AMD Radeon R7 M445 vs NVIDIA GeForce 940MX Comparison

Head-to-Head Benchmarks

The database records only one direct head-to-head benchmark between these two mobile graphics processors, and it is a clear victory for the AMD Radeon R7 M445. In the Geekbench OpenCL test, the AMD part scores 5358 points against the NVIDIA GeForce 940MX's 4939 points, giving the R7 M445 an 8.5% advantage. This is a meaningful margin in the entry-level mobile GPU space, where single-digit percentage differences often separate otherwise comparable parts.

Context from the rivals list reinforces the picture. The R7 M445's 5358 OpenCL score places it essentially level with the Intel UHD Graphics P630 (5370, a 0.2% gap against the AMD part), and it edges out the NVIDIA GeForce 840M (5322, 0.7% ahead) and the NVIDIA GeForce 930A (5317, 0.8% ahead). Interestingly, the database also shows the R7 M445 slightly ahead of the NVIDIA GeForce GTX 980M (5308) by 0.9%, though that delta is within noise for a single benchmark. The 940MX, by contrast, sits at 4939 for OpenCL, which is 0.2% behind the GeForce GTX 560M (4855), 0.5% behind the AMD Radeon R6 M255DX (4867), and 1% behind the GeForce GTS 450 (4893). The 940MX also has a separate Vulkan score of 4749, which the R7 M445 does not have recorded.

The percentile rankings summarize the broader picture. The R7 M445 lands at the 31st percentile among all GPUs, while the 940MX sits at the 28th percentile. These are close positions, but the AMD part holds a small but consistent edge in the recorded compute workloads. The 8.5% OpenCL win is the only head-to-head data point, so it carries the full weight of the comparison in this section.

Where Each One Wins

The AMD Radeon R7 M445 wins the only benchmark where both parts have comparable scores. In OpenCL compute, it delivers 5358 points versus 4939 for the 940MX, an 8.5% lead. This suggests the R7 M445 has a slight edge in general-purpose compute tasks that leverage OpenCL, such as video encoding filters, physics simulations, or image processing workloads that are offloaded to the GPU. The 320 shading units on the AMD part, paired with a 920 MHz boost clock, produce 588.8 GFLOPS of FP32 throughput, which is the raw compute figure behind that OpenCL result.

The NVIDIA GeForce 940MX, however, has its own advantages that do not show up in the single head-to-head test. It records a Vulkan score of 4749, which is a data point the AMD part lacks entirely. That means the 940MX has at least a functional Vulkan path that is measured, while the R7 M445's Vulkan support is not quantified in the database. For gaming scenarios that use Vulkan, the 940MX can point to a concrete recorded score, whereas the R7 M445 has no comparable number. The 940MX also carries a higher texture rate (27.55 GTexel/s versus 18.40 GTexel/s) and more TMUs (32 versus 20), which should help in texture-bound workloads even if the compute benchmark does not capture that advantage.

In practical terms, the R7 M445 is the pick for raw compute throughput in OpenCL, while the 940MX is the safer choice for Vulkan-based applications and for tasks that depend on texture filtering and fill rate. The pixel rates are close (7.360 GPixel/s for AMD versus 6.888 GPixel/s for NVIDIA), so neither part dominates in basic rasterization throughput.

Architecture Differences

The two GPUs come from different architectural lineages. The AMD Radeon R7 M445 uses the Meso chip based on GCN 3.0, built on a 28 nm process at TSMC. The NVIDIA GeForce 940MX uses the GM107 chip based on Maxwell, also on 28 nm at TSMC. Both are end-of-life products from 2016, with the AMD part releasing on May 14, 2016, and the NVIDIA part on June 27, 2016. The process node and foundry are identical, but the underlying designs diverge significantly.

Transistor counts differ: the R7 M445 packs 1,550 million transistors on a 125 mm² die, while the 940MX has 1,870 million transistors on a 148 mm² die. Density is nearly the same (12.4M per mm² versus 12.6M per mm²), but NVIDIA's chip is physically larger and carries more logic. The AMD part relies on 320 shading units, 20 TMUs, and 8 ROPs. The NVIDIA part has 512 shading units, 32 TMUs, and 8 ROPs. That is a 60% advantage in shader count and a 60% advantage in TMUs for NVIDIA, yet the AMD part still wins the OpenCL benchmark, which suggests GCN 3.0's compute efficiency offsets some of the raw resource deficit.

Memory configurations also differ. The R7 M445 has 4 GB of GDDR5 on a 64-bit bus, yielding 32.00 GB/s of bandwidth. The 940MX has 2 GB of GDDR5 on the same 64-bit bus, but with higher memory clocks (1253 MHz base, 5 Gbps effective) it reaches 40.10 GB/s. So the NVIDIA part has 25% more memory bandwidth, while the AMD part has double the capacity. For frame buffering at higher resolutions or with larger textures, the extra 2 GB on the R7 M445 matters; for bandwidth-sensitive workloads, the 940MX has the edge.

