AMD Radeon Pro 460 vs NVIDIA GeForce GTX 660 Ti Comparison

The Verdict

The AMD Radeon Pro 460 and the NVIDIA GeForce GTX 660 Ti are both end-of-life products, but they serve entirely different purposes based on the recorded data. The Radeon Pro 460, with an average benchmark score of 17509, sits 2.4 points ahead of the GTX 660 Ti’s 15063 average. The Pro 460 wins both recorded head-to-head tests, making it the clear choice for compute workloads and modern API performance. The GTX 660 Ti, however, still holds ground in raw throughput metrics like pixel rate and texture rate, which may appeal to users running legacy applications that rely on those specific capabilities.

For a user deciding between these two today, the data points toward the Radeon Pro 460 for almost any scenario involving current software. Its 61st percentile ranking versus the GTX 660 Ti’s 57th percentile shows it sits higher in the overall GPU population. The GTX 660 Ti’s only real advantage comes from its higher memory bandwidth and faster clock speeds, which benefit older, less API-efficient workloads. If you are constrained to a system that only accepts a dual-slot card with 2x 6-pin power connectors and can handle a 150 W TDP, the GTX 660 Ti is the only option. Otherwise, the Radeon Pro 460’s integrated form factor and lower power draw make it the more practical pick for modern builds.

Where Each One Wins

The Radeon Pro 460 wins both head-to-head benchmarks recorded in the database. In Geekbench OpenCL, it scores 15284 against the GTX 660 Ti’s 15113, a 1.1% lead. In Geekbench Vulkan, the margin widens dramatically: the Pro 460 scores 16816 versus 15012, a 12% advantage. This suggests the Pro 460’s architecture handles modern compute and graphics APIs much more efficiently. The GTX 660 Ti does not have a recorded Metal benchmark, so the Pro 460’s 20426 Metal score stands alone, indicating strong performance in Apple’s ecosystem.

The GTX 660 Ti wins in several traditional specification-based metrics. Its pixel rate of 27.44 GPixel/s is nearly double the Pro 460’s 14.51 GPixel/s. Its texture rate of 109.8 GTexel/s also crushes the Pro 460’s 58.05 GTexel/s. Its memory bandwidth of 144.2 GB/s is significantly higher than the Pro 460’s 81.28 GB/s. These figures suggest the GTX 660 Ti is better suited for fill-rate-bound or bandwidth-bound legacy applications that do not leverage modern API features. However, none of these advantages translate into a benchmark win in the recorded tests.

Architecture Differences

The two GPUs come from fundamentally different eras and design philosophies. The AMD Radeon Pro 460 uses the Baffin chip on GCN 4.0 architecture, built on a 14 nm process at GlobalFoundries. It packs 3,000 million transistors into a 123 mm² die, yielding a transistor density of 24.4M per mm². The NVIDIA GeForce GTX 660 Ti uses the GK104 chip on Kepler architecture, built on a 28 nm process at TSMC. It contains 3,540 million transistors across a much larger 294 mm² die, with a lower density of 12.0M per mm².

The Pro 460 has 1024 shading units, 64 texture mapping units, and 16 ROPs. The GTX 660 Ti has 1344 shading units, 112 TMUs, and 24 ROPs. Despite having fewer units, the Pro 460 achieves a higher average benchmark score, indicating that architectural efficiency matters more than raw unit counts. The Pro 460 also supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3, while the GTX 660 Ti only reaches DirectX 12 (11_0) and Vulkan 1.2.175, though both support OpenGL 4.6.

The Pro 460 features 4 GB of GDDR5 memory on a 128-bit bus, while the GTX 660 Ti has only 2 GB on a 192-bit bus. The Pro 460’s memory operates at 5.1 Gbps effective, while the GTX 660 Ti runs at 6 Gbps effective. The GTX 660 Ti’s higher bus width gives it the bandwidth advantage, but the Pro 460’s larger capacity allows for bigger textures and datasets. The Pro 460 also supports FP16 at a 1:1 ratio with FP32, while the GTX 660 Ti has no recorded FP16 capability.

FAQ

Q: Which GPU is faster in compute workloads?

A: The AMD Radeon Pro 460 wins the Geekbench OpenCL test with 15284 points versus the GTX 660 Ti’s 15113, a 1.1% lead. It also wins Vulkan with 16816 versus 15012, a 12% margin.

Q: Does the GTX 660 Ti have any benchmark win?

A: No. The database records two head-to-head benchmarks, and the Radeon Pro 460 wins both. The GTX 660 Ti’s advantages are limited to specification-based metrics like pixel rate, texture rate, and memory bandwidth.

Q: Which card supports newer APIs?

A: The Radeon Pro 460 supports DirectX 12 (12_0) and Vulkan 1.3. The GTX 660 Ti is limited to DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Q: How much memory does each card have?

