AMD Radeon R9 M380 vs AMD Radeon RX 5500M Comparison

FAQ

Q: How much faster is the AMD Radeon RX 5500M than the AMD Radeon R9 M380 in Geekbench Metal?

A: The RX 5500M scores 50,359 points in Geekbench Metal, while the R9 M380 scores 18,476 points. The RX 5500M is 63.3% ahead in this test, a substantial margin.

Q: Which GPU has the higher average benchmark score?

A: The R9 M380 has a higher average benchmark score of 15,521 points compared to the RX 5500M’s 13,356 points. This is because the R9 M380 has two recorded Geekbench results, while the RX 5500M’s average includes several lower-scoring Passmark tests.

Q: What is the process node difference between the two cards?

A: The R9 M380 is built on a 28 nm process at TSMC, while the RX 5500M uses a 7 nm process at TSMC. The RX 5500M also packs 6,400 million transistors versus 2,080 million on the R9 M380.

Q: Do both cards support DirectX 12?

A: Yes, both support DirectX 12, but with different feature levels. The R9 M380 supports DirectX 12 (12_0), while the RX 5500M supports DirectX 12 (12_1), a higher feature level.

Q: What is the memory bandwidth difference?

A: The R9 M380 has 96.00 GB/s of bandwidth with GDDR5 memory, while the RX 5500M has 224.0 GB/s with GDDR6 memory. Both use a 128 bit bus.

Q: Which card has more shading units?

A: The RX 5500M has 1,408 shading units, while the R9 M380 has 768 shading units. This is a significant architectural difference that contributes to the performance gap.

Architecture Differences

The R9 M380 uses the Strato chip based on GCN 2.0 architecture, part of the Gem System (R9 M300) generation. It is manufactured on a 28 nm process at TSMC, with 2,080 million transistors on a 160 mm² die. The transistor density is 13.0M per mm². This is a mature design from 2015, with a base clock of 900 MHz and a boost clock of 1000 MHz. Memory runs at 1500 MHz with 6 Gbps effective speed, using 4 GB of GDDR5 on a 128 bit bus.

The RX 5500M uses the Navi 14 chip based on RDNA 1.0 architecture, part of the Navi Mobile (RX 5000M) generation. It is built on a 7 nm process at TSMC, with 6,400 million transistors on a 158 mm² die. The transistor density is 40.5M per mm², more than three times higher than the R9 M380. Base clock is 1375 MHz, boost clock is 1645 MHz, and a game clock of 1448 MHz is specified. Memory runs at 1750 MHz with 14 Gbps effective speed, using 4 GB of GDDR6 on a 128 bit bus.

The RDNA 1.0 architecture introduces a new compute unit design compared to GCN 2.0. The RX 5500M has 1,408 shading units, 88 texture mapping units, and 32 ROPs. The R9 M380 has 768 shading units, 48 TMUs, and 16 ROPs. The RX 5500M also supports FP16 with a 2:1 ratio at 9.265 TFLOPS, while the R9 M380 has no FP16 data recorded. The RX 5500M supports PCIe 4.0 x8, while the R9 M380 uses PCIe 3.0 x16. The RX 5500M’s API support includes Vulkan 1.4, while the R9 M380 supports Vulkan 1.2.170. Both support DirectX 12 and OpenGL 4.6, but the RX 5500M’s DirectX feature level is 12_1 versus 12_0 for the R9 M380.

Power characteristics differ significantly. The RX 5500M has a TDP of 85 W and requires no power connectors, while the R9 M380 has no TDP recorded. The RX 5500M’s display outputs are portable device dependent, and the R9 M380 has no display output data recorded.

Head-to-Head Benchmarks

The database contains two direct head-to-head benchmark comparisons between these cards, both in Geekbench tests. The RX 5500M wins both, with the R9 M380 recording zero wins.

In Geekbench Metal, the RX 5500M scores 50,359 points against the R9 M380’s 18,476 points. The delta is 63.3% in favor of the RX 5500M. This is a massive difference, indicating that the newer RDNA architecture handles Metal workloads far more efficiently. The R9 M380’s score of 18,476 places it near the NVIDIA GeForce GTX 1080 Ti (15,548 points, 0.2% difference) and AMD Radeon Pro W5500 (15,679 points, 1% difference) in the database’s average ranking, which shows its average score is competitive with those desktop-class parts in the aggregated data.

In Geekbench OpenCL, the RX 5500M scores 38,725 points versus 12,565 points for the R9 M380. The delta is 67.6% in favor of the RX 5500M. This is an even larger margin than the Metal test. The R9 M380’s OpenCL result is notably lower than its Metal result, suggesting the older GCN architecture is less optimized for OpenCL compute workloads.

The RX 5500M also has additional benchmark data beyond the head-to-head tests. In Passmark tests, it scores 5,848 in G3D, 2,316 in GPU compute, 414 in G2D, 100 in DirectX 9, 39 in DirectX 10, 35 in DirectX 11, and 28 in DirectX 12. The R9 M380 has no Passmark results recorded. This means the RX 5500M’s average benchmark score of 13,356 is pulled down by these Passmark scores, while the R9 M380’s average of 15,521 is based only on its two Geekbench results.

Looking at the nearest rivals in the database, the R9 M380’s average score of 15,521 puts it just 0.2% below the NVIDIA GeForce GTX 1080 Ti (15,548) and 1% below the AMD Radeon Pro W5500 (15,679). It is 1.5% above the NVIDIA GeForce RTX 2060 (15,290) and 1.6% above the NVIDIA GeForce GTX 580 (15,283). The RX 5500M’s average score of 13,356 puts it essentially tied with the AMD FirePro M6100 (13,354, 0% delta), 0.1% above the AMD Radeon HD 8950M (13,376), 0.3% above the AMD Radeon Pro 555X (13,321), and 0.4% below the AMD Radeon Pro 555 (13,407).

