NVIDIA GeForce RTX 2060 Linux Performance From Gaming To TensorFlow & Compute

Written by Michael Larabel in Graphics Cards on 8 January 2019 at 09:30 AM EST. Page 9 of 9. 35 Comments.
NVIDIA GeForce RTX 2060 Linux Benchmarks

When running OctaneBench 4.0, this proprietary CUDA renderer is a bit odd with for example the GTX 980 Ti hitting close to the GTX 1080 and GTX 1070 Ti performance. Anyhow, of the RTX 2060 performance, it did come in 10% faster than the GTX 1080.

NVIDIA GeForce RTX 2060 Linux Benchmarks

With the NAMD scientific simulations running on the GPUs via CUDA, the RTX 2060 performance came in between the GTX 1080 and 1080 Ti.

NVIDIA GeForce RTX 2060 Linux Benchmarks

When running the Folding@Home FAHBench, the RTX 2060 performance hit mid-way between the GTX 1080 and GTX 1080 Ti performance.

NVIDIA GeForce RTX 2060 Linux Benchmarks
NVIDIA GeForce RTX 2060 Linux Benchmarks

And with the Turing power efficiency for Folding@Home continuing to be well in front of Pascal GPUs and certainly for Maxwell.

NVIDIA GeForce RTX 2060 Linux Benchmarks

Over the course of the various GPU compute benchmarks carried out, the RTX 2060 had an average temperature of 60 degrees for the core temperature and a peak of 71 degrees, similar to the thermal performance while gaming.

NVIDIA GeForce RTX 2060 Linux Benchmarks

Under all these compute tests, the GeForce RTX 2060 with the Core i9 9900K had an average AC power draw of 164 Watts and a peak of 304 Watts. Meanwhile, the GTX 1080 came in slightly higher with a 167 Watt average and a peak of 316 Watts.

Overall, the GeForce RTX 2060 is an interesting graphics card at the $349 USD mark. For Linux gaming, FP32 compute, and other scenarios the performance is generally on-par with the GeForce GTX 1080 and similar to or beating the Radeon RX Vega 64. The RTX 2060 has the advantage of the ray-tracing capabilities though at the moment there isn't any noteworthy Vulkan ray-tracing games or even interesting demos for Linux users. For the tensor cores on the RTX 2060, the performance was certainly quite good with TensorFlow FP16 testing and in several of the benchmarks approaching the GeForce GTX 1080 Ti. The power efficiency of the RTX 2060 is certainly great as shown by those many data points. As I've only been testing this graphics card the past 24 hours now, I'll refrain from drawing any other conclusions but overall the performance is looking very good so far. Stay tuned for additional testing of the GeForce RTX 2060 and these other Turing graphics cards with more benchmarks to come.

If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.


Related Articles
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.