NVIDIA GeForce RTX 2060 vs NVIDIA Tesla M2090 Comparison

Where Each One Wins

The recorded data splits these two NVIDIA cards into completely separate usage worlds. The GeForce RTX 2060 is the clear winner in every head-to-head comparison available in the database, taking 1 win against 0 for the Tesla M2090. That single win, however, is enormous in magnitude and tells the story of two different compute philosophies.

The RTX 2060 dominates in general-purpose compute workloads. Its Geekbench OpenCL score of 65014 versus the Tesla M2090's 13075 represents a 397.2% advantage, a more than fourfold gap. This makes the RTX 2060 the obvious choice for any task that relies on OpenCL acceleration, from rendering to data processing. The architecture is simply far more efficient at executing parallel workloads.

The Tesla M2090, by contrast, has no recorded wins in any benchmark. Its only entry in the database is that same Geekbench OpenCL test, where it falls far behind. This is not a card designed for modern consumer workloads. Its Fermi 2.0 architecture, from the Tesla generation, was built for a different era of compute. The data shows it is thoroughly outclassed by the much newer Turing-based card.

In terms of overall standing, the RTX 2060 sits at the 58th percentile among all GPUs in the database, while the Tesla M2090 sits at the 53rd percentile. The average benchmark score for the RTX 2060 is 15290, against 13075 for the Tesla M2090. The RTX 2060 also has a much richer benchmark profile, with ten recorded tests spanning DirectX 9 through DirectX 12, Vulkan, OpenCL, and 2D performance. The Tesla M2090 has just one. For a buyer looking at raw capability today, the RTX 2060 is the only viable option. For a collector or a legacy system, the Tesla M2090 has historical interest, but the numbers show no performance advantage.

Architecture Differences

The architectural gap between these two cards spans nearly a decade of GPU evolution. The RTX 2060 uses the TU106 chip built on Turing architecture, produced on a 12 nm process at TSMC. The Tesla M2090 uses the GF110 chip built on Fermi 2.0 architecture, produced on a 40 nm process, also at TSMC. The process node difference alone, 12 nm versus 40 nm, explains much of the efficiency and performance gap.

Transistor counts tell a stark story. The RTX 2060 packs 10,800 million transistors on a 445 mm² die, giving a transistor density of 24.3M per mm². The Tesla M2090 has just 3,000 million transistors on a larger 520 mm² die, yielding a density of only 5.8M per mm². The RTX 2060 crams nearly four times as many transistors into a smaller physical area.

The memory subsystems are fundamentally different. The RTX 2060 uses 6 GB of GDDR6 memory on a 192-bit bus, delivering 336.0 GB/s of bandwidth. The Tesla M2090 also has 6 GB, but it is GDDR5 on a wider 384-bit bus, delivering just 177.4 GB/s. The newer memory type and higher effective clock speed, 14 Gbps effective versus 3.7 Gbps effective, give the RTX 2060 nearly double the bandwidth despite the narrower bus.

Compute resources are massively skewed toward the RTX 2060. It has 1920 shading units, 120 texture mapping units, and 48 raster output units. The Tesla M2090 has only 512 shading units, 64 TMUs, and 48 ROPs. The RTX 2060 also includes dedicated hardware that the Tesla M2090 lacks entirely: 30 RT cores for ray tracing and 240 tensor cores for AI acceleration. The Fermi card has neither.

Clock behavior differs as well. The RTX 2060 has a base clock of 1365 MHz and a boost clock of 1680 MHz. The Tesla M2090 has no recorded base or boost clocks in the database, suggesting it ran at fixed frequencies typical of early Tesla compute cards. Pixel rate and texture rate reflect the resource gap: the RTX 2060 achieves 80.64 GPixel/s and 201.6 GTexel/s, while the Tesla M2090 manages only 20.83 GPixel/s and 41.66 GTexel/s. Floating-point performance is 6.451 TFLOPS for the RTX 2060 versus 1,332.2 GFLOPS (roughly 1.33 TFLOPS) for the Tesla M2090. The RTX 2060 also supports FP16 at 12.90 TFLOPS with a 2:1 ratio; the Tesla M2090 has no FP16 support recorded.

Power and connectivity also separate them. The RTX 2060 has a TDP of 160 W, a single 8-pin power connector, and a suggested PSU of 450 W. The Tesla M2090 draws 250 W, requires both a 6-pin and an 8-pin connector, and needs a 600 W PSU. The bus interface is PCIe 3.0 x16 on the RTX 2060 versus PCIe 2.0 x16 on the Tesla. Display outputs are another major split: the RTX 2060 has 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C. The Tesla M2090 has no display outputs at all, as it is a compute-only accelerator.

Head-to-Head Benchmarks

The database contains one direct comparison between these two cards, and it is a landslide. In Geekbench OpenCL, the RTX 2060 scores 65014 against the Tesla M2090's 13075. That is a delta of 397.2%, meaning the RTX 2060 is almost five times faster in this compute test. This is not a marginal win; it is a generational obliteration.

