AMD Radeon R7 M380 vs NVIDIA GeForce GTX 560 Comparison

The AMD Radeon R7 M380 and NVIDIA GeForce GTX 560 represent two very different generations of GPU design, separated by four years of architectural evolution. The benchmark data shows a narrow but definitive victory for the newer AMD part, with the R7 M380 scoring 9,313 in Geekbench OpenCL against the GTX 560’s 9,058. This 2.8% delta in favor of the AMD card is modest, yet the underlying specifications reveal starkly different design philosophies: the R7 M380 is a 28 nm mobile-oriented chip with 640 shading units and 4 GB of DDR3, while the GTX 560 is a 40 nm desktop part with 336 cores and 1 GB of GDDR5. The results indicate that architectural efficiency can overcome raw memory bandwidth disadvantages, but the GTX 560 still holds its own in specific throughput metrics.

Head-to-Head Benchmarks

The only direct benchmark comparison available is the Geekbench OpenCL test, and it is a close contest. The AMD Radeon R7 M380 scores 9,313, while the NVIDIA GeForce GTX 560 scores 9,058. That puts the AMD part 2.8% ahead, a margin that is statistically significant but not transformative in real-world terms. For context, the R7 M380’s nearest rival in the broader database is the NVIDIA GeForce GTX 850M at 9,302, a difference of just 0.1%, meaning the R7 M380 is essentially performance-identical to that mobile GPU. Meanwhile, the GTX 560’s closest competitor is the NVIDIA TITAN V CEO Edition at 9,037, which the GTX 560 beats by 0.2%; it also trails the AMD Radeon 890M by 1.7%, showing the older Fermi card is still competitive against much newer integrated solutions.

When examining the components of that OpenCL score, the two cards take opposite approaches to achieving similar results. The R7 M380 relies on a higher shader count (640 vs. 336) and a much higher pixel rate of 14.64 GPixel/s compared to the GTX 560’s 11.34 GPixel/s, a 29% advantage in fill-rate-bound workloads. However, the GTX 560 counters with a texture rate of 45.36 GTexel/s, which is 24% higher than the R7 M380’s 36.60 GTexel/s, suggesting it would win in texture-heavy scenes. The FP32 compute figures are close: the R7 M380 delivers 1,171.2 GFLOPS, while the GTX 560 produces 1,088.6 GFLOPS, a 7.6% edge for AMD. These divergent strengths explain why the overall benchmark delta is so narrow—each card wins different sub-tests, and the final score reflects a balance of the two.

The memory subsystem is where the largest gap appears. The GTX 560 has 128.0 GB/s of bandwidth thanks to a 256-bit bus and GDDR5, while the R7 M380 is limited to 32.00 GB/s over a 128-bit bus with DDR3. That is a 4x bandwidth deficit for the AMD card, which should theoretically cripple it in bandwidth-sensitive compute tasks. Yet the OpenCL score shows the R7 M380 still comes out ahead, indicating that the newer GCN architecture is far more efficient in how it utilizes available memory bandwidth. This is a classic case where architectural improvements outweigh raw specifications, and benchmark results confirm that the R7 M380’s design compensates for its slower memory.

FAQ

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

A: The AMD Radeon R7 M380 scores 9,313, which is 2.8% higher than the NVIDIA GeForce GTX 560’s 9,058. The AMD card wins the head-to-head comparison with a delta of 2.8%.

Q: How does the R7 M380 compare to its closest rival, the GTX 850M?

A: The R7 M380 scores 9,313, while the GTX 850M averages 9,302. That is a 0.1% difference, meaning the two are effectively tied in performance, with the AMD card holding a negligible edge.

Q: What is the GTX 560’s position relative to the AMD Radeon 890M?

A: The GTX 560 scores 9,058, while the AMD Radeon 890M averages 9,210. This gives the 890M a 1.7% advantage, making it the only rival in the GTX 560’s nearest list that beats it.

Q: Does the R7 M380 have a higher pixel fill rate than the GTX 560?

A: Yes. The R7 M380 achieves 14.64 GPixel/s, while the GTX 560 manages 11.34 GPixel/s. This gives the AMD card a 29% advantage in pixel throughput.

Q: Which GPU has higher texture fill rate?

A: The GTX 560 wins this metric with 45.36 GTexel/s, compared to the R7 M380’s 36.60 GTexel/s. That is a 24% advantage for the NVIDIA card.

Q: How do the memory bandwidth figures compare?

A: The GTX 560 has 128.0 GB/s of bandwidth, which is 4x the R7 M380’s 32.00 GB/s. The NVIDIA card uses a 256-bit bus with GDDR5, while the AMD card uses a 128-bit bus with DDR3.

Architecture Differences

The two GPUs come from fundamentally different architectural eras. The R7 M380 is built on GCN 1.0, using the Tropo chip, and is part of the Gem System generation within the R7 M300 series. It is fabricated on a 28 nm process at TSMC, with 1,500 million transistors packed into a 123 mm² die. That yields a transistor density of 12.2M per mm², reflecting the more advanced manufacturing node. The architecture is designed for efficiency, with 640 shading units, 40 TMUs, and 16 ROPs. It supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170, making it compatible with modern API expectations.

