Update EIP-7691: add update fraction specification

[adietrichs]

wip

todo:

• add rationale to EIP
• list and compare alternative values for update fraction here, to inform a decision
• double-check the update fraction calculations

[eth-bot]

✅ All reviewers have approved.

[nerolation]

Thanks for opening this one. To add some context, as agreed on ACD, we may want to increase the update base fee fraction in 7691 in order to ensure the base fee scales appropriately.

usually we would do something like "target_blob_gas / ln(1.125)" but since we're moving away from symmetric base fee scaling we might want to lower the update fraction relative to the target blob gas we currently use. I'm fine with the value as-is in this PR.

[adietrichs]

Added rationale to EIP:

The original target and max values from EIP-4844 were at a 1:2 ratio. As a consequence, responsiveness to full and empty blob sections was symmetrical:

• full blobs: basefee increases by ~12.5%
• no blobs: basefee decreases by ~11.1%

The new target and max values from this EIP are at a 2:3 ratio, which breaks that symmetry.As a consequence, the basefee becomes significantly more responsive to empty blob sections (that are 6 blobs under target) than to full ones (that are 3 blobs over target). This is by design, as it takes two blocks with full blobs in a row to make up for a single block with no blobs. However, it creates the challenge of finding a good compromise base fee sensitivity level.

The BLOB_BASE_FEE_UPDATE_FRACTION_ELECTRA value in this EIP is chosen as the mid-point between keeping the responsiveness to full blobs and no blobs constant:

• full blobs: basefee increases by ~8.2%
• no blobs: basefee decreases by ~14.5%

The alternative options we would have:

Keep update fraction at today's value:

• full blobs: basefee increases by ~12.5%
• no blobs: basefee decreases by ~21.0%

Keep update fraction proportional to target (i.e. double it):

• full blobs: basefee increases by ~6.1%
• no blobs: basefee decreases by ~11.1%

I'd argue the mid-point value chosen in the PR is the simplest available compromise, but happy to change it if people disagree.

DISCLAIMER: I haven't double checked any of these numbers yet, there could be a mistake somewhere.

[github-actions]

The commit f8da514 (as a parent of cc53043) contains errors.
Please inspect the Run Summary for details.

[g11tech]

Added rationale to EIP:

yes rationale looks good to me

[nerolation]

Add space between "symmetry" and "As a consequence"

[eth-bot]

All Reviewers Have Approved; Performing Automatic Merge...