pull down to refresh

WEEK 2 MINING POOL RESULTS:

Most profitable:

+2.98% | MARA Pool (4.55% hashrate)
-0.61% | Solo CK (0.1% hashrate)
-2.40% | Foundry USA (26% hashrate)

Least profitable:

-24.76% | SBI Crypto (0.39% hashrate)
-20.17% | OCEAN (3.29% hashrate)
-8.95% | Spiderpool (9.01% hashrate)

Leading censors:

79.02% | SBI Crypto (0.39% hashrate)
81% | Poolin (0.1% hashrate)
83.76% | OCEAN (3.29% hashrate)

What is a block's "Health" score calculated from?

Every time a block is mined, mempool.space takes a snapshot of the most profitable block template that could have been constructed by the miner. Whenever a pool excludes a high fee paying transaction, their "health" metric is damaged. A pool with a low health score means they are either censoring, or their node is not well connected to the rest of the network.

How do you deal with template distribution timeliness in your "censorship" score?

reply
150 sats \ 3 replies \ @Kruw OP 3 Aug

I'm not quite sure what you mean. Are you saying that mempool's health scores may be inaccurate because the pool's node may not receive the same bids in the same order due to propagation latency?

reply

The pool's node doesn't do the hashes. Stratum sends a template every x seconds which may or may not have been switched to by the miner consuming the template. The only time a miner is guaranteed to have done that hard work reset is when the previous blockhash changed (otherwise you'd be mining an orphan.)

There's a field in the stratum message where the pool requests reset but even if that is not followed by the miner and the resulting block found is valid, you'd still publish it or you'd be throwing away moneys.

So I think that censorship can only be determined by transactions that were known at the time of the previous block, must have been in the new template (assuming the pool is well-connected) and still not included in the published block found.

reply
150 sats \ 1 reply \ @Kruw OP 3 Aug
Stratum sends a template every x seconds which may or may not have been switched to by the miner consuming the template. The only time a miner is guaranteed to have done that hard work reset is when the previous blockhash changed (otherwise you'd be mining an orphan.)

This is still somewhat under the pool's control since they can penalize miners who submit shares of stale templates.

So I think that censorship can only be determined by transactions that were known at the time of the previous block, must have been in the new template (assuming the pool is well-connected) and still not included in the published block found.

Mempool offers a short grace period so that recently broadcasted transactions will not penalize block health.

reply

Good you say that word. Mempool tracks the stratums too (or at least used to), so they could actually do something there!

reply