AMD Radeon Pro 460 vs NVIDIA GeForce RTX 2060 Comparison

Head-to-Head Benchmarks

The recorded data includes two direct comparisons between the AMD Radeon Pro 460 and the NVIDIA GeForce RTX 2060, and in both cases the NVIDIA part dominates decisively. In Geekbench OpenCL, the RTX 2060 scores 65,014 against the Radeon Pro 460's 15,284, a delta of -76.5% from the AMD card's perspective. That is not a marginal gap; the RTX 2060 delivers roughly 4.25 times the raw compute throughput in this workload. The Geekbench Vulkan result tells a similar story: the RTX 2060 posts 65,846 versus 16,816 for the Radeon Pro 460, a -74.5% delta. Across both head-to-head tests, the NVIDIA GPU wins every round, and the database shows a 2-0 win tally in its favor.

These are not close contests by any measure. The OpenCL gap is slightly larger than the Vulkan gap, which suggests the RTX 2060's advantage is consistent across different API front-ends rather than being workload-specific. The Radeon Pro 460's best showing in either test is its Vulkan score, which is still less than 26% of the RTX 2060's Vulkan result. The data implies that any compute-heavy task, whether it uses OpenCL or Vulkan, will see a massive performance differential in favor of the NVIDIA card.

It is notably the average benchmark score in the database also reflects this disparity, though the averages are computed over different test sets. The Radeon Pro 460 has an average score of 17,509, while the RTX 2060 averages 15,290. This seems contradictory at first glance, but the explanation lies in the benchmark composition: the AMD card's average is derived from three Geekbench tests (Metal, OpenCL, Vulkan), all of which are compute-oriented, whereas the RTX 2060's average includes a broader mix including PassMark DirectX tests, some of which score very low (e.g., 53 in DirectX 12 and 98 in DirectX 10). The RTX 2060's PassMark G3D score of 14,111 and its Geekbench scores above 65,000 show that its compute potential is far higher, but its overall average is dragged down by legacy DirectX tests that are not representative of modern workloads.

FAQ

Q: Which GPU wins in Geekbench OpenCL?

A: The NVIDIA GeForce RTX 2060 wins decisively with a score of 65,014 versus 15,284 for the AMD Radeon Pro 460, a -76.5% delta for the AMD part.

Q: Does the AMD Radeon Pro 460 win any benchmark in the head-to-head comparison?

A: No. The database records zero wins for the AMD card across the two shared tests. The RTX 2060 takes both Geekbench OpenCL and Geekbench Vulkan.

Q: How does the RTX 2060's Vulkan performance compare to the Radeon Pro 460's?

A: The RTX 2060 scores 65,846 in Vulkan, while the Radeon Pro 460 scores 16,816, resulting in a -74.5% delta for the AMD card. The NVIDIA GPU is roughly 3.9 times faster.

Q: What is the average benchmark score for each GPU, and what does that indicate?

A: The Radeon Pro 460 has an average score of 17,509 across its tested benchmarks, while the RTX 2060 averages 15,290. However, the RTX 2060's average is pulled down by low PassMark DirectX results (e.g., 53 in DirectX 12), while its Geekbench scores are far higher. The averages are not directly comparable due to different test sets.

Q: Which GPU has a higher percentile ranking among all GPUs?

A: The Radeon Pro 460 sits at the 61st percentile, while the RTX 2060 is at the 58th percentile. This is counterintuitive given the head-to-head results, but it reflects the different benchmark suites used for each card's average.

Q: Does the RTX 2060 support newer DirectX features than the Radeon Pro 460?

A: Yes. The RTX 2060 supports DirectX 12 Ultimate (12_2), while the Radeon Pro 460 supports DirectX 12 (12_0). The NVIDIA card also supports Vulkan 1.4 versus 1.3 for the AMD card.

Where Each One Wins

Based on the head-to-head data, the NVIDIA GeForce RTX 2060 is the clear winner in compute-heavy applications that leverage OpenCL or Vulkan. Its 65,014 OpenCL score and 65,846 Vulkan score indicate strong performance in general-purpose GPU computing, which includes tasks like rendering, physics simulation, and certain machine learning workloads. The RTX 2060 also brings hardware ray tracing cores (30) and tensor cores (240), which the Radeon Pro 460 lacks entirely, making it suitable for applications that offload ray tracing or AI inference to the GPU. Its 6 GB of GDDR6 memory on a 192-bit bus provides 336.0 GB/s of bandwidth, which is advantageous for large datasets or high-resolution textures.

The AMD Radeon Pro 460, despite losing both head-to-head tests, is not without a niche. Its 35 W TDP and integrated form factor (labeled IGP, meaning it is designed for portable devices) make it a low-power solution for thin-and-light laptops. The database shows its transistor density (24.4M / mm²) is nearly identical to the RTX 2060's (24.3M / mm²), but the AMD chip is far smaller (123 mm² die size versus 445 mm²) and uses only 3,000 million transistors versus 10,800 million. For scenarios where power consumption and physical space are the primary constraints, the Radeon Pro 460 is the only viable option between the two, as the RTX 2060 requires a dual-slot cooler, a 1x 8-pin power connector, and a 450 W suggested PSU.

In terms of benchmark diversity, the RTX 2060 also shows strengths in DirectX workloads. Its PassMark G3D score of 14,111 is substantial, and its DirectX 11 score of 110 is its best PassMark DirectX result. The Radeon Pro 460 has no PassMark data recorded, so the comparison is limited to Geekbench tests. The RTX 2060's Vulkan score is nearly identical to its OpenCL score (65,846 versus 65,014), indicating consistent performance across different compute APIs, whereas the Radeon Pro 460 shows a modest uplift in Vulkan (16,816) over OpenCL (15,284), a 10% improvement that suggests better optimization for Vulkan on the AMD side, though still far behind.

Specification Differences

The two GPUs differ in nearly every measurable specification. The Radeon Pro 460 uses 4 GB of GDDR5 memory on a 128-bit bus, yielding 81.28 GB/s of bandwidth. The RTX 2060 uses 6 GB of GDDR6 on a 192-bit bus, delivering 336.0 GB/s, which is over four times the bandwidth. Clock speeds also diverge: the AMD card runs at 850 MHz base and 907 MHz boost, while the NVIDIA card runs at 1365 MHz base and 1680 MHz boost. Memory clocks are 1270 MHz (5.1 Gbps effective) for the AMD part versus 1750 MHz (14 Gbps effective) for the NVIDIA part.

Compute resources show a similar gap. The Radeon Pro 460 has 1,024 shading units, 64 TMUs, and 16 ROPs. The RTX 2060 has 1,920 shading units (87.5% more), 120 TMUs (87.5% more), and 48 ROPs (200% more). Pixel rate is 14.51 GPixel/s for the AMD card versus 80.64 GPixel/s for the NVIDIA card. Texture rate is 58.05 GTexel/s versus 201.6 GTexel/s. FP32 throughput is 1.858 TFLOPS versus 6.451 TFLOPS, and FP16 is 1.858 TFLOPS (1:1) for AMD but 12.90 TFLOPS (2:1) for NVIDIA. The RTX 2060 also has 30 RT cores and 240 tensor cores, which the Radeon Pro 460 does not list. TDP is 35 W versus 160 W. The RTX 2060 is a dual-slot card with 1x 8-pin power and a 450 W suggested PSU, while the Radeon Pro 460 is an IGP with no power connectors. Bus interface is PCIe 3.0 x8 for AMD and PCIe 3.0 x16 for NVIDIA.

Architecture Differences

The architectural divide is stark. The Radeon Pro 460 is built on GCN 4.0 using a 14 nm process from GlobalFoundries, with a 123 mm² die containing 3,000 million transistors. The RTX 2060 is built on Turing using a 12 nm process from TSMC, with a 445 mm² die containing 10,800 million transistors. Transistor density is nearly identical (24.4M / mm² versus 24.3M / mm²), meaning the NVIDIA chip's advantage comes from sheer size rather than design efficiency. The AMD card is part of the Radeon Pro Mac (400 Series) generation, while the NVIDIA card is from the GeForce 20 series, with a predecessor in GeForce 10 and a successor in GeForce 30.

The RTX 2060 introduces hardware features that the GCN 4.0 architecture lacks: dedicated RT cores for ray tracing and tensor cores for AI acceleration. It also supports DirectX 12 Ultimate (12_2), which includes features like variable rate shading and mesh shaders, whereas the Radeon Pro 460 only supports DirectX 12 (12_0). Vulkan support differs as well: 1.4 for NVIDIA versus 1.3 for AMD. The memory technology is also different (GDDR6 versus GDDR5), and the NVIDIA card has a wider memory bus. The RTX 2060's FP16 throughput is 2:1 relative to FP32, meaning it can double its rate for half-precision workloads, while the AMD card runs FP16 at a 1:1 ratio, offering no such acceleration. Display outputs also differ: the Radeon Pro 460 is "Portable Device Dependent" (no fixed outputs), while the RTX 2060 has 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C.

The Verdict

The data points to a straightforward conclusion for most users: the NVIDIA GeForce RTX 2060 is dramatically faster in every direct benchmark recorded. Its OpenCL score is 4.25 times higher, and its Vulkan score is 3.9 times higher. For anyone running compute workloads in OpenCL or Vulkan, the RTX 2060 is the only rational choice. It also offers ray tracing and tensor cores, which the Radeon Pro 460 cannot match, and its 6 GB GDDR6 memory with 336.0 GB/s bandwidth is better suited for modern game textures and data-intensive tasks.

The AMD Radeon Pro 460's case rests entirely on its 35 W TDP and integrated form factor. It is designed for portable Mac systems where a discrete dual-slot card with a 450 W PSU requirement is physically and electrically impossible. If the system must be a thin-and-light laptop with no external power connectors, the Radeon Pro 460 is the only option between these two. Its 61st percentile ranking versus the RTX 2060's 58th is an artifact of different benchmark suites, not an indication of real-world performance equivalence.

The verdict is clear: choose the RTX 2060 for performance, but only if the system can accommodate its 229 mm length, dual-slot width, and 160 W TDP. Choose the Radeon Pro 460 if power and space constraints are absolute, and accept that its compute performance is far below the NVIDIA part. There is no middle ground in this comparison; the two GPUs serve completely different market segments, and the benchmark data reflects that chasm.