Add EIP: Add time-weighted averaging to the base fee #7378
Conversation
github-actions
bot
added
c-new
|
✅ All reviewers have approved. |
eth-bot
changed the title
eth-bot
added
e-consensus
Spelling and git handle
Add number Co-authored-by: lightclient <[email protected]>
github-actions
bot
added
w-ci
|
The commit bcf0d2f (as a parent of 303b18e) contains errors. |
eth-bot
changed the title
Added spaces around section headers
EIPS/eip-7378.md
Outdated
| ## Abstract | ||
|
|
||
| The current formula for setting the base fee follows the original [EIP-1559](./eip-1559.md): | ||
| $$b[i+1]\triangleq b[i] \cdot \left( 1+\frac{1}{8} \cdot \frac{s[i]-s^* }{s^* }\right)$$ where $i$ enumerates the blocks, $s^*$ is a predetremined block size (the "desired" size, aka gas amount), $b[i]$ is the base fee at block $i$, and $s[i]$ is the size of the block at height $i$. This formula considers only the last block size $s[i]$. This mechanism might lead to incentives for users to bribe miners in order to reduce the base fee, or (in analogous manner) for miners to intiate a collusion with sophisticated users for the benefit of both. (See Motivation section.) |
There was a problem hiding this comment.
The base fee itself is an accumulator of the past since current base fee is calculated on parents, and with induction on all its parent. So saying it considers only the last block size might not be true
There was a problem hiding this comment.
Although the sentence "This formula considers only the last block size
There was a problem hiding this comment.
using base fee update clarifies this i think, so even though base fee already is an accumulator, base fee update which is the center of this proposal is just based off last block size.
Clarifying Miners==Proposers
| --- | ||
|
|
||
| ## Abstract | ||
|
|
There was a problem hiding this comment.
I think the abstract can be shorter something along the lines of (rest should be moved in specification)
To prevent proposers/miners and/or users from colluding and gaming the base fee in the scenario of consecutive proposals this EIP proposes a new formula to update the base fee by considering geometric weighted past block sizes instead of just the last block size.
There was a problem hiding this comment.
The motivation builds on the reader familiarity with the formula. I therefore, find it useful to have it right at the beginning. If you feel strongly on the matter, I can rearrange the presentation.
|
|
||
| Mainly, to reduce bribe motivation (in particular in case the demand for blockspace is high -- see Incentive Considerations section). Additionaly, to reduce oscillations and have a more stable fee setting mechanism. | ||
|
|
||
| Proposers use a mechanism described in EIP-1559 to determine which messages to include in a block. This mechanism includes a "base fee": a portion of the transaction fee that is burned. The base fee varies according to the fill rate of blocks. A target block size is defined. If a block exceeds the target size, the base fee increases, and if it is smaller, the base fee lowers. |
There was a problem hiding this comment.
can this be shortned assuming the audience is a bit aware of 1559
There was a problem hiding this comment.
It could certainly be omitted assuming the reader is familiar with the base fee mechanism. However, I believe that dedicating a few lines (approximately three) is reasonable to accommodate readers who may be less familiar with the topic.
Co-authored-by: g11tech <[email protected]>
Added paragraph about MEV
remark about alpha=1
remark re. 'max change denominator'
Co-authored-by: g11tech <[email protected]>
Co-authored-by: g11tech <[email protected]>
delete redundant part from incentive considerations
When opening a pull request to submit a new EIP, please use the suggested template: https://github.com/ethereum/EIPs/blob/master/eip-template.md
We have a GitHub bot that automatically merges some PRs. It will merge yours immediately if certain criteria are met: