AMD Radeon RX 5700 XT vs NVIDIA GeForce RTX 2060 Comparison

The AMD Radeon RX 5700 XT and NVIDIA GeForce RTX 2060 represent two distinct philosophies from the same era, and the benchmark data shows a clear split in their strengths. The RX 5700 XT dominates the majority of the head-to-head tests, winning 8 out of 10 comparisons, while the RTX 2060 secures two notable victories. This analysis breaks down the numbers to show where each card excels, what architectural choices drive those results, and which use cases favor which GPU.

Head-to-Head Benchmarks

The most dramatic divergence appears in compute-oriented workloads. In the 3DMark Steel Nomad DX12 test, the RX 5700 XT scores 2134 against the RTX 2060’s 1242, a commanding 71.8% advantage. This is the single largest delta in the entire comparison and underscores the AMD card’s raw throughput advantage in modern DirectX 12 titles. The gap narrows but remains substantial in the PassMark GPU Compute test, where the RX 5700 XT posts 7650 versus 5488, a 39.4% lead. These results align with the card’s higher FP32 throughput and suggest that shader-heavy workloads will consistently favor the AMD part.

The RX 5700 XT also wins decisively in legacy API tests. In PassMark DirectX 9, it scores 226 against 190, a 18.9% margin. DirectX 11 shows a 5.5% lead (116 vs 110), and DirectX 12 expands to a 20.8% advantage (64 vs 53). While the absolute scores in these PassMark tests are low, the consistent pattern indicates that the RX 5700 XT maintains its edge across a wide range of graphics APIs, not just the newest ones. The 2D performance also favors AMD, with a PassMark G2D score of 922 versus 745, a 23.8% difference, though this is less relevant for gaming.

The Geekbench Vulkan test provides a closer contest. The RX 5700 XT wins with 68310 against 65846, but only by 3.7%. This suggests that in Vulkan-specific workloads, the two cards are nearly peers, with the AMD card holding a modest edge. The PassMark G3D score tells a similar story at a higher level: the RX 5700 XT scores 16277 versus 14111, a 15.3% win. This is a solid, but not overwhelming, victory in a general-purpose 3D benchmark.

The RTX 2060’s wins are concentrated in two specific areas. The Geekbench OpenCL score is a massive outlier: the RTX 2060 posts 65014 while the RX 5700 XT manages only 9524, a staggering 85.4% deficit for the AMD card. This result is anomalous compared to every other test and likely reflects a driver or optimization quirk in the OpenCL path, but it is nonetheless a significant data point for compute workloads that rely on that API. The other win is marginal: PassMark DirectX 10 shows the RTX 2060 ahead by a single point, 98 versus 97, a 1% difference that is effectively a tie in practical terms.

Looking at the aggregate data, the RX 5700 XT has an average benchmark score of 16361, which places it in the 59th percentile of all GPUs. The RTX 2060 averages 15290, sitting in the 58th percentile. The 1071-point gap between their averages is meaningful, and the RX 5700 XT’s nearest rivals include the NVIDIA RTX PRO 6000 Blackwell at 16408 (a 0.3% deficit) and the AMD Radeon PRO W7500 at 16415 (also a 0.3% deficit). The RTX 2060, by contrast, sits just 0.1% above the AMD Radeon 680M and 0.6% below the NVIDIA GeForce RTX 3050 OEM. This places the two cards in different performance tiers despite their similar release timing.

Architecture Differences

The fundamental divergence starts with the manufacturing process. The RX 5700 XT uses a 7 nm node at TSMC, while the RTX 2060 uses a 12 nm node, also from TSMC. This process advantage is reflected in transistor density: the RX 5700 XT packs 41.0 million transistors per square millimeter, versus 24.3 million for the RTX 2060. The RX 5700 XT’s die is also physically smaller at 251 mm² compared to 445 mm² for the RTX 2060, yet it contains 10,300 million transistors versus 10,800 million. The smaller, denser die allows the RX 5700 XT to achieve higher clock speeds while maintaining a lower power draw per unit of performance.

