AMD Radeon RX 560X vs NVIDIA GeForce RTX 2070 Comparison

The GeForce RTX 2070 and Radeon RX 560X occupy vastly different tiers, yet their average benchmark scores place them within 1% of each other in the database’s aggregate ranking. This paradox is explained by the sparse benchmark data available for the RX 560X, which only includes two compute-oriented tests. The RTX 2070’s average score of 18789 sits just 0.9% above the RX 560X’s 18626, but the head-to-head results tell a far more lopsided story. In the two shared tests, the RTX 2070 dominates by margins that dwarf the aggregate delta, revealing that the RX 560X’s inclusion in the same performance neighborhood is an artifact of limited data points rather than genuine competitiveness.

Head-to-Head Benchmarks

The GeForce RTX 2070 delivers its most emphatic victory in the Geekbench OpenCL test, scoring 79966 against the RX 560X’s 17020. That is a 369.8% advantage, meaning the RTX 2070 achieves nearly 4.7 times the raw compute throughput of the AMD card in this workload. The gap is so wide that the RX 560X’s score represents only about 21% of the RTX 2070’s result. For context, the RTX 2070’s nearest rival in the overall database, the NVIDIA Tesla K80, averages 18866, which is 0.4% higher than the RTX 2070’s average—yet the Tesla K80 would still trail the RTX 2070 by a similar magnitude in this specific test, as the delta between the two cards here is far larger than any inter-rival spread.

In the Geekbench Vulkan test, the pattern repeats with slightly less severity. The RTX 2070 scores 82521, while the RX 560X manages 20231, yielding a 307.9% advantage for NVIDIA. This translates to the RTX 2070 delivering roughly 4.1 times the Vulkan performance of the RX 560X. The absolute scores are higher for both cards in Vulkan compared to OpenCL, but the proportional gap narrows only modestly. The RTX 2070’s Vulkan result also exceeds its OpenCL score by about 3.2%, whereas the RX 560X’s Vulkan score is 18.9% higher than its OpenCL result—suggesting the AMD card scales better with the Vulkan API, even though it remains hopelessly outmatched.

Across the two head-to-head benchmarks, the RTX 2070 wins 2 out of 2, with an average delta of roughly 339% across both tests. No single benchmark in the shared set favors the RX 560X, and the smallest margin of victory for NVIDIA is still over 300%. The aggregate scores from the nearestRivals data confirm that the RTX 2070’s 18789 average is 0.9% above the RX 560X’s 18626, but this narrow gap is misleading: the RX 560X’s average is computed from only two benchmarks, both of which are catastrophic losses. If the RX 560X had been tested in the same suite as the RTX 2070—which includes PassMark DirectX 9/10/11/12, 3DMark Steel Nomad, and G2D/G3D tests—the average delta would likely stretch into the hundreds of percent.

Where Each One Wins

The RTX 2070 wins every scenario where compute throughput, memory bandwidth, or API feature support matters. In OpenCL and Vulkan workloads—which cover general-purpose GPU compute, machine learning inference, and modern cross-platform rendering—the RTX 2070’s 7.465 TFLOPS FP32 throughput and 448.0 GB/s memory bandwidth crush the RX 560X’s 2.611 TFLOPS and 112.0 GB/s. The RTX 2070 also brings hardware ray tracing cores (36) and tensor cores (288), which are entirely absent from the RX 560X, making the NVIDIA card the only choice for ray-traced effects or DLSS-style acceleration. The RTX 2070’s DirectX 12 Ultimate support (12_2) versus the RX 560X’s DirectX 12 (12_0) means the newer card can handle mesh shaders, variable rate shading, and other next-gen features, while the RX 560X is limited to baseline DX12 functionality.

The RX 560X has no clear winning category in the data. Its only potential advantages are qualitative: it draws 75 W versus the RTX 2070’s 175 W, requires no external power connectors, and fits in a shorter 170 mm length compared to 229 mm. For a low-profile or legacy system with a 250 W power supply, the RX 560X is the only feasible option. It also supports Vulkan 1.3, which is a newer API revision than the RTX 2070’s Vulkan 1.4—though the RTX 2070’s higher raw performance makes this a moot point in practice. The RX 560X’s 14 nm process node and 3,000 million transistors on a 123 mm² die give it a transistor density of 24.4M / mm², nearly identical to the RTX 2070’s 24.3M / mm², but that parity does not translate into competitive performance.

Architecture Differences

The two GPUs come from fundamentally different architectural eras. The RTX 2070 uses the TU106 chip on TSMC’s 12 nm process, packing 10,800 million transistors into a 445 mm² die. The RX 560X uses the Polaris 21 chip on GlobalFoundries’ 14 nm process, with just 3,000 million transistors on a 123 mm² die. The RTX 2070 is built on NVIDIA’s Turing architecture, which introduced dedicated ray tracing and tensor cores—36 and 288 of them, respectively—alongside 2304 shading units, 144 texture mapping units, and 64 render output units. The RX 560X is based on GCN 4.0, with 1024 shading units, 64 TMUs, and only 16 ROPs. The RTX 2070’s pixel rate of 103.7 GPixel/s is over 5 times the RX 560X’s 20.40 GPixel/s, and its texture rate of 233.3 GTexel/s is nearly 3 times the RX 560X’s 81.60 GTexel/s.

