AMD Radeon RX 5700 XT vs NVIDIA GeForce RTX 2060 SUPER Comparison

FAQ

Q: Which card is faster in the 3DMark Steel Nomad DX12 test?

A: The AMD Radeon RX 5700 XT scores 2134, while the NVIDIA GeForce RTX 2060 SUPER scores 2011. That gives AMD a 5.8% lead in this modern DX12 workload.

Q: How does the NVIDIA card perform in OpenCL compute workloads?

A: The GeForce RTX 2060 SUPER scores 76957 in Geekbench OpenCL, which is a massive 708% advantage over the Radeon RX 5700 XT's 9524. This is the single largest delta recorded between the two cards.

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce RTX 2060 SUPER records an average benchmark score of 18093, placing it at the 62nd percentile of all GPUs. The AMD card averages 16361, which puts it at the 59th percentile.

Q: What is the difference in texture processing capability?

A: The Radeon RX 5700 XT delivers a texture rate of 304.8 GTexel/s, compared to 224.4 GTexel/s for the GeForce RTX 2060 SUPER. This is a notable advantage for AMD in fill-rate-bound scenarios.

Q: Are both cards equipped with the same memory configuration?

A: Yes, both cards feature 8 GB of GDDR6 memory on a 256-bit bus, with identical bandwidth of 448.0 GB/s and the same 1750 MHz memory clock (14 Gbps effective).

Q: Which card supports hardware ray tracing?

A: Only the NVIDIA GeForce RTX 2060 SUPER includes dedicated ray tracing cores, with 34 RT cores present. The AMD Radeon RX 5700 XT has no RT cores listed in the database.

Architecture Differences

The two cards represent fundamentally different architectural approaches. The NVIDIA GeForce RTX 2060 SUPER is built on the Turing architecture with the TU106 chip, fabricated on TSMC's 12 nm process. The AMD Radeon RX 5700 XT uses the Navi 10 chip with the RDNA 1.0 architecture, produced on a more advanced 7 nm process from the same foundry. This process difference is reflected in transistor density: the AMD chip packs 41.0M transistors per mm², while the NVIDIA chip achieves 24.3M per mm². The die sizes tell the story clearly, the NVIDIA chip measures 445 mm² with 10,800 million transistors, while the AMD chip is much smaller at 251 mm² with 10,300 million transistors.

The compute configuration differs substantially. The Radeon RX 5700 XT fields 2560 shading units, 160 texture mapping units, and 64 raster operation units. The GeForce RTX 2060 SUPER counters with 2176 shading units, 136 TMUs, and 64 ROPs. Despite the lower shading unit count, the NVIDIA card includes specialized hardware that AMD lacks: 34 RT cores and 272 tensor cores. These dedicated units enable hardware-accelerated ray tracing and AI-enhanced features, capabilities absent from the RDNA 1.0 design.

Clock speeds also reflect the architectural split. The AMD card runs at a 1605 MHz base clock and boosts to 1905 MHz, with a game clock of 1755 MHz. The NVIDIA card operates at a 1470 MHz base and 1650 MHz boost. The higher clocks on the AMD card contribute to its raw throughput advantages, though the NVIDIA design compensates with its specialized cores.

Power characteristics differ notably. The Radeon RX 5700 XT has a TDP of 225 W and requires a 550 W suggested PSU with a 1x 6-pin plus 1x 8-pin power connector arrangement. The GeForce RTX 2060 SUPER draws 175 W, needs a 450 W suggested PSU, and uses a single 8-pin connector. The bus interface also differs: AMD supports PCIe 4.0 x16, while NVIDIA sticks with PCIe 3.0 x16.

DirectX support shows a generational gap. The NVIDIA card supports DirectX 12 Ultimate with feature level 12_2, while the AMD card is limited to DirectX 12 with feature level 12_1. Both cards support OpenGL 4.6 and Vulkan 1.4. Display outputs differ as well: the NVIDIA card includes 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C, while the AMD card offers 1x HDMI 2.0b and 3x DisplayPort 1.4a.

Head-to-Head Benchmarks

The benchmark results reveal a split decision, with each card winning exactly 5 of 10 recorded tests. The AMD Radeon RX 5700 XT takes the modern DX12 crown in 3DMark Steel Nomad, scoring 2134 versus 2011 for the NVIDIA card, a 5.8% margin. This result suggests the RDNA architecture handles contemporary rendering workloads efficiently.

The NVIDIA GeForce RTX 2060 SUPER dominates compute-oriented tests. In Geekbench OpenCL, the NVIDIA card scores 76957 against AMD's 9524, an extraordinary 708% advantage. This gap likely stems from the tensor cores and the maturity of NVIDIA's compute stack in OpenCL environments. The Vulkan test also favors NVIDIA, with 77402 versus 68310, a 13.3% lead.

