AMD Radeon R5 M420 vs NVIDIA GeForce 920MX Comparison

The NVIDIA GeForce 920MX and AMD Radeon R5 M420 are both end-of-life mobile graphics solutions from 2016, aimed at entry-level laptops. The data available for direct comparison is limited to a single OpenCL benchmark, but it provides a clear, if narrow, picture of their relative performance. The GeForce 920MX edges out the Radeon R5 M420, securing the only win in the head-to-head comparison.

Head-to-Head Benchmarks

The only direct benchmark comparison available is in Geekbench OpenCL, a compute-oriented test that measures raw GPU throughput for general-purpose workloads. In this test, the NVIDIA GeForce 920MX scores 4068 points, while the AMD Radeon R5 M420 scores 3956 points. This gives the NVIDIA part a 2.8% performance advantage, a modest but definitive lead in this specific workload.

This narrow margin is interesting, as the two GPUs have significantly different hardware configurations. The AMD Radeon R5 M420 boasts a higher peak FP32 throughput of 544.0 GFLOPS compared to the GeForce 920MX's 508.4 GFLOPS. Despite having ~7% more theoretical compute horsepower, the AMD chip loses in the actual benchmark. This suggests that the NVIDIA architecture, or its driver stack, is more efficient at translating raw FLOPs into real-world performance in this test. The 2.8% delta is small, but it is a clear win for NVIDIA.

Looking at the broader context from the `nearestRivals` data, both GPUs are clustered in a very tight performance band. The GeForce 920MX's average benchmark score of 3528 is just 1.8% behind the NVIDIA GeForce GT 545 (score 3594) and 2.4% behind the GeForce GT 735M (score 3616). Similarly, the Radeon R5 M420's average score of 3956 is virtually identical to the NVIDIA GeForce 830M (score 3957, 0% delta) and the GeForce GT 745M (score 3953, 0.1% delta). This data indicates that neither GPU is a performance outlier; they are both firmly situated in the entry-level segment, with the R5 M420's average score placing it at the 23rd percentile of all GPUs, slightly ahead of the 920MX's 21st percentile.

The fact that the R5 M420 has a higher average benchmark score (3956 vs 3528) than the 920MX is a critical nuance. While the 920MX wins the single head-to-head OpenCL test, the R5 M420 has a higher overall score based on a different set of benchmarks. This implies that in other, untested workloads, the AMD part may have an advantage, or that its single benchmark result is more representative of its average performance.

FAQ

Q: Which GPU is faster in the direct benchmark comparison?

A: The NVIDIA GeForce 920MX is faster, winning the Geekbench OpenCL test with a score of 4068 compared to the AMD Radeon R5 M420's 3956, a 2.8% difference.

Q: Does the AMD Radeon R5 M420 have a higher theoretical compute power?

A: Yes. The Radeon R5 M420 has a higher FP32 rating of 544.0 GFLOPS compared to the GeForce 920MX's 508.4 GFLOPS, yet it still lost the OpenCL benchmark.

Q: How does the average benchmark score compare between the two?

A: The AMD Radeon R5 M420 has a higher average benchmark score of 3956, while the NVIDIA GeForce 920MX has an average score of 3528. This is despite the 920MX winning the direct head-to-head test.

Q: What is the performance percentile for each GPU?

A: The NVIDIA GeForce 920MX sits in the 21st percentile of all GPUs, while the AMD Radeon R5 M420 is slightly higher, in the 23rd percentile.

Q: Which GPU is closer to the performance of the NVIDIA GeForce 830M?

A: The AMD Radeon R5 M420 is essentially tied with the GeForce 830M, showing a 0% delta in average benchmark scores (3956 vs 3957).

Q: Do both GPUs support DirectX 12?

A: Yes, both support DirectX 12, but at different feature levels. The GeForce 920MX supports DirectX 12 (11_0), while the Radeon R5 M420 supports DirectX 12 (11_1).

Architecture Differences

The two GPUs are built on fundamentally different architectures from their respective manufacturers. The NVIDIA GeForce 920MX is based on the Maxwell architecture, using the GM108S chip. In contrast, the AMD Radeon R5 M420 is based on the older GCN 1.0 architecture, using the Jet chip. This architectural divide explains many of their differing characteristics.

Both are manufactured on the same 28 nm process node at TSMC, but the chips themselves are quite different. The NVIDIA chip is larger, with a die size of 77 mm² and containing 1,020 million transistors, translating to a transistor density of 13.2M / mm². The AMD chip is smaller, at 56 mm², with 690 million transistors and a density of 12.3M / mm². This means the NVIDIA chip packs more transistors into a larger area, which is a sign of a more complex design.

