase in the gas. Unfortunately this hit a bit of a speed bump as Core Devs discovered that raising it over 40m was actually not safe due to a constant that has to be changed in the clients first (ethresear.ch/t/on-increasin…).
However, I think now there is a lot of messaging agreeing that 36m is a good "first step" and clients are releasing updates to make this the default (so validators don't have to manually use command line flags to vote for this). I think it is likely that some big pools will join in the next few weeks and we will see this increase.
Different from almost all other protocol matters, the gas limit is directly controlled by the validators. With every block, a validator votes whether to raise or lower the gas limit. So, very unlike the EIP process, there is no "standard process" to raise the gas limit.
Now, the gas limit has *never* been increased since the merge. However, before the merge, it was an established practice that miners would not use this power arbitrarily, but typically only after core devs signalled it was ok to change it.
Recently, some of us have started signalling that we think it is time for an increase in the gas. Unfortunately this hit a bit of a speed bump as Core Devs discovered that raising it over 40m was actually not safe due to a constant that has to be changed in the clients first (ethresear.ch/t/on-increasin…).
However, I think now there is a lot of messaging agreeing that 36m is a good "first step" and clients are releasing updates to make this the default (so validators don't have to manually use command line flags to vote for this). I think it is likely that some big pools will join in the next few weeks and we will see this increase.
Now on the one hand, I think it is really cool that we have this way to raise the gas limit even outside of a hard fork because hard forks take a really long time to coordinate.
On the other hand, I am kind of sad by what has now become a very small "consensus" increase to only 36m and I don't think it's reasonable to ask validators to change this again in a few months when clients have fixed the bug.
Personally I think a good compromise would be to set the clients to vote for an exponential increase according to Moore's and Nielsen's laws. If that becomes too much, the vote can still be used to intervene and set a lower limit.
On the other hand, I am kind of sad by what has now become a very small "consensus" increase to only 36m and I don't think it's reasonable to ask validators to change this again in a few months when clients have fixed the bug.
Personally I think a good compromise would be to set the clients to vote for an exponential increase according to Moore's and Nielsen's laws. If that becomes too much, the vote can still be used to intervene and set a lower limit.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
