NVIDIA GeForce MX350 vs NVIDIA Tesla M2090 Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark between the NVIDIA Tesla M2090 and the NVIDIA GeForce MX350 is the Geekbench OpenCL test, and the results are decisive. The Tesla M2090 scores 13075, while the MX350 scores 8689, giving the Tesla a 50.5% advantage. This is a substantial margin, placing the older compute-oriented card firmly ahead in raw OpenCL compute performance.

Context from the database's nearest rival comparisons reinforces this gap. The Tesla M2090's score sits within 1% of the NVIDIA GeForce GTX 1660 SUPER (12986, a 0.7% difference) and the AMD Radeon RX 580 (12928, a 1.1% difference). It also edges out the NVIDIA GeForce RTX 3050 Ti Mobile (12940) by 1%. The MX350, by contrast, lands in a much lower tier: it is 0.7% ahead of the AMD Radeon Pro 450 (10804) and 1.1% ahead of the NVIDIA Quadro K2200 (10761), while trailing the NVIDIA GeForce GTX 1650 SUPER (11047) by 1.5% and the AMD Radeon RX 550 (11075) by 1.7%. The MX350's average benchmark score across all recorded tests is 10883, compared to the Tesla M2090's 13075, a difference of roughly 20% in the Tesla's favor.

The MX350 does have one counterpoint in the broader benchmark suite: its Geekbench Vulkan score is 13077, which is nearly identical to the Tesla M2090's OpenCL result. However, the Tesla M2090 has no recorded Vulkan score, so this cannot be treated as a head-to-head comparison. Within the single shared test, the OpenCL result stands as the definitive measurement: the Tesla M2090 wins, and it wins by a wide margin.

Where Each One Wins

The Tesla M2090 wins in compute-heavy OpenCL workloads, as evidenced by its 50.5% lead in the Geekbench OpenCL benchmark. This aligns with its design as a high-throughput compute card, and its nearest rival positioning (within 1.1% of cards like the GTX 1660 SUPER and RX 580) suggests it holds its own against much newer mainstream GPUs in this specific workload. The database records one win for the Tesla M2090 in the head-to-head comparison.

The GeForce MX350 wins nowhere in the direct head-to-head comparison; the database records zero wins for it. However, its strengths lie elsewhere. It has a recorded Vulkan score of 13077, which is a strong result and matches the Tesla M2090's OpenCL score almost exactly, though no direct comparison is available. The MX350 also carries a much lower power footprint (20 W TDP versus 250 W for the Tesla), making it suitable for portable devices, as its display outputs are described as "Portable Device Dependent." Its base clock of 1354 MHz and boost clock of 1468 MHz are far higher than the Tesla's unlisted clock speeds, which suggests better real-world responsiveness in lighter tasks, though no benchmark data confirms this.

In terms of architectural efficiency, the MX350 packs 640 shading units into a 132 mm² die on a 14 nm process, while the Tesla M2090 fits 512 shading units into a 520 mm² die on a 40 nm process. The MX350's transistor density of 25.0M per mm² is over four times that of the Tesla's 5.8M per mm², indicating a much more modern and power-efficient design. But efficiency does not translate into compute dominance in the recorded data: the Tesla M2090's FP32 throughput is 1,332.2 GFLOPS versus the MX350's 1.879 TFLOPS, which at face value appears to favor the MX350, yet the OpenCL benchmark decisively favors the Tesla. The likely explanation is memory bandwidth: the Tesla's 177.4 GB/s versus the MX350's 56.06 GB/s is a 3.2x advantage, and its 384-bit bus versus 64-bit bus allows far more data to flow through compute-heavy workloads.

The Verdict

The data points to a clear conclusion: the NVIDIA Tesla M2090 is the stronger performer in the shared benchmark, and by a wide margin. Its 50.5% lead in Geekbench OpenCL is not a marginal difference; it is a dominant one. The Tesla M2090's score of 13075 places it among much newer GPUs like the GTX 1660 SUPER and RX 580, while the MX350's 8689 score places it alongside entry-level mobile parts like the Radeon Pro 450 and Quadro K2200.

For users who prioritize raw compute performance in OpenCL workloads, the Tesla M2090 is the obvious choice. Its 6 GB of GDDR5 memory, 384-bit bus, and 177.4 GB/s bandwidth give it a structural advantage that the MX350 cannot overcome. The Tesla also supports DirectX 12 (11_0) and OpenGL 4.6, matching the MX350's OpenGL support, though the MX350 adds Vulkan 1.4 support and a higher DirectX feature level (12_1).

For users who need a low-power, portable solution, the MX350 is the only realistic option. Its 20 W TDP requires no power connectors, while the Tesla M2090 demands a 250 W TDP, a 1x 6-pin plus 1x 8-pin power connector setup, and a 600 W suggested PSU. The Tesla has no display outputs at all, making it unsuitable for any system that needs to drive a monitor directly. The MX350, by contrast, is designed for portable devices.