Clock speeds are close. The R7 M445 runs at 780 MHz base and 920 MHz boost. The 940MX runs at 795 MHz base and 861 MHz boost. The AMD part boosts higher, which helps its compute score. The NVIDIA part has a lower boost but more shaders. Power draw is only listed for the 940MX at 23 W; the R7 M445 has no TDP figure in the database. The 940MX is an MXM module, while the R7 M445 is an integrated-style IGP, which implies different thermal and form-factor expectations.

API support differs in one notable way. Both support DirectX 12 and OpenGL 4.6, but the R7 M445 supports DirectX 12 (12_0) and Vulkan 1.2.170, while the 940MX supports DirectX 12 (11_0) and Vulkan 1.4. The higher Vulkan version on the NVIDIA part is a feature advantage, even though the AMD part has a higher DirectX feature level. Both use PCIe 3.0 x8 and have portable-device-dependent display outputs.

FAQ

Q: Which GPU has the higher OpenCL score?

A: The AMD Radeon R7 M445 scores 5358 in Geekbench OpenCL, which is 8.5% higher than the NVIDIA GeForce 940MX's 4939.

Q: Does the NVIDIA GeForce 940MX have any benchmark advantage?

A: The 940MX has a recorded Geekbench Vulkan score of 4749, which the R7 M445 lacks entirely. No direct Vulkan comparison exists in the database.

Q: How much memory does each GPU have?

A: The AMD Radeon R7 M445 has 4 GB of GDDR5, while the NVIDIA GeForce 940MX has 2 GB of GDDR5. Both use a 64-bit memory bus.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce 940MX has 512 shading units, compared to 320 on the AMD Radeon R7 M445. The NVIDIA part also has 32 TMUs versus 20 on the AMD part.

Q: What are the memory bandwidth figures?

A: The AMD Radeon R7 M445 delivers 32.00 GB/s, while the NVIDIA GeForce 940MX delivers 40.10 GB/s, a 25% advantage for NVIDIA.

Q: What is the power draw for each?

A: Only the NVIDIA GeForce 940MX has a TDP listed at 23 W. The AMD Radeon R7 M445 has no TDP recorded in the database.

The Verdict

The data points to a narrow but real win for the AMD Radeon R7 M445 in compute performance. Its 8.5% OpenCL lead over the 940MX is the only direct head-to-head metric, and it aligns with the percentile rankings (31st versus 28th). For users who run OpenCL-heavy workloads, the R7 M445 is the better choice. It also offers twice the memory capacity (4 GB versus 2 GB), which can be decisive for larger textures or multiple displays.

The NVIDIA GeForce 940MX, however, is not without merit. It has a measured Vulkan score, which the AMD part cannot match because no such score exists for it. It also has higher memory bandwidth (40.10 GB/s versus 32.00 GB/s), more shading units (512 versus 320), and more TMUs (32 versus 20), plus a lower TDP at 23 W. For Vulkan-based games or applications, and for scenarios where bandwidth matters more than raw compute, the 940MX is the more versatile part.

The choice depends on the workload. If the primary use is OpenCL compute or if memory capacity is a hard requirement, the R7 M445 wins. If Vulkan support, bandwidth, or lower power draw are priorities, the 940MX is the safer pick. Neither part is a performance leader by modern standards, but within this 2016-era matchup, the AMD part holds the compute crown while the NVIDIA part offers broader API coverage and better memory throughput.

Specification Differences

| Specification | AMD Radeon R7 M445 | NVIDIA GeForce 940MX |

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

| Chip | Meso | GM107 |

| Architecture | GCN 3.0 | Maxwell |

| Process Node | 28 nm | 28 nm |

| Transistors | 1,550 million | 1,870 million |

| Die Size | 125 mm² | 148 mm² |

| Transistor Density | 12.4M / mm² | 12.6M / mm² |

| Base Clock | 780 MHz | 795 MHz |

| Boost Clock | 920 MHz | 861 MHz |

| Memory Clock | 1000 MHz, 4 Gbps effective | 1253 MHz, 5 Gbps effective |

| Memory Size | 4 GB | 2 GB |

| Memory Type | GDDR5 | GDDR5 |

| Memory Bus Width | 64 bit | 64 bit |

| Memory Bandwidth | 32.00 GB/s | 40.10 GB/s |

| Shading Units | 320 | 512 |

| TMUs | 20 | 32 |

| ROPs | 8 | 8 |

| Pixel Rate | 7.360 GPixel/s | 6.888 GPixel/s |

| Texture Rate | 18.40 GTexel/s | 27.55 GTexel/s |

| FP32 | 588.8 GFLOPS | 881.7 GFLOPS |

| FP16 | 588.8 GFLOPS (1:1) | Not listed |

| TDP | Not listed | 23 W |

| Slot Width | IGP | MXM Module |

| Power Connectors | Not listed | None |

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

| OpenGL | 4.6 | 4.6 |

| Vulkan | 1.2.170 | 1.4 |

| Release Date | May 14, 2016 | June 27, 2016 |

| Production Status | End-of-life | End-of-life |