These data sets were all recorded on the author's workstation, a 2GHz Core2 Duo E2180, with 1MB cache/processor and 2GB RAM, running Fedora 9 Linux. In all cases 1000 loops through the Dhrystone benchmark was used. The figures presented are the average of at least 6 runs.
Total processor time for elaboration was 1.78 s and for simulation was 796.84 s. Net simulation performance was 1.48kHz.
Note | |
---|---|
Throughout this application note, results will be given showing:
Cycle accurate models are typically used in scenarios, where the model is created once and used many times, so the run-time performance is the critical figure. |