NVIDIA GeForce GTX 1060 6 GB vs NVIDIA Tesla K80 Comparison

The Verdict

The data splits this comparison cleanly. The NVIDIA GeForce GTX 1060 6 GB is the more balanced, everyday card: it wins the OpenCL benchmark by a massive 73.6%, and its average benchmark score of 21856 puts it in the 67th percentile of all GPUs. The Tesla K80, despite its larger memory and older compute-oriented design, trails with an average score of 18866 (63rd percentile). For general compute workloads, the GTX 1060 is the obvious pick. The Tesla K80 only takes the Vulkan test, and even then by a slim 2.4% margin. Choose the GTX 1060 unless you specifically need the K80's 12 GB frame buffer or its unique dual-GPU compute layout for a narrow set of tasks.

Where Each One Wins

GTX 1060 6 GB wins decisively in OpenCL. Its score of 32322 dwarfs the Tesla K80's 18620, a 73.6% advantage. This is the strongest single result in the comparison and reflects the Pascal architecture's much higher clock speeds (1506 MHz base versus 562 MHz) and better per-shader efficiency. The GTX 1060 also leads in pixel rate (82.03 GPixel/s versus 42.85 GPixel/s) and FP32 throughput (4.375 TFLOPS versus 4.113 TFLOPS), despite having fewer shading units (1280 versus 2496).

Tesla K80 wins narrowly in Vulkan. Its score of 19111 edges out the GTX 1060's 18659, a 2.4% difference. This is a small victory, but it shows the K80's older Kepler architecture can still handle Vulkan workloads acceptably. The K80 also has clear hardware advantages in raw resources: 2496 shading units, 208 TMUs, a 384-bit memory bus, and 12 GB of GDDR5 versus the GTX 1060's 192-bit bus and 6 GB. Its texture rate of 171.4 GTexel/s is higher than the GTX 1060's 136.7 GTexel/s.

Architecture Differences

The two cards come from different generations and design philosophies. The GTX 1060 uses the GP106 chip on TSMC's 16 nm process, packing 4,400 million transistors into a 200 mm² die (transistor density of 22.0M per mm²). The Tesla K80 uses the GK210 chip on the older 28 nm process, with 7,100 million transistors across a much larger 561 mm² die (density of 12.7M per mm²). This explains the GTX 1060's efficiency advantage: it delivers more performance per transistor and per watt.

The GTX 1060 is a consumer gaming card with display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a) and a 120 W TDP. The Tesla K80 is a compute accelerator with no display outputs, a 300 W TDP, and a single 8-pin power connector (versus a single 6-pin on the GTX 1060). The K80's architecture is Kepler 2.0, which supports DirectX 12 (11_1), while the GTX 1060's Pascal supports DirectX 12 (12_1). Vulkan support differs: the GTX 1060 reports version 1.4, the K80 reports 1.2.175.

Clock speeds tell a big story. The GTX 1060 runs at 1506 MHz base and 1709 MHz boost, while the K80 sits at 562 MHz base and 824 MHz boost. Memory clocks differ too: the GTX 1060 uses 2002 MHz (8 Gbps effective), the K80 uses 1253 MHz (5 Gbps effective). Despite the K80's higher memory bandwidth (240.6 GB/s versus 192.2 GB/s), the GTX 1060's faster core clocks drive its benchmark superiority.

FAQ

Q: Which card has higher average benchmark scores?

A: The GTX 1060 6 GB scores 21856 on average, versus 18866 for the Tesla K80. This places the GTX 1060 in the 67th percentile of all GPUs and the K80 in the 63rd.

Q: Is the Tesla K80 better in any benchmark?

A: Yes, it wins the geekbench_vulkan test with a score of 19111 against the GTX 1060's 18659, a 2.4% margin. That is its only benchmark win.

Q: How much memory does each card have?

A: The Tesla K80 has 12 GB of GDDR5 on a 384-bit bus, while the GTX 1060 has 6 GB of GDDR5 on a 192-bit bus. The K80's bandwidth is 240.6 GB/s versus 192.2 GB/s.

Q: What is the power draw difference?

A: The GTX 1060 has a 120 W TDP and requires a 300 W power supply. The Tesla K80 has a 300 W TDP and suggests a 700 W power supply.

Q: Do both cards support the same APIs?

A: No. The GTX 1060 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The Tesla K80 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.175.

Q: Which card is smaller?

A: The GTX 1060 is 250 mm long (9.8 inches), while the Tesla K80 is 267 mm (10.5 inches). Both are dual-slot designs.

Head-to-Head Benchmarks

The OpenCL result is the headline. The GTX 1060 scores 32322 versus the K80's 18620, a 73.6% advantage. This is not a marginal win; it is a generational gap. The GTX 1060's FP32 throughput of 4.375 TFLOPS exceeds the K80's 4.113 TFLOPS, and its pixel rate of 82.03 GPixel/s nearly doubles the K80's 42.85 GPixel/s. Even though the K80 has more shading units (2496 vs 1280), the GTX 1060's higher clocks and Pascal efficiency more than compensate.

The Vulkan test flips the result, but barely. The Tesla K80 scores 19111, just 2.4% ahead of the GTX 1060's 18659. This is within noise and unlikely to matter in real workloads. The K80's higher texture rate (171.4 GTexel/s versus 136.7 GTexel/s) and larger memory bus may help in specific Vulkan compute tasks, but the margin is too small to call it a strength.

Looking at nearest rivals puts these scores in context. The GTX 1060's average sits between the Quadro RTX 5000 (21629, 1% lower) and the RX 6700S (22154, 1.3% higher). The Tesla K80's average is close to the RTX 2070 (18789, 0.4% lower) and the RTX 2000 Ada Generation (18954, 0.5% higher). So the GTX 1060 competes with newer mid-range cards, while the K80 aligns with older high-end parts.

Specification Differences

The key differences are process node, chip size, memory configuration, and power. The GTX 1060 uses a 16 nm TSMC process with 4,400 million transistors on a 200 mm² die. The Tesla K80 uses a 28 nm process with 7,100 million transistors on a 561 mm² die. Transistor density is 22.0M per mm² for the GTX 1060 versus 12.7M for the K80.

Memory: the K80 has 12 GB GDDR5 on a 384-bit bus with 240.6 GB/s bandwidth; the GTX 1060 has 6 GB GDDR5 on a 192-bit bus with 192.2 GB/s. Shading units: 1280 (GTX 1060) versus 2496 (K80). TMUs: 80 versus 208. ROPs are equal at 48.

Clocks: the GTX 1060 runs at 1506 MHz base and 1709 MHz boost; the K80 at 562 MHz base and 824 MHz boost. Memory clocks are 2002 MHz (8 Gbps effective) versus 1253 MHz (5 Gbps effective). Power: 120 W TDP with a 6-pin connector and 300 W suggested PSU versus 300 W TDP with an 8-pin connector and 700 W suggested PSU.

Outputs: the GTX 1060 has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a; the K80 has no display outputs. API support: DirectX 12 (12_1) versus 12 (11_1), OpenGL 4.6 on both, Vulkan 1.4 versus 1.2.175. The GTX 1060 is 250 mm long; the K80 is 267 mm. Release dates differ: the GTX 1060 launched in July 2016, the K80 in November 2014. The GTX 1060 has a launch MSRP of 299 USD; the K80 has no recorded launch MSRP.