fix: ganache simulation fails with out of gas about convex-shutdown-simulation HOT 4 CLOSED

1 point here that likely make up the difference between foundry and tenderly: we remove the base tx cost + calldata cost, i cant imagine tenderly does. This likely accounts for roughly 21064 of our difference wrt tenderly

from convex-shutdown-simulation.

description updated with gas usage

from convex-shutdown-simulation.

Hey @brockelmore, curious why you remove base + calldata cost, what do you achieve with that?

from convex-shutdown-simulation.

we remove the base cost because it's just noise especially when trying to gas optimize, for example, what may normally be an internal function, but during gas testing they are testing it explicitly. We remove calldata for a similar reason, but slightly different. since we test with smart contracts, we always have an extra 64 gas that doesn't change regardless of what you do internally. so the long and short of it is, we want to report execution cost of tests as pure as possible and let people not have to do redundant math in their head

from convex-shutdown-simulation.