When we added the "Miso Buyer" module to the server's wholesale market, the intent was to make the simulation more realistic. In general, I think we've succeeded; now the buyers in our city are competing with buyers across a region more than 10X the size of our city. On the other hand, the intent was not to raise prices, and I think that's what happened, because the expansion of the price curve in the quantity dimension is interacting with a change contributed a couple years ago by Daniel Urieli at UT Austin. A detailed discussion of what we are seeing is given in the writeup for Issue #929.
I have implemented, tested, and committed a change to the genco module that makes the location of the "knee" in the curve a configurable parameter, as well as the slope to the right of the knee. I looked at some games from the first trial to get an idea of where the knee should be. Before this change, it was down around the 1st decile of overall demand; for now, I have moved it just above the 3rd quartile, which should mean that 75% of the time, wholesale prices will be lower and less volatile. When overall demand exceeds the threshold, prices will be nearly as volatile as they are now, but lower in general.
The question is whether we should use this version for the next trial. If you have objections to this change, please let us know in the next day or two. Otherwise we'll run the trial with 1.4.2-SNAPSHOT, and we'll do a 1.4.2 release before the qualifying round.
Thanks. As always, please let us know if you have questions or problems.