Partial and semipartial correlation

Hello everybody,
I am confused when it comes to semipartial and partial correlation.
Partial correlation helps us to discover Which PROPORTION of the variance of Y not explained by the other IVs, is uniquely explained by a specific IV. However when I read the definition of semipartial, I feel like if they serve to discover the same thing... The result of partial correlation indicates the contribution of a IV on y... so why do we need then to compute a semipartial correlation?

Thank you very much for your time.
Best Regards,