AMD Radeon R5 M320 vs NVIDIA GeForce 940MX Comparison

The benchmark data splits the verdict cleanly between these two end-of-life mobile graphics parts. The NVIDIA GeForce 940MX takes the Vulkan workload decisively, while the AMD Radeon R5 M320 counters with a narrow win in OpenCL. Neither GPU is a powerhouse — both sit near the bottom of the global percentile rankings — but the specific test results reveal distinct strengths that matter for different use cases.

Head-to-Head Benchmarks

The most significant gap between these two chips appears in the Geekbench Vulkan test. The NVIDIA GeForce 940MX scores 4,749 against the AMD Radeon R5 M320’s 4,262. That is a delta of 11.4 percent in NVIDIA’s favor, making this the clearest single-test victory in the comparison. For any workload leveraging the Vulkan API, the 940MX is the substantially stronger option.

The OpenCL result flips the script, though the margin is far smaller. AMD’s Radeon R5 M320 posts 5,051 points, edging out NVIDIA’s 4,939 by just 2.2 percent. This is a competitive result, not a dominant one. The two chips are effectively within striking distance in OpenCL, which suggests that compute-heavy applications that use this API will see comparable performance, with a slight lean toward AMD.

Looking at the broader average benchmark score, the two GPUs remain close. NVIDIA’s 940MX averages 4,844 across its tested workloads, while AMD’s R5 M320 averages 4,657. That is a 3.9 percent lead for NVIDIA in overall aggregate performance, driven almost entirely by the Vulkan result. The head-to-head tally ends at one win apiece, but the magnitude of NVIDIA’s Vulkan victory outweighs AMD’s narrow OpenCL edge.

Context from the nearest rival data reinforces this picture. The 940MX’s average score of 4,844 places it just 0.2 percent behind the NVIDIA GeForce GTX 560M (4,855) and 0.5 percent behind the AMD Radeon R6 M255DX (4,867). It also sits 1.1 percent ahead of the NVIDIA GeForce RTX 3080 12 GB (4,791) in this specific benchmark aggregate, a reminder that these synthetic tests do not translate directly to real-world gaming performance across generations. The R5 M320’s average of 4,657 is exactly tied with the AMD Radeon RX 9060 XT 16 GB (4,657) and sits 0.6 percent behind the NVIDIA Quadro P400 (4,684).

Architecture Differences

The two GPUs come from entirely different architectural lineages. NVIDIA’s 940MX is built on the Maxwell architecture using the GM107 chip, a 28 nm design fabricated by TSMC. AMD’s R5 M320 uses the GCN 1.0 architecture with the Jet chip, also on a 28 nm TSMC process. The process node is identical, but the underlying designs diverge sharply.

The transistor counts tell that story. NVIDIA packs 1,870 million transistors into a 148 mm² die, yielding a transistor density of 12.6 million per square millimeter. AMD’s Jet chip is far smaller at 56 mm² and holds 690 million transistors, for a density of 12.3 million per square millimeter. The NVIDIA chip is 2.6 times larger in die area and has 2.7 times more transistors, which explains the substantial difference in compute resources.

Those resources translate into a clear specification advantage for NVIDIA. The 940MX fields 512 shading units, 32 texture mapping units, and 8 ROPs. The R5 M320 offers 320 shading units, 20 TMUs, and the same 8 ROPs. That gives NVIDIA a 60 percent advantage in shading units and a 60 percent advantage in TMUs, while ROP counts match.

Clock speeds are nearly identical. The 940MX runs at a base of 795 MHz with an 861 MHz boost. The R5 M320 operates at 780 MHz base and 855 MHz boost. The differences are negligible, so the architectural and unit-count advantages are not offset by clock frequency.

Memory configuration is where the two diverge most dramatically. NVIDIA uses 2 GB of GDDR5 on a 64-bit bus, running at 1253 MHz (5 Gbps effective), producing 40.10 GB/s of bandwidth. AMD offers 4 GB of DDR3 on the same 64-bit bus, at 1000 MHz (2 Gbps effective), yielding only 16.00 GB/s. NVIDIA’s memory bandwidth is 2.5 times higher, a critical advantage for texture-heavy workloads, even though AMD doubles the capacity.

The resulting throughput figures reflect these differences. NVIDIA’s pixel rate is 6.888 GPixel/s versus AMD’s 6.840 GPixel/s, nearly identical. But texture rate favors NVIDIA heavily at 27.55 GTexel/s against AMD’s 17.10 GTexel/s. FP32 compute also goes to NVIDIA at 881.7 GFLOPS versus 547.2 GFLOPS — a 61 percent lead. API support differs slightly: both support DirectX 12, but NVIDIA is at 11_0 while AMD is at 11_1. OpenGL is 4.6 for both, and Vulkan support is 1.4 on NVIDIA versus 1.2.170 on AMD.

Where Each One Wins

NVIDIA’s GeForce 940MX is the pick for Vulkan-based workloads. The 11.4 percent lead in the Geekbench Vulkan test is the largest performance gap between the two in any metric. The 940MX also wins on raw compute throughput, with 881.7 GFLOPS of FP32 performance against AMD’s 547.2 GFLOPS, and on texture rate at 27.55 GTexel/s versus 17.10 GTexel/s. The 40.10 GB/s of GDDR5 memory bandwidth is a decisive advantage for any application that streams large textures or geometry.

