ASUS Turbo-RTX2060-6G Blower-Style Performance Review


ASUS Turbo-RTX2060-6G Blower Specifications

Here are the key specs for the ASUS Turbo-RTX2060-6G:

  • Graphics Processor: TU106-200A-KA-A1
  • Architecture: Turing
  • Process Size: 12nm
  • Memory: 6GB GDDR6
  • Memory Bus: 192 bit
  • Bandwidth: 336.0 GB/s
  • GPU Clock: 1365 MHz
  • Boost Clock: 1680 MHz
  • Memory Clock: 1750 MHz
  • Shading Units: 1920
  • TMUs: 120
  • ROPs: 48
  • SM Count: 30
  • Tensor Cores: 240
  • RT Cores: 30
  • Pixel Rate: 80.64 GPixel/s
  • Texture Rate: 201.6 GTexel/s
  • FP32 (float): 6,451 GFLOPS
  • FP64 (float): 201.6 GFLOPS
  • Slot Width: Dual-Slot
  • Length: 10.54 inches
  • TDP: 160W
  • Cooling: Blower Style

Like the previous-generation GTX 1060, the NVIDIA RTX 2060 utilizes only 6GB of memory, albeit GDDR6 in this generation. We would have liked to have seen NVIDIA bump the memory size more in this generation.

With a TDP of only 160W, the NVIDIA RTX 2060 is well suited to not just workstations, but also departmental servers. An example may be a tower server deployed at the edge like the Dell EMC PowerEdge T340. If you need low cost and low power compute at the edge, this may indeed make sense. We have our NVIDIA GTX 1160 Ti 6GB review in-progress as well so we will compare both soon.

Testing the ASUS Turbo-RTX2060-6G

We will use the same setup that we have used in the past for our testing needs.

• Motherboard: ASUS WS C621E SAGE Motherboard
• CPU: 2x Intel Xeon Gold 6134 (8 core / 16 Threads)
• GPU: ASUS Turbo-RTX2060-6G
• Cooling: Noctua NH-U14S DX-3647 LGA3647
• RAM: 12x Micron 16GB DDR4 Low Profile RDIMMs
• SSD: Samsung PM961 1TB
• OS: Windows 10 Pro

ASUS Turbo RTX2060 GPUz
ASUS Turbo RTX2060 GPUz

GPU-Z shows the primary stats of our Turbo-RTX2060-6G. The Turing GPU clocks in at 1395 MHz and GDDR6 Memory at 1365 MHz, Boost speed goes up the 1710 MHz. Pixel Fillrates run at 82.1 GPixels/s, and Texture Fillrate comes in at 205.2 GTexel/s.

The RTX2060-6G GPU features 1920 active shader/stream cores, and the RT core count is at 30 and packs in 240 Tensor cores.

Let us move on and start our testing with computing-related benchmarks.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.