NVIDIA GeForce GTX 870M vs NVIDIA GeForce MX350 Comparison

The NVIDIA GeForce MX350 and NVIDIA GeForce GTX 870M are both end-of-life mobile graphics solutions, but they represent vastly different technical eras. The data shows a clear overall performance victor: the older GTX 870M holds a commanding lead in raw compute benchmarks, despite the MX350 being a much newer and more power-efficient design. In the sole head-to-head benchmark available, the GeForce GTX 870M outperforms the MX350 by 31.2% in Geekbench OpenCL, scoring 12630 versus 8689. This result places the GTX 870M in a higher performance tier, with an average benchmark score of 9959 compared to the MX350's 10883, a reversal that highlights how the newer card's score is buoyed by its Vulkan result.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and the results are decisive. The NVIDIA GeForce GTX 870M scores 12630, while the NVIDIA GeForce MX350 scores 8689. This represents a 31.2% advantage for the GTX 870M, a substantial margin that places it in a different performance class for compute workloads. This is a significant win for the older chip, as it demonstrates that its larger memory bus and higher raw shading power overcome its older architecture.

In the broader benchmark context, the MX350's average score of 10883 is higher than the GTX 870M's 9959. This is primarily due to the MX350's strong showing in the Geekbench Vulkan test, where it scores 13077. The GTX 870M has no comparable Vulkan result in the data, and its only other benchmark is a Geekbench Metal score of 7288, which is not directly comparable to the OpenCL results. The data suggests that in modern, low-level graphics APIs, the newer Pascal-based MX350 can achieve higher scores, but in the more universal OpenCL compute test, the older Kepler card's sheer hardware resources win out.

When placed against their respective peer groups, the performance picture becomes more nuanced. The MX350's average score of 10883 places it 0.7% ahead of the AMD Radeon Pro 450 and 1.1% ahead of the NVIDIA Quadro K2200, but it trails the NVIDIA GeForce GTX 1650 SUPER by 1.5% and the AMD Radeon RX 550 by 1.7%. The GTX 870M, with its average of 9959, sits 0.5% behind the AMD Radeon Pro 5300M and 0.8% behind the NVIDIA Quadro K5100M, while leading the AMD Radeon R9 M375 by 1.1% and the NVIDIA Quadro 6000 by 1.1%. This shows that while the GTX 870M loses the direct comparison to the MX350 in average score, it is still a competitive part in its own generation.

Architecture Differences

The architectural gap between these two GPUs is the primary driver of their performance and feature differences. The MX350 is built on the Pascal architecture using the GP107S chip, fabricated on a 14 nm process at Samsung. In contrast, the GTX 870M uses the older Kepler architecture with the GK104 chip, built on a 28 nm process at TSMC. This process advantage gives the MX350 a transistor density of 25.0M / mm², more than double the GTX 870M's 12.0M / mm², even though the GTX 870M has more total transistors (3,540 million versus 3,300 million) and a much larger die size (294 mm² versus 132 mm²).

The compute configurations diverge significantly. The GTX 870M packs 1344 shading units, 112 texture mapping units (TMUs), and 24 raster output units (ROPs), while the MX350 is equipped with only 640 shading units, 32 TMUs, and 16 ROPs. This hardware disparity explains the GTX 870M's higher peak fill rates: 27.08 GPixel/s pixel rate and 108.3 GTexel/s texture rate, compared to the MX350's 23.49 GPixel/s and 46.98 GTexel/s. The GTX 870M also holds a significant edge in FP32 compute, delivering 2.599 TFLOPS versus the MX350's 1.879 TFLOPS. The MX350 does have a dedicated FP16 rate of 29.36 GFLOPS, while the GTX 870M has no listed FP16 capability.

Memory architecture is another major differentiator. The GTX 870M uses a 192-bit memory bus with 3 GB of GDDR5 memory, providing a bandwidth of 120.0 GB/s. The MX350 is limited to a 64-bit bus and 2 GB of GDDR5, resulting in a bandwidth of just 56.06 GB/s. This 2.1x bandwidth advantage for the GTX 870M is critical for memory-intensive workloads. The MX350 does operate at a higher memory clock (1752 MHz versus 1250 MHz), but the narrower bus severely limits its overall throughput.

Where Each One Wins

Based on the benchmark and specification data, each GPU has distinct use cases where it excels. The NVIDIA GeForce GTX 870M is the clear winner for compute-heavy applications and traditional gaming scenarios. Its 31.2% lead in OpenCL performance, combined with its larger 3 GB frame buffer and much higher memory bandwidth, makes it the superior choice for tasks that rely on raw shading power, high-resolution textures, and heavy data throughput. Its higher pixel and texture rates also suggest it can handle higher resolutions and more complex geometry without bottlenecking.

