NVIDIA GeForce GTX 970M vs NVIDIA GeForce GTX 980M Comparison

The NVIDIA GeForce GTX 980M and GTX 970M are both Maxwell 2.0 era mobile GPUs from the same generation and family, yet the benchmark data shows a clear and consistent performance hierarchy. The GTX 980M wins 9 out of 10 head-to-head comparisons, establishing itself as the substantially faster part, while the GTX 970M manages a single narrow victory in one specific API test.

Head-to-Head Benchmarks

The most decisive victory for the GTX 980M comes in the 3DMark Steel Nomad DX12 test, where it scores 649 against the GTX 970M's 472, a delta of 37.5%. This is the largest performance gap recorded between the two, indicating a significant advantage in modern DirectX 12 workloads. In DirectX 11, the GTX 980M scores 57 versus 42, a 35.7% lead, showing that the gap persists across both the older and newer graphics APIs.

The pattern continues across legacy DirectX tests. In DirectX 9, the GTX 980M scores 125 against 99, a 26.3% advantage. DirectX 10 shows a 25% lead (35 vs. 28), and DirectX 12 shows a 29.2% lead (31 vs. 24). These results demonstrate that the GTX 980M is consistently faster regardless of the API generation, from the older DirectX 9 titles to the modern DirectX 12 workloads.

Compute performance also favors the GTX 980M. In Geekbench OpenCL, it scores 23832 against 18946, a 25.8% advantage. Passmark GPU Compute shows a 23% lead, with scores of 2816 versus 2289. The GTX 980M also wins the synthetic gaming benchmarks decisively: Passmark G3D shows a 28.6% lead (7338 vs. 5704), and Passmark G2D shows the same 28.6% margin (490 vs. 381).

The single win for the GTX 970M comes in Geekbench Vulkan, where it scores 18292 against the GTX 980M's 17703, a 3.2% advantage. This is the only test where the GTX 970M outperforms its sibling, and the margin is relatively small compared to the GTX 980M's leads elsewhere. The Vulkan result is an outlier against the broader trend, suggesting that the GTX 970M benefits from an optimization or workload characteristic specific to that API, but it does not offset the overall performance deficit.

The Verdict

The data is unambiguous: the GTX 980M is the superior GPU in nearly every measurable way. Its average benchmark score of 5308 places it ahead of the GTX 970M's 4628, and its percentile rank of 31 versus 27 confirms that it sits higher in the overall GPU distribution. The GTX 980M wins 9 of 10 comparisons, with deltas ranging from 23% to 37.5% in its favor.

For users prioritizing raw performance, especially in DirectX 12 and compute-heavy workloads, the GTX 980M is the clear choice. The 37.5% lead in 3DMark Steel Nomad DX12 and 35.7% lead in DirectX 11 make it the better option for modern gaming and demanding applications. The GTX 970M's only advantage is in Vulkan, where it is 3.2% faster, but this single narrow win does not compensate for its substantial losses elsewhere.

The GTX 970M should be considered only when the Vulkan API is the primary workload, or when the reduced specifications are acceptable for less demanding tasks. For all general-purpose gaming, compute, and synthetic benchmark scenarios, the GTX 980M delivers measurably higher performance.

FAQ

Q: Which GPU wins the most benchmark comparisons?

A: The NVIDIA GeForce GTX 980M wins 9 out of 10 head-to-head benchmarks against the GTX 970M, with the GTX 970M winning only the Geekbench Vulkan test.

Q: What is the largest performance difference between the two?

A: The largest difference is in the 3DMark Steel Nomad DX12 test, where the GTX 980M scores 649 versus 472, a 37.5% advantage.

Q: Is the GTX 970M faster in any test?

A: Yes, the GTX 970M scores 18292 in Geekbench Vulkan compared to the GTX 980M's 17703, a 3.2% lead.

Q: How do their average benchmark scores compare?

A: The GTX 980M has an average benchmark score of 5308, while the GTX 970M averages 4628.