The GTX 560, by contrast, uses the GF114 chip based on Fermi 2.0, part of the GeForce 500 generation. It is built on a 40 nm process, also at TSMC, but with 1,950 million transistors on a much larger 332 mm² die. That gives it a transistor density of only 5.9M per mm², showing how much less compact the older process was. The Fermi architecture employs 336 shading units, 56 TMUs, and 32 ROPs. It supports DirectX 12 (11_0) and OpenGL 4.6, but notably lacks Vulkan support, which is listed as null. The card also uses a PCIe 2.0 x16 interface, whereas the R7 M380 uses PCIe 3.0 x16, doubling the theoretical bus bandwidth.

The memory architecture differences are stark. The R7 M380 uses 4 GB of DDR3 on a 128-bit bus, while the GTX 560 uses 1024 MB of GDDR5 on a 256-bit bus. The GDDR5 memory runs at 4 Gbps effective, doubling the DDR3’s 2 Gbps effective rate. This combination of wider bus and faster memory gives the GTX 560 its 4x bandwidth advantage. However, the R7 M380’s newer architecture compensates by requiring less bandwidth per operation, as evidenced by its higher OpenCL score despite the massive memory deficit.

Specification Differences

The two cards diverge on nearly every major specification. The R7 M380 has a base clock of 900 MHz and a boost clock of 915 MHz, while the GTX 560 has no base or boost clock listed. The memory clock differs as well: the R7 M380 runs at 1000 MHz (2 Gbps effective), while the GTX 560 also runs at 1000 MHz but achieves 4 Gbps effective due to GDDR5’s double data rate.

Memory capacity and type are completely different: the R7 M380 has 4 GB of DDR3, while the GTX 560 has 1024 MB of GDDR5. The bus width is 128-bit for AMD and 256-bit for NVIDIA, leading to bandwidths of 32.00 GB/s and 128.0 GB/s respectively. Shading units are 640 vs. 336 in favor of AMD, while TMUs are 40 vs. 56 and ROPs are 16 vs. 32 in favor of NVIDIA.

The GTX 560 has a listed TDP of 150 W, while the R7 M380 has no TDP specified. The NVIDIA card is dual-slot with 2x 6-pin power connectors and a suggested PSU of 450 W, while the AMD card has no power specifications listed. The GTX 560 measures 210 mm (8.3 inches) in length and has display outputs of 2x DVI and 1x mini-HDMI 1.3a; the R7 M380 has no dimensions or display outputs listed. The bus interface differs: PCIe 3.0 x16 for AMD vs. PCIe 2.0 x16 for NVIDIA.

Release dates are four years apart: the R7 M380 launched on May 4, 2015, while the GTX 560 launched on May 16, 2011. The GTX 560 has a launch MSRP of 199 USD, which can be stated once as historical context. The R7 M380 has no launch MSRP listed. Both are end-of-life products, with the AMD card succeeding the Solar System generation and preceding Polaris Mobile, while the GTX 560 follows GeForce 400 and precedes GeForce 600.

The Verdict

The data presents a clear but nuanced choice. For raw compute performance in OpenCL, the AMD Radeon R7 M380 is the winner, scoring 2.8% higher than the GTX 560. Its 640 shading units and higher FP32 throughput of 1,171.2 GFLOPS give it an edge in general-purpose compute tasks. It also offers 4 GB of memory versus 1 GB, which is advantageous for workloads that require larger datasets. However, the GTX 560 is not obsolete: its 128.0 GB/s bandwidth and 45.36 GTexel/s texture rate make it superior in bandwidth-bound and texture-heavy scenarios, and its 32 ROPs provide higher fill-rate potential in certain rendering paths.

The GTX 560 is the better choice for users who prioritize memory bandwidth and texture throughput, or who need a card with a defined 150 W TDP and a 450 W PSU recommendation for system planning. It also has a known physical footprint of 210 mm and dual-slot design, making it easier to fit in a chassis. The R7 M380, meanwhile, is the pick for those who want newer API support (Vulkan 1.2.170 vs. none) and a more efficient 28 nm process that delivers competitive performance with far fewer transistors per mm². Its PCIe 3.0 interface also ensures better compatibility with modern motherboards.

In terms of competitive positioning, the R7 M380 sits just 0.1% above the GTX 850M, making it a solid mid-range mobile-class performer. The GTX 560, despite its age, still beats the TITAN V CEO Edition by 0.2% and only trails the AMD Radeon 890M by 1.7%, showing it remains relevant against much newer hardware. Ultimately, the R7 M380 wins the benchmark comparison, but the GTX 560’s strengths in memory and texture performance mean it is not a clear-cut recommendation for every use case. Choose the R7 M380 for balanced compute and API support; choose the GTX 560 for bandwidth-driven workloads and predictable desktop integration.