Board Thread:New on My Singing Monsters Wiki/@comment-7326455-20150810064315/@comment-190001-20150901032809

Thanks. You probably did try using floor, but you would have had a hard time relating shards and diamonds /directly/ since both are truncations involving the food amount. I know because I was trying to do the same thing, at first. :-)

As for the tables, I think there are arguments to be made both for and against all of them, so the real questions are about how strong those arguments are and what the right balance to strike is. When I wonder about their value, I think of a few use cases:


 * Validate the formulae by comparing generated data points with in-game data.  I just did this to level 101, but I'd be happy for anyone who goes higher to confirm those levels.  The feed cost by level table enables this.  This isn't an everyday scenario, though, so it might make most sense to move this table to a subpage.


 * Determine what might be a resonable level to expect of a non-shard (i.e. newer) player.  This is enabled by the cumulative non-shard cost to level table.  This data is currently in the same table as the cumulative shards, but I think it's much easier to say that this part of the data need not go past 50.  If we already had some sort of a "Full Tables" subpage, that could include more of this, if we wanted.


 * Determine how high of a level you're willing to feed to.  Determine how much it cost a tribe-mate to get to the level they did.  Determine how much extra it would cost you to close the Y-level gap to 1500 by yourself.  The cumulative shards to level table enables these scenarios.  Given that most tribes will include some non-endgame players who have trouble getitng to 50, the rest of the players in the tribe will have to pick up the slack by going past 50.  For this reason, I think this data should be available well past 50 and probably even past 100.  I'm considering that maybe we should streamline the in-page cumulative table by removing the single-level deltas (i.e. four feedings) and show only cumulative costs to levels evenly divisible by five.  Again, we could have /all/ of the data shown in a "Full Tables" subpage.

Thoughts?