NVIDIA GeForce 940M vs NVIDIA GeForce GTX 980M Comparison

NVIDIA’s GeForce GTX 980M and GeForce 940M are both Maxwell-generation mobile parts from the GeForce 900M lineup, but the benchmark data separates them into entirely different performance tiers. The GTX 980M holds a decisive lead in every shared test, while the 940M trails by wide margins. Both cards sit at the 31st percentile of all GPUs, yet their average benchmark scores—5308 for the GTX 980M and 5284 for the 940M—are nearly identical despite the GTX 980M’s far larger silicon. This discrepancy highlights how average scores can obscure massive differences in specific workloads.

Where Each One Wins

The GTX 980M wins every head-to-head benchmark recorded. In Geekbench OpenCL, it scores 23832 against the 940M’s 6018, a 296% advantage. In Geekbench Vulkan, the GTX 980M posts 17703 versus 4549, a 289.2% lead. There are no benchmark categories where the 940M comes out ahead—its two listed scores are both far lower than the GTX 980M’s corresponding results.

The 940M’s only statistical claim is its average benchmark score of 5284, which actually sits slightly above the GTX 980M’s 5308 average when rounded—a difference of 0.5% in the 940M’s favor according to the GTX 980M’s nearest rivals list. This is a quirk of averaging across different test sets, not a real performance win. The GTX 980M has ten benchmark entries spanning DirectX 9 through 12, 2D, 3D, and compute workloads, while the 940M only has two Geekbench entries. For any task that stresses raw GPU throughput—compute, Vulkan rendering, or high-resolution textures—the GTX 980M is the clear choice.

Architecture Differences

Both chips are built on TSMC’s 28 nm process, but the silicon underneath differs substantially. The GTX 980M uses the GM204 chip with 5,200 million transistors on a 398 mm² die, yielding a transistor density of 13.1M per mm². The 940M uses the smaller GM107 chip with 1,870 million transistors on a 148 mm² die, at 12.6M per mm². The GTX 980M’s die is nearly 2.7 times larger in area and holds roughly 2.8 times more transistors.

The GTX 980M’s architecture is Maxwell 2.0, while the 940M is simply labeled Maxwell. This distinction matters for feature support: the GTX 980M supports DirectX 12 (12_1), whereas the 940M only reaches DirectX 12 (11_0). Both cards support OpenGL 4.6 and Vulkan 1.4, so API compatibility is otherwise equal.

Memory configurations diverge completely. The GTX 980M carries 8 GB of GDDR5 on a 256-bit bus, delivering 160.4 GB/s of bandwidth. The 940M has 2 GB of DDR3 on a 64-bit bus, with just 14.40 GB/s. That’s an 11-fold difference in memory bandwidth, which directly impacts texture-heavy scenes and high-resolution rendering. The GTX 980M’s memory clock runs at 1253 MHz (5 Gbps effective), while the 940M’s memory operates at 900 MHz (1800 Mbps effective).

Compute resources are similarly lopsided. The GTX 980M has 1536 shading units, 96 TMUs, and 64 ROPs. The 940M has 512 shading units, 32 TMUs, and 16 ROPs. That’s exactly three times the shaders and TMUs, and four times the ROPs on the GTX 980M. Pixel rate on the GTX 980M is 72.13 GPixel/s versus 17.57 GPixel/s on the 940M; texture rate is 108.2 GTexel/s versus 35.14 GTexel/s. FP32 throughput is 3.462 TFLOPS for the GTX 980M and 1,124.4 GFLOPS for the 940M—a 3.08x gap.

Head-to-Head Benchmarks

The two shared benchmarks tell a consistent story of overwhelming GTX 980M dominance. In Geekbench OpenCL, the GTX 980M scores 23832, which is 296% higher than the 940M’s 6018. This test stresses general-purpose compute, and the GTX 980M’s triple shader count and wider memory bus allow it to process far more data per cycle. The 940M’s smaller GM107 chip, with its 64-bit DDR3 interface, becomes a bottleneck for any memory-intensive compute workload.

Geekbench Vulkan shows a similar pattern: 17703 for the GTX 980M versus 4549 for the 940M, a 289.2% difference. Vulkan’s low-overhead API exposes raw hardware capabilities, and the GTX 980M’s Maxwell 2.0 architecture with DirectX 12_1 support handles modern draw calls more efficiently. The 940M’s DirectX 11_0-level feature set limits its Vulkan performance, even though the API version is the same.

