AMD Radeon R9 M380 vs NVIDIA GeForce RTX 3060 Ti Comparison

The GeForce RTX 3060 Ti and the Radeon R9 M380 represent two entirely different eras of GPU design, separated by five years of architectural evolution. The data shows a decisive performance gap, but the R9 M380’s profile as a mobile-oriented part from the GCN 2.0 generation makes it a curious comparison piece rather than a direct competitor. This analysis breaks down the benchmark results, architectural chasms, and practical use cases strictly from the provided data.

Head-to-Head Benchmarks

The only shared benchmark between the two cards is Geekbench OpenCL, and the results are not close. The RTX 3060 Ti scores 78,927 points, while the R9 M380 manages 12,565 points. That is a delta of 528.1% in favor of the NVIDIA card. To put it plainly, the RTX 3060 Ti delivers over six times the OpenCL compute performance of the older AMD part. This is a generational stomping, not a competitive race.

Looking at the broader average benchmark scores reinforces this gap. The RTX 3060 Ti holds an average benchmark score of 16,129 across its suite of tests, placing it in the 59th percentile of all GPUs. The R9 M380’s average score is 15,521, which lands it in the 58th percentile. The percentile difference appears small because the R9 M380’s average is pulled from only two tests (Geekbench Metal and OpenCL), whereas the RTX 3060 Ti’s average comes from ten tests including Passmark suites and 3DMark. When comparing the single overlapping test, the RTX 3060 Ti’s lead is overwhelming.

The RTX 3060 Ti’s nearest rivals in the data are the AMD Radeon RX 9060 (0.7% faster), the AMD Radeon Pro 5600M (1.4% faster), and the AMD Radeon RX 5700 XT (1.4% faster). It also edges out the AMD Radeon R9 370X by 1.7%. These are all close calls, suggesting the RTX 3060 Ti sits in a competitive mid-range bracket for its generation. Meanwhile, the R9 M380’s nearest rivals include the NVIDIA GeForce GTX 1080 Ti (0.2% faster), the AMD Radeon Pro W5500 (1% faster), and it leads the NVIDIA GeForce RTX 2060 by 1.5% and the GTX 580 by 1.6%. That the R9 M380 trades blows with a GTX 1080 Ti in average score is a statistical artifact of the limited test set, not a reflection of real-world parity.

Architecture Differences

The RTX 3060 Ti is built on NVIDIA’s Ampere architecture, using the GA104 chip fabricated on Samsung’s 8 nm process. It packs 17,400 million transistors into a 392 mm² die, yielding a transistor density of 44.4M per mm². The R9 M380 uses AMD’s GCN 2.0 architecture on the Strato chip, built on TSMC’s 28 nm node. It holds just 2,080 million transistors on a 160 mm² die, with a density of 13.0M per mm². The process node difference alone explains a significant portion of the performance chasm.

Core counts tell a similar story. The RTX 3060 Ti has 4,864 shading units, 152 texture mapping units, and 80 ROPs. It also features 38 dedicated ray tracing cores and 152 tensor cores, enabling hardware-accelerated ray tracing and AI workloads. The R9 M380 has 768 shading units, 48 TMUs, and 16 ROPs, with no ray tracing or tensor cores present. The NVIDIA card’s FP32 throughput is 16.20 TFLOPS, while the AMD card manages 1.536 TFLOPS. The RTX 3060 Ti also supports FP16 at a 1:1 ratio (16.20 TFLOPS), whereas the R9 M380’s FP16 capability is listed as null, indicating no comparable hardware path.

Memory subsystems diverge sharply. The RTX 3060 Ti uses 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s of bandwidth. The R9 M380 uses 4 GB of GDDR5 on a 128-bit bus, with 96.00 GB/s. That is a 352 GB/s bandwidth advantage for NVIDIA. The RTX 3060 Ti’s memory clock is 1750 MHz (14 Gbps effective), while the R9 M380 runs at 1500 MHz (6 Gbps effective). Pixel and texture rates follow suit: the RTX 3060 Ti hits 133.2 GPixel/s and 253.1 GTexel/s, versus 16.00 GPixel/s and 48.00 GTexel/s for the R9 M380.

The RTX 3060 Ti is a desktop card with a 200 W TDP, dual-slot cooler, and a 1x 12-pin power connector, requiring a 550 W suggested PSU. The R9 M380 has no TDP, slot width, or power connector data listed, which aligns with its mobile-oriented "Gem System" generation. The NVIDIA card uses PCIe 4.0 x16, while the AMD card uses PCIe 3.0 x16. API support also differs: the RTX 3060 Ti supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the R9 M380 tops out at DirectX 12 (12_0) and Vulkan 1.2.170. Both support OpenGL 4.6.

FAQ

Q: Which card has higher raw compute performance?

A: The RTX 3060 Ti delivers 16.20 TFLOPS of FP32 compute, compared to 1.536 TFLOPS for the R9 M380. In the Geekbench OpenCL test, the RTX 3060 Ti scores 78,927 versus 12,565 for the R9 M380, a 528.1% advantage.

Q: Does the R9 M380 support ray tracing?

