Ideally, I'd lie to be able to optimize to find the most delta-v for a given maximum mass / TWR / burn time / part count / cost. Currently the only way to do so is to optimize for mass and iterate minimum delta-v until you get something close to the maximum mass.
Ideally, I'd lie to be able to optimize to find the most delta-v for a given maximum mass / TWR / burn time / part count / cost. Currently the only way to do so is to optimize for mass and iterate minimum delta-v until you get something close to the maximum mass.