What is block density? Pools currently before the Vasil upgrade are saying switch from 1.35.x back to 1.34.x to increase “block density”.
2
-
Bonus question, any idea why the current 1.35.x pre Vasil in testnet node seems to be dropping blocks, and reducing chain density?– TheStopheCommented Jul 28, 2022 at 14:44
-
Looks like it’s an issue checking minFee between different era versions: github.com/input-output-hk/cardano-node/issues/…– TheStopheCommented Jul 28, 2022 at 16:20
Add a comment
|
2 Answers
Each epoch is supposed to have about 21600 blocks per epoch, where an epoch is 5 days long. If every epoch had 21600 blocks, the density would be 100%. However blocks occasionally get skipped so the density will be less than 100%.
This is not the exact way that "chain density" is calculated (I forget the exact way it is calculated) but it should be enough to understand the concept.
answered Jul 28, 2022 at 2:30
Just building on top of answer from Erik, a simple way to look at it would be : (number of blocks minted so far in current epoch) / (number of slots so far in current epoch)
In ideal scenario, the density should match the active slot co-efficient (0.05 on testnet and mainnet, i.e. on average - 5% of chances for a random slot to have a block minted).
Is there a risk to protocol if density is too low? While it's far from ideal / healthy, current consensus network should work fine as long as there is 1 block in stability window , viz : 3k/f (i.e., 32160/0.05, which corresponds to ~ 1.5 days on testnet/mainnet).
