AMD Radeon R9 M380 vs NVIDIA RTX A2000 Mobile Comparison

FAQ

Q: How do the two GPUs compare in raw compute performance?

A: The NVIDIA RTX A2000 Mobile delivers 8.637 TFLOPS of FP32 performance, while the AMD Radeon R9 M380 delivers 1.536 TFLOPS. That is a 5.6x difference in raw shader throughput.

Q: Which GPU has more memory bandwidth?

A: The RTX A2000 Mobile has 192.0 GB/s of bandwidth, exactly double the 96.00 GB/s of the R9 M380. Both use a 128-bit bus, but the A2000 uses GDDR6 at 12 Gbps effective, while the M380 uses GDDR5 at 6 Gbps effective.

Q: Are there any benchmark scores they share?

A: Yes, both have a Geekbench OpenCL score. The RTX A2000 Mobile scores 56,518, while the R9 M380 scores 12,565, a delta of -77.8% for the AMD part.

Q: What is the transistor density difference?

A: The RTX A2000 Mobile packs 43.5 million transistors per mm², while the R9 M380 has 13.0 million per mm². The A2000's 8 nm Samsung process enables 8,700 million transistors on a 200 mm² die, versus 2,080 million on a 160 mm² die for the M380.

Q: Do both support the same DirectX version?

A: No. The RTX A2000 Mobile supports DirectX 12 Ultimate (12_2), while the R9 M380 supports DirectX 12 (12_0). The A2000 also adds Vulkan 1.4 support versus Vulkan 1.2.170 for the M380.

Q: Which GPU has the higher overall percentile ranking?

A: The R9 M380 sits at the 58th percentile of all GPUs, while the RTX A2000 Mobile is at the 55th percentile. Despite the A2000's higher raw scores in individual tests, its average benchmark score of 13,021 is lower than the M380's 15,521 average.

Architecture Differences

The two GPUs come from completely different architectural generations. The AMD Radeon R9 M380 is built on GCN 2.0 architecture, using the Strato chip from the R9 M300 series. The NVIDIA RTX, by contrast, uses the Ampere architecture with the GA107 chip. The M380 belongs to the "Gem System (R9 M300)" generation, while the A2000 belongs to the "Ampere-MW (Ax000)" generation, released roughly six years later. The M380 launched in May 2015, and the A2000 in April 2021.

The manufacturing processes differ substantially. The M380 uses a 28 nm TSMC process with 2,080 million transistors on a 160 mm² die. The A2000 uses an 8 nm Samsung process with 8,700 million transistors on a 200 mm² die, giving it a transistor density of 43.5M per mm², more than triple the density of the M380.

Memory architecture is a major departure. Both use a 128-bit memory bus and 4 GB of memory, but the M380 uses GDDR5 with 96.00 GB/s bandwidth, whereas the A2000 uses GDDR6 at 192.0 GB/s bandwidth. The A2000's memory runs at 12 Gbps effective, versus 6 Gbps effective. The A2000 also adds dedicated hardware that the M380 lacks entirely: 20 ray tracing cores and 20 RT cores, alongside 80 tensor cores. The A2000 has 2560 shading units, 80 TMUs, and 48 ROPs, dwarfing the M380's 768 shading units, 48 TMUs, and 16 ROPs.

The A2000 supports PCIe 4.0 x16, while the M380 is limited to PCIe 8 x16, and the A2000's API support includes DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.2 for the A2000. The M380 tops out at DirectX 12 (12_0), Vulkan 1.2.170, OpenGL 4.6, and PCIe 3.0 x16. Both are end-of-life products, and the M380 lists no TDP data, while the A2000 lists a 95 W TDP with no power connectors required.

The A2000's pixel rate reaches 80.98 GPixel/s, texture rate of 135.0 GTexel/s, FP32 of 8.637 TFLOPS, and FP16 of 8.637 TFLOPS (1:1). The M380 manages 16.00 GPixel/s pixel rate, 48.00 GTexel/s texture rate, and 1.536 TFLOPS FP32, with no FP16 figure recorded. The A2000's display outputs are listed as "portable device dependent," while the M380 has no display output data.

Head-to-Head Benchmarks

The two GPUs share only one benchmark test in the database: Geekbench OpenCL. The NVIDIA RTX A2000 Mobile wins this head-to-head by a decisive margin. The A2000 scores 56,518, while the R9 M380 scores 12,565, giving a delta of -77.8% for the AMD part. Put in perspective, the M380 achieves only about 22.2% of the A2000's OpenCL score. This is not a close contest; the A2000's Ampere architecture with its 8,700 million transistors and 8.637 TFLOPS FP32 simply overwhelms the older GCN 2.0 chip.

The R9 M380 has no wins in any head-to-head benchmark in the database, while the A2000 has one win. The broader benchmark data reinforces the gap. The A2000's average benchmark score of 13,021 is actually lower than the M380's 15,521, but this is because the A2000 is tested across a wider range of workloads. The A2000's Passmark G3D score of 9,611, its Passmark DirectX 11 score of 68, its Passmark DirectX 12 score of 47, and its Passmark GPU compute score of 4,334, all far exceed the M380's single other score outside OpenCL of 12,565.

The M380's nearest rivals include the NVIDIA GeForce GTX 1080 Ti at 15,548 average score, just 0.2% higher, and the AMD Radeon Pro W5500 at 15,679 average score, 1% higher. The RTX 2060 scores 15,290 average, 1.5% higher than the M380, and the GTX 580 scores 15,283 average, 1.6% higher. The A2000's nearest rivals include the AMD Radeon 660M at 13,812 average score, just 0.1% higher, the AMD Radeon RX 570X at 13,871 average, 0.4% higher, the AMD Radeon RX 7900 XT at 13,745 average, 0.6% higher, and the NVIDIA Tesla K10 at 14,029 average, 1.5% lower.

