4 ms·
Starting with the GBA games, Fire Emblem's hit and critical chances take the average of two random numbers. So what shows as a 90% hit rate is actually 98% (tha
by jtolmar 8y ago
Starting with the GBA games, Fire Emblem's hit and critical chances take the average of two random numbers. So what shows as a 90% hit rate is actually 98% (that is, (2d100)/2 <= 90).
I think this is a pretty neat trick; it's more in line with what people feel like probabilities should be than the reality of those probabilities. It also subtly biases in the player's favor, since in Fire Emblem the heroes usually have high accuracy while enemies more often go for high damage and low accuracy.
- IggleSniggle 8y agoThat is brilliant, but I have mixed feelings about misinforming the player this way. I like the way Brogue trains you on odds. You roll the dice A LOT, but it goes by super fast, so you get averages in one battle that conform to player expectations without misleading the player. At the same time, you can still achieve high stakes moments...but generally the player has walked themselves into those high stakes moments by repeatedly playing poor odds, so it feels deserved rather than like a rage quit situation.
- Bartweiss 8y agoI've seen the claim that the Japanese versions of Fire Emblem don't do this, though I'm not sure if it's true. It wouldn't surprise me if people's past gaming experiences did a lot to vary their expectations for random number generation. I know XCOM was sometimes accused of fudging numbers against the player, though experimentation shows the percentage chances are correct (excluding bugs). I wonder if that's tied to expectations set by games like Fire Emblem.