AMD Radeon R5 M420 vs NVIDIA GeForce 930M Comparison

Where Each One Wins

The benchmark data splits these two mobile graphics parts into clear, distinct roles. The NVIDIA GeForce 930M takes the outright performance crown in the only head-to-head test recorded, while the AMD Radeon R5 M420 finds its advantage in memory capacity and raw compute characteristics that do not translate into a benchmark victory.

In the Geekbench OpenCL test, the only benchmark where both chips appear, the NVIDIA GeForce 930M scores 5046 against the AMD Radeon R5 M420's 3956. That is a 27.6% delta in favor of NVIDIA, and it is the sole recorded win for either side. The AMD part has no benchmark wins in the database. For workloads that stress OpenCL compute, whether that is GPGPU acceleration, video encoding helpers, or OpenCL-based filters in creative applications, the NVIDIA chip is the one to pick.

However, the AMD Radeon R5 M420 counters with a specification-level advantage that matters for certain tasks: it ships with 4 GB of DDR3 memory compared to the 930M's 2 GB. For workloads that are memory-capacity bound, such as holding larger textures or datasets in VRAM, the AMD part has the edge in headroom, even if its raw throughput is lower. The AMD chip also posts a higher base clock of 780 MHz (boost 850 MHz) versus NVIDIA's flat 549 MHz, and a higher memory bandwidth of 16.00 GB/s against 12.80 GB/s. Yet in the recorded benchmark, those advantages do not materialize into a score win.

The percentile data reinforces this split. The GeForce 930M sits at the 26th percentile of all GPUs, while the Radeon R5 M420 sits at the 23rd percentile. Both are firmly entry-level mobile parts, but NVIDIA holds a small but consistent edge in the aggregate performance ranking.

Architecture Differences

The two chips come from different architectural families and different generations. The NVIDIA GeForce 930M is built on the Maxwell architecture, specifically the GM108S chip, and belongs to the GeForce 900M generation. The AMD Radeon R5 M420 uses the GCN 1.0 architecture with the Jet chip, and is part of the Gem System (R5 M400) generation. Both are fabricated by TSMC on a 28 nm process, so the manufacturing node is identical.

The transistor counts differ notably. NVIDIA's GM108S packs 1,020 million transistors on a 77 mm² die, giving a transistor density of 13.2 million per square millimeter. AMD's Jet chip contains 690 million transistors on a 56 mm² die, with a density of 12.3 million per square millimeter. NVIDIA's chip is physically larger and denser, which helps explain its higher compute output.

The compute resources are also asymmetric. The GeForce 930M has 384 shading units, 24 texture mapping units, and 8 ROPs. The Radeon R5 M420 has 320 shading units, 20 TMUs, and 8 ROPs. NVIDIA's advantage in shader and TMU counts gives it a theoretical peak of 421.6 GFLOPS FP32 and 13.18 GTexel/s texture rate. AMD, despite fewer units, posts a higher theoretical peak of 544.0 GFLOPS FP32 and 17.00 GTexel/s, thanks to its higher clocks. The pixel rates tell a similar story: AMD reaches 6.800 GPixel/s while NVIDIA manages 4.392 GPixel/s.

Memory configurations differ in capacity and bandwidth. Both use DDR3 on a 64-bit bus, but AMD pairs that bus with 4 GB and a 1000 MHz memory clock (2 Gbps effective) for 16.00 GB/s. NVIDIA uses 2 GB with an 800 MHz clock (1600 Mbps effective) for 12.80 GB/s. The AMD part has 25% more bandwidth and double the capacity.

In API support, NVIDIA lists DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. AMD lists DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Both are end-of-life products, with NVIDIA releasing on 2015-03-12 and AMD on 2016-05-14. The NVIDIA part's predecessor is GeForce 800M and successor is GeForce 10 Mobile; AMD's predecessor is Solar System and successor is Polaris Mobile. Both use a PCIe 3.0 x8 bus interface and are integrated graphics parts (IGP) with portable-device-dependent display outputs.

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is Geekbench OpenCL. Here the NVIDIA GeForce 930M scores 5046 against the AMD Radeon R5 M420's 3956, a delta of 27.6%. That is a decisive win for NVIDIA in compute performance.

