NVIDIA GeForce GTX 870M vs NVIDIA GeForce MX330 Comparison

Head-to-Head Benchmarks

The recorded database contains one direct head-to-head comparison between these two mobile GPUs, and it is decisively one-sided. In the Geekbench OpenCL test, the NVIDIA GeForce GTX 870M scores 12,630 points against 7,896 for the NVIDIA GeForce MX330. That is a 60% advantage for the older, larger chip. The delta is substantial enough that no amount of driver optimization or thermal headroom on the MX330 side could close the gap in raw compute throughput.

The GTX 870M’s victory in OpenCL is consistent with its broader benchmark profile. Its average benchmark score across all recorded tests is 9,959, while the MX330 averages 8,458. That places the GTX 870M at the 48th percentile of all GPUs in the database, versus the 43rd percentile for the MX330. The five-point percentile gap understates the performance difference because the GTX 870M’s single OpenCL result is far above its average, while the MX330’s OpenCL score is below its average, dragged down by a weaker Vulkan result of 9,019.

Looking at the nearest rivals in the database provides additional context. The GTX 870M sits within 1.1% of the NVIDIA Quadro 6000 (which scores 9,846) and trails the AMD Radeon Pro 5300M by only 0.5% (10,013). It effectively trades blows with professional workstation GPUs from its era. The MX330, by contrast, is statistically tied with the AMD Radeon HD 8870M (8,462, a 0% delta) and the AMD Radeon 880M (8,436, a 0.3% delta). The MX330 also edges out the NVIDIA GeForce GTX 675MX by 0.4% (8,427) but falls 1.2% short of the Intel Arc A380 (8,558). In short, the MX330 competes in a lower performance tier entirely.

The GTX 870M’s 60% lead in OpenCL is not a marginal win; it is a generational chasm in compute throughput. The MX330 does not win a single recorded head-to-head test in the database. The wins tally stands at 1 for the GTX 870M and 0 for the MX330. For any workload that relies on OpenCL acceleration, the GTX 870M is the clear choice, despite being six years older in release date.

The Verdict

The data points to one unambiguous conclusion: the NVIDIA GeForce GTX 870M outperforms the NVIDIA GeForce MX330 in every recorded benchmark comparison. The 60% OpenCL advantage is the only direct head-to-head measurement available, and it is decisive. The GTX 870M also holds a higher average benchmark score (9,959 versus 8,458) and a higher percentile rank (48th versus 43rd).

Who should pick the GTX 870M? Anyone whose workload involves OpenCL compute, which includes many GPU-accelerated applications in video encoding, physics simulation, and general-purpose computing. The GTX 870M also offers more memory (3 GB versus 2 GB) and a wider memory bus (192-bit versus 64-bit), which translates to significantly higher memory bandwidth (120.0 GB/s versus 56.06 GB/s). For texture-heavy workloads, the GTX 870M’s texture rate of 108.3 GTexel/s dwarfs the MX330’s 38.26 GTexel/s.

Who should pick the MX330? The primary argument is not performance but power efficiency. The MX330 has a 10 W TDP, which is one-tenth of the GTX 870M’s 100 W TDP. It is an integrated-class part (IGP slot width) versus the GTX 870M’s MXM Module form factor. For thin-and-light laptops where battery life and thermals matter more than raw compute, the MX330 is the rational choice. It also supports newer API features: DirectX 12 (12_1) versus 12 (11_0), and Vulkan 1.4 versus 1.2.175.

Neither card is current; both are end-of-life. The GTX 870M launched in March 2014, the MX330 in February 2020. The MX330 is the newer part by six years, but the GTX 870M remains the faster one in compute-bound tasks. If the workload fits in the GTX 870M’s 3 GB frame buffer and the system can supply 100 W, the older chip wins. If the priority is a low-power, passively coolable solution for light graphics and display output, the MX330 is the data-backed pick.

Architecture Differences

The two GPUs come from different architectural generations and are built on different process nodes. The GTX 870M uses the GK104 chip based on Kepler architecture, manufactured by TSMC on a 28 nm process. The MX330 uses the GP108B chip based on Pascal architecture, manufactured by Samsung on a 14 nm process. The node shrink explains why the MX330 achieves a much higher transistor density: 24.3 million transistors per square millimeter versus 12.0 million for the GTX 870M. The GTX 870M has more total transistors (3,540 million versus 1,800 million) and a much larger die (294 mm² versus 74 mm²), but the MX330 packs its transistors more tightly.

The GTX 870M has 1,344 shading units, 112 texture mapping units, and 24 raster output units. The MX330 has 384 shading units, 24 TMUs, and 16 ROPs. The GTX 870M’s raw compute advantage comes from having 3.5 times the shading units and 4.7 times the TMUs. The GTX 870M’s pixel rate is 27.08 GPixel/s versus 25.50 GPixel/s for the MX330, a modest 6% lead. Its texture rate advantage is far larger: 108.3 GTexel/s versus 38.26 GTexel/s, a 183% lead.