The NVIDIA GeForce MX350 wins in the field of modern API compatibility and efficiency. It supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 870M is limited to DirectX 12 (11_0) and Vulkan 1.2.175. The MX350's strong Vulkan score of 13077 indicates it is well-optimized for modern game engines that leverage low-level APIs, potentially offering smoother performance in newer titles despite its lower raw compute. Its 20 W TDP versus the GTX 870M's 100 W TDP also makes it vastly more suitable for slim, lightweight laptops where battery life and thermal management are paramount, whereas the GTX 870M is an MXM module designed for larger, gaming-focused machines.

FAQ

Q: Which GPU is faster in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 870M is significantly faster, scoring 12630 compared to the MX350's 8689, a 31.2% advantage.

Q: Does the newer MX350 have a higher average benchmark score?

A: Yes, the MX350 has an average benchmark score of 10883, while the GTX 870M's average is 9959. This is due to the MX350's strong Vulkan performance.

Q: What are the key memory specification differences?

A: The GTX 870M has 3 GB of GDDR5 on a 192-bit bus with 120.0 GB/s bandwidth. The MX350 has 2 GB on a 64-bit bus with 56.06 GB/s bandwidth.

Q: How do their power requirements compare?

A: The MX350 has a TDP of 20 W, while the GTX 870M has a TDP of 100 W, making the MX350 far more power-efficient.

Q: Which GPU supports a newer version of DirectX?

A: The MX350 supports DirectX 12 (12_1), while the GTX 870M supports DirectX 12 (11_0).

Q: What is the process node difference between the two chips?

A: The MX350 is built on a 14 nm process, while the GTX 870M uses a 28 nm process, giving the MX350 a significant manufacturing advantage.

Specification Differences

• Chip: GP107S (MX350) versus GK104 (GTX 870M)
• Architecture: Pascal (MX350) versus Kepler (GTX 870M)
• Process Node: 14 nm (MX350) versus 28 nm (GTX 870M)
• Foundry: Samsung (MX350) versus TSMC (GTX 870M)
• Transistors: 3,300 million (MX350) versus 3,540 million (GTX 870M)
• Die Size: 132 mm² (MX350) versus 294 mm² (GTX 870M)
• Base Clock: 1354 MHz (MX350) versus 941 MHz (GTX 870M)
• Boost Clock: 1468 MHz (MX350) versus 967 MHz (GTX 870M)
• Memory Size: 2 GB (MX350) versus 3 GB (GTX 870M)
• Memory Bus Width: 64 bit (MX350) versus 192 bit (GTX 870M)
• Memory Bandwidth: 56.06 GB/s (MX350) versus 120.0 GB/s (GTX 870M)
• Shading Units: 640 (MX350) versus 1344 (GTX 870M)
• TMUs: 32 (MX350) versus 112 (GTX 870M)
• ROPs: 16 (MX350) versus 24 (GTX 870M)
• Pixel Rate: 23.49 GPixel/s (MX350) versus 27.08 GPixel/s (GTX 870M)
• Texture Rate: 46.98 GTexel/s (MX350) versus 108.3 GTexel/s (GTX 870M)
• FP32 Performance: 1.879 TFLOPS (MX350) versus 2.599 TFLOPS (GTX 870M)
• TDP: 20 W (MX350) versus 100 W (GTX 870M)
• Bus Interface: PCIe 3.0 x4 (MX350) versus MXM-B (3.0) (GTX 870M)
• Vulkan Version: 1.4 (MX350) versus 1.2.175 (GTX 870M)
• DirectX Version: 12 (12_1) (MX350) versus 12 (11_0) (GTX 870M)
• Release Date: 2020-02-09 (MX350) versus 2014-03-11 (GTX 870M)

The Verdict

The data leads to a clear conclusion: the NVIDIA GeForce GTX 870M is the superior choice for raw performance and demanding graphical workloads. Its 31.2% lead in OpenCL, larger memory pool, and substantially higher bandwidth make it the better option for gaming and compute-heavy tasks, provided the system can accommodate its 100 W TDP. Benchmark results indicate that the older Kepler architecture still delivers more brute-force capability.

The NVIDIA GeForce MX350 is the better choice for users prioritizing modern software support and efficiency. Its 20 W TDP, support for DirectX 12_1 and Vulkan 1.4, and higher average benchmark score driven by Vulkan performance make it ideal for thin-and-light laptops focused on battery life and contemporary game engines. For users who need maximum frame rates in a traditional gaming laptop, the GTX 870M is the clear winner from this data. For those who value portability and future-proofing through API support, the MX350 is the more sensible option.