AMD Radeon R7 M380

The AMD Radeon R7 M380 is a mobile graphics solution built on the 28 nm GCN 1.0 architecture, using the Tropo chip manufactured by TSMC. It was released in May 2015 as part of the Gem System (R7 M300) generation, positioned between the Solar System predecessor and the Polaris Mobile successor. The GPU has since reached end-of-life production status. Its benchmark results place it at the 44th percentile among all GPUs, indicating a mid-range standing that is closely contested by several rivals.

Memory Subsystem

The R7 M380 is equipped with 4 GB of DDR3 memory, which is a notable quantity for a mobile part of its generation, but the memory type itself imposes significant performance limitations. The memory interface is 128 bit wide, and the memory clock runs at 1000 MHz, translating to 2 Gbps effective. This combination yields a memory bandwidth of 32.00 GB/s, a figure that is modest by contemporary standards.

For high-resolution workloads, this bandwidth becomes the primary constraint. At 1080p, the 4 GB frame buffer can accommodate most textures and assets, but the 32.00 GB/s throughput will likely bottleneck the GPU when rendering complex scenes with high-resolution textures. The data suggests that the R7 M380 is better suited to 1080p gaming with moderate settings rather than pushing into 1440p or 4K territory, where the gap between raw compute and memory throughput widens. At higher resolutions, the GPU's shading units may still process geometry, but the memory subsystem will struggle to feed data fast enough, leading to frame pacing issues and reduced overall smoothness. In contrast, the bus width and bandwidth figures indicate that the card is closer to entry-level performance in memory-bound scenarios.

Ray Tracing and Feature Set

The R7 M380 does not include dedicated ray tracing cores or tensor cores, as its GCN 1.0 architecture predates the hardware acceleration for these features. The GPU's feature set relies on its 640 shading units, 40 texture mapping units, and 16 render output units to handle rendering tasks. The pixel rate is 14.64 GPixel/s, and the texture rate is 36.60 GTexel/s, which provides a baseline for its rasterization capabilities.

In terms of API support, the R7 M380 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. This is a mixed bag: the DirectX 12 support is at the 11_1 feature level, which means it can run DX12 titles but may miss some advanced features available on newer hardware. Vulkan 1.2.170 support is relatively modern, allowing the GPU to access low-level optimizations in games that implement the Vulkan API. OpenGL 4.6 is fully supported, which is useful for legacy titles and certain productivity applications. However, without RT cores, any ray tracing effects must be computed via compute shaders, which will be significantly slower than on dedicated hardware. The absence of tensor cores also means no hardware-accelerated DLSS or similar AI-based upscaling, leaving users reliant on traditional rendering methods or software-based solutions.

Power and Cooling

The FACT PACK does not provide a TDP figure, a suggested PSU rating, or power connector requirements for the R7 M380. The absence of these specifications means that power consumption must be inferred from the chip's characteristics rather than measured data. The GPU is fabricated on a 28 nm process with 1,500 million transistors on a 123 mm² die, with a transistor density of 12.2M per mm². These figures suggest a relatively modest power draw for a mobile part, but without a TDP number, no concrete wattage can be stated.

The lack of a slot width, power connector, and suggested PSU information indicates that this GPU is designed primarily for laptop integration, where the system manufacturer determines the cooling solution and power delivery. In a mobile chassis, the thermal design is typically handled by the OEM, meaning end-users will not need to select a separate cooler or PSU. The 28 nm process node, while older, generally produced chips with moderate heat output that could be managed by standard laptop cooling solutions. The data implies that the R7 M380 is not a power-hungry part, but the absence of explicit numbers prevents a more precise assessment of its thermal envelope.

How It Compares

The R7 M380 sits in a tight cluster of GPUs with nearly identical average benchmark scores. Its nearest rival is the NVIDIA GeForce GTX TITAN BLACK, which posts an average score of 9385, a mere 0.8% higher than the R7 M380's 9313. This is a surprisingly small gap given the vastly different market positions of the two cards, but the score reflects a single benchmark metric.

The NVIDIA Quadro K5000 is the next competitor, with an average score of 9235, which is 0.8% lower than the R7 M380. The margin is negligible, indicating that the R7 M380 holds a slight edge over this professional workstation GPU in the measured workload. The performance difference is unlikely to be perceptible in real-world usage.

The AMD Radeon RX 5500M follows closely, scoring 9233, which is 0.9% below the R7 M380. This is notable because the RX 5500M is a much newer part, yet the benchmark data shows the older R7 M380 slightly ahead, suggesting that the gap in this specific test is not representative of broader gaming performance.

Finally, the AMD Radeon Vega 8, an integrated graphics solution, scores 9215, which is 1.1% lower than the R7 M380. The R7 M380's discrete design gives it a small advantage, but the margin is slim enough that it would not be noticeable in most applications.

Benchmark Performance

The single benchmark available for the R7 M380 is the Geekbench OpenCL test, where it achieves a score of 9313. This score, which also serves as the average benchmark score, places the GPU at the 44th percentile among all GPUs, meaning that 56% of the GPUs in the database perform better. The percentile ranking indicates that the R7 M380 is in the lower half of the performance distribution, which aligns with its positioning as an entry-to-mid-range mobile part.

Comparing the raw score to the nearest rivals reveals how tightly grouped these GPUs are. The R7 M380 is 0.8% behind the GTX TITAN BLACK, a difference of 72 points. It is 0.8% ahead of the Quadro K5000, a margin of 78 points. Against the RX 5500M, the R7 M380 leads by 0.9%, or 80 points. The largest gap is against the Vega 8, where the R7 M380 is 1.1% ahead, a difference of 98 points. These deltas are all within a narrow band of roughly ±1%, indicating that in this particular OpenCL workload, all five GPUs deliver effectively identical performance. The practical implication is that the R7 M380's compute performance is indistinguishable from its closest rivals in this test, and any real-world differences would depend on drivers, thermals, and the specific application.

Who Should Consider It

Based on the benchmark data, the R7 M380 is suited for users who prioritize basic compute tasks and light gaming at 1080p resolution. The 44th percentile ranking and the 32.00 GB/s memory bandwidth suggest that the GPU can handle older titles and less demanding esports games at medium settings, but it will struggle with modern AAA releases at high settings. The 4 GB VRAM is adequate for 1080p textures, but the DDR3 memory type and relatively low bandwidth will limit performance in memory-intensive scenarios.

For users who play at 1080p and are willing to lower settings to achieve playable frame rates, the R7 M380 is a viable option. The 640 shading units provide enough compute throughput for basic rendering, and the DirectX 12 (11_1) support ensures compatibility with most contemporary games. However, users who plan to engage in ray tracing or use AI-based upscaling will find the hardware lacking, as there are no dedicated cores for these features.

The R7 M380 is not recommended for high-resolution gaming beyond 1080p, as the memory subsystem will become a bottleneck. Users with 1440p or 4K displays should look elsewhere, as the bandwidth figures are too low to sustain smooth performance at those resolutions. Similarly, professionals who require workstation-grade compute performance may find the R7 M380 adequate for light workloads, but the slim margin over the Quadro K5000 suggests that neither card offers a significant advantage in OpenCL tasks. The R7 M380 is best suited for budget-conscious mobile users who need a discrete GPU for everyday computing, light photo editing, and casual gaming at 1080p with modest settings.