AMD Radeon R9 M295X vs NVIDIA GeForce GTX 980 Ti Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R9 M295X holds a slight edge in average benchmark score at 28580, compared to the NVIDIA GeForce GTX 980 Ti's 28020. That is a 2% advantage for the AMD part, and both cards sit within 0.6% of their nearest rivals in the database.

Q: How big is the performance gap in the head-to-head Metal test?

A: The AMD Radeon R9 M295X wins the Geekbench Metal test decisively, scoring 33790 against the NVIDIA GeForce GTX 980 Ti's 19520. That is a 73.1% delta in AMD's favor, the single largest margin in any shared benchmark.

Q: Where does the NVIDIA card outperform AMD?

A: The NVIDIA GeForce GTX 980 Ti wins in both OpenCL and Vulkan. In Geekbench OpenCL, it scores 43513 versus 22858, a 47.5% advantage. In Geekbench Vulkan, it scores 47724 versus 29091, a 39% advantage.

Q: What are the percentile rankings for each GPU?

A: The AMD Radeon R9 M295X sits at the 74th percentile among all GPUs, while the NVIDIA GeForce GTX 980 Ti is at the 73rd percentile. Despite the R9 M295X winning the average score, the two are effectively adjacent in overall standing.

Q: Which card has more memory and a wider bus?

A: The NVIDIA GeForce GTX 980 Ti has 6 GB of GDDR5 on a 384-bit bus, yielding 336.6 GB/s bandwidth. The AMD Radeon R9 M295X has 4 GB of GDDR5 on a 256-bit bus, providing 160.0 GB/s. The NVIDIA card more than doubles the bandwidth.

Q: Do both cards have the same power draw?

A: Yes, both the AMD Radeon R9 M295X and the NVIDIA GeForce GTX 980 Ti are rated at 250 W TDP. However, the NVIDIA card requires a 600 W suggested PSU and uses 1x 6-pin plus 1x 8-pin power connectors, while the AMD card is an MXM module with no power connectors and no suggested PSU listed.

Architecture Differences

The AMD Radeon R9 M295X is built on GCN 3.0 architecture with the Amethyst chip, fabricated on a 28 nm process at TSMC. It packs 5,000 million transistors on a 366 mm² die, resulting in a transistor density of 13.7M per mm². The NVIDIA GeForce GTX 980 Ti uses Maxwell 2.0 architecture with the GM200 chip, also on a 28 nm TSMC process, but with 8,000 million transistors on a much larger 601 mm² die, yielding a slightly lower density of 13.3M per mm².

The NVIDIA chip has more raw compute resources: 2816 shading units, 176 texture mapping units, and 96 ROPs. The AMD chip counters with 2048 shading units, 128 TMUs, and only 32 ROPs. That ROP disparity is stark — the NVIDIA card has three times the render output units. This shows up directly in pixel throughput: the GTX 980 Ti reaches 103.3 GPixel/s while the R9 M295X manages just 23.14 GPixel/s.

The memory subsystems differ fundamentally. The AMD part uses a 256-bit bus with 4 GB of GDDR5 at 5 Gbps effective, delivering 160.0 GB/s. The NVIDIA card uses a 384-bit bus with 6 GB of GDDR5 at 7 Gbps effective, delivering 336.6 GB/s. The NVIDIA card has a 110% bandwidth advantage, which is critical for high-resolution textures and compute workloads.

In terms of feature support, the AMD card supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.2.170. The NVIDIA card supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The higher DirectX feature level and newer Vulkan version on the NVIDIA side are notable for API-level compatibility. Neither card has ray tracing or tensor cores.

The form factors are entirely different. The AMD R9 M295X is an MXM Module with an MXM-B (3.0) bus interface, designed for laptops and portable devices, with display outputs described as "Portable Device Dependent." The NVIDIA GTX 980 Ti is a dual-slot desktop card with PCIe 3.0 x16, measuring 267 mm in length, 111 mm in height, and 40 mm in width, with 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2 outputs.

Head-to-Head Benchmarks

The three shared benchmarks tell a story of workload-dependent dominance. In Geekbench Metal, the AMD Radeon R9 M295X wins outright with 33790 points against the NVIDIA GeForce GTX 980 Ti's 19520 points. That 73.1% delta is the largest of any comparison, and it is a clear verdict for AMD's Metal driver stack on this mobile part.

The NVIDIA card flips the script in Geekbench OpenCL. It scores 43513 points versus AMD's 22858, a 47.5% margin. This result aligns with the NVIDIA card's higher shading unit count and much higher memory bandwidth — OpenCL workloads often scale with those resources.

In Geekbench Vulkan, the NVIDIA GTX 980 Ti again takes the win, scoring 47724 points against the AMD part's 29091 points, a 39% advantage. The NVIDIA card's newer Vulkan 1.4 support likely contributes to this lead, as does its overall compute throughput.

