> Uncorrelated error [...] accumulates as RSS Uncorrelated average error, yes. Uncorrelated maximum error accumulates as a linear sum, as he calculated. But y

> Uncorrelated error [...] accumulates as RSS Uncorrelated average error, yes. Uncorrelated maximum error accumulates as a linear sum, as he calculated. But yes, you're absolutely correct in that it tends to be nonlinear effects that cause issues.