Q: What do the percentile rankings indicate?

A: The GTX 980M ranks in the 31st percentile of all GPUs, while the GTX 970M ranks in the 27th percentile, showing the GTX 980M performs better relative to the broader GPU population.

Q: Which GPU is better for compute workloads?

A: The GTX 980M leads in both Geekbench OpenCL (23832 vs. 18946, 25.8% faster) and Passmark GPU Compute (2816 vs. 2289, 23% faster).

Specification Differences

The two GPUs share the same chip (GM204), architecture (Maxwell 2.0), manufacturing process (28 nm), and foundry (TSMC), but their internal configurations differ significantly. The GTX 980M has 1536 shading units, 96 texture mapping units, and 64 raster output units, while the GTX 970M is configured with 1280 shading units, 80 TMUs, and 48 ROPs. This translates to higher pixel and texture rates for the GTX 980M: 72.13 GPixel/s versus 49.82 GPixel/s, and 108.2 GTexel/s versus 83.04 GTexel/s. The FP32 compute throughput is also higher on the GTX 980M at 3.462 TFLOPS compared to 2.657 TFLOPS.

Clock speeds differ as well. The GTX 980M runs at a base clock of 1038 MHz with a boost of 1127 MHz, while the GTX 970M operates at 924 MHz base and 1038 MHz boost. The memory subsystem is another differentiator: the GTX 980M features 8 GB of GDDR5 on a 256-bit bus with 160.4 GB/s bandwidth, while the GTX 970M has 6 GB of GDDR5 on a 192-bit bus with 120.3 GB/s bandwidth. Both use the same memory clock of 1253 MHz (5 Gbps effective).

Architecture Differences

Both GPUs are built on the Maxwell 2.0 architecture using the GM204 chip, and they share identical fundamental characteristics: 5,200 million transistors, a 398 mm² die size, and a transistor density of 13.1M per mm². They are both manufactured on TSMC's 28 nm process and offer the same API support: DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Neither GPU has dedicated ray tracing or tensor cores, as these are not part of the Maxwell 2.0 design.

The architectural differences are purely in the number of active execution units. The GTX 980M uses a fuller configuration of the GM204 die, with 1536 shading units, 96 TMUs, and 64 ROPs, whereas the GTX 970M is a cut-down variant with 1280 shading units, 80 TMUs, and 48 ROPs. This means the GTX 980M has 20% more shading units, 20% more TMUs, and 33% more ROPs than the GTX 970M. The extra ROPs contribute directly to the higher pixel fill rate, and the additional shading units and TMUs drive the superior texture and compute throughput. Both GPUs use the same MXM-B (3.0) bus interface and are packaged as MXM modules with no power connectors, with display outputs dependent on the portable device.

Where Each One Wins

The GTX 980M is the dominant choice for virtually every workload measured. Its 37.5% lead in 3DMark Steel Nomad DX12 makes it the better option for modern DirectX 12 games and future-proofing. The 35.7% advantage in DirectX 11 covers the vast majority of current PC game libraries. In compute applications, the 25.8% lead in Geekbench OpenCL and 23% lead in Passmark GPU Compute make it the stronger choice for GPGPU tasks, rendering, and other compute-heavy workloads. The 28.6% margins in both Passmark G3D and G2D reinforce its superiority in general gaming and 2D acceleration.

The GTX 970M wins only in Geekbench Vulkan, where its 18292 score edges out the GTX 980M's 17703 by 3.2%. This makes it the slightly better option for applications that rely heavily on the Vulkan API, though the margin is small. For users with a specific Vulkan-based workload, the GTX 970M offers a marginal advantage, but this single data point does not outweigh the GTX 980M's comprehensive wins across DirectX 9, 10, 11, 12, OpenCL, and both Passmark graphics tests. The GTX 980M is the recommended GPU for all general-purpose use, with the GTX 970M reserved for niche Vulkan-centric scenarios.