Unrealistic results are generated from miner changes

[GriffGreen]

PhilH, [16.05.21 10:37]
It doesn’t help that some extremely light variations on one parameter can have devastating impacts on the final score. I bumped into a configuration where 1% variation in the contributions to the funding pool results into +250 points difference in the final outcome (flipping it from “you saved the world” to “all hope is gone”). Other neighboring configurations do not have at all the same impact. There’s no way to understand why, but also, one can’t help suspecting that such swerve is an artifact due to the model or to the simulation engine, rather than a realistic forecast of what would happened in the real world.