Intel Core i9-10900K in the versatility test. What are the strengths and weaknesses of the last 14-nm bolide?

1 - Introduction and test setup 2 - Graphic benchmarks: Gaming 3 - Graphic benchmarks: Workstation Real-Time 3D 4 - Graphic benchmarks: Workstation Composite 5 - CPU benchmarks: Composite und I/O 6 - CPU benchmarks: Rendering 7 - CPU benchmarks: Compute and HPC 8 - Power Consumption 9 - Summary and Conclusion

Now the final crunch is done, even if it gets hot up to the EPS cables. Whereby at 7-Zip again the 2066-Club and the 2066-Club will have a swim

Calculix is based on a free finite element program for three-dimensional structural calculations and relies on a large core team.

Convolution is a benchmark for a subarea of function analysis, where a new, third function is created as the product of two functions (convolution, convolution). Here, clock and core number scale equally well. Again an advantage for AMD and the 2066s have something to celebrate again.

The FFTW test looks a bit strange, because Intel is gallantly preferred here. SSE only for AMD? One should take a closer look at this. But the cores are in charge again. And how … well, unless they were supplied by AMD.

The underground structure in the earth can be determined by seismic processing. One of the four basic steps is the Kirchhoff migration, in order to model an image from the existing data by means of mathematical operations. The benchmark used, including the calculations contained in it, meets AMD’s requirements and the 2066 old rockers are allowed to cheer themselves up once again.

LAMMPS stands for Large-scale Atomic/Molecular Massively Parallel Simulator and the program is also a 2066 domain. And how! Even the nuclear mass and the clock rate don’t help the Ryzen, if the compiler has changed to the enemy camp.

.

NAMD is a benchmark for high-performance simulation of large biomolecular systems

The Poisson equation is an elliptic, partial differential equation of second order, which is used as part of boundary value problems in many areas of physics. Intel’s new 10-core is still quite well in the race, while the two i9-9900(KS) weaken inexplicably but reproducibly.

In Python, like in Math Lab, you rely on Intel’s Math Kernel Library (MKL) in many areas. Especially NumPy suffered a little in the past. That’s why the Core i9-10900K does comparatively well.

The Pre-Euler3D CFD test (Ccomputational Fluid Dynamics Benchmark) runs preferably broadly diversified on as many cores as possible as well as a high-performance platform with fast memory connection. Exactly that is what you see on the bars.