AMD Radeon R9 M360 vs NVIDIA GeForce 945M Comparison

The AMD Radeon R9 M360 and NVIDIA GeForce 945M are two end-of-life mobile graphics solutions from 2015 that land in the same performance tier, with the data placing them within a hair's breadth of each other in raw compute benchmarks. Despite their similar average scores, the architecture and memory configurations tell a more nuanced story about where each chip excels, and the benchmark results reveal a narrow but measurable edge for the AMD part in the single test where they directly compete.

Head-to-Head Benchmarks

The only direct benchmark comparison available is the Geekbench OpenCL test, and it shows a razor-thin victory for the AMD Radeon R9 M360. The AMD card scores 8,211 points against the NVIDIA GeForce 945M's 8,099 points, a delta of just 1.4 percent. This is a marginal lead, but it is consistent with the broader average benchmark scores for both parts. The R9 M360's average benchmark score is 8,129, while the 945M sits slightly lower at 8,099, reinforcing that the AMD part holds a small but real advantage in general compute workloads.

Contextualizing these scores against their nearest rivals clarifies the competitive landscape. The AMD R9 M360's nearest rival, the NVIDIA GeForce GTX 950M, posts an average score of 8,135, which is only 0.1 percent higher than the R9 M360's average — essentially a statistical tie. The GeForce 945M's own nearest rival list includes the same GTX 950M, which beats it by 0.4 percent, and the AMD R9 M360, which leads it by 0.4 percent from the NVIDIA card's perspective. Interestingly, both cards also appear near the NVIDIA GRID K2, with the R9 M360 leading it by 0.6 percent and the 945M leading it by 0.2 percent, showing that all four GPUs are clustered within a performance band of roughly one percent.

The bigger picture is that neither card dominates its peer group. The R9 M360's 1.4 percent win in the head-to-head OpenCL test is the largest delta observed, but it is still a small margin. The data suggests that in day-to-day compute tasks, users would be hard-pressed to notice a difference between the two. However, the AMD part's lead is consistent across both its individual benchmark score and its average, while the NVIDIA card lags slightly in both metrics. The GeForce 945M does not have a Vulkan benchmark score listed, whereas the R9 M360 posts 8,047 points in that test, suggesting the AMD card has broader benchmark coverage and a second data point that confirms its compute capability.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R9 M360 has an average benchmark score of 8,129, while the NVIDIA GeForce 945M averages 8,099, giving the AMD part a 0.4 percent advantage.

Q: What is the exact performance difference in the OpenCL test?

A: In the Geekbench OpenCL test, the AMD Radeon R9 M360 scores 8,211, beating the NVIDIA GeForce 945M's 8,099 by a 1.4 percent margin.

Q: How do these GPUs compare to the NVIDIA GeForce GTX 950M?

A: The GTX 950M has an average score of 8,135, which is 0.1 percent higher than the R9 M360 and 0.4 percent higher than the 945M, showing all three are effectively in the same performance class.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce 945M has 640 shading units, while the AMD Radeon R9 M360 has only 512, meaning the NVIDIA part has a 25 percent higher count despite losing in raw compute scores.

Q: What memory type does each GPU use?

A: The AMD Radeon R9 M360 uses 4 GB of GDDR5 memory with 72.00 GB/s bandwidth, while the NVIDIA GeForce 945M uses 2 GB of DDR3 memory with only 28.80 GB/s bandwidth.

Q: Which GPU supports a higher DirectX version?

A: The AMD Radeon R9 M360 supports DirectX 12 (11_1), while the NVIDIA GeForce 945M supports DirectX 12 (11_0), giving the AMD part a slightly more recent feature level.

Architecture Differences

The two GPUs represent fundamentally different design philosophies from their respective manufacturers. The AMD Radeon R9 M360 is built on the Tropo chip using the GCN 1.0 architecture, fabricated on a 28 nm process at TSMC. It packs 1,500 million transistors onto a 123 mm² die, yielding a transistor density of 12.2M per mm². The NVIDIA GeForce 945M, in contrast, uses the GM107 chip with the Maxwell architecture, also on a 28 nm TSMC process, but with a larger 148 mm² die containing 1,870 million transistors and a slightly higher density of 12.6M per mm².

The compute resources differ significantly. The NVIDIA card has 640 shading units, 40 texture mapping units, and 16 render output units, while the AMD card has 512 shading units, 32 TMUs, and 16 ROPs. Despite having fewer shaders and TMUs, the R9 M360 achieves a pixel rate of 14.80 GPixel/s and a texture rate of 29.60 GTexel/s, whereas the 945M posts 16.32 GPixel/s and 40.80 GTexel/s. In raw fillrate terms, the NVIDIA part is faster, with 10.3 percent higher pixel throughput and 37.8 percent higher texture throughput. However, the AMD card's FP32 compute is 947.2 GFLOPS compared to the 945M's 1,305.6 GFLOPS — a 37.8 percent advantage for NVIDIA — yet the AMD card still wins the OpenCL benchmark, suggesting that memory bandwidth plays a crucial role in real-world compute performance.

