ASUS Turbo-RTX2080-8G Blower-Style GeForce RTX 2080 Performance Review

ASUS Turbo-RTX2080-8G Compute Related Benchmarks

We are going to run the ASUS Turbo-RTX2080-8G through a number of benchmarks to see if it is a worthy compute successor to the GeForce GTX 1080. We have a number of readers wondering how this will compare to the Pascal generation and the AMD offerings as compute-focused cards.

Geekbench 4

Geekbench 4 measures the compute performance of your GPU using workloads ranging from image processing to computer vision to number crunching.

Our first compute benchmark we see the Turbo-RTX2080-8G nails Geekebench on both OpenCL and almost double the CUDA scores we have seen before on other cards.

LuxMark OpenCL Benchmark

LuxMark is an OpenCL benchmark tool based on LuxRender.

The Turbo-RTX2080-8G takes the lead on OpenCL scores generated by Luxmark. This pushes performance beyond the AMD Radeon Vega Frontier Edition 16GB cards which tend to be good in OpenCL performance.

AIDA64 GPGPU

These benchmarks are designed to measure GPGPU computing performance via different OpenCL workloads. The first two focus on classic floating point operations per second:

• Single-Precision FLOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as FLOPS (Floating-Point Operations Per Second), with single-precision (32-bit, “float”) floating-point data.
• Double-Precision FLOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as FLOPS (Floating-Point Operations Per Second), with double-precision (64-bit, “double”) floating-point data.

Here you can see that in raw double precision FLOPS, the AMD cards tend to perform extremely well. On the single precision side, the ASUS Turbo-RTX2080-8GB edges a small victory over the AMD Radeon Pro WX 8200 and a larger one over the GTX 1080.

The next set of benchmarks from AIDA64 measure integer operation speed:

• 24-bit Integer IOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as IOPS (Integer Operations Per Second), with 24-bit integer (“int24”) data. This particular data type defined in OpenCL on the basis that many GPUs are capable of executing int24 operations via their floating-point units.
• 32-bit Integer IOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as IOPS (Integer Operations Per Second), with 32-bit integer (“int”) data.
• 64-bit Integer IOPS: Measures the classic MAD (Multiply-Addition) performance of the GPU, otherwise known as IOPS (Integer Operations Per Second), with 64-bit integer (“long”) data. Most GPUs do not have dedicated execution resources for 64-bit integer operations, so instead, they emulate the 64-bit integer operations via existing 32-bit integer execution units.

The ASUS Turbo-RTX2080-8G shows strong numbers in AIDA64 GPGPU benchmark. This shows some of the architectural improvements that NVIDIA’s Turing architecture brings to the table over Pascal generations.

SPECviewperf 12

SPECviewperf measures the 3D graphics performance of systems running under the OpenGL and Direct X application programming interfaces.

With SPECviewperf 12 the ASUS Turbo-RTX2080-8G falls in the middle of the pack. Let us move on to the newer version. This is important because software and hardware do not always advance at the same rate. Much has been made about the ray tracing capabilities and tensor cores on the RTX series, however, there are a lot of older code bases that do not support newer features.

SPECviewperf 13

There is a newer version of SPECviewperf 13 that we are starting to run with this card.

Version 13 adds improvements to the benchmarks it runs. We will use Version 13 from now on.

We will have more results to share here as we run additional cards through the test suite.

Next, we will have SPECworkstation 3 and 3DMark results before moving onto power consumption, thermals, and our final thoughts.