To put that score in context, the RTX 2060's nearest rivals in the database include the NVIDIA GeForce GTX 580 with an average score of 15283 (a 0% delta), the AMD Radeon 680M at 15270 (0.1% ahead), the NVIDIA GeForce RTX 3050 OEM at 15199 (0.6% ahead), and the AMD Radeon RX 7600 at 15171 (0.8% ahead). The RTX 2060's average benchmark score of 15290 places it just above all of these. The Tesla M2090, with its average score of 13075, sits near the NVIDIA GeForce GTX 1660 SUPER at 12986 (0.7% behind), the NVIDIA GeForce GTX 950 at 13189 (0.9% ahead), the NVIDIA GeForce RTX 3050 Ti Mobile at 12940 (1% behind), and the AMD Radeon RX 580 at 12928 (1.1% behind). In other words, the Tesla M2090's compute score is roughly comparable to a mid-range GTX 950, while the RTX 2060 is racing well above that tier.

The RTX 2060's other benchmark scores reinforce its superiority. In Passmark G3D, it scores 14111. In Passmark GPU Compute, it scores 5488. In Geekbench Vulkan, it scores 65846. In 3DMark Steel Nomad DX12, it scores 1242. It also has scores for older DirectX versions: 98 in Passmark DirectX 10, 110 in DirectX 11, 53 in DirectX 12, and 190 in DirectX 9. The 2D score is 745. None of these tests exist for the Tesla M2090 in the database, so no direct comparison is possible, but the single OpenCL result is decisive enough.

The verdict from the data is unambiguous. The RTX 2060 wins the only head-to-head test by a factor of five. There is no scenario in the recorded benchmarks where the Tesla M2090 comes out ahead.

FAQ

Q: Which card has better overall performance in the database?

A: The NVIDIA GeForce RTX 2060 has an average benchmark score of 15290, while the Tesla M2090 has an average score of 13075. The RTX 2060 also holds a higher percentile ranking at 58 versus 53 for the Tesla.

Q: How much faster is the RTX 2060 in OpenCL compute?

A: In Geekbench OpenCL, the RTX 2060 scores 65014 versus the Tesla M2090's 13075, a 397.2% advantage. This means the RTX 2060 is nearly five times faster in that test.

Q: What are the key architectural differences?

A: The RTX 2060 uses Turing architecture on a 12 nm process with 10,800 million transistors, while the Tesla M2090 uses Fermi 2.0 on a 40 nm process with 3,000 million transistors. The RTX 2060 also has 1920 shading units, 30 RT cores, and 240 tensor cores; the Tesla has 512 shading units and no RT or tensor cores.

Q: Does the Tesla M2090 support modern graphics APIs?

A: The Tesla M2090 supports DirectX 12 (11_0) and OpenGL 4.6, but it has no Vulkan support recorded. The RTX 2060 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: Can the Tesla M2090 output video to a display?

A: No. The Tesla M2090 has no display outputs, as it is a compute-only accelerator. The RTX 2060 includes 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C.

Q: Which card is more power-efficient?

A: The RTX 2060 has a TDP of 160 W and requires a 450 W PSU, while the Tesla M2090 has a TDP of 250 W and requires a 600 W PSU. The RTX 2060 delivers far higher performance while drawing less power.

Specification Differences

The two cards differ in nearly every measurable specification. The RTX 2060 uses a 12 nm process; the Tesla M2090 uses 40 nm. Transistor count is 10,800 million versus 3,000 million. Die size is 445 mm² versus 520 mm². Transistor density is 24.3M per mm² versus 5.8M per mm².

Memory differs in type and bandwidth: the RTX 2060 has 6 GB of GDDR6 on a 192-bit bus with 336.0 GB/s, while the Tesla M2090 has 6 GB of GDDR5 on a 384-bit bus with 177.4 GB/s. The RTX 2060's memory clock is 1750 MHz (14 Gbps effective); the Tesla's is 924 MHz (3.7 Gbps effective).

Compute resources: the RTX 2060 has 1920 shading units, 120 TMUs, and 48 ROPs. The Tesla M2090 has 512 shading units, 64 TMUs, and 48 ROPs. The RTX 2060 has 30 RT cores and 240 tensor cores; the Tesla has none. Pixel rate is 80.64 GPixel/s versus 20.83 GPixel/s. Texture rate is 201.6 GTexel/s versus 41.66 GTexel/s. FP32 is 6.451 TFLOPS versus 1,332.2 GFLOPS. FP16 is 12.90 TFLOPS for the RTX 2060; the Tesla has no FP16 data.

Power and physical specs: TDP is 160 W versus 250 W. The RTX 2060 uses a single 8-pin connector; the Tesla uses 1x 6-pin plus 1x 8-pin. Suggested PSU is 450 W versus 600 W. Bus interface is PCIe 3.0 x16 versus PCIe 2.0 x16. The RTX 2060 is 229 mm long, 113 mm tall, and 35 mm wide. The Tesla M2090 is 248 mm long; height and width are not recorded.

API support: the RTX 2060 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Tesla M2090 supports DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan. Display outputs: the RTX 2060 has 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C. The Tesla has no outputs.

Release timing also separates them: the RTX 2060 was released on 2019-01-06, while the Tesla M2090 came out on 2011-07-24. The RTX 2060 has a launch MSRP of 349 USD. The Tesla M2090 has no recorded launch MSRP. Both are end-of-life products, but they belong to different generations entirely: the RTX 2060 is GeForce 20-series with a successor of GeForce 30, while the Tesla M2090 is Tesla Fermi with a successor of Tesla Kepler.