The core configurations are starkly different. The RX 5700 XT features 2560 shading units, 160 texture mapping units, and 64 ROPs. The RTX 2060 has 1920 shading units, 120 TMUs, and 48 ROPs. These numbers explain the RX 5700 XT’s higher pixel rate of 121.9 GPixel/s versus 80.64 GPixel/s, and its texture rate of 304.8 GTexel/s versus 201.6 GTexel/s. The FP32 compute performance follows the same pattern: 9.754 TFLOPS for the AMD card versus 6.451 TFLOPS for the NVIDIA card.

The RTX 2060 compensates for its lower raw compute with dedicated hardware. It includes 30 RT cores and 240 tensor cores, neither of which the RX 5700 XT possesses. This is the defining architectural difference: the RTX 2060 is built on the Turing architecture with hardware acceleration for ray tracing and AI workloads, while the RX 5700 XT uses RDNA 1.0 without such dedicated units. The FP16 performance tells a similar story, with the RTX 2060 achieving 12.90 TFLOPS (2:1) versus 19.51 TFLOPS (2:1) for the RX 5700 XT, but the tensor cores give the NVIDIA card an edge in AI-specific tasks that do not rely purely on FP16 throughput.

Memory configurations also differ significantly. The RX 5700 XT has 8 GB of GDDR6 on a 256-bit bus, yielding 448.0 GB/s of bandwidth. The RTX 2060 has 6 GB of GDDR6 on a 192-bit bus, providing 336.0 GB/s. Both run their memory at 1750 MHz with 14 Gbps effective speed, so the bandwidth difference comes entirely from the wider bus on the AMD card. This gives the RX 5700 XT a 33.3% bandwidth advantage, which is critical for high-resolution textures and memory-intensive workloads.

The bus interface differs as well: the RX 5700 XT uses PCIe 4.0 x16, while the RTX 2060 uses PCIe 3.0 x16. The power requirements follow the performance gap, with the RX 5700 XT rated at 225 W TDP and the RTX 2060 at 160 W. The AMD card requires a 1x 6-pin plus 1x 8-pin power connector and a 550 W suggested PSU, while the NVIDIA card needs only a single 8-pin connector and a 450 W PSU.

FAQ

Q: Which card wins in DirectX 12 gaming workloads?

A: The RX 5700 XT is clearly ahead. In the 3DMark Steel Nomad DX12 test, it scores 2134 versus 1242 for the RTX 2060, a 71.8% advantage. The PassMark DirectX 12 test also favors the AMD card, with a 20.8% lead (64 vs 53).

Q: Is the RTX 2060 better at compute tasks?

A: It depends on the API. The RTX 2060 wins the Geekbench OpenCL test by a massive margin (65014 vs 9524, an 85.4% difference). However, the RX 5700 XT wins the PassMark GPU Compute test (7650 vs 5488, a 39.4% lead) and the Geekbench Vulkan test (68310 vs 65846, a 3.7% edge). Compute advantage is therefore workload-dependent.

Q: How do the memory configurations compare?

A: The RX 5700 XT has 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth. The RTX 2060 has 6 GB of GDDR6 on a 192-bit bus with 336.0 GB/s bandwidth. The AMD card offers 33.3% more bandwidth and 2 GB more capacity.

Q: Does the RTX 2060 support ray tracing?

A: Yes. The RTX 2060 includes 30 RT cores and 240 tensor cores, which are dedicated to ray tracing and AI workloads. The RX 5700 XT has no equivalent hardware, as its RDNA 1.0 architecture lacks these units.

Q: What is the overall performance ranking of each card?

A: The RX 5700 XT has an average benchmark score of 16361, placing it in the 59th percentile of all GPUs. The RTX 2060 averages 15290, in the 58th percentile. The RX 5700 XT is 0.3% behind the NVIDIA RTX PRO 6000 Blackwell, while the RTX 2060 is 0.1% above the AMD Radeon 680M.