To put that score in context, the 930M's average benchmark score across all recorded tests is 4388, while the R5 M420's average is exactly its single OpenCL score of 3956. The 930M's nearest rivals include the NVIDIA GeForce GT 645M (average 4411, delta -0.5%), Intel Iris Pro Graphics 5200 (average 4360, delta 0.7%), and NVIDIA GeForce RTX 4070 GDDR6 (average 4335, delta 1.2%). The presence of the RTX 4070 GDDR6 in that list, with a delta of only 1.2%, is a quirk of the database's score normalization, but it shows how tightly clustered the 930M's benchmark band is. The AMD R5 M420's nearest rivals are the NVIDIA GeForce 830M (average 3957, delta 0%), NVIDIA GeForce GT 745M (average 3953, delta 0.1%), NVIDIA Quadro K2000 (average 3964, delta -0.2%), and AMD Radeon HD 6850 X2 (average 3977, delta -0.5%). The R5 M420 sits almost exactly at parity with those parts.

The head-to-head delta of 27.6% is the single largest gap between the two chips in any recorded metric. NVIDIA wins the only direct comparison, and the R5 M420 has no countervailing benchmark win. However, the AMD chip's higher theoretical pixel rate (6.800 vs 4.392 GPixel/s) and texture rate (17.00 vs 13.18 GTexel/s) suggest that in rasterization-heavy workloads that are not captured by the OpenCL test, the AMD part could close the gap. The database does not include a gaming or DirectX benchmark for these two, so that remains speculative.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The NVIDIA GeForce 930M scores 5046, while the AMD Radeon R5 M420 scores 3956. That is a 27.6% advantage for NVIDIA.

Q: Does the AMD Radeon R5 M420 have more memory than the GeForce 930M?

A: Yes. The AMD Radeon R5 M420 has 4 GB of DDR3 memory on a 64-bit bus, while the GeForce 930M has 2 GB of DDR3 on the same bus width.

Q: Which GPU has the higher memory bandwidth?

A: The AMD Radeon R5 M420 has 16.00 GB/s of bandwidth, compared to 12.80 GB/s for the NVIDIA GeForce 930M.

Q: What are the shading unit counts for each GPU?

A: The NVIDIA GeForce 930M has 384 shading units, while the AMD Radeon R5 M420 has 320 shading units.

Q: How do the two GPUs compare in terms of their theoretical FP32 performance?

A: The AMD Radeon R5 M420 has a higher theoretical FP32 peak of 544.0 GFLOPS, versus 421.6 GFLOPS for the NVIDIA GeForce 930M, despite NVIDIA having more shading units.

Q: Which GPU has the higher percentile ranking among all GPUs?

A: The NVIDIA GeForce 930M is at the 26th percentile, while the AMD Radeon R5 M420 is at the 23rd percentile.

The Verdict

The recorded data makes a straightforward call: the NVIDIA GeForce 930M is the faster GPU in compute workloads, winning the only head-to-head benchmark by 27.6% and holding a higher average benchmark score (4388 vs 3956). Its higher percentile ranking (26th vs 23rd) and a larger transistor budget (1,020 million vs 690 million) support that position. For any user whose primary concern is OpenCL performance or general compute acceleration, the 930M is the clear choice.

The AMD Radeon R5 M420 is not without its own strengths, but they are not reflected in benchmark wins. It offers double the memory capacity (4 GB vs 2 GB), higher memory bandwidth (16.00 GB/s vs 12.80 GB/s), and higher theoretical pixel and texture rates. Those specs suggest it could handle certain memory-heavy tasks or rasterization workloads better than the 930M, but the database contains no benchmark that confirms that advantage. In the only test where both chips appear, AMD loses decisively.

For buyers facing these two options, the decision hinges on workload. If the task is OpenCL compute, pick the NVIDIA GeForce 930M. If the task requires more VRAM capacity, such as holding larger textures or datasets, the AMD Radeon R5 M420 provides that headroom, but with lower measured compute performance. The data does not support choosing AMD for raw speed. Both are end-of-life mobile parts, so the context is likely a used laptop or a legacy system, and in that context the 930M offers the better measured performance profile.