BOXX APEXX T3 Flagship AMD Ryzen Threadripper Workstation Review

BOXX APEXX T3 Performance

We are going to focus on the BOXX APEXX T3 CPU performance with the AMD Ryzen Threadripper 2990WX in this piece using our standard Linux-Bench2 tests. We are going to have an additional piece focusing on the AMD Radeon Pro WX 8200 GPU. Since this is STH, we utilize Linux as our testing foundation. Linux tends to have better thread scheduling to take advantage of the unique NUMA architecture of the Threadripper 2990WX that is in our BOXX APEXX T3 test system.

Python Linux 4.4.2 Kernel Compile Benchmark

This is one of the most requested benchmarks for STH over the past few years. The task was simple, we have a standard configuration file, the Linux 4.4.2 kernel from kernel.org, and make the standard auto-generated configuration utilizing every thread in the system. We are expressing results in terms of compiles per hour to make the results easier to read.

This is always an interesting test as it stresses a number of different aspects of CPU and memory performance. At the end of the day, the AMD Ryzen Threadripper 2990WX benefits from high clock speeds which push it past many of the server CPU configurations. If your users were previously running dual Intel Xeon E5-2690 V3 processors, one can see a speedup through the single 32-core Threadripper CPU in this BOXX APEXX T3.

c-ray 1.1 Performance

We have been using c-ray for our performance testing for years now. It is a ray tracing benchmark that is extremely popular to show differences in processors under multi-threaded workloads. Our previous generation benchmark suite uses a 4K resolution test. We normally do not publish this anymore since higher-end SKUs do not offer enough differentiation. Since our data set is larger with the 4K tests on the embedded side, we wanted to get some numbers out there.

C-ray on Linux and Cinebench on Windows show a similar picture. The AMD Ryzen Threadripper 2990WX is a blazingly fast chip. AMD’s Zen architectures do well here and so having 32 fast AMD Zen+ cores leads to an excellent result. There is not much more to say here except that there is an enormous gap even between this and the previous AMD Ryzen Threadripper 1950X flagship processor.

7-zip Compression Performance

7-zip is a widely used compression/ decompression program that works cross-platform. We started using the program during our early days with Windows testing. It is now part of Linux-Bench.

Our default is sorting on the decompression speed for this chart. You will note that our compression test shows the architectural limitation of leaving half of the cores another hop from main memory. To be clear, this is still an extremely fast chip, but it has the architectural limitation keeping it from blowing away just about anything else out there. This is one of the cases where we can see the impact clearly.

GROMACS STH Small AVX2/ AVX-512 Enabled

We have a small GROMACS molecule simulation we previewed in the first AMD EPYC 7601 Linux benchmarks piece. In Linux-Bench2 we are using a “small” test for single and dual socket capable machines. Our medium test is more appropriate for higher-end dual and quad socket machines. Our GROMACS test will use the AVX-512 and AVX2 extensions if available.

Here we see a solid performance, competing directly with dual Intel Xeon Gold 5119T setups. At the same time, this result is being limited by AVX2 and the Threadripper 2990WX memory connectivity situation. AMD does not have AVX-512 so if your applications are AAVX-512 heavy, the Intel Xeon Gold is a good option instead. Still, the BOXX APEXX T3 performs extremely well, even faster than more expensive AMD EPYC 32 core server processors due to the increased clock speed.

Chess Benchmarking

Chess is an interesting use case since it has almost unlimited complexity. Over the years, we have received a number of requests to bring back chess benchmarking. We have been profiling systems and are ready to start sharing results:

Getting this result was surprisingly challenging. This is one of the workloads that likes to assume that you have locally attached memory on each NUMA node. As such, it required a bit of script editing to get this to work on the Threadripper 2990WX. If you have applications that explicitly look for CPUs with directly attached memory, you may run into a challenge with the Threadripper 2990WX, but again, the 2950X is a great alternative in the BOXX APEXX T3 platform. The on-package connectivity does have an impact. At the same time, the Threadripper 2990WX is still the performance value leader.

Now that we have explored performance, let us take a look at the power consumption, noise, and give our final thoughts on the BOXX system.