AMD’s Radeon R5 M320 wins the OpenCL benchmark by 2.2 percent, which is modest but real. Its 4 GB of DDR3 memory doubles the capacity of the NVIDIA part, which can matter for workloads that fit larger datasets in VRAM, even if the bandwidth is far lower. The R5 M320 also has a slightly higher pixel rate at 6.840 GPixel/s versus NVIDIA’s 6.888 GPixel/s — a negligible difference, but the AMD part is not entirely without merit.

For gaming, the data points toward NVIDIA. The Vulkan API is increasingly common in modern game engines, and the 940MX’s substantial lead there, combined with higher texture rate and memory bandwidth, gives it the edge for most graphics-intensive tasks. The R5 M320’s OpenCL advantage is more relevant to compute-oriented applications like general-purpose GPU programming rather than gaming.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 940MX averages 4,844 across its tested workloads, while the AMD Radeon R5 M320 averages 4,657. That is a 3.9 percent lead for NVIDIA.

Q: How big is the Vulkan performance gap between the two?

A: The 940MX scores 4,749 in Geekbench Vulkan, which is 11.4 percent higher than the R5 M320’s 4,262.

Q: Does the AMD R5 M320 win any benchmark?

A: Yes, it wins the Geekbench OpenCL test with a score of 5,051 against NVIDIA’s 4,939, a 2.2 percent margin.

Q: What is the difference in memory bandwidth?

A: NVIDIA’s 940MX has 40.10 GB/s of bandwidth from GDDR5 memory, while AMD’s R5 M320 has 16.00 GB/s from DDR3. NVIDIA’s bandwidth is 2.5 times higher.

Q: How do the two compare in terms of shading units?

A: The 940MX has 512 shading units, while the R5 M320 has 320, giving NVIDIA a 60 percent advantage in this specification.

Q: Are the clock speeds similar between the two GPUs?

A: Yes. The 940MX runs at 795 MHz base and 861 MHz boost, while the R5 M320 runs at 780 MHz base and 855 MHz boost. The differences are minimal.

The Verdict

The data supports NVIDIA’s GeForce 940MX as the stronger overall performer. It wins the Vulkan benchmark by a wide 11.4 percent margin, holds a 3.9 percent lead in average benchmark score, and dominates in every key compute specification: FP32 throughput (881.7 GFLOPS versus 547.2 GFLOPS), texture rate (27.55 GTexel/s versus 17.10 GTexel/s), and memory bandwidth (40.10 GB/s versus 16.00 GB/s). For users who prioritize graphics performance, especially in Vulkan-based applications, the 940MX is the clear choice.

The AMD Radeon R5 M320 is not without its niche. Its OpenCL win, though narrow, indicates competitive compute performance in that API. The 4 GB memory capacity doubles NVIDIA’s 2 GB, which can be beneficial for specific workloads that require larger memory footprints. However, the DDR3 memory type and significantly lower bandwidth (16.00 GB/s versus 40.10 GB/s) undermine that capacity advantage in most practical scenarios.

In terms of market positioning, neither GPU is a high performer. The 940MX sits at the 28th percentile of all GPUs, while the R5 M320 sits at the 27th percentile. These are entry-level mobile parts, both end-of-life and built on the 28 nm process. The architectural gap is stark: NVIDIA’s GM107 die is nearly triple the size of AMD’s Jet chip, with 1,870 million transistors against 690 million. That investment shows up in the benchmark results.

For a user choosing between these two in a legacy laptop, the 940MX is the safer bet for gaming and graphics work. The R5 M320 only makes sense if the workload specifically favors OpenCL or requires more than 2 GB of VRAM, accepting the substantial bandwidth penalty. The verdict is not unanimous — AMD takes one benchmark — but the weight of evidence across the full specification list and benchmark data points firmly toward NVIDIA.

Specification Differences

| Specification | NVIDIA GeForce 940MX | AMD Radeon R5 M320 |

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

| Chip | GM107 | Jet |

| Architecture | Maxwell | GCN 1.0 |

| Generation | GeForce 900M | Gem System (R5 M300) |

| Transistors | 1,870 million | 690 million |

| Die Size | 148 mm² | 56 mm² |

| Transistor Density | 12.6M / mm² | 12.3M / mm² |

| Base Clock | 795 MHz | 780 MHz |

| Boost Clock | 861 MHz | 855 MHz |

| Memory Size | 2 GB | 4 GB |

| Memory Type | GDDR5 | DDR3 |

| Memory Clock | 1253 MHz (5 Gbps effective) | 1000 MHz (2 Gbps effective) |

| Memory Bandwidth | 40.10 GB/s | 16.00 GB/s |

| Shading Units | 512 | 320 |

| TMUs | 32 | 20 |

| ROPs | 8 | 8 |

| Pixel Rate | 6.888 GPixel/s | 6.840 GPixel/s |

| Texture Rate | 27.55 GTexel/s | 17.10 GTexel/s |

| FP32 | 881.7 GFLOPS | 547.2 GFLOPS |

| TDP | 23 W | Not specified |

| Slot Width | MXM Module | IGP |

| Power Connectors | None | Not specified |

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

| OpenGL | 4.6 | 4.6 |

| Vulkan | 1.4 | 1.2.170 |

| Release Date | 2016-06-27 | 2015-05-04 |

| Predecessor | GeForce 800M | Solar System |

| Successor | GeForce 10 Mobile | Polaris Mobile |