How does that work? Will blocks actually be released faster than the 40 second mark if they're solved? I didn't think that's how it worked.
Could you elaborate more on your second calculation?
40 seconds is the optimal expected time break the block.
If the hash power is large then this time is shortened. Then work the mechanism for calculating the difficulties and will increase the difficulty.
This time breaking blocks approaching the optimal (planned) in 40 seconds.
Difficulty Retarget: Every 60 blocks (40 minutes)
After every 60 blocks system calculate difficulty (based network hash power, current difficulty and planed block break time) and if hash power increases then difficulty up if is low power then difficulty down.
self-monitoring mechanism
I think the way it works, I'm not an expert.