AMD Radeon R7 M380 vs NVIDIA GeForce GTX 560 Ti Comparison

FAQ

Q: Which GPU has the higher average benchmark score in the database?

A: The NVIDIA GeForce GTX 560 Ti records an average score of 10,690, while the AMD Radeon R7 M380 scores 9,313. The GTX 560 Ti holds a 14.8% lead in the head-to-head OpenCL test.

Q: How do these two cards compare to their nearest rivals?

A: The GTX 560 Ti sits within 1.1% of its closest rivals, ranging from 0.6% ahead of the AMD Radeon RX 6600S to 1.1% behind the AMD Radeon Pro 450. The R7 M380 is similarly tight, with a 1% lead over the AMD Radeon Vega 8 and a 0.1% lead over the NVIDIA GeForce GTX 850M.

Q: What are the memory capacity and type differences?

A: The GTX 560 Ti comes with 1024 MB of GDDR5 on a 256-bit bus, delivering 128.3 GB/s of bandwidth. The R7 M380 has 4 GB of DDR3 on a 128-bit bus, providing 32.00 GB/s of bandwidth.

Q: Which GPU has a higher transistor density despite being smaller?

A: The AMD Radeon R7 M380, built on a 28 nm process, packs 1,500 million transistors into a 123 mm² die, yielding 12.2M transistors per mm². The NVIDIA GeForce GTX 560 Ti, on a 40 nm process, uses 1,950 million transistors across 332 mm², giving 5.9M per mm².

Q: What is the peak single-precision floating-point performance for each?

A: The GTX 560 Ti reaches 1,263.4 GFLOPS, while the R7 M380 computes 1,171.2 GFLOPS. The GTX 560 Ti is roughly 7.9% higher in raw FP32 throughput.

Q: Which card supports Vulkan, and which does not?

A: The AMD Radeon R7 M380 lists Vulkan 1.2.170 support. The NVIDIA GeForce GTX 560 Ti has no Vulkan entry recorded in the database.

The Verdict

The recorded data positions the NVIDIA GeForce GTX 560 Ti as the stronger performer in the single available benchmark. Its OpenCL score of 10,690 beats the AMD Radeon R7 M380's 9,313 by 14.8%. The GTX 560 Ti also lands at the 49th percentile among all GPUs, while the R7 M380 sits at the 46th percentile. For users prioritizing raw compute throughput, the GTX 560 Ti is the clear choice, especially given its higher FP32 rate of 1,263.4 GFLOPS versus 1,171.2 GFLOPS.

However, the AMD Radeon R7 M380 holds advantages in other measurable areas. It offers four times the memory capacity (4 GB versus 1024 MB), a newer process node (28 nm versus 40 nm), and more shading units (640 versus 384). It also supports Vulkan 1.2.170 and a newer DirectX feature level (12 (11_1) versus 12 (11_0)). For workloads that depend on memory capacity or modern API features, the R7 M380 may be the more suitable option.

The verdict depends on the workload. If the task is dominated by OpenCL compute and raw FP32 performance, the GTX 560 Ti wins. If the task needs larger memory allocations or Vulkan support, the R7 M380 is the better fit. Neither card is current production, as both are marked end-of-life. The GTX 560 Ti launched with a 249 USD MSRP, which should be noted only as a historical reference.

Head-to-Head Benchmarks

The database contains one direct comparison test: Geekbench OpenCL. In this test, the NVIDIA GeForce GTX 560 Ti scores 10,690, and the AMD Radeon R7 M380 scores 9,313. The delta is 14.8% in favor of the GTX 560 Ti. This is a substantial margin, larger than the typical gaps seen between each card and its nearest rivals.

To contextualize: the GTX 560 Ti's nearest rival, the AMD Radeon RX 6600S, scores 10,629, just 0.6% lower. The R7 M380's nearest rival, the NVIDIA GeForce GTX 850M, scores 9,302, only 0.1% higher. The 14.8% gap between the two compared cards is therefore not a reflection of benchmark noise; it represents a genuine performance tier difference in OpenCL compute.

Breaking down the score components helps explain the gap. The GTX 560 Ti has a 256-bit memory bus and GDDR5 memory, yielding 128.3 GB/s of bandwidth. The R7 M380 uses a 128-bit bus with DDR3, capping bandwidth at 32.00 GB/s. That fourfold difference in memory bandwidth is likely a major contributor to the GTX 560 Ti's higher OpenCL score, as many OpenCL workloads are memory-bandwidth sensitive.

The GTX 560 Ti also posts higher pixel and texture rates: 13.17 GPixel/s and 52.67 GTexel/s, respectively. The R7 M380 counters with 14.64 GPixel/s and 36.60 GTexel/s. The R7 M380 wins on pixel fill rate by about 11%, but loses on texture fill rate by about 30%. The GTX 560 Ti's higher texture rate, combined with its larger memory bandwidth, explains its overall compute lead.

In terms of shading units, the R7 M380 has 640 versus the GTX 560 Ti's 384. Despite the 67% more shading units, the R7 M380 still trails in FP32 throughput, indicating that the GTX 560 Ti's higher clock efficiency and memory subsystem more than compensate.

Specification Differences

The two cards differ across nearly every specification field recorded. The GTX 560 Ti uses a 40 nm process, while the R7 M380 uses 28 nm. The GTX 560 Ti has 1,950 million transistors on a 332 mm² die; the R7 M380 has 1,500 million on 123 mm². Transistor density is 5.9M per mm² for NVIDIA and 12.2M per mm² for AMD.