The FP32 compute figures tell the same story. The GTX 870M delivers 2.599 TFLOPS, while the MX330 delivers 1,224.2 GFLOPS (1.224 TFLOPS). The MX330 does have a recorded FP16 figure of 19.13 GFLOPS with a 1:64 ratio, but the GTX 870M has no FP16 entry in the database, reflecting Kepler’s lack of dedicated half-precision throughput. The MX330’s memory clock is higher (1,752 MHz versus 1,250 MHz, with 7 Gbps effective versus 5 Gbps effective), but its 64-bit bus limits bandwidth to 56.06 GB/s compared to the GTX 870M’s 192-bit bus and 120.0 GB/s.

The MX330 connects via PCIe 3.0 x4, while the GTX 870M uses MXM-B (3.0). The MX330’s slot width is listed as IGP (integrated graphics package), while the GTX 870M is an MXM Module. The GTX 870M has a base clock of 941 MHz and boost of 967 MHz; the MX330 has a base clock of 1,531 MHz and boost of 1,594 MHz. The MX330’s higher clocks partially compensate for its fewer cores, but not enough to overcome the core count deficit.

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The NVIDIA GeForce GTX 870M scores 12,630 in Geekbench OpenCL versus 7,896 for the MX330, a 60% advantage.

Q: Does the MX330 win any benchmark in the database?

A: No. The head-to-head record shows 1 win for the GTX 870M and 0 wins for the MX330.

Q: Which GPU has more memory and bandwidth?

A: The GTX 870M has 3 GB of GDDR5 on a 192-bit bus with 120.0 GB/s bandwidth. The MX330 has 2 GB of GDDR5 on a 64-bit bus with 56.06 GB/s bandwidth.

Q: Which GPU is more power-efficient?

A: The MX330 has a 10 W TDP versus the GTX 870M’s 100 W TDP, making the MX330 far more suitable for low-power systems.

Q: Which GPU supports newer graphics APIs?

A: The MX330 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 870M supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Q: How does the GTX 870M compare to its nearest rivals?

A: The GTX 870M trails the AMD Radeon Pro 5300M by 0.5% and the NVIDIA Quadro K5100M by 0.8%, and it leads the NVIDIA Quadro 6000 by 1.1% in average benchmark score.

Where Each One Wins

The GTX 870M wins in compute-heavy workloads. Its 60% OpenCL lead over the MX330 makes it the better choice for GPU-accelerated applications that leverage OpenCL, including many scientific, engineering, and media tools. Its larger memory pool (3 GB versus 2 GB) and 2.1 times the memory bandwidth (120.0 GB/s versus 56.06 GB/s) give it a clear edge in texture streaming and large dataset handling. The texture rate advantage (108.3 GTexel/s versus 38.26 GTexel/s) means it excels in fragment-heavy rendering. The FP32 throughput of 2.599 TFLOPS versus 1.224 TFLOPS positions it well for general compute tasks.

The MX330 wins in efficiency and portability. Its 10 W TDP is one-tenth of the GTX 870M’s 100 W, making it suitable for fanless or low-noise designs. As an IGP with PCIe 3.0 x4, it integrates into compact laptops without the space and cooling requirements of an MXM module. Its higher base and boost clocks (1,531 MHz and 1,594 MHz versus 941 MHz and 967 MHz) show it can execute per-core instructions quickly, which helps lightly threaded workloads. The MX330 also supports DirectX 12 (12_1) and Vulkan 1.4, which are more recent API versions than the GTX 870M’s DirectX 12 (11_0) and Vulkan 1.2.175. For modern titles that require these newer feature levels, the MX330 has a compatibility advantage.

The GTX 870M is the clear winner for any workload that fits within its power budget and memory capacity. The MX330 is the winner for battery-powered, thermally constrained systems where the GPU is not the primary performance bottleneck.

Specification Differences

| Specification | NVIDIA GeForce GTX 870M | NVIDIA GeForce MX330 |

|---|---|---|

| Architecture | Kepler | Pascal |

| Process Node | 28 nm (TSMC) | 14 nm (Samsung) |

| Transistors | 3,540 million | 1,800 million |

| Die Size | 294 mm² | 74 mm² |

| Transistor Density | 12.0M / mm² | 24.3M / mm² |

| Base Clock | 941 MHz | 1,531 MHz |

| Boost Clock | 967 MHz | 1,594 MHz |

| Memory Clock | 1,250 MHz (5 Gbps effective) | 1,752 MHz (7 Gbps effective) |

| Memory Size | 3 GB | 2 GB |

| Memory Bus | 192 bit | 64 bit |

| Memory Bandwidth | 120.0 GB/s | 56.06 GB/s |

| Shading Units | 1,344 | 384 |

| TMUs | 112 | 24 |

| ROPs | 24 | 16 |

| Pixel Rate | 27.08 GPixel/s | 25.50 GPixel/s |

| Texture Rate | 108.3 GTexel/s | 38.26 GTexel/s |

| FP32 | 2.599 TFLOPS | 1,224.2 GFLOPS |

| FP16 | Not recorded | 19.13 GFLOPS (1:64) |

| TDP | 100 W | 10 W |

| Slot Width | MXM Module | IGP |

| Bus Interface | MXM-B (3.0) | PCIe 3.0 x4 |

| DirectX | 12 (11_0) | 12 (12_1) |

| Vulkan | 1.2.175 | 1.4 |

| Release Date | 2014-03-11 | 2020-02-09 |