Stanislav Frolov

I'm not sure if my issue is related because I didn't dig into it too much but my observation is that text rendering is too expensive in ebiten. 1. Running...

There're two interconnected values in the vote transaction: target_hash and target_epoch. Both are set up in this way: ```C++ m_recommended_target_hash = block_index.GetBlockHash(); m_recommended_target_epoch = GetEpoch(block_index); ``` and then these values...

Ok I realized why do we have target_hash, and the intention of why vote recorder is implemented in this way: we slash double votes on different branches and ignore them...

There's also a lot of false ERROR messages happens during commits-full-sync. In the commits handler, we [record the vote](https://github.com/dtr-org/unit-e/blob/0a1d375198fdc2ad8be1fc461b2a52b9cc84df44/src/p2p/finalizer_commits_handler_impl.cpp#L355) which makes a check against [the current activeChain.Tip](https://github.com/dtr-org/unit-e/blob/0a1d375198fdc2ad8be1fc461b2a52b9cc84df44/src/finalization/vote_recorder.cpp#L22). In my opinion,...

> I think we should save votes which are 1 epoch in the future to make sure we don't drop them because of this "race condition" that the first block...

We also started to store commits in the CBlockIndex, and it increases its size significantly.

Do we want to add `-Wall -pedantic` as well? Personally, I like these flags but want to know what others think about it.

> Would be good if we could compile different targets with the different flags (maybe we would need to introduce more targets). Then we can be stricter in some "sub-folders"....