5. CPU tests
In order to measure the performance of the Shuttle XS 3510MA we used the latest version of some popular benchmarking suites, the FutureMark PCMark, the CineBench, the wPrime and of course Sisoft Sandra.
The latest Sisoft Sandra provides useful comparisons with other already tested systems. The system's Intel Atom D510 performed better than the Intel Atom 330 with 8.24GIPS vs 7.86GIPS. As it was was expected, the Intel Atom D525 clocked at 1.86GHz is slightly faster than our system with 8.84GIPS:
At the PCMark benchmark our system scored 1953 marks, after running the following sub-level tests:
You can see an online comparison of our Shuttle system with other systems at Futuremark's website .
The CineBench R11.5 is an OpenGL graphics benchmark and a CPU benchmark based on the popular MAXON’s CINEMA 4D. The test procedure consists of two main components - the graphics card performance test and the CPU performance test. The CPU test uses all of the system's processing power to render a 3D scene. This scene makes use of various algorithms to stress all available processor cores. The result is displayed in points (pts). The higher the number, the faster the processor.
The Graphics Card Performance (OpenGL) test uses a complex 3D scene which measures the performance of the graphics card in OpenGL mode. The performance depends on various factors, such as the GPU processor on the hardware, but also on the drivers used. The result given is measured in frames per second (fps). The higher the number, the faster your graphics card.
At the left picture you see the CPU results of our Atom D5190 compared to other CPUs. The right graph compares the GPU performance of our system (FPS) with other graphics cards. It is obvious that our system performs significantly lower than the Core i7 CPUs of the chart, as well as the ATI Radeons and Geforce series graphics cards that appear in the comparison charts:
The final benchmark we used was the wPrime 2.03. The software has been designed to use a highly multi-threaded approach to calculating the square-roots of large amounts of numbers (up to 32 billion at this stage).
Our system preformed a 32M calculation in 65.0333 seconds: