AMD Radeon RX 6800 vs NVIDIA GeForce RTX 2080 Ti Comparison

The AMD Radeon RX 6800 and NVIDIA GeForce RTX 2080 Ti are two end-of-life GPUs that occupy the same performance tier, separated by just 1% in average benchmark score. The RX 6800 edges ahead with an average score of 30095 against the RTX 2080 Ti's 29783, and it also holds a 6-4 win record in their direct head-to-head tests. However, the RTX 2080 Ti fights back with decisive victories in specific workloads, particularly compute and DirectX 12, making this a genuinely split decision depending on your application.

Head-to-Head Benchmarks

The most lopsided result in this matchup is Geekbench OpenCL, where the RTX 2080 Ti scores 128171 against the RX 6800's 24508 — a massive 80.9% advantage for NVIDIA. This is not a small gap; it is a category-level difference that points to the 2080 Ti's strength in general-purpose compute workloads. The RX 6800 has no answer here, and any application that relies heavily on OpenCL will strongly favor the older NVIDIA card.

In 3DMark Steel Nomad DX12, the RTX 2080 Ti again takes the win, scoring 3537 versus 3188, a 9.9% margin. This is a modern DirectX 12 test, and the 2080 Ti's 4352 shading units and 68 RT cores help it pull ahead despite the RX 6800's newer architecture. The Geekbench Vulkan result follows a similar pattern: 132930 for the 2080 Ti versus 115107 for the RX 6800, a 13.4% lead. PassMark DirectX 10 also goes NVIDIA's way, 153 to 128, a 16.3% difference.

The RX 6800's wins are smaller but more numerous. In PassMark DirectX 11, it scores 214 against 190, a 12.6% advantage. PassMark DirectX 9 shows 257 versus 235, a 9.4% lead. The RX 6800 also wins PassMark DirectX 12 (89 vs 84, 6% ahead), PassMark G2D (990 vs 927, 6.8% ahead), and PassMark GPU Compute (10864 vs 10056, 8% ahead). The closest contest is PassMark G3D, where the RX 6800 scores 22067 against 21548, just 2.4% ahead.

Looking at the overall picture, the RX 6800 wins the majority of tests, but its margins are modest — mostly in the 2-13% range. The RTX 2080 Ti's wins include the two largest deltas in the entire comparison (80.9% and 16.3%), plus two more in the 10-14% range. If you average the deltas across all ten tests, the RTX 2080 Ti's wins are more decisive, while the RX 6800's victories are narrower but more consistent.

Architecture Differences

The RX 6800 is built on RDNA 2.0 using TSMC's 7 nm process, packing 26,800 million transistors into a 520 mm² die. The RTX 2080 Ti uses Turing on a 12 nm process, with 18,600 million transistors spread across a much larger 754 mm² die. The process advantage shows in transistor density: 51.5M per mm² for AMD versus 24.7M per mm² for NVIDIA. Despite the smaller die, the RX 6800 has a higher boost clock at 2105 MHz versus 1545 MHz for the 2080 Ti.

Memory configurations differ significantly. The RX 6800 has 16 GB of GDDR6 on a 256-bit bus, delivering 512.0 GB/s of bandwidth. The RTX 2080 Ti has 11 GB on a wider 352-bit bus, achieving 616.0 GB/s. The 2080 Ti's higher bandwidth is notable, but the RX 6800's 5 GB extra capacity matters for high-resolution textures and large datasets.

Compute unit counts tell a mixed story. The RTX 2080 Ti has more shading units (4352 vs 3840), more TMUs (272 vs 240), and more RT cores (68 vs 60). It also has 544 tensor cores, which the RX 6800 lacks entirely. However, the RX 6800 has more ROPs (96 vs 88) and significantly higher pixel and texture rates: 202.1 GPixel/s and 505.2 GTexel/s versus 136.0 GPixel/s and 420.2 GTexel/s. In raw FP32 throughput, the RX 6800 leads with 16.17 TFLOPS against 13.45 TFLOPS.

Both cards support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. The RX 6800 uses PCIe 4.0 x16, while the 2080 Ti is limited to PCIe 3.0 x16. Both have the same 250 W TDP, dual-slot design, 2x 8-pin power connectors, and 600 W suggested PSU. The RX 6800 offers one additional DisplayPort output (2x DP 1.4a vs 3x DP 1.4a for NVIDIA) and both include HDMI and USB Type-C.

The Verdict

The data supports different choices depending on your primary workload. For general gaming and DirectX 11/9 titles, the RX 6800 is the stronger pick, winning PassMark DirectX 11, 12, and 9 tests. Its 16 GB VRAM also gives it a capacity advantage that the 2080 Ti cannot match, which matters for texture-heavy mods or future titles. The RX 6800's higher FP32 throughput and pixel rate suggest better raw rasterization performance.

