-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Account for L2 Gas in AA Benchmarks #6
Comments
thanks for pointing this out, and yes i 100% agree on this point and also added a calldata size calculation a while ago but didn't had time to look into how call data costs are actually calculated, do you have any good idea how to deal with it? my concern here was that L2 price usually depends on 1) calldata bytes(length and amount of 0s in bytes) 2) L1/L2 gas price 1 is calculated easily, as you pointed out we have calldata size any idea? |
Also, the fact that pvg calculation defers per bundler is one another issue 🤦 |
Maybe it could make sense to start by showing the calldata length and some indicator of 0 byte distribution for each smart account? Given that this repository aims to provide a comparison between different smart account implementations, I would assume L1 has fluctuations would affect each implementation proportionally to the calldata cost (post compression). For the sake of comparison we could maybe also assume that all ops are submitted by the same bundler. |
good. let's start with that I'll make a pr later this week :) or do you want to make a pr? |
Sure! Can take a stab at it. Please assign the issue to me |
done |
I suspect this will pretty much just come down what all gets packed in the signature field? |
Yup |
Also apologies for the rasio silence here, I should have something here this week |
Given that a lot of AA activity happens on rollups such as Optimism and Arbtirum, the compute gas cost might not be indicative of the actual fee paid for transaction execution since L1 gas is responsible for the majority of costs of transaction execution on these chains.
I propose that the benchmarks be updated to reflect the call data / L1 costs as well.
The text was updated successfully, but these errors were encountered: