NVIDIA GeForce GTX 1660 vs NVIDIA Tesla C2070 Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark between the NVIDIA GeForce GTX 1660 and the NVIDIA Tesla C2070 is the Geekbench OpenCL test, and the result is decisively one-sided. The GTX 1660 scores 47,850 points, while the Tesla C2070 manages 9,716 points. That is a 392.5% advantage for the newer card, meaning the GTX 1660 delivers nearly five times the compute performance in this particular workload.

Looking at the broader database picture, the GTX 1660's average benchmark score of 11,680 places it at the 51st percentile of all GPUs. Its nearest rivals in the database include the AMD Radeon RX 7800 XT at 11,627 (0.5% behind), the AMD Radeon RX 6500 XT at 11,842 (1.4% ahead), and the NVIDIA Tesla K20c at 11,479 (1.8% behind). The GTX 1660 sits in a tightly contested cluster where a few percentage points separate it from completely different product generations and price tiers.

The Tesla C2070, meanwhile, has an average benchmark score of 9,716 and sits at the 47th percentile. Its nearest rivals are close neighbors: the NVIDIA Tesla M10 at 9,724 (0.1% ahead), the NVIDIA Quadro P4000 at 9,665 (0.5% behind), and the AMD Radeon Pro WX 2100 at 9,653 (0.7% behind). Notably, the NVIDIA GeForce GTX 1070 scores 9,780, a 0.7% margin over the Tesla C2070. The C2070 is not dramatically outclassed by its immediate database neighbors, which suggests its age shows more acutely against modern consumer cards than against other professional or workstation parts from its own era.

The Geekbench OpenCL delta is the headline figure because it is the only head-to-head metric recorded. A 392.5% difference is not a marginal gap; it is a generational chasm. In raw FP32 throughput, the GTX 1660 delivers 5.027 TFLOPS against the C2070's 1,027.7 GFLOPS, reinforcing the compute gap observed in the benchmark. The GTX 1660 also leads in texture rate at 157.1 GTexel/s versus 32.14 GTexel/s, and pixel rate at 85.68 GPixel/s versus 16.07 GPixel/s. Every measured throughput metric tells the same story: the GTX 1660 is fundamentally faster across the board.

Where Each One Wins

The GTX 1660 wins in every category where data exists. In compute workloads, the Geekbench OpenCL score is 392.5% higher. In rasterization fundamentals, the 85.68 GPixel/s pixel rate and 157.1 GTexel/s texture rate dwarf the C2070's 16.07 GPixel/s and 32.14 GTexel/s. The GTX 1660 also supports a wider range of modern APIs: DirectX 12 (12_1) versus the C2070's DirectX 12 (11_0), Vulkan 1.4 versus no Vulkan support at all, and OpenGL 4.6 on both.

The Tesla C2070 does not win any of the recorded benchmarks. Its only advantage is structural rather than performance-based: it uses a 384-bit memory bus with 6 GB of GDDR5, the same capacity as the GTX 1660, but with a wider interface. However, the C2070's memory bandwidth of 143.4 GB/s is lower than the GTX 1660's 192.1 GB/s, because the GTX 1660 runs its GDDR5 at 8 Gbps effective versus 3 Gbps effective on the C2070. The wider bus cannot compensate for the much slower memory clock.

For use cases, the GTX 1660 is the clear choice for anything involving modern graphics APIs, gaming, or general compute acceleration. The C2070, with its 448 shading units and 56 TMUs, belongs to a different era of GPU design. Its 48 ROPs match the GTX 1660's 48 ROPs, but that is the only specification where the two cards are equal. The C2070's 1x DVI output limits it to a single legacy display, while the GTX 1660 offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. For multi-monitor setups or modern display connectivity, the GTX 1660 is the only viable option.

Architecture Differences

The two cards come from completely different architectural lineages. The GTX 1660 uses the TU116 chip built on Turing architecture, fabricated on TSMC's 12 nm process. It packs 6,600 million transistors into a 284 mm² die, achieving a transistor density of 23.2M per mm². The Tesla C2070 uses the GF100 chip based on Fermi architecture, built on TSMC's 40 nm process. It contains 3,100 million transistors across a much larger 529 mm² die, with a transistor density of just 5.9M per mm². The GTX 1660 packs more than twice the transistors into roughly half the silicon area.

The clock speeds reflect the architectural and process differences. The GTX 1660 has a base clock of 1530 MHz and a boost clock of 1785 MHz. The Tesla C2070 has no recorded base or boost clock in the database, but its memory runs at 747 MHz (3 Gbps effective), compared to the GTX 1660's memory clock of 2001 MHz (8 Gbps effective). The GTX 1660's memory advantage is substantial: 192.1 GB/s of bandwidth from a 192-bit bus, versus 143.4 GB/s from a 384-bit bus on the C2070. The newer card achieves higher bandwidth with a narrower bus, purely through faster memory technology.

Compute resources differ sharply. The GTX 1660 has 1,408 shading units and 88 TMUs. The C2070 has 448 shading units and 56 TMUs. FP32 performance is 5.027 TFLOPS on the GTX 1660 versus 1,027.7 GFLOPS on the C2070. The GTX 1660 also supports FP16 at 10.05 TFLOPS (2:1 ratio), while the C2070 has no recorded FP16 capability. Neither card has RT cores or tensor cores; the GTX 1660 is a raster-focused Turing implementation without ray tracing acceleration.

Power and connectivity further separate the two. The GTX 1660 has a 120 W TDP, requiring a single 8-pin power connector and a suggested 300 W PSU. The C2070 draws 238 W, needs a 6-pin plus an 8-pin connector, and demands a suggested 550 W PSU. Both are dual-slot cards, but the C2070 is longer at 248 mm (9.8 inches) versus 229 mm (9 inches) for the GTX 1660. The GTX 1660 also uses PCIe 3.0 x16, while the C2070 is limited to PCIe 2.0 x16.

The release dates underline the generational gap: the GTX 1660 launched on March 13, 2019, while the C2070 launched on July 24, 2011. Both are end-of-life products. The GTX 1660's predecessor is GeForce 10 and its successor is GeForce 20; the C2070's predecessor is Tesla and its successor is Tesla Kepler. The GTX 1660 had a launch MSRP of 219 USD.

FAQ

Q: Which GPU is faster in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 1660 scores 47,850 versus 9,716 for the NVIDIA Tesla C2070, a 392.5% advantage for the GTX 1660.

Q: Do both cards have the same amount of memory?

A: Yes, both have 6 GB of GDDR5, but the GTX 1660 delivers 192.1 GB/s of bandwidth over a 192-bit bus, while the C2070 delivers 143.4 GB/s over a 384-bit bus.

Q: Does the Tesla C2070 support Vulkan?

A: No, the C2070 has no Vulkan support recorded, while the GTX 1660 supports Vulkan 1.4.

Q: Which card has higher power consumption?

A: The Tesla C2070 has a 238 W TDP and requires a 550 W suggested PSU, while the GTX 1660 has a 120 W TDP and a 300 W suggested PSU.

Q: How do the two cards compare in transistor count and die size?

A: The GTX 1660 has 6,600 million transistors on a 284 mm² die, while the C2070 has 3,100 million transistors on a 529 mm² die.

Q: What is the DirectX support difference?

A: The GTX 1660 supports DirectX 12 (12_1), while the C2070 only supports DirectX 12 (11_0).

The Verdict

The data leaves no room for ambiguity. The NVIDIA GeForce GTX 1660 outperforms the NVIDIA Tesla C2070 in every recorded benchmark and every relevant throughput metric. The 392.5% lead in Geekbench OpenCL is the single most telling statistic: it is not a close contest, and no workload category favors the older card.

The GTX 1660 is the appropriate choice for any user needing modern graphics API support, higher memory bandwidth, lower power draw, or contemporary display outputs. Its 51st percentile ranking among all GPUs and its competitive position against cards like the AMD Radeon RX 6500 XT (1.4% behind) and AMD Radeon RX 7800 XT (0.5% ahead) show that it remains a relevant performer in the database's current landscape.

The Tesla C2070, at the 47th percentile, sits just a few points below the GTX 1660 in percentile terms, but that ranking is based on a much smaller benchmark footprint. Its only recorded benchmark is Geekbench OpenCL, and in that test it is thoroughly outclassed. The C2070's wider 384-bit memory bus and equal 6 GB capacity are its only redeeming specifications, but they do not translate into competitive performance.

For a professional compute workload that specifically requires Fermi-era characteristics, or for legacy system compatibility, the C2070 might still serve a purpose. But for any modern application, gaming workload, or general-purpose GPU compute, the GTX 1660 is the only rational pick from the recorded data. The GTX 1660 wins the head-to-head 1 to 0, and the margin of that win is so large that the verdict is effectively unanimous.