Darwinbots Forum

Code center => Suggestions => Topic started by: shvarz on June 21, 2006, 12:07:50 AM

Title: Dynamic costs
Post by: shvarz on June 21, 2006, 12:07:50 AM

Once again want to say that I like the feature a lot.  It needs just a bit more improvement though.
Some customization options I already mentioned.  Most important one is to allow either species-specific or hetero vs auto-troph-specific costs.

I also wander how the multiplier is calculated.  Seems like its change is proportional to the ratio between current population and desired one.  It would be nice to figure the rate of change of population into the equation somehow.  So that if the population grows quickly, then the costs should grow quickly and if it begins to stabilize, then the costs should stabilize or even go down a bit.

Title: Dynamic costs
Post by: EricL on June 21, 2006, 02:20:40 AM

Quote from: shvarz

Once again want to say that I like the feature a lot.  It needs just a bit more improvement though.
Some customization options I already mentioned.  Most important one is to allow either species-specific or hetero vs auto-troph-specific costs.

I also wander how the multiplier is calculated.  Seems like its change is proportional to the ratio between current population and desired one.  It would be nice to figure the rate of change of population into the equation somehow.  So that if the population grows quickly, then the costs should grow quickly and if it begins to stabilize, then the costs should stabilize or even go down a bit.

Thanks.  I plan to make the improvements you suggest for the next drop and indeed have already taken some steps in that direction.

BTW, the multiplier algorithm is described here (http://www.darwinbots.com/Forum/index.php?showtopic=1397).  It does take into consideration how far out of the target range the current population and takes some consideration of the direction of change, but not the actual rate.  Good idea though.  I'll play with it some.