Memory architecture amplifies the gap. The RTX 2070 uses 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s of bandwidth at 14 Gbps effective. The RX 560X uses 4 GB of GDDR5 on a 128-bit bus, with 112.0 GB/s at 7 Gbps. That is a 4x bandwidth difference, which directly impacts high-resolution textures and compute workloads. The RTX 2070 also supports FP16 at 14.93 TFLOPS via a 2:1 ratio, while the RX 560X offers only 2.611 TFLOPS at a 1:1 ratio—meaning the AMD card gets no throughput benefit from half-precision math. The RTX 2070’s PCIe 3.0 x16 interface provides twice the bandwidth of the RX 560X’s PCIe 3.0 x8, though this matters less for gaming than for data transfer in compute tasks.

The RTX 2070’s feature set extends to display outputs, with 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and an additional USB Type-C port, versus the RX 560X’s 1x DVI, 1x HDMI 2.0b, and 1x DisplayPort 1.4a. Both cards support OpenGL 4.6, but the RTX 2070’s Vulkan 1.4 and DirectX 12 Ultimate (12_2) outclass the RX 560X’s Vulkan 1.3 and DirectX 12 (12_0). The RTX 2070 requires a 450 W power supply and an 8-pin connector, while the RX 560X runs off the PCIe slot alone with a 250 W PSU recommendation. The RTX 2070 is also physically larger at 229 mm long, 113 mm tall, and 35 mm wide, whereas the RX 560X measures 170 mm in length with no listed height or width.

FAQ

Q: Which card has a higher average benchmark score?

A: The NVIDIA GeForce RTX 2070 averages 18789, which is 0.9% higher than the AMD Radeon RX 560X’s 18626. However, the RX 560X’s average is based on only two tests, both of which it loses by over 300%.

Q: How much faster is the RTX 2070 in OpenCL compute?

A: The RTX 2070 scores 79966 in Geekbench OpenCL versus the RX 560X’s 17020, a 369.8% advantage. This means the RTX 2070 delivers nearly 4.7 times the OpenCL compute performance.

Q: Does the RX 560X support ray tracing?

A: No. The RX 560X has no ray tracing cores (rtCores: null) and no tensor cores, while the RTX 2070 includes 36 ray tracing cores and 288 tensor cores. The RTX 2070 also supports DirectX 12 Ultimate (12_2), which includes ray tracing features, whereas the RX 560X is limited to DirectX 12 (12_0).

Q: What is the memory bandwidth difference?

A: The RTX 2070 has 448.0 GB/s of bandwidth from 8 GB of GDDR6 on a 256-bit bus, while the RX 560X has 112.0 GB/s from 4 GB of GDDR5 on a 128-bit bus. That is exactly a 4x difference in favor of the RTX 2070.

Q: Can the RX 560X run on a system without extra power connectors?

A: Yes. The RX 560X has no power connectors (powerConnectors: None) and a 75 W TDP, with a suggested 250 W power supply. The RTX 2070 requires a 1x 8-pin connector, 175 W TDP, and a 450 W power supply.

Q: Which card has a higher pixel fill rate?

A: The RTX 2070’s pixel rate is 103.7 GPixel/s, which is over 5 times the RX 560X’s 20.40 GPixel/s. The RTX 2070 also has 64 ROPs versus the RX 560X’s 16 ROPs.

Specification Differences

| Specification | NVIDIA GeForce RTX 2070 | AMD Radeon RX 560X |

|---|---|---|

| Chip | TU106 | Polaris 21 |

| Architecture | Turing | GCN 4.0 |

| Process Node | 12 nm | 14 nm |

| Foundry | TSMC | GlobalFoundries |

| Transistors | 10,800 million | 3,000 million |

| Die Size | 445 mm² | 123 mm² |

| Base Clock | 1410 MHz | 1175 MHz |

| Boost Clock | 1620 MHz | 1275 MHz |

| Memory Size | 8 GB | 4 GB |

| Memory Type | GDDR6 | GDDR5 |

| Memory Bus Width | 256 bit | 128 bit |

| Memory Bandwidth | 448.0 GB/s | 112.0 GB/s |

| Effective Memory Clock | 14 Gbps | 7 Gbps |

| Shading Units | 2304 | 1024 |

| TMUs | 144 | 64 |

| ROPs | 64 | 16 |

| RT Cores | 36 | None |

| Tensor Cores | 288 | None |

| Pixel Rate | 103.7 GPixel/s | 20.40 GPixel/s |

| Texture Rate | 233.3 GTexel/s | 81.60 GTexel/s |

| FP32 Performance | 7.465 TFLOPS | 2.611 TFLOPS |

| FP16 Performance | 14.93 TFLOPS (2:1) | 2.611 TFLOPS (1:1) |

| TDP | 175 W | 75 W |

| Power Connectors | 1x 8-pin | None |

| Suggested PSU | 450 W | 250 W |

| Bus Interface | PCIe 3.0 x16 | PCIe 3.0 x8 |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, 1x USB Type-C | 1x DVI, 1x HDMI 2.0b, 1x DisplayPort 1.4a |

| DirectX | 12 Ultimate (12_2) | 12 (12_0) |

| Vulkan | 1.4 | 1.3 |

| Length | 229 mm (9 inches) | 170 mm (6.7 inches) |

| Height | 113 mm (4.4 inches) | Not listed |

| Width | 35 mm (1.4 inches) | Not listed |

| Release Date | 2018-10-16 | 2018-04-10 |

| Predecessor | GeForce 10 | Polaris |

| Successor | GeForce 30 | Vega |

| Launch MSRP | 499 USD | Not listed |