AMD Radeon R7 M380 vs NVIDIA GeForce GTX 880M Comparison

The Verdict

The benchmark data presents a clear, though narrowly scoped, picture. In the single recorded head-to-head test, Geekbench OpenCL, the AMD Radeon R7 M380 decisively outperforms the NVIDIA GeForce GTX 880M. The AMD part scores 9313 points against 5622 for the NVIDIA part, a 65.7% advantage. This places the R7 M380 in a higher performance tier within the database. Its percentile ranking of 46 against all GPUs is ahead of the GTX 880M's 42nd percentile. The R7 M380 also averages closer to its nearest rivals, with scores of 9302, 9294, 9273, and 9221 for the GTX 850M, GTX 465, GTX 960, and Vega 8 respectively, all within a 1% delta. The GTX 880M, by contrast, is separated from its nearest rivals by a significant 0% to 0.5% delta, but those rivals are clustered around its lower average.

Who should pick which? For any task that relies on the OpenCL compute workload, the data indicates the AMD Radeon R7 M380 is the superior choice. It provides a substantial performance lead in compute throughput. The NVIDIA GeForce GTX 880M, however, has its own strengths. Its average benchmark score is 8040, and it has a separate Metal benchmark score of 10458, implying strong performance in that specific API. The GTX 880M also offers double the memory capacity and a wider memory bus, which are qualitative advantages for certain workloads. The choice depends on the user's priority: maximum raw compute performance per the OpenCL test favors the AMD, while the NVIDIA part may be preferable for applications that can leverage its specific API support and larger memory pool. The data suggests that for pure number-crunching, the R7 M380 is the winner, but the GTX 880M is not without its own merits in other measured areas.

Architecture Differences

The two mobile GPUs come from different architectural generations and design philosophies. The AMD Radeon R7 M380 is built on the GCN 1.0 architecture using the Tropo chip, while the NVIDIA GeForce GTX 880M uses the Kepler architecture with the GK104 chip. Both are manufactured on a 28 nm process at TSMC, but that is where the similarity ends.

The most striking difference is in the scale of the silicon. The GTX 880M's GK104 chip is a much larger and more complex design, housing 3,540 million transistors on a 294 mm² die. The R7 M380's Tropo chip is smaller, with 1,500 million transistors on a 123 mm² die. Interestingly, the transistor density is nearly identical, with the AMD part at 12.2M / mm² and the NVIDIA part at 12.0M / mm². This indicates that the architectural decisions, not the manufacturing process, drive the performance and feature differences.

The compute resources are heavily skewed toward the NVIDIA part. The GTX 880M has 1536 shading units, 128 texture mapping units, and 32 ROPs. In contrast, the R7 M380 is configured with 640 shading units, 40 TMUs, and only 16 ROPs. This is a 2.4x difference in shading units and a 3.2x difference in TMUs. The GTX 880M also has a higher base clock of 954 MHz and a boost clock of 993 MHz, compared to the R7 M380's 900 MHz base and 915 MHz boost. Despite this massive resource advantage, the R7 M380 still wins the OpenCL benchmark, suggesting that the GCN architecture is significantly more efficient in this specific workload, or that the GTX 880M's performance is constrained by other factors such as power limits or driver overhead in this test.

The memory subsystems are also fundamentally different. The GTX 880M uses 8 GB of GDDR5 memory on a 256-bit bus, yielding a bandwidth of 160.0 GB/s. The R7 M380 uses 4 GB of DDR3 memory on a 128-bit bus, providing just 32.00 GB/s of bandwidth. This is a 5x difference in memory bandwidth in favor of the NVIDIA part, and a 2x difference in memory capacity. This suggests the GTX 880M is designed for high-resolution textures and bandwidth-intensive tasks, while the R7 M380 is more modestly provisioned. The R7 M380's memory clock is listed as 1000 MHz with 2 Gbps effective, while the GTX 880M runs at 1250 MHz with 5 Gbps effective, further widening the bandwidth gap.

Head-to-Head Benchmarks

The database records a single head-to-head benchmark between these two parts, and it is a decisive win for AMD. In the Geekbench OpenCL test, the AMD Radeon R7 M380 scores 9313, while the NVIDIA GeForce GTX 880M scores 5622. The delta is 65.7%, meaning the AMD part is nearly two-thirds faster in this compute test. This is a significant margin that places the R7 M380 in a different performance class for OpenCL workloads.

To contextualize this, the R7 M380's score of 9313 is only 0.1% above the NVIDIA GeForce GTX 850M (9302) and 0.4% above the desktop GeForce GTX 960 (9273). This shows that the R7 M380's compute performance is comparable to a mid-range NVIDIA desktop part from a similar era. The GTX 880M's score of 5622, however, is far below these figures. It is, in fact, below the AMD Radeon Vega 8 (9221) by a huge margin. The GTX 880M's average benchmark score of 8040, which includes its Metal score of 10458, is closer to its nearest rivals, the NVIDIA Quadro P5000 (8039) and GeForce GTX 650 Ti (8053). This indicates that the GTX 880M's OpenCL score is an outlier on the low side, potentially due to driver or workload characteristics, while its overall average reflects a more balanced performance profile across multiple APIs.

