AMD Radeon Pro WX 5100 vs NVIDIA GeForce GTX 960M Comparison

# NVIDIA GeForce GTX 960M vs AMD Radeon Pro WX 5100

The NVIDIA GeForce GTX 960M and AMD Radeon Pro WX 5100 represent two very different approaches to mobile and workstation graphics, separated by roughly a year and a half of GPU evolution. The GTX 960M is a Maxwell-based mobile part from NVIDIA's GeForce 900M generation, targeting thin-and-light gaming notebooks, while the Radeon Pro WX 5100 is a Polaris-based workstation card from AMD's Radeon Pro lineup, designed for professional visualization workloads. Benchmark data shows the WX 5100 winning both head-to-head tests decisively, but the two cards occupy distinct performance tiers and use-case niches that go far beyond raw compute numbers.

Where Each One Wins

The AMD Radeon Pro WX 5100 wins every benchmark where the two cards are directly compared, and it wins by substantial margins. In Geekbench OpenCL, the WX 5100 scores 24,217 versus the GTX 960M's 11,045, a difference of 54.4 percent in favor of AMD. The Vulkan gap is even wider: 26,909 for the WX 5100 against 8,245 for the GTX 960M, a 69.4 percent deficit for NVIDIA. These are not close contests; the WX 5100 more than triples the GTX 960M's Vulkan output and more than doubles its OpenCL score.

The GTX 960M's strengths are not visible in the head-to-head data, but its architecture and positioning suggest where it fits. As a mobile GPU with a 75 W TDP in an MXM module, it was designed for gaming laptops where power efficiency and thermal headroom are critical. Its Maxwell architecture supports DirectX 12 (11_0) and Vulkan 1.4, making it competent for its era's game titles. The WX 5100, by contrast, is a single-slot workstation card with four DisplayPort 1.4a outputs, suggesting multi-monitor professional setups rather than portable gaming.

The WX 5100's additional benchmark suite—including PassMark DirectX 9, 10, 11, and 12 tests, plus G2D, G3D, and GPU compute tests—shows where workstation cards typically excel. Its PassMark G3D score of 5,496 and GPU compute score of 2,053 indicate strong general 3D rendering and compute workloads, while its DirectX 9 score of 88 outperforms its DirectX 11 score of 36 and DirectX 12 score of 26, a pattern common in professional cards optimized for legacy CAD and OpenGL applications.

Architecture Differences

The architectural gap between these two GPUs is generational and fundamental. The GTX 960M uses NVIDIA's GM107 chip built on a 28 nm process at TSMC, packing 1,870 million transistors into a 148 mm² die. The WX 5100 uses AMD's Ellesmere chip fabricated on a 14 nm process at GlobalFoundries, containing 5,700 million transistors across a 232 mm² die. The transistor density tells the story: 12.6 million transistors per square millimeter for Maxwell versus 24.6 million for GCN 4.0, a near-doubling of density thanks to the smaller node.

The compute resources diverge sharply. The GTX 960M has 640 shading units, 40 texture mapping units, and 16 ROPs. The WX 5100 more than doubles the shading units to 1,792, nearly triples TMUs to 112, and doubles ROPs to 32. This translates directly into theoretical throughput: the GTX 960M delivers 1.505 TFLOPS of FP32 compute, while the WX 5100 achieves 3.892 TFLOPS—and uniquely, the WX 5100 also offers FP16 at the same 3.892 TFLOPS (1:1 ratio), a feature the GTX 960M lacks entirely.

Memory architecture differs as much as compute. The GTX 960M uses 4 GB of GDDR5 on a 128-bit bus, yielding 80.19 GB/s of bandwidth. The WX 5100 doubles both capacity and bus width—8 GB on a 256-bit interface—for exactly double the bandwidth at 160.0 GB/s. Clock speeds tell a different story: the GTX 960M runs higher at 1,097 MHz base and 1,176 MHz boost, while the WX 5100's base clock is just 713 MHz but boosts to 1,086 MHz. The WX 5100's lower base clock is offset by far more execution units and wider memory.

Both cards share a 75 W TDP and require no power connectors, but the WX 5100 suggests a 250 W power supply while the GTX 960M lists none. The WX 5100 also supports newer API features: DirectX 12 (12_0) versus the GTX 960M's DirectX 12 (11_0), and Vulkan 1.3 versus 1.4—though the NVIDIA card actually lists a newer Vulkan version.

Head-to-Head Benchmarks

The two head-to-head benchmarks available are both Geekbench tests, and both show overwhelming AMD victories. In Geekbench OpenCL, the WX 5100's 24,217 score dwarfs the GTX 960M's 11,045. That is a 54.4 percent delta in favor of AMD, meaning the WX 5100 delivers roughly 2.2 times the OpenCL compute performance. For context, the GTX 960M's average benchmark score of 9,645 places it near the NVIDIA Quadro K5000 (9,637, just 0.1 percent behind) and AMD Radeon Pro WX 2100 (9,653, 0.1 percent ahead). Its 46th percentile ranking among all GPUs puts it in the lower-middle range.

