80% power (1-β=0.8) is the standard target — meaning 20% chance of missing a real effect.
For d=0.5 effect, α=0.05, 80% power: need n=64 per group in a two-sample t-test.
For a medium effect (d=0.5), α=0.05, two-tailed t-test: ~64 per group (128 total).