The memory subsystems are where the most dramatic divergence appears. The AMD R9 M360 uses 4 GB of GDDR5 memory on a 128-bit bus, delivering 72.00 GB/s of bandwidth, while the NVIDIA 945M uses just 2 GB of DDR3 on the same 128-bit bus, yielding only 28.80 GB/s. That is a 150 percent bandwidth advantage for the AMD part, which likely explains why it wins compute tests despite lower theoretical FLOPs and fillrates. Clock speeds also favor NVIDIA: the 945M runs at 928 MHz base and 1020 MHz boost, while the R9 M360 runs at 900 MHz base and 925 MHz boost. The memory clocks further diverge, with the AMD card at 1125 MHz (4.5 Gbps effective) versus the NVIDIA card's 900 MHz (1800 Mbps effective).

The Verdict

The benchmark data paints a clear picture: for general compute and OpenCL workloads, the AMD Radeon R9 M360 is the better performer, albeit by a slim margin. Its 1.4 percent lead in the head-to-head test and 0.4 percent lead in average score are small but consistent, and its victory is likely attributable to the massive memory bandwidth advantage that GDDR5 provides over DDR3. The R9 M360 also offers twice the VRAM capacity at 4 GB versus 2 GB, which could matter in memory-constrained scenarios. However, the NVIDIA GeForce 945M counters with higher raw compute throughput — 1,305.6 GFLOPS versus 947.2 GFLOPS — and superior fillrates, suggesting it might handle certain graphics-heavy tasks better, though the OpenCL data does not support that hypothesis directly.

For users prioritizing compute performance in a portable device, the AMD R9 M360 is the data-backed choice, as it wins the only available benchmark. For those who might value NVIDIA's architecture features, such as its higher shading unit count or the newer Vulkan 1.4 API support (versus AMD's Vulkan 1.2.170), the 945M has merits, but the benchmark evidence does not show it winning any performance tests. The 945M also has a specified TDP of 75 W and uses an MXM module form factor, while the R9 M360 has no listed TDP or slot width, making power comparisons impossible from the provided data. Ultimately, the data supports the AMD Radeon R9 M360 as the stronger overall compute performer, but the margin is so thin that real-world differences would be negligible for most users.

Specification Differences

| Specification | AMD Radeon R9 M360 | NVIDIA GeForce 945M |

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

| Architecture | GCN 1.0 | Maxwell |

| Chip | Tropo | GM107 |

| Transistors | 1,500 million | 1,870 million |

| Die Size | 123 mm² | 148 mm² |

| Transistor Density | 12.2M / mm² | 12.6M / mm² |

| Base Clock | 900 MHz | 928 MHz |

| Boost Clock | 925 MHz | 1020 MHz |

| Memory Clock | 1125 MHz (4.5 Gbps effective) | 900 MHz (1800 Mbps effective) |

| Memory Size | 4 GB | 2 GB |

| Memory Type | GDDR5 | DDR3 |

| Memory Bus | 128 bit | 128 bit |

| Memory Bandwidth | 72.00 GB/s | 28.80 GB/s |

| Shading Units | 512 | 640 |

| TMUs | 32 | 40 |

| ROPs | 16 | 16 |

| Pixel Rate | 14.80 GPixel/s | 16.32 GPixel/s |

| Texture Rate | 29.60 GTexel/s | 40.80 GTexel/s |

| FP32 | 947.2 GFLOPS | 1,305.6 GFLOPS |

| TDP | Not listed | 75 W |

| Slot Width | Not listed | MXM Module |

| Power Connectors | Not listed | None |

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

| Display Outputs | Not listed | Portable Device Dependent |

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

| Vulkan | 1.2.170 | 1.4 |

| Release Date | 2015-05-04 | 2015-10-26 |

| Predecessor | Solar System | GeForce 800M |

| Successor | Polaris Mobile | GeForce 10 Mobile |

Where Each One Wins

The AMD Radeon R9 M360 wins in the compute benchmark arena, taking the OpenCL test by 1.4 percent and posting a higher average score. Its 4 GB GDDR5 memory with 72.00 GB/s bandwidth gives it a decisive advantage in memory-intensive workloads, and the data shows that this bandwidth more than compensates for its lower shader count and FP32 throughput. The R9 M360 also supports DirectX 12 (11_1), a slightly newer feature level than the 945M's DirectX 12 (11_0), and it has a Vulkan benchmark score of 8,047, indicating it can handle that API with measurable performance. For users running OpenCL compute tasks, scientific workloads, or any application that benefits from high memory bandwidth, the R9 M360 is the clear winner.

The NVIDIA GeForce 945M wins on paper in raw graphics throughput. Its 640 shading units, 40 TMUs, and higher clocks produce a texture rate of 40.80 GTexel/s — 37.8 percent higher than the AMD card — and a pixel rate of 16.32 GPixel/s, which is 10.3 percent higher. Its FP32 compute of 1,305.6 GFLOPS is 37.8 percent above the R9 M360's 947.2 GFLOPS. These specs suggest that in pure rasterization or shader-bound scenarios, the 945M could outperform the AMD part, though no benchmark data confirms this. The 945M also supports Vulkan 1.4, a newer API version than the R9 M360's Vulkan 1.2.170, and it is specified with a 75 W TDP, making its power envelope a defined factor. For users who prioritize NVIDIA's driver ecosystem or need a known power draw, the 945M has structural advantages, but the only performance test in the data favors AMD.