Specification Differences

The table below lists only the fields where the two cards differ, based on the recorded data.

| Specification | AMD Radeon R9 M380 | AMD Radeon RX 5500M |

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

| Series | None | Radeon RX 5000 series |

| Chip | Strato | Navi 14 |

| Architecture | GCN 2.0 | RDNA 1.0 |

| Generation | Gem System (R9 M300) | Navi Mobile (RX 5000M) |

| Process Node | 28 nm | 7 nm |

| Transistors | 2,080 million | 6,400 million |

| Die Size | 160 mm² | 158 mm² |

| Transistor Density | 13.0M / mm² | 40.5M / mm² |

| Base Clock | 900 MHz | 1375 MHz |

| Boost Clock | 1000 MHz | 1645 MHz |

| Game Clock | None | 1448 MHz |

| Memory Clock | 1500 MHz, 6 Gbps effective | 1750 MHz, 14 Gbps effective |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bandwidth | 96.00 GB/s | 224.0 GB/s |

| Shading Units | 768 | 1,408 |

| TMUs | 48 | 88 |

| ROPs | 16 | 32 |

| Pixel Rate | 16.00 GPixel/s | 52.64 GPixel/s |

| Texture Rate | 48.00 GTexel/s | 144.8 GTexel/s |

| FP32 | 1.536 TFLOPS | 4.632 TFLOPS |

| FP16 | None | 9.265 TFLOPS (2:1) |

| TDP | None | 85 W |

| Power Connectors | None | None |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

| Display Outputs | None | Portable Device Dependent |

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

| Vulkan | 1.2.170 | 1.4 |

| Release Date | 2015-05-04 | 2019-10-06 |

| Predecessor | Solar System | Polaris Mobile |

| Successor | Polaris Mobile | None |

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

The Verdict

The benchmark data clearly favors the RX 5500M in raw performance. It wins both head-to-head tests by margins of 63.3% and 67.6%. The RX 5500M’s FP32 compute of 4.632 TFLOPS is roughly three times the R9 M380’s 1.536 TFLOPS. Its pixel rate of 52.64 GPixel/s and texture rate of 144.8 GTexel/s are similarly far ahead of the R9 M380’s 16.00 GPixel/s and 48.00 GTexel/s.

However, the average benchmark score tells a more nuanced story. The R9 M380’s average of 15,521 is higher than the RX 5500M’s 13,356, because the latter’s average includes several lower Passmark scores. The R9 M380’s average score places it near the NVIDIA GeForce GTX 1080 Ti in the database’s ranking, while the RX 5500M’s average is tied with the AMD FirePro M6100. This discrepancy suggests that the R9 M380 performs better in Geekbench workloads relative to its raw specifications, while the RX 5500M’s Passmark results are comparatively weaker.

For users prioritizing Geekbench Metal or OpenCL performance, the RX 5500M is the clear choice. Its scores are more than double the R9 M380’s in both tests. The RX 5500M also offers modern features such as FP16 support, PCIe 4.0, and a higher DirectX feature level. For users who rely on the specific workloads captured in Geekbench, the R9 M380’s higher average score might be misleading, as that average is based on only two tests versus ten for the RX 5500M.

The R9 M380’s 28 nm process and 2015 release date make it an older design, while the RX 5500M’s 7 nm process and 2019 release date represent a much newer architecture. Both are end-of-life products, but the RX 5500M has the advantage of newer technology.

Where Each One Wins

The RX 5500M wins decisively in every head-to-head benchmark where both cards have recorded data. In Geekbench Metal, it is 63.3% ahead. In Geekbench OpenCL, it is 67.6% ahead. For anyone running Metal-based applications or OpenCL compute workloads, the RX 5500M is the superior choice by a wide margin. Its 224.0 GB/s memory bandwidth, 1,408 shading units, and 4.632 TFLOPS FP32 performance provide a strong foundation for these tasks.

The R9 M380’s advantages are more subtle. It has a higher average benchmark score of 15,521 versus 13,356, driven by its strong Geekbench results. In the database’s nearest rivals ranking, the R9 M380 sits near the NVIDIA GeForce GTX 1080 Ti, a desktop flagship, while the RX 5500M sits near mobile workstation parts like the AMD FirePro M6100. For users who only consider the average score, the R9 M380 appears to be the stronger card, but this is an artifact of the different benchmark suites recorded for each.

The R9 M380 also has a larger die size relative to its transistor count, which may indicate different thermal characteristics, though no TDP data is recorded for it. The RX 5500M has a recorded TDP of 85 W and requires no power connectors, making it suitable for portable devices. The R9 M380’s power requirements are unknown from the data.

For gaming workloads, the RX 5500M’s Passmark DirectX scores provide some insight, with its highest score in DirectX 9 (100) and progressively lower scores in DirectX 10 (39), DirectX 11 (35), and DirectX 12 (28). The R9 M380 has no comparable data. The RX 5500M’s higher DirectX feature level (12_1) suggests better support for modern graphics APIs.

Ultimately, the choice depends on the workload. For Geekbench-class compute and Metal performance, the RX 5500M is overwhelmingly faster. For users who value the higher average benchmark score, the R9 M380 may appear more attractive, but that score is based on limited data. The RX 5500M is the more capable GPU in nearly every measurable performance category.