The GTX 980M’s other benchmark scores, while not directly compared to the 940M, reinforce its position. It hits 7338 in Passmark G3D, 2816 in Passmark GPU Compute, and 125 in Passmark DirectX 9. The 940M has no entries for these tests, so no direct percentage comparison is possible from the data. But the GTX 980M’s 72.13 GPixel/s pixel rate and 108.2 GTexel/s texture rate dwarf the 940M’s 17.57 GPixel/s and 35.14 GTexel/s, so any DirectX workload would follow the same trend.

FAQ

Q: Which GPU has higher memory bandwidth?

A: The GTX 980M has 160.4 GB/s of bandwidth from 8 GB of GDDR5 on a 256-bit bus. The 940M has 14.40 GB/s from 2 GB of DDR3 on a 64-bit bus.

Q: Do both cards support the same DirectX version?

A: No. The GTX 980M supports DirectX 12 (12_1), while the 940M supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4.

Q: What is the transistor count difference?

A: The GTX 980M’s GM204 chip has 5,200 million transistors, while the 940M’s GM107 has 1,870 million.

Q: Which card has more shading units?

A: The GTX 980M has 1536 shading units. The 940M has 512, exactly one-third as many.

Q: What is the average benchmark score for each card?

A: The GTX 980M averages 5308, and the 940M averages 5284. The 940M’s nearest rivals list shows a 0.5% delta in its favor over the GTX 980M, but this is based on incomplete test coverage.

Q: Which card has a higher pixel rate?

A: The GTX 980M achieves 72.13 GPixel/s, compared to the 940M’s 17.57 GPixel/s.

Specification Differences

| Specification | GTX 980M | 940M |

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

| Chip | GM204 | GM107 |

| Architecture | Maxwell 2.0 | Maxwell |

| Transistors | 5,200 million | 1,870 million |

| Die Size | 398 mm² | 148 mm² |

| Transistor Density | 13.1M / mm² | 12.6M / mm² |

| Base Clock | 1038 MHz | 1020 MHz |

| Boost Clock | 1127 MHz | 1098 MHz |

| Memory Clock | 1253 MHz (5 Gbps effective) | 900 MHz (1800 Mbps effective) |

| Memory Size | 8 GB | 2 GB |

| Memory Type | GDDR5 | DDR3 |

| Memory Bus | 256 bit | 64 bit |

| Memory Bandwidth | 160.4 GB/s | 14.40 GB/s |

| Shading Units | 1536 | 512 |

| TMUs | 96 | 32 |

| ROPs | 64 | 16 |

| Pixel Rate | 72.13 GPixel/s | 17.57 GPixel/s |

| Texture Rate | 108.2 GTexel/s | 35.14 GTexel/s |

| FP32 | 3.462 TFLOPS | 1,124.4 GFLOPS |

| TDP | Not specified | 75 W |

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

| Release Date | 2014-10-06 | 2015-03-12 |

The Verdict

The data leaves no ambiguity: the GTX 980M outperforms the 940M in every measurable way. Its 296% Geekbench OpenCL lead and 289.2% Vulkan lead are not marginal differences—they represent a completely different performance class. The GTX 980M’s 8 GB GDDR5 memory with 160.4 GB/s bandwidth versus the 940M’s 2 GB DDR3 at 14.40 GB/s means the GTX 980M will handle large textures and high resolutions without stuttering, while the 940M will struggle with anything beyond light workloads. The 940M’s only advantage is its lower 75 W TDP, which could suit thinner laptops, but the GTX 980M’s MXM module form factor and lack of power connectors suggest it was designed for larger chassis. For gaming, compute, or any DirectX 12_1 workload, the GTX 980M is the only choice. The 940M is adequate for basic 2D tasks or older DirectX 9 titles, as indicated by its absence from those benchmark lists—but the GTX 980M’s 125 Passmark DirectX 9 score, 57 DirectX 11 score, and 31 DirectX 12 score show it handles legacy APIs with ease too. The 940M’s 31st percentile ranking matches the GTX 980M, but that ranking masks the fact that the GTX 980M’s average is dragged down by its broader test coverage across ten benchmarks, while the 940M only reports two Geekbench scores. When the same test runs on both cards, the GTX 980M wins by nearly 300%.