A: No. The R9 M380 has no ray tracing cores listed. The RTX 3060 Ti includes 38 dedicated RT cores, enabling hardware-accelerated ray tracing in supported titles.

Q: How do their memory bandwidth figures compare?

A: The RTX 3060 Ti has 448.0 GB/s of bandwidth from 8 GB of GDDR6 on a 256-bit bus. The R9 M380 has 96.00 GB/s from 4 GB of GDDR5 on a 128-bit bus. The NVIDIA card has over four times the bandwidth.

Q: Which card is newer and what process node does each use?

A: The RTX 3060 Ti was released on 2020-11-30 and uses Samsung’s 8 nm process. The R9 M380 was released on 2015-05-04 and uses TSMC’s 28 nm process. The RTX 3060 Ti also has a much higher transistor density at 44.4M per mm² versus 13.0M per mm².

Q: Are these cards comparable in their likely use cases?

A: Not really. The RTX 3060 Ti is a dual-slot desktop card with a 200 W TDP and PCIe 4.0 support, designed for modern gaming and compute. The R9 M380 lacks published power or slot data, and its GCN 2.0 architecture with only 4 GB of VRAM targets older, lower-resolution workloads. The benchmark data shows a 528.1% performance gap in OpenCL.

Q: What is the average benchmark score difference?

A: The RTX 3060 Ti has an average benchmark score of 16,129 across ten tests, while the R9 M380 has an average of 15,521 across two tests. The RTX 3060 Ti ranks in the 59th percentile of all GPUs, and the R9 M380 ranks in the 58th percentile.

The Verdict

The data is unambiguous: the RTX 3060 Ti is the far more capable card. Its 528.1% lead in the shared OpenCL benchmark dwarfs any other consideration. The RTX 3060 Ti offers 8 GB of GDDR6 memory versus 4 GB of GDDR5, a 256-bit bus versus 128-bit, and 16.20 TFLOPS versus 1.536 TFLOPS of FP32 compute. It also brings modern features like ray tracing cores and tensor cores, which the R9 M380 lacks entirely. For anyone choosing between these two, the RTX 3060 Ti is the only sensible option for gaming, content creation, or compute workloads.

The R9 M380’s only statistical advantage is its percentile ranking, which is effectively tied at 58th versus 59th. That tie is misleading because the R9 M380’s average score is based on just two benchmarks, one of which (Geekbench Metal) is not tested on the NVIDIA card. In practice, the R9 M380 is a legacy part from the 28 nm era, best suited for basic display output in an old laptop. It is not a competitor to the RTX 3060 Ti in any meaningful sense.

Specification Differences

• Process Node: 8 nm (Samsung) vs 28 nm (TSMC)
• Transistors: 17,400 million vs 2,080 million
• Die Size: 392 mm² vs 160 mm²
• Transistor Density: 44.4M / mm² vs 13.0M / mm²
• Base Clock: 1410 MHz vs 900 MHz
• Boost Clock: 1665 MHz vs 1000 MHz
• Memory Size: 8 GB vs 4 GB
• Memory Type: GDDR6 vs GDDR5
• Memory Bus: 256 bit vs 128 bit
• Memory Bandwidth: 448.0 GB/s vs 96.00 GB/s
• Shading Units: 4864 vs 768
• Texture Mapping Units: 152 vs 48
• ROPs: 80 vs 16
• RT Cores: 38 vs None
• Tensor Cores: 152 vs None
• Pixel Rate: 133.2 GPixel/s vs 16.00 GPixel/s
• Texture Rate: 253.1 GTexel/s vs 48.00 GTexel/s
• FP32 Performance: 16.20 TFLOPS vs 1.536 TFLOPS
• TDP: 200 W vs Not listed
• Bus Interface: PCIe 4.0 x16 vs PCIe 3.0 x16
• DirectX Support: 12 Ultimate (12_2) vs 12 (12_0)
• Vulkan Support: 1.4 vs 1.2.170
• Release Date: 2020-11-30 vs 2015-05-04
• Launch MSRP: 399 USD vs Not listed

Where Each One Wins

The RTX 3060 Ti wins every measurable category in the data. It is faster in compute, has more memory and bandwidth, supports newer APIs, and includes dedicated hardware for ray tracing and tensor operations. Its 16.20 TFLOPS FP32 throughput and 448.0 GB/s bandwidth make it suitable for 1440p gaming, GPU-accelerated rendering, and machine learning inference. The 38 RT cores allow for hardware ray tracing in supported games, a feature the R9 M380 cannot offer at any setting.

The R9 M380’s only win is its lower transistor count and die size, which historically would imply lower power draw, but no TDP data is provided to confirm that. Its 2,080 million transistors on a 160 mm² die mean it is a smaller, simpler chip. In a purely legacy context, it could drive a 1080p display for basic tasks, but its 4 GB VRAM and 96.00 GB/s bandwidth severely limit modern workloads. The R9 M380’s nearest rival list includes the GTX 1080 Ti and RTX 2060, but those are average score artifacts, not real-world comparisons.

For a builder in 2024 or later, the RTX 3060 Ti is the clear pick. For anyone still running an R9 M380, the data suggests an upgrade is overdue. The performance gap is not incremental; it is transformational.