The Vulkan test is even more lopsided. The WX 5100 scores 26,909 versus 8,245 for the GTX 960M, a delta of 69.4 percent. The WX 5100 more than triples the GTX 960M's Vulkan performance. This suggests the WX 5100's GCN 4.0 architecture handles modern low-level graphics APIs far more efficiently than Maxwell, which was designed before Vulkan's release and treats it as an afterthought via driver translation.

The WX 5100's average benchmark score of 8,863 actually sits slightly below the GTX 960M's 9,645, which seems paradoxical given its head-to-head dominance. That is because the WX 5100's benchmark pool includes many PassMark tests with low scores—DirectX 9 at 88, DirectX 10 at 29, DirectX 11 at 36, DirectX 12 at 26—which drag down its average. Its percentile ranking of 44 is also slightly below the GTX 960M's 46. The WX 5100's nearest rivals include the AMD Radeon R9 M265X (8,851, 0.1 percent behind) and NVIDIA GeForce RTX 3050 A Mobile (8,746, 1.3 percent ahead), positioning it in a similar performance tier despite its workstation pedigree.

The GTX 960M's nearest rivals cluster tightly around its average score: the Quadro K5000 at 9,637, the Radeon Pro WX 2100 at 9,653, and the Quadro P4000 at 9,665. All are within 0.2 percent of the GTX 960M, indicating a crowded performance band. The WX 5100's rivals span a wider range, from the R9 M265X at 0.1 percent behind to the GeForce GTX 460 v2 at 1.4 percent behind.

FAQ

Q: Which GPU has higher raw compute performance?

A: The AMD Radeon Pro WX 5100 delivers 3.892 TFLOPS of FP32 compute, more than double the GTX 960M's 1.505 TFLOPS. It also offers FP16 at the same 3.892 TFLOPS rate, which the GTX 960M does not support.

Q: How do the memory configurations compare?

A: The WX 5100 has 8 GB of GDDR5 on a 256-bit bus with 160.0 GB/s bandwidth. The GTX 960M has 4 GB of GDDR5 on a 128-bit bus with 80.19 GB/s bandwidth. The WX 5100 doubles capacity, bus width, and bandwidth.

Q: Which card performs better in Vulkan workloads?

A: The WX 5100 scores 26,909 in Geekbench Vulkan versus 8,245 for the GTX 960M, a 69.4 percent advantage for AMD. The GTX 960M lists Vulkan 1.4 support while the WX 5100 lists Vulkan 1.3.

Q: Are both cards suitable for the same power envelope?

A: Yes, both have a 75 W TDP and require no power connectors. However, the WX 5100 suggests a 250 W power supply, while the GTX 960M lists no suggested PSU.

Q: What are the physical form factor differences?

A: The GTX 960M is an MXM module using an MXM-B (3.0) interface, while the WX 5100 is a single-slot card using PCIe 3.0 x16. The WX 5100 measures 173 mm in length and 112 mm in height.

Q: Which card has better DirectX feature support?

A: The WX 5100 supports DirectX 12 (12_0), while the GTX 960M supports DirectX 12 (11_0). This indicates the WX 5100 supports additional DX12 features like bindless resources and enhanced barrier semantics.

Specification Differences

| Specification | NVIDIA GeForce GTX 960M | AMD Radeon Pro WX 5100 |

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

| Architecture | Maxwell | GCN 4.0 |

| Process node | 28 nm | 14 nm |

| Foundry | TSMC | GlobalFoundries |

| Transistors | 1,870 million | 5,700 million |

| Die size | 148 mm² | 232 mm² |

| Transistor density | 12.6M / mm² | 24.6M / mm² |

| Base clock | 1097 MHz | 713 MHz |

| Boost clock | 1176 MHz | 1086 MHz |

| Memory size | 4 GB | 8 GB |

| Memory bus width | 128 bit | 256 bit |

| Memory bandwidth | 80.19 GB/s | 160.0 GB/s |

| Shading units | 640 | 1792 |

| TMUs | 40 | 112 |

| ROPs | 16 | 32 |

| Pixel rate | 18.82 GPixel/s | 34.75 GPixel/s |

| Texture rate | 47.04 GTexel/s | 121.6 GTexel/s |

| FP32 compute | 1.505 TFLOPS | 3.892 TFLOPS |

| FP16 compute | None | 3.892 TFLOPS (1:1) |

| Slot width | MXM Module | Single-slot |

| Bus interface | MXM-B (3.0) | PCIe 3.0 x16 |

| Display outputs | Portable Device Dependent | 4x DisplayPort 1.4a |

| DirectX version | 12 (11_0) | 12 (12_0) |

| Vulkan version | 1.4 | 1.3 |

| Suggested PSU | None | 250 W |

| Dimensions | Not specified | 173 mm × 112 mm |

| Launch MSRP | None | 499 USD |

The specification table reveals a GPU that is larger in every dimension of silicon complexity—more transistors, bigger die, higher density—and a card that doubles memory capacity and bandwidth. The GTX 960M counters with higher clock speeds and a more compact mobile form factor, but the WX 5100's architectural advantages in shading units, texture units, and ROPs overwhelm any clock-speed deficit. The release dates reinforce the generational gap: the GTX 960M arrived in March 2015, while the WX 5100 launched in November 2016, both now end-of-life products in their respective lineups.