Specification Differences

The specification sheet separates the two GPUs clearly along architectural lines. The AMD Radeon R9 M380 uses the GCN 4.0 architecture, 28 nm process, TSMC foundry, 2,080 million transistors, and a 160 mm² die size with 13.0M/mm² transistor density. The NVIDIA RTX A2000 Mobile uses Ampere, 8 nm process, Samsung foundry, 8,700 million transistors, and a 200 mm² die with 43.5M/mm² density.

Clock speeds differ significantly. The M380 has a base clock of 900 MHz and boost of 1000 MHz. The A2000 has a base of 1215 MHz and a boost of 1687 MHz. Memory clocks also differ: both list a memory clock of 1500 MHz, but the M380's effective rate is 6 Gbps, while the A2000's is 12 Gbps. The M380 has 4 GB of GDDR5 on a 128-bit bus with 96.00 GB/s bandwidth. The A2000 has 4 GB of GDDR6 on a 128-bit bus with 192.0 GB/s bandwidth.

The compute resources are starkly different. The M380 has 768 shading units, 48 TMUs, and 16 ROPs. The A2000 has 2560 shading units, 80 TMUs, and 48 ROPs, plus 20 RT cores and 80 tensor cores. The A2000's pixel rate is 80.98 GPixel/s, texture rate 135.0 GTexel/s, FP32 8.637 TFLOPS, and FP16 8.637 TFLOPS (1:1). The M380's pixel rate is 16.00 GPixel/s, texture rate 48.00 GTexel/s, FP32 1.536 TFLOPS, and no FP16 figure.

The A2000 lists a 95 W TDP, an IGP slot width, no power connectors, and PCIe 4.0 x16. The M380 lists no TDP, no slot width, no power connectors, and PCIe 3.0 x16. The A2000's display outputs are "portable device dependent," while the M380 has none recorded. The A2000 supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6. The M380 supports DirectX 12 (12_0), Vulkan 1.2.170, and OpenGL 4.6.

Release timing also differs: the M380 launched in May 2015, the A2000 in April 2021. The M380's predecessor is listed as "Solar System" and successor as "Polaris Mobile." The A2000's predecessor is "Quadro Turing-M" and successor is "Ada-MW."

Where Each One Wins

The NVIDIA RTX A2000 Mobile wins in every raw compute scenario captured by the database. Its Geekbench OpenCL score of 56,518 is more than four times the M380's 12,565. The A2000's FP32 throughput of 8.637 TFLOPS, texture rate of 135.0 GTexel/s, and pixel rate of 80.98 GPixel/s indicate workloads that stress shader math, texture fill, and rasterization will see a massive advantage. The A2000 also carries 20 RT cores and 80 tensor cores, which the M380 lacks entirely, so any workload using ray tracing or tensor operations is exclusive to the A2000. Its 192.0 GB/s bandwidth is double the M380's, benefiting memory-heavy tasks. The A2000 also supports DirectX 12 Ultimate and Vulkan 1.4, while the M380 is limited to DirectX 12 (12_0) and Vulkan 1.2.170.

The AMD Radeon R9 M380 wins on the aggregate benchmark average. Its average benchmark score of 15,521 is higher than the A2000's 13,021, and its 58th percentile ranking across all GPUs beats the A2000's 55th percentile. In the nearest rival comparisons, the M380 sits just 0.2% below the GeForce GTX 1080 Ti, while the A2000 sits just 0.1% above the AMD Radeon 660M, a much lower-tier comparison. The M380's 2,080 million transistors on a 160 mm² die is far more efficient in terms of die size per performance in its benchmark pool, though the A2000 has a higher transistor density of 43.5M/mm².

The Verdict

The data points to a clear performance hierarchy, but with a nuance in aggregate scores. If the workload is OpenCL compute, the NVIDIA RTX A2000 Mobile is the definitive choice: its 56,518 Geekbench OpenCL score utterly eclipses the R9 M380's 12,565, a 77.8% gap. The A2000 also brings modern features like ray tracing, tensor cores, 8.637 TFLOPS FP32, and 192.0 GB/s bandwidth, all of which the M380 cannot match. Any user selecting a GPU for modern API support, high-bandwidth tasks, or ray-traced workloads should pick the A2000.

The AMD Radeon R9 M380 still holds a place for users prioritizing the database's aggregate ranking. Its average benchmark score of 15,521 beats the A2000's 13,021, and its 58th percentile placement is above the A2000's 55th. The M380's nearest rivals include the GTX 1080 Ti at essentially parity, while the A2000's nearest rival is the Radeon 660M. This suggests the M380's overall score profile is more consistent across the tests it runs, even though its single shared test with the A2000 is a loss.

The choice depends on the workload. For modern compute, ray tracing, tensor acceleration, and high memory bandwidth, the RTX A2000 Mobile is the clear winner. For users looking at the database's overall benchmark average and percentile ranking, the R9 M380 appears stronger. The A2000 is the more capable GPU by raw specifications, but the M380's aggregate score suggests it holds up better in the broader benchmark pool. The A2000's 95 W TDP and PCIe 4.0 interface also indicate a more modern platform fit, while the M380's lack of TDP data and PCIe 3.0 interface reflect its older design. Ultimately, the RTX A2000 Mobile is the superior choice for any performance-critical task, while the R9 M380 remains viable only in contexts where the aggregate benchmark profile matters more than raw compute power.