Experimental Tests
Euler 3D Fluid DynamicsThis fluid dynamics simulation is very CPU and memory intensive. From the benchmark source website:
The benchmark testcase is the AGARD 445.6 aeroelastic test wing. The wing uses a NACA 65A004 airfoil section and has a panel aspect ratio of 1.65, a taper ratio of 0.66, and a 45 degree quarter-chord sweep angle. This AGARD wing was tested at the NASA Langley Research Center in the 16-foot Transonic Dynamics Tunnel and is a standard aeroelastic test case used for validation of unsteady, compressible CFD codes. Figure 1 shows the CFD predicted Mach contours for a freestream Mach number of 0.960.
The benchmark CFD grid contains 1.23 million tetrahedral elements
and 223 thousand nodes. The benchmark executable advances the Mach 0.50
AGARD flow solution. Our benchmark score is reported as a CFD cycle frequency in
Hertz.
Just like we saw in our 7-Zip testing, only the Core i7 processors saw the move to 8 threads of computing as a good step to take for obvious reasons. The gap between the i7-975 and the i7-870 is pretty large here though: 13% due to the memory needs of this particular test. When looking at 4 thread performance though the Nehalem/Lynnfield bunch jump way ahead.
Microsoft Image Composite Engine
This application takes any number of images and attempts to “stitch” them together to make a single panoramic image.
These are VERY odd results for sure – the Lynnfield testing seems to indicate that something in the P55 chipset might be the difference maker here. It is likely then that the driver for the P55 controller on Windows 7 x64 RTM is more robust than that for the X58 chipset at this time and the Intel X25-M SSD was just more fully utilized there.
Hyper PI (Super PI)
From Wikipedia: Super PI is a computer program that calculates pi to a specified number of digits after the decimal point – up to a maximum of 32 million. It uses Gauss-Legendre algorithm and is a Windows port of the program used by Yasumasa Kanada in 1995 to compute Pi to 232 digits.
Our Hyper PI / Super PI results scale fairly nicely but there is at least one interesting data point to showcase here that highlights the differences between HyperThreading and 2 threads on two different cores in terms of performance. Look at the i7-870 and i5-750 results for 2 threads: by default the Hyper PI applications puts the two threads on the first two virtual cores which are on the i7-870’s case, actually two HyperThreaded cores on one physical core. Then look at the result of the i5-750 where HyperThreading is disabled – i5-750 is about 23% faster on the two thread result.
