AMD Radeon R9 M380 vs AMD Radeon RX 560X Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon RX 560X has an average benchmark score of 18626, while the AMD Radeon R9 M380 scores 15521. That puts the RX 560X roughly 20% higher in overall measured performance.

Q: How do the two GPUs compare in the only direct head-to-head test available?

A: In Geekbench OpenCL, the RX 560X scores 17020 versus 12565 for the R9 M380. The recorded delta is 35.5% in favor of the RX 560X.

Q: Which GPU sits higher in the database percentile ranking?

A: The RX 560X ranks in the 62nd percentile among all GPUs, while the R9 M380 ranks in the 58th percentile. The gap is small but consistent with the average score difference.

Q: What architecture generation does each GPU use?

A: The RX 560X uses GCN 4.0 with a Polaris 21 chip, while the R9 M380 uses GCN 2.0 with a Strato chip. The RX 560X is built on a 14 nm process, the R9 M380 on 28 nm.

Q: Do both GPUs support DirectX 12 and OpenGL 4.6?

A: Yes, both support DirectX 12 (12_0) and OpenGL 4.6. For Vulkan, the RX 560X supports version 1.3, while the R9 M380 supports version 1.2.170.

Q: What is the memory configuration of each card?

A: Both have 4 GB of GDDR5 on a 128-bit bus. The RX 560X has a bandwidth of 112.0 GB/s, while the R9 M380 has 96.00 GB/s. The RX 560X memory clock is listed at 1750 MHz (7 Gbps effective), versus 1500 MHz (6 Gbps effective) for the R9 M380.

Architecture Differences

The architectural divide between these two GPUs is substantial. The RX 560X is built on GCN 4.0, the fourth iteration of AMD's Graphics Core Next design, implemented with a Polaris 21 chip on a 14 nm process from GlobalFoundries. The R9 M380 uses the older GCN 2.0 architecture, based on the Strato chip and fabricated by TSMC on a 28 nm process. The node difference alone explains much of the performance and efficiency gap: 14 nm allows for significantly higher transistor density.

Transistor counts reflect this generational shift. The RX 560X packs 3,000 million transistors into a die size of 123 mm², yielding a density of 24.4 million transistors per square millimeter. The R9 M380 has 2,080 million transistors on a larger 160 mm² die, resulting in just 13.0 million transistors per square millimeter. The RX 560X achieves nearly double the density, which translates directly into more compute resources at a smaller physical footprint.

Core counts diverge sharply. The RX 560X has 1024 shading units, 64 texture mapping units, and 16 ROPs. The R9 M380 has 768 shading units, 48 TMUs, and also 16 ROPs. The RX 560X therefore has 33% more shaders and TMUs, while ROP count stays equal. Clock speeds follow the same pattern: the RX 560X runs at a base of 1175 MHz with a boost of 1275 MHz, while the R9 M380 is at 900 MHz base and 1000 MHz boost. This means the RX 560X has both more cores and higher clocks, a compound advantage.

Memory architecture is similar in width but not in speed. Both use 4 GB of GDDR5 on a 128-bit bus, but the RX 560X memory runs at 1750 MHz (7 Gbps effective) versus 1500 MHz (6 Gbps effective) for the R9 M380. The resulting bandwidth difference is 112.0 GB/s versus 96.00 GB/s, a roughly 17% advantage for the newer card. Pixel rate and texture rate scale accordingly: the RX 560X delivers 20.40 GPixel/s and 81.60 GTexel/s, while the R9 M380 manages 16.00 GPixel/s and 48.00 GTexel/s.

The FP32 compute throughput tells the same story. The RX 560X is rated at 2.611 TFLOPS, while the R9 M380 is at 1.536 TFLOPS. That is a 70% advantage in raw floating-point performance. The RX 560X also supports FP16 at a 1:1 ratio (2.611 TFLOPS), while the R9 M380 has no listed FP16 capability.

Interface and power delivery differ as well. The RX 560X uses PCIe 3.0 x8, while the R9 M380 uses PCIe 3.0 x16. The RX 560X is a dual-slot card with no external power connectors and a suggested PSU of 250 W. The R9 M380 has no listed TDP, slot width, power connectors, or PSU recommendation in the database. The RX 560X also has a defined physical length of 170 mm (6.7 inches), while the R9 M380 dimensions are not recorded.

API support is nearly identical on the DirectX and OpenGL fronts, but Vulkan compatibility differs slightly: the RX 560X supports Vulkan 1.3, while the R9 M380 supports Vulkan 1.2.170. Both are end-of-life products. The RX 560X was released in April 2018, with the R9 M380 released in May 2015. The RX 560X predecessor is listed as Polaris and its successor as Vega, while the R9 M380 predecessor is Solar System and successor is Polaris Mobile.

Head-to-Head Benchmarks

The only direct benchmark comparison recorded between these two GPUs is Geekbench OpenCL. In that test, the RX 560X scores 17020 against 12565 for the R9 M380, a delta of 35.5% in favor of the RX 560X. That is the single largest measured gap between the two, and it reflects both the core count and clock speed advantages discussed above.