Q: Which card has a higher power requirement?

A: The RX 5700 XT has a 225 W TDP and requires a 550 W PSU, using a 1x 6-pin plus 1x 8-pin connector. The RTX 2060 has a 160 W TDP and requires a 450 W PSU, using only a single 8-pin connector.

Specification Differences

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

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

| Architecture | RDNA 1.0 | Turing |

| Process Node | 7 nm | 12 nm |

| Transistors | 10,300 million | 10,800 million |

| Die Size | 251 mm² | 445 mm² |

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

| Base Clock | 1605 MHz | 1365 MHz |

| Boost Clock | 1905 MHz | 1680 MHz |

| Game Clock | 1755 MHz | N/A |

| Memory Size | 8 GB | 6 GB |

| Memory Bus Width | 256 bit | 192 bit |

| Memory Bandwidth | 448.0 GB/s | 336.0 GB/s |

| Shading Units | 2560 | 1920 |

| TMUs | 160 | 120 |

| ROPs | 64 | 48 |

| RT Cores | N/A | 30 |

| Tensor Cores | N/A | 240 |

| Pixel Rate | 121.9 GPixel/s | 80.64 GPixel/s |

| Texture Rate | 304.8 GTexel/s | 201.6 GTexel/s |

| FP32 Performance | 9.754 TFLOPS | 6.451 TFLOPS |

| FP16 Performance | 19.51 TFLOPS (2:1) | 12.90 TFLOPS (2:1) |

| TDP | 225 W | 160 W |

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

| Suggested PSU | 550 W | 450 W |

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

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

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

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

| Height | 111 mm (4.4 inches) | 113 mm (4.4 inches) |

| Width | 36 mm (1.4 inches) | 35 mm (1.4 inches) |

| Release Date | 2019-07-06 | 2019-01-06 |

Where Each One Wins

The RX 5700 XT is the clear winner for traditional rasterization-heavy gaming. Its 8 wins out of 10 head-to-head tests span DirectX 9, 11, and 12, Vulkan, and general 3D performance. The 71.8% lead in 3DMark Steel Nomad DX12 is the standout result, indicating that modern DirectX 12 titles will run substantially faster on the AMD card. The 15.3% lead in PassMark G3D reinforces this conclusion for a broad range of 3D applications. Gamers prioritizing raw frame rates in standard rendering pipelines should favor the RX 5700 XT. Its 8 GB memory capacity and 448.0 GB/s bandwidth also provide more headroom for high-resolution textures and future game requirements.

The RTX 2060 wins in two specific scenarios. The first is OpenCL compute, where its 65014 score versus 9524 is a decisive victory. This makes it the better choice for applications that leverage OpenCL acceleration, such as certain physics simulations or scientific computing tasks. The second is DirectX 10, where it edges out the AMD card by 1% (98 vs 97), though this is negligible in practice. More importantly, the RTX 2060 offers hardware ray tracing through its 30 RT cores and AI acceleration through its 240 tensor cores. For users who want to experiment with ray-traced effects or use AI-based features like DLSS, the RTX 2060 provides capabilities that the RX 5700 XT simply cannot match.

Power efficiency is another differentiator. The RTX 2060 draws 160 W versus 225 W for the RX 5700 XT, and requires a 450 W PSU instead of 550 W. This makes the NVIDIA card easier to integrate into smaller or older systems with less robust power delivery. The RTX 2060 is also physically shorter at 229 mm versus 272 mm, which may be a consideration for compact cases. The RX 5700 XT counters with PCIe 4.0 support, which is forward-looking but offers no immediate benefit on PCIe 3.0 platforms.

In summary, the RX 5700 XT is the superior card for pure rasterization performance, winning the majority of benchmarks with margins ranging from 3.7% to 71.8%. The RTX 2060 is the better choice for users who need OpenCL compute performance or want access to ray tracing and tensor core features. The data shows two cards optimized for different priorities: AMD for raw throughput, NVIDIA for specialized hardware acceleration.