For compute-heavy tasks, particularly OpenCL workloads, the RTX 2080 Ti is the clear winner by an enormous 80.9% margin. It also performs better in modern DirectX 12 benchmarks like 3DMark Steel Nomad and in Vulkan. The 2080 Ti's 544 tensor cores make it the only one of these two with dedicated tensor hardware, though the benchmark data does not isolate tensor performance.

The overall average scores are nearly identical — 30095 versus 29783, a 1% difference. Both cards sit at the 75th percentile among all GPUs. The RX 6800's nearest rivals include the RTX 3070 Ti (29945, 0.5% behind) and RTX 5070 Mobile (29928, 0.6% behind), while the 2080 Ti is also close to those same cards. In practice, neither card has a meaningful aggregate advantage, so the deciding factor should be the specific benchmarks that matter for your use case.

Specification Differences

| Specification | AMD Radeon RX 6800 | NVIDIA GeForce RTX 2080 Ti |

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

| Architecture | RDNA 2.0 | Turing |

| Process node | 7 nm | 12 nm |

| Transistors | 26,800 million | 18,600 million |

| Die size | 520 mm² | 754 mm² |

| Transistor density | 51.5M / mm² | 24.7M / mm² |

| Base clock | 1700 MHz | 1350 MHz |

| Boost clock | 2105 MHz | 1545 MHz |

| Memory size | 16 GB | 11 GB |

| Memory bus | 256 bit | 352 bit |

| Memory bandwidth | 512.0 GB/s | 616.0 GB/s |

| Shading units | 3840 | 4352 |

| TMUs | 240 | 272 |

| ROPs | 96 | 88 |

| RT cores | 60 | 68 |

| Tensor cores | None | 544 |

| Pixel rate | 202.1 GPixel/s | 136.0 GPixel/s |

| Texture rate | 505.2 GTexel/s | 420.2 GTexel/s |

| FP32 | 16.17 TFLOPS | 13.45 TFLOPS |

| FP16 | 32.33 TFLOPS (2:1) | 26.90 TFLOPS (2:1) |

| Bus interface | PCIe 4.0 x16 | PCIe 3.0 x16 |

| Display outputs | 1x HDMI 2.1, 2x DP 1.4a, 1x USB-C | 1x HDMI 2.0, 3x DP 1.4a, 1x USB-C |

| Height | 120 mm | 116 mm |

| Width | 40 mm | 35 mm |

| Launch MSRP | 579 USD | 999 USD |

FAQ

Q: Which card has more VRAM?

A: The AMD Radeon RX 6800 has 16 GB of GDDR6, while the NVIDIA GeForce RTX 2080 Ti has 11 GB. The RX 6800's 5 GB advantage is significant for high-resolution textures.

Q: Is the RTX 2080 Ti faster in compute workloads?

A: Yes, by a wide margin. In Geekbench OpenCL, the RTX 2080 Ti scores 128171 versus 24508 for the RX 6800, an 80.9% difference. The 2080 Ti also wins PassMark GPU Compute in the overall comparison, though the RX 6800 wins that specific head-to-head test (10864 vs 10056).

Q: Which card has higher memory bandwidth?

A: The NVIDIA GeForce RTX 2080 Ti, with 616.0 GB/s across a 352-bit bus. The RX 6800 delivers 512.0 GB/s on a 256-bit bus.

Q: Do both cards support DirectX 12 Ultimate?

A: Yes, both the RX 6800 and RTX 2080 Ti support DirectX 12 Ultimate (12_2), as well as OpenGL 4.6 and Vulkan 1.4.

Q: What is the power consumption of these cards?

A: Both have a 250 W TDP, use dual-slot designs with 2x 8-pin power connectors, and recommend a 600 W PSU.

Q: Which card wins in PassMark G3D?

A: The AMD Radeon RX 6800 wins with a score of 22067 against 21548 for the RTX 2080 Ti, a 2.4% margin.

Where Each One Wins

Choose the AMD Radeon RX 6800 for: DirectX 11 and DirectX 9 gaming, where it leads by 12.6% and 9.4% respectively. It also wins DirectX 12 in PassMark, G2D, G3D, and GPU Compute tests. The 16 GB VRAM capacity is a practical advantage for texture-heavy workloads, and its higher FP32 throughput (16.17 TFLOPS) supports rasterization-heavy applications. The RX 6800's smaller die and 7 nm process make it a more dense design, though the benchmarks do not directly test efficiency.

Choose the NVIDIA GeForce RTX 2080 Ti for: OpenCL compute workloads, where its 128171 score is nearly five times the RX 6800's. It also wins in 3DMark Steel Nomad DX12 by 9.9%, Geekbench Vulkan by 13.4%, and PassMark DirectX 10 by 16.3%. The 2080 Ti's 544 tensor cores provide dedicated tensor hardware absent from the RX 6800, and its higher memory bandwidth (616.0 GB/s) benefits bandwidth-sensitive tasks. Its larger die and higher shading unit count (4352 vs 3840) contribute to its compute advantage, despite lower clocks.