The data shows a clear winner in the only direct comparison available. The R7 M380 leads by 65.7% in the OpenCL test. It is important to note that this is the only head-to-head data point; there is no record of a DirectX or Vulkan test between these two. Therefore, the conclusion that the R7 M380 is the better overall GPU cannot be drawn from this data alone. The GTX 880M's higher average score, driven by its strong Metal performance, suggests that the two parts may be better suited to different tasks.

Specification Differences

The two GPUs differ in nearly every major specification category, reflecting their different designs and target markets.

• Architecture and Chip: AMD Radeon R7 M380 uses GCN 1.0 with the Tropo chip. NVIDIA GeForce GTX 880M uses Kepler with the GK104 chip.
• Transistors: The GTX 880M has 3,540 million transistors, more than double the R7 M380's 1,500 million.
• Die Size: The GTX 880M's die is 294 mm², compared to 123 mm² for the R7 M380.
• Transistor Density: Nearly identical, at 12.2M / mm² for AMD and 12.0M / mm² for NVIDIA.
• Base Clock: 900 MHz for the R7 M380, 954 MHz for the GTX 880M.
• Boost Clock: 915 MHz for the R7 M380, 993 MHz for the GTX 880M.
• Memory Clock: 1000 MHz (2 Gbps effective) for the R7 M380, 1250 MHz (5 Gbps effective) for the GTX 880M.
• Memory Size: 4 GB for the R7 M380, 8 GB for the GTX 880M.
• Memory Type: DDR3 for the R7 M380, GDDR5 for the GTX 880M.
• Memory Bus Width: 128 bit for the R7 M380, 256 bit for the GTX 880M.
• Memory Bandwidth: 32.00 GB/s for the R7 M380, 160.0 GB/s for the GTX 880M.
• Shading Units: 640 for the R7 M380, 1536 for the GTX 880M.
• TMUs: 40 for the R7 M380, 128 for the GTX 880M.
• ROPs: 16 for the R7 M380, 32 for the GTX 880M.
• Pixel Rate: 14.64 GPixel/s for the R7 M380, 31.78 GPixel/s for the GTX 880M.
• Texture Rate: 36.60 GTexel/s for the R7 M380, 127.1 GTexel/s for the GTX 880M.
• FP32 Performance: 1,171.2 GFLOPS for the R7 M380, 3.050 TFLOPS for the GTX 880M.
• TDP: Not listed for the R7 M380, 122 W for the GTX 880M.
• Slot Width: Not listed for the R7 M380, MXM Module for the GTX 880M.
• Power Connectors: Not listed for the R7 M380, None for the GTX 880M.
• Bus Interface: PCIe 3.0 x16 for the R7 M380, MXM-B (3.0) for the GTX 880M.
• Display Outputs: Not listed for the R7 M380, Portable Device Dependent for the GTX 880M.
• DirectX Support: 12 (11_1) for the R7 M380, 12 (11_0) for the GTX 880M.
• OpenGL Support: 4.6 for both.
• Vulkan Support: 1.2.170 for the R7 M380, 1.2.175 for the GTX 880M.
• Release Date: 2015-05-04 for the R7 M380, 2014-03-11 for the GTX 880M.
• Predecessor: Solar System for the R7 M380, GeForce 700M for the GTX 880M.
• Successor: Polaris Mobile for the R7 M380, GeForce 900M for the GTX 880M.
• Production Status: End-of-life for both.

FAQ

Q: Which GPU is faster in the Geekbench OpenCL test?

A: The AMD Radeon R7 M380 is significantly faster, scoring 9313 compared to the NVIDIA GeForce GTX 880M's 5622. This represents a 65.7% advantage for the AMD part.

Q: Does the NVIDIA GeForce GTX 880M have any benchmark advantages?

A: Yes. The GTX 880M has a Geekbench Metal score of 10458, which is higher than its OpenCL score. Its average benchmark score, which includes both tests, is 8040, placing it at the 42nd percentile of all GPUs.

Q: How does the memory configuration differ between the two?

A: The GTX 880M uses 8 GB of GDDR5 memory on a 256-bit bus, providing 160.0 GB/s of bandwidth. The R7 M380 uses 4 GB of DDR3 memory on a 128-bit bus, providing 32.00 GB/s of bandwidth.

Q: What are the transistor counts for each GPU?

A: The NVIDIA GeForce GTX 880M has 3,540 million transistors on a 294 mm² die. The AMD Radeon R7 M380 has 1,500 million transistors on a 123 mm² die. Their transistor densities are nearly identical at 12.0M / mm² and 12.2M / mm² respectively.

Q: Which GPU has a higher shading unit count?

A: The NVIDIA GeForce GTX 880M has 1536 shading units, while the AMD Radeon R7 M380 has 640. The GTX 880M also has more TMUs (128 vs 40) and ROPs (32 vs 16).

Q: What is the performance percentile for each GPU?

A: The AMD Radeon R7 M380 is at the 46th percentile against all GPUs, while the NVIDIA GeForce GTX 880M is at the 42nd percentile. The R7 M380's nearest rivals are within a 0.1% to 1% delta, while the GTX 880M's are within a 0% to -0.5% delta.