gRR comments on The limited predictor problem