Eduard S.

> @ed255 I created this PR even if this conflicts with #730 just for tracking purposes. If the circuit optimizations gets merged, then I can reflect these changes for the...

> I'm planning to add an `initial_value` advice column (which will be the commited value at the beginning of the block for `Rw::Account` rows and the commited value at the...

Currently the state circuit has an `initial_value` column that I think fulfills the issue: https://github.com/privacy-scaling-explorations/zkevm-circuits/blob/878e5e61a9d8a52f7b10571c6f89ef9933a71bb5/zkevm-circuits/src/state_circuit.rs#L52-L55 So I believe this issue is complete. @z2trillion could you confirm this?

Currently we indicate the license of our crates in the `Cargo.toml` file. See https://github.com/privacy-scaling-explorations/zkevm-circuits/blob/7cd18d19453447bee58c8b774a34c62f579e3651/zkevm-circuits/Cargo.toml#L6 Our chosen license is "MIT OR Apache-2.0". I agree that we can add the license documents...

I believe we have 2 cases of randomness in tests: 1. The randomness value for RLC. I think we could have a fixed constant for this used across all tests....

> We have now the CHallenge-API. Are we still considering this an issue? Or should I close it?? cc: @ed255 @ChihChengLiang It makes sense to me to close this issue...

> I agreee @ed255 but we're still using randomness in a lot of tests. I'd mark this as a `good_first issue` and leave it without a milestone for now. WDYT?...

Thanks for the analysis! The EVM circuit degree was recently increased in this PR https://github.com/privacy-scaling-explorations/zkevm-circuits/pull/718 where I explain a bit the rationale behind going to a non-optimal gate degree: the...

> Couldn’t instead q_step be constrained by q_usable? > > ``` > if q_usable == 0 then q_step = 0 > ``` > > Then q_usable won’t need to multiply...

@icemelon could you do a second review to this PR? It's a small change that gives some performance benefits (and I think it's useful for the time until we bring...