RPT received criticism in its early launch because it could only detect performance bottlenecks in areas that were specifically tested. If a bottleneck existed in an untested part of the system, RPT would not find it - this was a significant limitation for comprehensive performance assessment.