Roman Khimov

> but this proposal makes them stateful, so recheck is needed during block proposal/acceptance We just need to check for conflicts (kinda similar to UTXO double spend check) and it's...

> One map store active contracts and another for destroyed contracts? Well, there could be some deactivation flag stored, although we already block destroyed contracts now. Not sure about that,...

> I think that we can remove Pong and send ping only if we spend X seconds in send a block. Well, we need to remember *why* these ping-pongs were...

Certainly not, from what I see in the code we're still misusing pings and sending one ping per persisted block.

It's not just that, it's also important to stop misusing pings for things that are to be done with invs.

I'm not an NFT or E-letter chain expert, but I've tried to take a look at this out of curiosity. The batching part of EIP-1155 really is irrelevant for Neo,...

Neo's dBFT is as eco-friendly as it could possibly be, so maybe it's time to close this issue.

Let me show you two pictures: The second one performs `HasTransaction` check somewhat like five times faster, so it may seem that it should work much better. But real TPS...

Speaking of `inmemory` or MemoryStore, btw. Now that we can test it, there are some other pictures to share:    Aside from MemoryStore scalability issues (it obviously wasn't...

> @roman-khimov Really love your benchmark result. May I know the name of the tool that you use to conduct it? ...........:arrow_down: :arrow_right: [neo-bench](https://github.com/nspcc-dev/neo-bench/) :arrow_left: ...........:arrow_up: (`plot.py` might need to...