It could be very extreme in case (2) if for some reason you think that the worse suffering is a million times worse than the best happiness (maybe you are imagining severe torture) but I agree that this seems implausibly extreme. Re how to weigh the different possibilities, it depends whether you: 1) scale it as +1 vs 1M, 2) scale it as +1 vs 1/β1M, or 3) give both models equal vote in a moral parliament.
It could be very extreme in case (2) if for some reason you think that the worse suffering is a million times worse than the best happiness (maybe you are imagining severe torture) but I agree that this seems implausibly extreme. Re how to weigh the different possibilities, it depends whether you: 1) scale it as +1 vs 1M, 2) scale it as +1 vs 1/β1M, or 3) give both models equal vote in a moral parliament.