When blocks come in super close together a miner will accept the block **header** of the latest and start mining before actually downloading the entire block and verifying all the data.

However, by only using the header, they can’t put any transactions in the block, because they don’t know which ones to remove without checking the last one. Sometimes they find a valid hash in that small period of time where they don’t yet know what they can actually put into the block… and thus, it stays empty.

Reply to this note

Please Login to reply.

Discussion

Learning is cool 😎

I wonder if you could include a bunch of low fee txns & still be relatively certain the block would be accepted & grab a little more in fees 🤔

Maybe a single 1 sat/byte fee txn is actually a miners transaction that got included in the last block because they wanted to mine their own and didn’t care about the fee. If so, then you risk losing 6.25 BTC for a 112 sat fee.

Maybe it works 99.9% of the time. But you’d need to do it *millions* of times for it to make economic sense. Until fees overtake the reward, the empty blocks will continue 😆

You will notice that essentially every single time this happens, the blocks are within seconds of each other. The loss of those few seconds of mining cost way more than worrying about the fees. Block reward is worth a lot more. The empty blocks even existing is essentially proof that it’s not worth waiting for all the data to begin mining, because those 5 seconds of waiting could literally cost you 6.25 BTC

great breakdown

Interesting. I thought it might have to do with just pure speed, but the time between 780,997 and 780,997 was 26 seconds and the time between 781,001 and 781,002 was only 2 seconds more and both those blocks were full. Even so, I’ve seen blocks mined sub-10 seconds apart both with transactions. Bitcoin is weird as shit.

If the same miner or pool mined both blocks then they know which transactions to remove. Or it’s possible that in certain circumstances they can simply do the verification fast enough.

Does the blocks coming super close phenomenon only happen with a congested mempool?

Yeah Tatum… what he said.

Duhhh.