AMD Radeon Pro 560X vs NVIDIA GeForce GTX 660 Ti Comparison

The Verdict

The benchmark data shows two end-of-life graphics cards that land at nearly identical overall performance, with the AMD Radeon Pro 560X holding an average benchmark score of 15,082 against the NVIDIA GeForce GTX 660 Ti’s 15,063 — a difference of just 0.1%. Both cards sit at the 57th percentile of all GPUs, meaning the data places them in the same performance tier. The AMD card wins the Geekbench Vulkan test by 12%, while the NVIDIA card wins the Geekbench OpenCL test by 36.1%. Given this split, the choice comes down to workload: the Radeon Pro 560X is the pick for Vulkan-based applications, while the GTX 660 Ti is the pick for OpenCL-heavy tasks. The AMD card also offers double the memory (4 GB versus 2 GB), which matters for textures and datasets that exceed 2 GB, though the NVIDIA card has a wider memory bus and higher bandwidth. The GTX 660 Ti carries a launch MSRP of 299 USD, while the Radeon Pro 560X has no listed launch MSRP. For users constrained by power, the Radeon Pro 560X draws 75 W versus the GTX 660 Ti’s 150 W, making the AMD part the more energy-efficient option in the data.

Architecture Differences

The two cards come from different process generations and design philosophies. AMD’s Radeon Pro 560X uses the Polaris 21 chip built on GCN 4.0 architecture, manufactured on a 14 nm process at GlobalFoundries. The GTX 660 Ti uses NVIDIA’s GK104 chip based on Kepler architecture, manufactured on a 28 nm process at TSMC. The process node difference is stark: 14 nm versus 28 nm, which explains why the AMD chip packs 3,000 million transistors into a 123 mm² die, while the NVIDIA chip uses 3,540 million transistors across a much larger 294 mm² die. Transistor density tells the story — the Radeon Pro 560X achieves 24.4 million transistors per square millimeter, while the GTX 660 Ti manages 12.0 million.

The compute configurations differ as well. The Radeon Pro 560X has 1,024 shading units, 64 texture mapping units, and 16 ROPs. The GTX 660 Ti has 1,344 shading units, 112 TMUs, and 24 ROPs. Despite having fewer shading units, the AMD card’s FP32 throughput is 2.056 TFLOPS, while the NVIDIA card reaches 2.634 TFLOPS. The AMD card also lists FP16 performance at 2.056 TFLOPS with a 1:1 ratio, while the GTX 660 Ti has no listed FP16 capability — a notable architectural difference for workloads using half-precision math. The GTX 660 Ti’s clock speeds are listed at 915 MHz base and 980 MHz boost, while the Radeon Pro 560X has no base or boost clocks listed in the data. Memory clocks show the NVIDIA card running at 1502 MHz (6 Gbps effective) versus the AMD card’s 1470 MHz (5.9 Gbps effective). API support also differs: the Radeon Pro 560X supports DirectX 12 (12_0) and Vulkan 1.3, while the GTX 660 Ti supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Where Each One Wins

The head-to-head benchmark results split cleanly between the two cards. The GTX 660 Ti dominates in Geekbench OpenCL, scoring 15,113 against the Radeon Pro 560X’s 9,663 — a 36.1% advantage. That is a decisive margin, suggesting the NVIDIA card is substantially stronger in OpenCL compute workloads. For users running OpenCL-based applications, the data clearly favors the GTX 660 Ti.

The Radeon Pro 560X wins the Geekbench Vulkan test, scoring 16,819 against the GTX 660 Ti’s 15,012 — a 12% lead. This indicates the AMD card has a meaningful advantage in Vulkan-based rendering and compute tasks. The AMD card also carries a memory capacity advantage at 4 GB versus 2 GB, which can be decisive for workloads that exceed the NVIDIA card’s memory ceiling, even if the GTX 660 Ti has higher raw bandwidth (144.2 GB/s versus 94.08 GB/s).

The power draw data also favors the AMD card. The Radeon Pro 560X is rated at 75 W, while the GTX 660 Ti is rated at 150 W. The AMD card uses no power connectors and is listed as an integrated graphics processor (IGP) slot width, whereas the GTX 660 Ti is a dual-slot card requiring two 6-pin power connectors and a suggested 450 W power supply. For compact systems or low-power builds, the Radeon Pro 560X is the only viable choice in the data.

FAQ

Q: Which card has the higher average benchmark score?

A: The AMD Radeon Pro 560X has an average benchmark score of 15,082, which is 0.1% higher than the NVIDIA GeForce GTX 660 Ti’s 15,063.

Q: How do the two cards compare in Vulkan performance?

A: The Radeon Pro 560X scores 16,819 in Geekbench Vulkan, beating the GTX 660 Ti’s 15,012 by 12%.

Q: How do they compare in OpenCL performance?

A: The GTX 660 Ti scores 15,113 in Geekbench OpenCL, beating the Radeon Pro 560X’s 9,663 by 36.1%.

Q: What is the memory configuration difference?

A: The Radeon Pro 560X has 4 GB of GDDR5 on a 128-bit bus with 94.08 GB/s bandwidth, while the GTX 660 Ti has 2 GB of GDDR5 on a 192-bit bus with 144.2 GB/s bandwidth.