A: The Radeon Pro 460 has 4 GB of GDDR5 on a 128-bit bus. The GTX 660 Ti has 2 GB of GDDR5 on a 192-bit bus.

Q: What are the power requirements?

A: The Radeon Pro 460 has a 35 W TDP with no power connectors and is an integrated form factor. The GTX 660 Ti has a 150 W TDP, requires 2x 6-pin power connectors, and is a dual-slot card with a suggested 450 W PSU.

Q: Which card is better for a modern gaming PC?

A: Based on the Vulkan benchmark, the Radeon Pro 460 is 12% ahead of the GTX 660 Ti, suggesting better performance in modern APIs. The Pro 460 also has double the memory capacity and a higher overall percentile ranking.

Head-to-Head Benchmarks

The Geekbench Vulkan test is the clearest differentiator between these two GPUs. The Radeon Pro 460 scores 16816, while the GTX 660 Ti manages only 15012. That is a 12% advantage for the Pro 460, which is substantial for a single architectural generation gap. This result indicates that the Pro 460’s GCN 4.0 architecture handles Vulkan’s low-level draw calls and compute shaders far more efficiently than Kepler, which was designed for an era when DirectX 11 dominated.

The Geekbench OpenCL test is much closer. The Radeon Pro 460 scores 15284, and the GTX 660 Ti scores 15113, a 1.1% difference. This narrow margin suggests that in pure compute throughput, the two cards are nearly equivalent, despite the GTX 660 Ti having more shading units and higher FP32 performance. The Pro 460’s 1.858 TFLOPS FP32 is lower than the GTX 660 Ti’s 2.634 TFLOPS, yet it still wins the OpenCL test. This points to better driver optimization or architectural efficiency in the Pro 460.

The Radeon Pro 460 also has a Geekbench Metal score of 20426, which the GTX 660 Ti cannot match because it lacks Metal support in the database. This score is significantly higher than the Pro 460’s own OpenCL and Vulkan scores, indicating that Metal is the card’s strongest API. For users in Apple ecosystems, this is the definitive advantage.

The GTX 660 Ti’s specification-based strengths do not translate to benchmark wins. Its 27.44 GPixel/s pixel rate and 109.8 GTexel/s texture rate are nearly double the Pro 460’s figures, yet the recorded tests show it losing. This suggests that the benchmarks are not fill-rate-limited, but rather compute or API-overhead limited, where the Pro 460’s newer architecture excels.

Specification Differences

The Radeon Pro 460 and GTX 660 Ti differ across nearly every recorded specification. The Pro 460 uses a 14 nm process at GlobalFoundries, while the GTX 660 Ti uses 28 nm at TSMC. The Pro 460 has 3,000 million transistors on a 123 mm² die; the GTX 660 Ti has 3,540 million on a 294 mm² die. Transistor density is 24.4M per mm² for the Pro 460 versus 12.0M per mm² for the GTX 660 Ti.

Clock speeds favor the GTX 660 Ti: it runs at 915 MHz base and 980 MHz boost, while the Pro 460 runs at 850 MHz base and 907 MHz boost. Memory clocks also favor the GTX 660 Ti at 6 Gbps effective versus 5.1 Gbps effective. The GTX 660 Ti has a 192-bit bus and 144.2 GB/s bandwidth; the Pro 460 has a 128-bit bus and 81.28 GB/s bandwidth. Memory capacity favors the Pro 460 at 4 GB versus 2 GB.

The GTX 660 Ti has more shading units (1344 vs 1024), more TMUs (112 vs 64), and more ROPs (24 vs 16). Its pixel rate is 27.44 GPixel/s versus 14.51 GPixel/s, and its texture rate is 109.8 GTexel/s versus 58.05 GTexel/s. FP32 performance is 2.634 TFLOPS for the GTX 660 Ti and 1.858 TFLOPS for the Pro 460. The Pro 460 supports FP16 at 1.858 TFLOPS, while the GTX 660 Ti has no FP16 figure.

Power and physical specifications differ sharply. The Pro 460 has a 35 W TDP, an integrated form factor, and no power connectors. The GTX 660 Ti has a 150 W TDP, a dual-slot design, 2x 6-pin power connectors, and a suggested 450 W PSU. The GTX 660 Ti is 241 mm long, while the Pro 460 has no recorded length. The GTX 660 Ti uses PCIe 3.0 x16; the Pro 460 uses PCIe 3.0 x8. The GTX 660 Ti has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2 outputs, while the Pro 460’s outputs are portable device dependent.

The GTX 660 Ti has a launch MSRP of 299 USD. The Radeon Pro 460 has no recorded launch MSRP. Both are end-of-life. The GTX 660 Ti was released in 2012 and has a predecessor (GeForce 500) and successor (GeForce 700). The Pro 460 was released in 2016 and belongs to the Radeon Pro Mac 400 Series, with no recorded predecessor or successor.