This paper describes current work in assessing the microburst recognition performance of the Terminal Doppler Weather Radar (TDWR) operational testbed. The paper is divided into three main sections: microburst recognition algorithm, performance assessment methodology and results. The first section provides an overview of the prototype TDWR microburst recognition algorithm The algorithm uses radar data from both surface scans and scans aloft to identify microburst events. The surface scan is used to identify microburst outflows, and the scans aloft provide information concerning reflectivity and velocity structures associated with microbursts to improve recognition rate and timeliness. The second section of the paper describes the methodology for assessing the recognition performance of the system. The performance of the testbed system is addressed from two viewpoints: radar detectability and pattern recognition capability. The issue of radar detectability is examined by comparing radar and mesonet data to determine if any events observed by the mesonet fail to be observed by the radar. The issue of pattern recognition performance is assessed by comparing microburst recognition algorithm outputs with truth as determined by expert radar meteorologists. The final section of the paper provides performance results for data collected by the testbed radar at Huntsville, AL and Denver, CO.