Looking at the broader benchmark context, the average scores confirm the head-to-head result. The RX 560X averages 18626 across its recorded tests, which include Geekbench OpenCL at 17020 and Geekbench Vulkan at 20231. The R9 M380 averages 15521, based on Geekbench Metal at 18476 and Geekbench OpenCL at 12565. The OpenCL score for the R9 M380 is notably its weakest recorded result, while its Metal score is actually higher than the RX 560X OpenCL score. However, the RX 560X Vulkan score of 20231 is the single highest recorded number in this comparison.

The delta of 35.5% in OpenCL is consistent with the raw specification differences. The RX 560X has 33% more shading units, a 30% higher boost clock, and 17% more memory bandwidth. The combination of these factors produces a performance margin that is larger than any single spec difference, which is typical when multiple resources scale together.

The nearest rival data puts these scores in perspective. The RX 560X average of 18626 sits within 0.3% of the AMD Radeon Pro 5700 XT (18685), within 0.5% of the AMD FirePro D500 (18533), and within 0.9% of the NVIDIA GeForce RTX 2070 (18789). The R9 M380 average of 15521 is within 0.2% of the NVIDIA GeForce GTX 1080 Ti (15548), within 1% of the AMD Radeon Pro W5500 (15679), and within 1.5% of the NVIDIA GeForce RTX 2060 (15290). Notably, the R9 M380 is competitive with much newer high-end GPUs in this aggregate metric, which suggests its Metal benchmark score carries significant weight in the average.

The wins tally is one to zero in favor of the RX 560X. There is no benchmark category where the R9 M380 records a higher score than the RX 560X in the head-to-head data. The only question is whether the margin would hold across a wider test suite, but based on the recorded measurements, the RX 560X is the clear winner.

Specification Differences

The two GPUs differ across nearly every recorded specification field. The most consequential differences are process node, architecture, core counts, and clocks.

• Process node: 14 nm (GlobalFoundries) for the RX 560X, versus 28 nm (TSMC) for the R9 M380.
• Transistors: 3,000 million versus 2,080 million.
• Die size: 123 mm² versus 160 mm².
• Transistor density: 24.4M / mm² versus 13.0M / mm².
• Architecture: GCN 4.0 versus GCN 2.0.
• Chip: Polaris 21 versus Strato.
• Base clock: 1175 MHz versus 900 MHz.
• Boost clock: 1275 MHz versus 1000 MHz.
• Memory clock: 1750 MHz (7 Gbps effective) versus 1500 MHz (6 Gbps effective).
• Memory bandwidth: 112.0 GB/s versus 96.00 GB/s.
• Shading units: 1024 versus 768.
• Texture mapping units: 64 versus 48.
• Pixel rate: 20.40 GPixel/s versus 16.00 GPixel/s.
• Texture rate: 81.60 GTexel/s versus 48.00 GTexel/s.
• FP32 throughput: 2.611 TFLOPS versus 1.536 TFLOPS.
• FP16 throughput: 2.611 TFLOPS (1:1) versus not listed.
• TDP: 75 W versus not listed.
• Slot width: Dual-slot versus not listed.
• Power connectors: None versus not listed.
• Suggested PSU: 250 W versus not listed.
• Bus interface: PCIe 3.0 x8 versus PCIe 3.0 x16.
• Display outputs: 1x DVI, 1x HDMI 2.0b, 1x DisplayPort 1.4a versus not listed.
• Vulkan version: 1.3 versus 1.2.170.
• Dimensions: 170 mm (6.7 inches) length versus not listed.
• Release date: April 2018 versus May 2015.

The ROP count is identical at 16, as is memory size (4 GB), memory type (GDDR5), memory bus width (128 bit), DirectX support (12_0), OpenGL support (4.6), and production status (end-of-life).

The Verdict

The data points decisively toward the AMD Radeon RX 560X. It wins the only direct head-to-head benchmark by 35.5%, holds a 20% higher average benchmark score, and ranks four percentile points higher in the database. Its architectural advantages are systemic: more shading units, more texture units, higher clocks, faster memory, and a more advanced process node.

The RX 560X is the choice for any workload that stresses OpenCL compute, based on the 17020 versus 12565 score. Its Vulkan score of 20231 also indicates strong API-level performance, though no direct R9 M380 Vulkan result is recorded for comparison. Users who prioritize raw FP32 throughput will find the RX 560X delivers 2.611 TFLOPS versus 1.536 TFLOPS for the R9 M380, a 70% gap.

The R9 M380 is not without merit in the aggregate. Its average score of 15521 places it within 0.2% of the NVIDIA GeForce GTX 1080 Ti, which is a remarkable result for a 2015 mobile GPU. The Geekbench Metal score of 18476 suggests that in Metal-based workloads, the R9 M380 can outperform the RX 560X OpenCL result, though not the RX 560X Vulkan result. For users confined to Apple's Metal API, the R9 M380 may be a viable option.

However, the overall direction is clear. The RX 560X is newer by nearly three years, built on a more efficient process, and dominates in every recorded compute metric. The R9 M380 wins no benchmark categories in the head-to-head data. The verdict for most users is straightforward: the RX 560X offers measurably higher performance across the board, with the only caveat being that the R9 M380 has a strong showing in Metal-specific tests. Those who require Metal compatibility should consider the R9 M380, but for OpenCL and Vulkan workloads, the RX 560X is the superior part based on the recorded measurements.