The win tally is 2-1 in NVIDIA's favor. However, the AMD card's single Metal win is by a much larger margin (73.1%) than either of NVIDIA's wins (47.5% and 39%). This matters for use-case decisions: if your workload is Metal-centric, the AMD part is overwhelmingly better; if it leans OpenCL or Vulkan, the NVIDIA card has a solid but less extreme edge.

Specification Differences

| Specification | AMD Radeon R9 M295X | NVIDIA GeForce GTX 980 Ti |

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

| Architecture | GCN 3.0 | Maxwell 2.0 |

| Chip | Amethyst | GM200 |

| Transistors | 5,000 million | 8,000 million |

| Die Size | 366 mm² | 601 mm² |

| Transistor Density | 13.7M / mm² | 13.3M / mm² |

| Base Clock | Not listed | 1000 MHz |

| Boost Clock | Not listed | 1076 MHz |

| Memory Clock | 1250 MHz / 5 Gbps | 1753 MHz / 7 Gbps |

| Memory Size | 4 GB | 6 GB |

| Memory Bus | 256 bit | 384 bit |

| Memory Bandwidth | 160.0 GB/s | 336.6 GB/s |

| Shading Units | 2048 | 2816 |

| TMUs | 128 | 176 |

| ROPs | 32 | 96 |

| Pixel Rate | 23.14 GPixel/s | 103.3 GPixel/s |

| Texture Rate | 92.54 GTexel/s | 189.4 GTexel/s |

| FP32 | 2.961 TFLOPS | 6.060 TFLOPS |

| FP16 | 2.961 TFLOPS (1:1) | Not listed |

| Slot Width | MXM Module | Dual-slot |

| Power Connectors | None | 1x 6-pin + 1x 8-pin |

| Suggested PSU | Not listed | 600 W |

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

| Display Outputs | Portable Device Dependent | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2 |

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

| Vulkan | 1.2.170 | 1.4 |

| Release Date | 2014-11-22 | 2015-06-01 |

| Launch MSRP | Not listed | 649 USD |

The NVIDIA card doubles the FP32 throughput (6.060 TFLOPS vs 2.961 TFLOPS), has 37.5% more shading units, 37.5% more TMUs, and 200% more ROPs. It also has 50% more memory and 110% more bandwidth. The AMD card has a smaller die, lower transistor count, and a 1:1 FP16 ratio that the NVIDIA card does not list.

The Verdict

The data points to a clear split: the NVIDIA GeForce GTX 980 Ti is the more powerful GPU in raw compute terms and wins the majority of shared benchmarks (2 out of 3). Its FP32 throughput is 6.060 TFLOPS versus 2.961 TFLOPS, and its memory bandwidth is 336.6 GB/s versus 160.0 GB/s. For any workload that leverages OpenCL or Vulkan, the GTX 980 Ti is the decisive choice.

However, the AMD Radeon R9 M295X wins the Metal benchmark by an enormous 73.1% margin, and it edges the NVIDIA card in average benchmark score (28580 vs 28020). For macOS or iOS development environments where Metal is the primary API, the AMD part is the one to pick. Its 74th percentile ranking versus the NVIDIA card's 73rd percentile confirms that, on average, the AMD card is slightly better positioned in the overall GPU hierarchy.

The form factor difference cannot be overstated. The AMD card is an MXM module for portable devices, while the NVIDIA card is a dual-slot desktop card requiring a 600 W PSU. These are not interchangeable products. The choice is dictated by the platform first, then by the workload. In a laptop, the R9 M295X is your only option. In a desktop, the GTX 980 Ti offers superior compute and memory performance across most APIs.

Where Each One Wins

AMD Radeon R9 M295X wins in:

• Metal API workloads: 33790 vs 19520, a 73.1% advantage that is unmatched in any other test.
• Average benchmark score: 28580 vs 28020, a modest but consistent lead.
• Overall percentile ranking: 74th vs 73rd, placing it slightly higher among all GPUs.
• Mobile and portable use cases: as an MXM module, it fits in laptop form factors where the NVIDIA card physically cannot go.

NVIDIA GeForce GTX 980 Ti wins in:

• OpenCL workloads: 43513 vs 22858, a 47.5% lead that reflects its 2x FP32 throughput.
• Vulkan workloads: 47724 vs 29091, a 39% margin supported by its newer Vulkan 1.4 support.
• Memory-intensive tasks: 336.6 GB/s bandwidth and 6 GB capacity versus 160.0 GB/s and 4 GB.
• Rasterization-heavy scenarios: 103.3 GPixel/s pixel rate versus 23.14 GPixel/s, driven by 96 ROPs versus 32.
• Desktop configurations: dual-slot PCIe card with standard display outputs and a 600 W PSU requirement.