3DMark2001 was run at its default settings of 1024×768, 32-bit color and no AA.
3DMark05 was also run at its default settings of 1024×768, 32-bit color and no AA or AF.
AquaMark3 was run with default settings and the CPU result was recorded. The GFX and Total scores were ignored for relevance purposes.
CineBench renders graphical images both for hardware and software rendering. This benchmark also measures CPU time in seconds used toward these operations. With the Intel Pentium 4 HT technology present on our system, CineBench also measures CPU time for multiple CPU systems. The Single CPU and Multi CPU results were recorded.
These synthetic benchmarks are just another step closer to real-world performance. The memory bandwidth results on the previous page largely showed theoretical results. While these synthetic benchmarks also show theoretical performance, we are not just stressing the memory, but also the CPU and graphical data transfer capabilities of the entire system. The use of an X600 graphics card prevented our graphical numbers from being the highest possible, but we still see a decent discrepency on the 3DMarks and AquaMark results to show us the ASUS P5GD2 may have a slight gaming advantage over the Intel reference board.
Now CineBench is a little different than the other benchmarks in this category. While it’s synthetic, it also provides much more stress on the CPU than 3DMark or AquaMark, which tend to add a lot of stress to the GPU as well as the CPU. The results of CineBench could be a preview of what other CPU benchmark results will yield in our Professional section. This benchmark shows the “light weight” features of the Intel reference board, giving faster CPU times than the “fully loaded” P5GD2 Premium.