The compute unit layouts also differ significantly. The GeForce 920MX has 256 shading units, 24 texture mapping units (TMUs), and 8 render output units (ROPs). The Radeon R5 M420 has more shading units at 320, but fewer TMUs at 20, and the same number of ROPs at 8. This configuration explains the performance nuances: the AMD part has more ALUs for raw shader work, but the NVIDIA part has more TMUs for texture-heavy workloads. The pixel rate of the 920MX is 7.944 GPixel/s, higher than the R5 M420's 6.800 GPixel/s, while the texture rate is also higher for NVIDIA at 23.83 GTexel/s versus 17.00 GTexel/s.

Feature support also diverges. The GeForce 920MX supports Vulkan 1.4, while the Radeon R5 M420 supports the older Vulkan 1.2.170. Both support OpenGL 4.6. These differences in API support can impact compatibility with newer software titles.

Specification Differences

Beyond the architecture, the two GPUs have distinct specifications that are important to note. The most obvious difference is in memory configuration. The NVIDIA GeForce 920MX comes with 2 GB of DDR3 memory, while the AMD Radeon R5 M420 offers double the capacity at 4 GB of DDR3. However, the memory bus width is the same for both at 64 bit. This results in a slightly higher memory bandwidth for the AMD part: 16.00 GB/s compared to the 920MX's 14.40 GB/s. The memory clock speeds differ, with the 920MX running at 900 MHz (1800 Mbps effective) and the R5 M420 at 1000 MHz (2 Gbps effective).

Clock speeds are also a point of difference. The NVIDIA GeForce 920MX has a higher base clock of 965 MHz and a boost clock of 993 MHz. The AMD Radeon R5 M420 operates at a lower 780 MHz base clock and an 850 MHz boost clock. This significant clock speed advantage for NVIDIA helps explain its win in the OpenCL benchmark despite the AMD part's higher theoretical FP32.

The form factor and power characteristics are also different. The GeForce 920MX is listed as an MXM Module with a 16 W TDP and no power connectors. The Radeon R5 M420 is an IGP (Integrated Graphics Processor) with no listed TDP. This suggests the AMD part is designed for lower-power, integrated solutions, while the NVIDIA part is a discrete module. Both use a PCIe 3.0 x8 bus interface.

The Verdict

Based strictly on the data, the NVIDIA GeForce 920MX is the winner in the direct head-to-head comparison, taking the only benchmark with a 2.8% lead. If the sole criterion is the Geekbench OpenCL score, the 920MX is the better choice. However, the data also shows that the AMD Radeon R5 M420 has a higher average benchmark score (3956 vs 3528) and a better percentile ranking (23rd vs 21st). This suggests that the R5 M420 may perform better in a broader range of tasks not covered by the single head-to-head test.

The Radeon R5 M420 also offers double the memory capacity (4 GB vs 2 GB). In scenarios where memory capacity is a limiting factor, such as running applications that require large framebuffers, the AMD part could have an advantage. The higher memory bandwidth of the R5 M420 (16.00 GB/s vs 14.40 GB/s) also supports this potential edge in memory-intensive workloads.

Therefore, the verdict is nuanced. For a specific, compute-focused OpenCL workload, the NVIDIA GeForce 920MX is demonstrably faster. For general-purpose use and applications that might benefit from more memory, the AMD Radeon R5 M420's higher average score and larger memory pool make it a more balanced option. The data does not support a definitive overall winner; it supports a choice based on workload.

Where Each One Wins

NVIDIA GeForce 920MX: This GPU wins in the compute benchmark tested. Its higher clock speeds (965 MHz base, 993 MHz boost) and higher texture rate (23.83 GTexel/s) give it a clear advantage in the Geekbench OpenCL test. It is the better option for users who prioritize raw compute performance in applications that leverage OpenCL. Its higher pixel rate (7.944 GPixel/s) also suggests it may be slightly better at fill-rate-bound tasks. The 920MX is the pick for a specific, performance-oriented workload.

AMD Radeon R5 M420: This GPU wins on aggregate metrics. Its higher average benchmark score (3956) and superior percentile ranking (23rd) indicate it is generally more capable across a wider array of tests. The 4 GB memory capacity is double that of the 920MX, which is a significant advantage for multitasking or using applications with large memory footprints. Its higher FP32 rating (544.0 GFLOPS) also gives it a theoretical edge in pure shader math. The R5 M420 is the better choice for a user who wants a more well-rounded performer with more memory, based on the broader benchmark data.