power and sample size question

Should sample size computations for power analysis be based on "one
sample, one test". That is, if I have two primary outcomes, and the
necessary sample size for the first is N=300, and the necessary sample
size for the second is 150, will I need to have a sample of 450, or
300, since such a sample would be adequately large for both tests?

Thanks,


John Zawisa

.