The verdict from the database is straightforward: if compute performance is the sole criterion, the Tesla M2090 wins outright. If portability, power efficiency, and modern API support matter more, the MX350 is the practical choice, but it cannot match the Tesla in the one benchmark where both are measured.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The NVIDIA Tesla M2090 scores 13075, while the NVIDIA GeForce MX350 scores 8689. The Tesla M2090 leads by 50.5%.

Q: How does the Tesla M2090 compare to its nearest rivals?

A: The Tesla M2090 is 0.7% ahead of the GeForce GTX 1660 SUPER (12986), 1% ahead of the GeForce RTX 3050 Ti Mobile (12940), and 1.1% ahead of the AMD Radeon RX 580 (12928). It trails the GeForce GTX 950 (13189) by 0.9%.

Q: How does the MX350 compare to its nearest rivals?

A: The MX350 is 0.7% ahead of the AMD Radeon Pro 450 (10804) and 1.1% ahead of the NVIDIA Quadro K2200 (10761). It trails the GeForce GTX 1650 SUPER (11047) by 1.5% and the AMD Radeon RX 550 (11075) by 1.7%.

Q: Does the MX350 have any benchmark advantage over the Tesla M2090?

A: The MX350 has a recorded Geekbench Vulkan score of 13077, which is close to the Tesla M2090's OpenCL score of 13075, but the Tesla has no recorded Vulkan score, so no direct comparison exists. In the only shared benchmark, OpenCL, the Tesla wins by 50.5%.

Q: What are the memory specifications of each GPU?

A: The Tesla M2090 has 6 GB of GDDR5 memory on a 384-bit bus with 177.4 GB/s bandwidth. The MX350 has 2 GB of GDDR5 memory on a 64-bit bus with 56.06 GB/s bandwidth.

Q: What are the power requirements for each GPU?

A: The Tesla M2090 has a TDP of 250 W, requires a 1x 6-pin plus 1x 8-pin power connector, and a suggested PSU of 600 W. The MX350 has a TDP of 20 W and requires no power connectors.

Architecture Differences

The Tesla M2090 and GeForce MX350 differ fundamentally in architecture, process technology, and intended use case. The Tesla M2090 is built on the Fermi 2.0 architecture with the GF110 chip, manufactured on a 40 nm process at TSMC. It contains 3,000 million transistors on a 520 mm² die, yielding a transistor density of 5.8M per mm². The MX350 uses the Pascal architecture with the GP107S chip, manufactured on a 14 nm process at Samsung. It contains 3,300 million transistors on a 132 mm² die, yielding a transistor density of 25.0M per mm². The MX350's design is over four times denser, reflecting a much more modern manufacturing process.

The compute resources differ in configuration. The Tesla M2090 has 512 shading units, 64 texture mapping units, and 48 render output units. The MX350 has 640 shading units, 32 texture mapping units, and 16 render output units. Despite having more shading units, the MX350 delivers lower FP32 throughput: 1.879 TFLOPS versus the Tesla's 1,332.2 GFLOPS. The MX350 also has a listed FP16 performance of 29.36 GFLOPS (1:64), while the Tesla has no FP16 listing. Pixel and texture rates favor the MX350 slightly: 23.49 GPixel/s and 46.98 GTexel/s versus the Tesla's 20.83 GPixel/s and 41.66 GTexel/s.

Memory architecture is a major differentiator. The Tesla M2090 uses 6 GB of GDDR5 on a 384-bit bus with 177.4 GB/s bandwidth and a memory clock of 924 MHz (3.7 Gbps effective). The MX350 uses 2 GB of GDDR5 on a 64-bit bus with 56.06 GB/s bandwidth and a memory clock of 1752 MHz (7 Gbps effective). The Tesla's memory bandwidth is 3.2x higher, which explains its dominance in the OpenCL benchmark despite lower shading unit count.

Power and physical specifications diverge sharply. The Tesla M2090 has a 250 W TDP, is dual-slot, measures 248 mm (9.8 inches) in length, and requires a 1x 6-pin plus 1x 8-pin power connector with a 600 W suggested PSU. The MX350 has a 20 W TDP, needs no power connectors, and has no listed dimensions. The Tesla uses a PCIe 2.0 x16 interface, while the MX350 uses PCIe 3.0 x4. The Tesla has no display outputs, while the MX350's outputs are listed as "Portable Device Dependent."

API support shows the generational gap. The Tesla supports DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan support listed. The MX350 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The production status for both is end-of-life, with the Tesla released in 2011 as part of the Tesla Fermi generation and the MX350 released in 2020 as part of the GeForce MX (3xx) generation.