DirectX legacy tests show NVIDIA's strength in older APIs. The PassMark DirectX 10 test gives NVIDIA a 14.4% win (111 versus 97), and DirectX 11 shows a 12.1% advantage (130 versus 116). In the DirectX 9 test, AMD edges ahead with 226 versus 218, a 3.5% margin. DirectX 12 results are close: AMD wins 64 to 61, a 4.7% difference.

The PassMark G3D test, which measures overall 3D graphics performance, narrowly favors NVIDIA at 16462 versus 16277, a 1.1% margin. This result aligns with the average benchmark scores, where NVIDIA holds an advantage of 18093 versus 16361. The 2D test goes to AMD with 922 versus 854, a 7.4% win.

Compute performance in PassMark shows AMD's strength. The GPU Compute test records 7650 for AMD versus 6721 for NVIDIA, a 12.1% advantage. This contrasts sharply with the Geekbench OpenCL result, indicating that workload characteristics heavily influence compute outcomes.

Specification Differences

The recorded data shows several key specification differences between the two cards:

| Specification | NVIDIA GeForce RTX 2060 SUPER | AMD Radeon RX 5700 XT |

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

| Process Node | 12 nm | 7 nm |

| Die Size | 445 mm² | 251 mm² |

| Transistor Density | 24.3M / mm² | 41.0M / mm² |

| Base Clock | 1470 MHz | 1605 MHz |

| Boost Clock | 1650 MHz | 1905 MHz |

| Game Clock | Not specified | 1755 MHz |

| Shading Units | 2176 | 2560 |

| TMUs | 136 | 160 |

| RT Cores | 34 | None |

| Tensor Cores | 272 | None |

| Pixel Rate | 105.6 GPixel/s | 121.9 GPixel/s |

| Texture Rate | 224.4 GTexel/s | 304.8 GTexel/s |

| FP32 Performance | 7.181 TFLOPS | 9.754 TFLOPS |

| FP16 Performance | 14.36 TFLOPS (2:1) | 19.51 TFLOPS (2:1) |

| TDP | 175 W | 225 W |

| Power Connectors | 1x 8-pin | 1x 6-pin + 1x 8-pin |

| Suggested PSU | 450 W | 550 W |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x16 |

| DirectX Support | 12 Ultimate (12_2) | 12 (12_1) |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 2x DP 1.4a, 1x USB-C | 1x HDMI 2.0b, 3x DP 1.4a |

| Length | 229 mm (9 inches) | 272 mm (10.7 inches) |

Both cards share identical memory specifications: 8 GB GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth. The dimensions are similar in height (113 mm versus 111 mm) and width (35 mm versus 36 mm), but the AMD card is significantly longer at 272 mm versus 229 mm. Both are dual-slot designs.

Where Each One Wins

The AMD Radeon RX 5700 XT excels in scenarios that demand raw compute throughput and modern API efficiency. Its 12.1% win in PassMark GPU Compute and 5.8% lead in 3DMark Steel Nomad indicate suitability for compute-heavy workloads and current-generation DX12 gaming. The higher texture rate of 304.8 GTexel/s and pixel rate of 121.9 GPixel/s make it well-suited for resolution-heavy rendering tasks. The FP32 performance of 9.754 TFLOPS provides substantial raw compute capability, and the 7 nm process enables these results within a smaller die. The card's PCIe 4.0 interface offers additional bandwidth headroom for systems that support it.

The NVIDIA GeForce RTX 2060 SUPER wins decisively in OpenCL compute, with a 708% advantage that makes it the clear choice for applications leveraging that API. Its 13.3% Vulkan lead and wins in DirectX 10 and DirectX 11 show strong legacy API performance. The dedicated RT cores and tensor cores provide capabilities the AMD card cannot match, making the NVIDIA card the only option for hardware-accelerated ray tracing and AI-enhanced features. The lower TDP of 175 W and single 8-pin connector make it easier to integrate into existing systems, and the shorter 229 mm length offers greater chassis compatibility. The DirectX 12 Ultimate support ensures forward compatibility with the latest feature sets.

The overall average benchmark scores favor NVIDIA at 18093 versus 16361, a difference that reflects the NVIDIA card's strength across a broader range of tests. The percentile ranking confirms this, with NVIDIA at the 62nd percentile versus AMD's 59th. However, the workload-dependent nature of the results means neither card is universally superior, and the choice depends heavily on the specific applications and API environments being targeted.