Q: Which card has a higher transistor density?

A: The Radeon Pro 560X has a transistor density of 24.4 million per square millimeter, compared to the GTX 660 Ti’s 12.0 million per square millimeter.

Q: Do both cards support the same DirectX version?

A: No. The Radeon Pro 560X supports DirectX 12 (12_0), while the GTX 660 Ti supports DirectX 12 (11_0).

Head-to-Head Benchmarks

The two benchmark tests in the data show a clear split in strengths. The Geekbench OpenCL test is the biggest win for the NVIDIA card. The GTX 660 Ti scores 15,113, while the Radeon Pro 560X scores 9,663. The delta is -36.1% from the AMD card’s perspective, meaning the NVIDIA card is more than a third faster in this test. This is a substantial margin that suggests the GTX 660 Ti’s Kepler architecture handles OpenCL workloads significantly better than AMD’s GCN 4.0 implementation in this particular card. The GTX 660 Ti’s higher FP32 throughput (2.634 TFLOPS versus 2.056 TFLOPS) and higher texture rate (109.8 GTexel/s versus 64.26 GTexel/s) may explain part of this gap, as OpenCL workloads often stress raw compute throughput.

The Geekbench Vulkan test flips the result. The Radeon Pro 560X scores 16,819, beating the GTX 660 Ti’s 15,012 by 12%. This win is notable because the AMD card achieves it despite having lower raw FP32 throughput and lower texture fill rates. The Vulkan advantage likely stems from the newer architecture and deeper Vulkan API support — the Radeon Pro 560X supports Vulkan 1.3, while the GTX 660 Ti supports Vulkan 1.2.175. The AMD card also supports FP16 at a 1:1 ratio, which can benefit certain Vulkan compute paths. The 12% margin is meaningful but not overwhelming; it indicates a real architectural advantage for AMD in Vulkan workloads rather than a minor edge.

Looking at the broader benchmark picture, the average scores are almost identical — 15,082 for the AMD card versus 15,063 for the NVIDIA card — which places them within 0.1% of each other. The nearest rival data confirms this tight grouping: the AMD Radeon RX 7600 averages 15,171 (0.6% above the Pro 560X), the NVIDIA GeForce RTX 3050 OEM averages 15,199 (0.8% above), and the AMD Radeon 680M averages 15,270 (1.2% above). The GTX 660 Ti sees similar deltas against these same rivals: -0.7% versus the RX 7600, -0.9% versus the RTX 3050 OEM, and -1.4% versus the Radeon 680M. Both cards sit in the same performance neighborhood, with neither showing a decisive overall advantage in aggregate scoring.

Specification Differences

The two cards differ across nearly every major specification category. The process node is the most fundamental split: the Radeon Pro 560X uses a 14 nm process at GlobalFoundries, while the GTX 660 Ti uses a 28 nm process at TSMC. The AMD chip contains 3,000 million transistors on a 123 mm² die, while the NVIDIA chip contains 3,540 million transistors on a 294 mm² die. Transistor density is 24.4 million per square millimeter for AMD versus 12.0 million for NVIDIA.

Memory configurations differ significantly. The Radeon Pro 560X has 4 GB of GDDR5 on a 128-bit bus, delivering 94.08 GB/s bandwidth. The GTX 660 Ti has 2 GB of GDDR5 on a 192-bit bus, delivering 144.2 GB/s bandwidth. The NVIDIA card has higher bandwidth, but the AMD card has double the capacity. Memory clocks are close: 1470 MHz (5.9 Gbps effective) for AMD versus 1502 MHz (6 Gbps effective) for NVIDIA.

Compute unit counts show the NVIDIA card with more hardware resources: 1,344 shading units, 112 TMUs, and 24 ROPs, versus the AMD card’s 1,024 shading units, 64 TMUs, and 16 ROPs. Pixel rates follow suit: the GTX 660 Ti achieves 27.44 GPixel/s versus the Radeon Pro 560X’s 16.06 GPixel/s. Texture rates are 109.8 GTexel/s versus 64.26 GTexel/s. FP32 throughput is 2.634 TFLOPS for NVIDIA versus 2.056 TFLOPS for AMD. The AMD card lists FP16 at 2.056 TFLOPS (1:1), while the NVIDIA card has no FP16 listing.

Power and physical specifications diverge sharply. The Radeon Pro 560X is rated at 75 W with no power connectors and an IGP slot width, making it suitable for integrated or low-power configurations. The GTX 660 Ti is rated at 150 W, requires two 6-pin power connectors, a suggested 450 W power supply, and occupies a dual-slot form factor with a length of 241 mm (9.5 inches). The bus interface also differs: PCIe 3.0 x8 for AMD versus PCIe 3.0 x16 for NVIDIA. Display outputs are portable-device-dependent for the AMD card, while the NVIDIA card offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The NVIDIA card has a listed base clock of 915 MHz and boost clock of 980 MHz; the AMD card has no base or boost clocks listed. Release dates are also far apart: the GTX 660 Ti launched on 2012-08-15, while the Radeon Pro 560X launched on 2018-07-15. The GTX 660 Ti has a launch MSRP of 299 USD; the Radeon Pro 560X has no listed launch MSRP.