AMD Radeon RX 6750 XT vs NVIDIA GeForce RTX 2080 SUPER Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon RX 6750 XT leads with an average benchmark score of 26011, while the NVIDIA GeForce RTX 2080 SUPER scores 24170. This places the AMD card in the 71st percentile of all GPUs, versus the 69th percentile for the NVIDIA card.

Q: How do the two cards compare in the 3DMark Steel Nomad DX12 test?

A: The AMD Radeon RX 6750 XT wins decisively with a score of 2548 versus 1882 for the NVIDIA GeForce RTX 2080 SUPER. That is a 35.4% advantage for the AMD card in this modern DirectX 12 workload.

Q: In which benchmark does the NVIDIA card show its largest margin of victory?

A: The GeForce RTX 2080 SUPER dominates the Geekbench OpenCL test, scoring 99226 compared to the Radeon's 16360. The NVIDIA card is 83.5% ahead in this compute-oriented benchmark, its biggest win across all tests.

Q: What is the memory configuration difference between the two?

A: The AMD Radeon RX 6750 XT features 12 GB of GDDR6 memory on a 192-bit bus with 432.0 GB/s bandwidth. The NVIDIA GeForce RTX 2080 SUPER has 8 GB of GDDR6 on a 256-bit bus with 495.9 GB/s bandwidth.

Q: Do both cards support the same DirectX and Vulkan versions?

A: Yes, both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Their API feature sets are identical in this regard.

Q: Which card won more head-to-head benchmark comparisons?

A: The AMD Radeon RX 6750 XT won 7 of the 10 head-to-head tests, while the NVIDIA GeForce RTX 2080 SUPER won 3. The AMD card's victories include the 3DMark test, PassMark G3D, and PassMark GPU Compute.

Architecture Differences

The AMD Radeon RX 6750 XT is built on the Navi 22 chip using RDNA 2.0 architecture, fabricated on a 7 nm process at TSMC. It packs 17,200 million transistors into a 335 mm² die, achieving a transistor density of 51.3M per mm². The NVIDIA GeForce RTX 2080 SUPER uses the TU104 chip with Turing architecture, built on a larger 12 nm process. It contains 13,600 million transistors spread across a much larger 545 mm² die, resulting in a lower density of 25.0M per mm².

The shading resources differ substantially. The AMD card has 2560 shading units, 160 texture mapping units, and 64 ROPs. The NVIDIA card has more: 3072 shading units, 192 TMUs, and 64 ROPs. However, the AMD card's higher clock speeds compensate. Its base clock is 2150 MHz with a boost of 2600 MHz and a game clock of 2495 MHz. The NVIDIA card runs at a 1650 MHz base and 1815 MHz boost.

Ray tracing hardware is present on both, but with different counts. The AMD card has 40 RT cores, while the NVIDIA card has 48. The NVIDIA card also includes 384 tensor cores, which the AMD card lacks entirely. This is a key architectural divergence — Turing's tensor cores enable features that RDNA 2.0 does not provide.

The memory subsystems are designed differently. The AMD card uses 12 GB of GDDR6 on a 192-bit bus, while the NVIDIA card uses 8 GB on a wider 256-bit bus. Interestingly, the AMD card's memory runs at 2250 MHz (18 Gbps effective) versus 1937 MHz (15.5 Gbps effective) for the NVIDIA card. Despite the narrower bus, the AMD card's pixel rate of 166.4 GPixel/s and texture rate of 416.0 GTexel/s both exceed the NVIDIA card's 116.2 GPixel/s and 348.5 GTexel/s.

The interface generation differs: the AMD card uses PCIe 4.0 x16, while the NVIDIA card is limited to PCIe 3.0 x16. Both cards are dual-slot designs with identical 250 W TDP and the same power connector configuration (1x 6-pin + 1x 8-pin), plus the same 600 W suggested PSU.

Where Each One Wins

The AMD Radeon RX 6750 XT establishes clear superiority in modern DirectX 12 gaming workloads. Its 35.4% lead in the 3DMark Steel Nomad DX12 test is the largest margin in any benchmark — the AMD card is simply better suited for contemporary game engines. It also wins the PassMark G3D test (20700 vs 19490, +6.2%) and the PassMark GPU Compute test (9550 vs 8290, +15.2%), indicating stronger overall graphics throughput.