Memory capacity differs: 1024 MB versus 4 GB. Memory type is GDDR5 versus DDR3. Bus width is 256 bit versus 128 bit. Bandwidth is 128.3 GB/s versus 32.00 GB/s. The memory clock is 1002 MHz (4 Gbps effective) for the GTX 560 Ti, and 1000 MHz (2 Gbps effective) for the R7 M380.

The GTX 560 Ti has no base or boost clock recorded; the R7 M380 runs at 900 MHz base and 915 MHz boost. Shading units are 384 versus 640. TMUs are 64 versus 40. ROPs are 32 versus 16. Pixel rate is 13.17 GPixel/s versus 14.64 GPixel/s. Texture rate is 52.67 GTexel/s versus 36.60 GTexel/s. FP32 is 1,263.4 GFLOPS versus 1,171.2 GFLOPS.

Power and physical specs also differ. The GTX 560 Ti has a 170 W TDP, a dual-slot design, 2x 6-pin power connectors, and a suggested 450 W PSU. The R7 M380 has no TDP, slot width, power connector, or PSU data recorded. The GTX 560 Ti is 229 mm (9 inches) long; the R7 M380 has no dimensions recorded. Bus interface is PCIe 2.0 x16 for NVIDIA and PCIe 3.0 x16 for AMD.

Display outputs: the GTX 560 Ti offers 2x DVI and 1x mini-HDMI 1.3a. The R7 M380 has no display output data. API support: both list OpenGL 4.6 and DirectX 12, but the GTX 560 Ti is at feature level 11_0 while the R7 M380 reaches 11_1. Vulkan is present only on the R7 M380 (1.2.170).

Release dates are 2011-01-24 for the GTX 560 Ti and 2015-05-04 for the R7 M380. Both are end-of-life. The GTX 560 Ti's predecessor is GeForce 400 and successor is GeForce 600. The R7 M380's predecessor is Solar System and successor is Polaris Mobile.

Architecture Differences

The NVIDIA GeForce GTX 560 Ti uses the GF114 chip built on the Fermi 2.0 architecture. The AMD Radeon R7 M380 uses the Tropo chip based on GCN 1.0. These are fundamentally different GPU designs from different generations, and the architecture differences explain much of the performance and feature divergence.

Fermi 2.0 was designed around a 40 nm process at TSMC, with a large die of 332 mm². GCN 1.0, by contrast, was built on a 28 nm process, also at TSMC, but with a much smaller 123 mm² die. The smaller die allowed AMD to fit 1,500 million transistors, resulting in a higher transistor density of 12.2M per mm² versus NVIDIA's 5.9M per mm².

The core configurations reflect different design philosophies. Fermi 2.0 emphasizes fewer, wider units: 384 shading units, 64 TMUs, and 32 ROPs. GCN 1.0 uses more but narrower units: 640 shading units, 40 TMUs, and 16 ROPs. This explains why the R7 M380 has higher pixel rate (14.64 GPixel/s) despite fewer ROPs, and why the GTX 560 Ti dominates texture rate (52.67 GTexel/s) with more TMUs.

Cache hierarchies are not recorded in the database, but the memory architecture differences are clear. Fermi 2.0 pairs with a 256-bit GDDR5 interface, while GCN 1.0 uses a 128-bit DDR3 interface. This is a generation-defining difference: the GTX 560 Ti's memory bandwidth is four times that of the R7 M380.

Feature support also differs by architecture. GCN 1.0 natively supports Vulkan 1.2.170 and DirectX 12 (11_1). Fermi 2.0 supports DirectX 12 (11_0) and OpenGL 4.6, but has no Vulkan support recorded. For modern API-dependent workloads, GCN 1.0 is clearly more capable.

Power efficiency is not directly comparable due to missing TDP data for the R7 M380. The GTX 560 Ti's 170 W TDP is recorded, but without the R7 M380's figure, no efficiency conclusion can be drawn from the data.

Where Each One Wins

The NVIDIA GeForce GTX 560 Ti wins in compute-heavy scenarios. Its OpenCL score of 10,690 is 14.8% higher than the R7 M380's 9,313. Its FP32 throughput of 1,263.4 GFLOPS exceeds the R7 M380's 1,171.2 GFLOPS. Its texture rate of 52.67 GTexel/s is substantially higher than 36.60 GTexel/s. Memory bandwidth is the clearest advantage: 128.3 GB/s versus 32.00 GB/s. Any workload that stresses memory bandwidth, such as large texture fetches or compute kernels with high data movement, will favor the GTX 560 Ti.

The AMD Radeon R7 M380 wins in memory-capacity-bound scenarios. Its 4 GB frame buffer is four times larger than the GTX 560 Ti's 1024 MB. For workloads that require large datasets resident in VRAM, such as high-resolution textures or multi-pass rendering, the R7 M380 can hold more data locally. Its pixel rate of 14.64 GPixel/s is also higher, giving it an edge in fill-rate-bound tasks at lower texture complexity.

The R7 M380 also wins on API modernity. It supports Vulkan 1.2.170, which the GTX 560 Ti lacks entirely. It also reaches DirectX feature level 11_1 versus 11_0. For software that leverages Vulkan, the R7 M380 is the only option of the two.

Process technology favors the R7 M380. The 28 nm node with 12.2M transistors per mm² indicates a denser, more modern design. However, without TDP data for the R7 M380, the efficiency advantage cannot be quantified.

The GTX 560 Ti wins on raw compute and memory bandwidth. The R7 M380 wins on memory capacity, pixel fill rate, and API support. The database records 1 win for the GTX 560 Ti and 0 for the R7 M380 in direct comparisons, but the choice ultimately depends on which of these factors matters more for the intended workload.