The AMD card also wins across legacy DirectX APIs. It leads in PassMark DirectX 9 (263 vs 227, +15.9%), DirectX 11 (180 vs 165, +9.1%), and DirectX 12 (80 vs 75, +6.7%). Even in 2D workloads, the AMD card edges ahead with a PassMark G2D score of 972 versus 920 (+5.7%). For gaming across multiple API generations, the AMD card is the more consistent performer.

The NVIDIA GeForce RTX 2080 SUPER wins in compute-heavy OpenCL workloads, where it scores 99226 versus 16360 — an 83.5% margin. This is a massive difference that suggests NVIDIA's Turing architecture has a significant advantage in raw compute tasks that leverage OpenCL. The NVIDIA card also wins the Vulkan benchmark (111284 vs 98089, +11.9%) and PassMark DirectX 10 (140 vs 125, +10.7%).

For users prioritizing OpenCL compute performance — such as certain professional or scientific workloads — the NVIDIA card is clearly superior. The Vulkan win also matters for games and applications that use this cross-platform API. However, the NVIDIA card's wins are more narrow in scope, whereas the AMD card's victories span a broader range of graphics tests.

Specification Differences

| Specification | AMD Radeon RX 6750 XT | NVIDIA GeForce RTX 2080 SUPER |

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

| Architecture | RDNA 2.0 | Turing |

| Process Node | 7 nm | 12 nm |

| Transistors | 17,200 million | 13,600 million |

| Die Size | 335 mm² | 545 mm² |

| Transistor Density | 51.3M / mm² | 25.0M / mm² |

| Base Clock | 2150 MHz | 1650 MHz |

| Boost Clock | 2600 MHz | 1815 MHz |

| Memory Size | 12 GB | 8 GB |

| Memory Bus Width | 192 bit | 256 bit |

| Memory Bandwidth | 432.0 GB/s | 495.9 GB/s |

| Shading Units | 2560 | 3072 |

| TMUs | 160 | 192 |

| RT Cores | 40 | 48 |

| Tensor Cores | N/A | 384 |

| Pixel Rate | 166.4 GPixel/s | 116.2 GPixel/s |

| Texture Rate | 416.0 GTexel/s | 348.5 GTexel/s |

| FP32 | 13.31 TFLOPS | 11.15 TFLOPS |

| FP16 | 26.62 TFLOPS (2:1) | 22.30 TFLOPS (2:1) |

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

| Display Outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a | 1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C |

| Height | 110 mm | 116 mm |

| Width | 40 mm | 35 mm |

| Release Date | 2022-03-02 | 2019-07-22 |

| Launch MSRP | 549 USD | 699 USD |

Head-to-Head Benchmarks

The 3DMark Steel Nomad DX12 test delivers the most dramatic divergence. The AMD Radeon RX 6750 XT scores 2548 versus 1882 for the NVIDIA card, a 35.4% advantage. This modern DirectX 12 workload showcases the AMD card's architectural efficiency — its smaller die and higher clocks translate into substantially better performance in current-generation gaming scenarios.

The Geekbench OpenCL result flips the picture entirely. The NVIDIA card scores 99226, while the AMD card manages only 16360. The NVIDIA card is 83.5% ahead, which is the single largest delta in the entire comparison. This indicates a fundamental difference in how each architecture handles OpenCL compute workloads — Turing's design clearly favors this API.

The Geekbench Vulkan test also favors NVIDIA, with the RTX 2080 SUPER scoring 111284 versus 98089 for the AMD card, an 11.9% lead. This matters for Vulkan-based games and applications. However, the AMD card's wins in the PassMark suite tell a different story. In PassMark G3D, the AMD card scores 20700 versus 19490 (+6.2%). In PassMark GPU Compute, it scores 9550 versus 8290 (+15.2%).

The legacy DirectX tests show consistent AMD advantages. PassMark DirectX 9: 263 vs 227 (+15.9%). PassMark DirectX 11: 180 vs 165 (+9.1%). PassMark DirectX 12: 80 vs 75 (+6.7%). Even PassMark G2D favors AMD at 972 vs 920 (+5.7%). Only PassMark DirectX 10 goes to NVIDIA, at 140 vs 125 (+10.7%).

The overall pattern is clear: the AMD card wins in 7 of 10 tests, with its largest victory in a modern DX12 workload. The NVIDIA card wins 3 tests, with its largest victory in OpenCL compute. The AMD card's wins are spread across gaming, compute, and legacy APIs, while the NVIDIA card's wins are concentrated in specific compute and API scenarios.

The Verdict

The data paints a clear picture for gaming performance. The AMD Radeon RX 6750 XT wins the modern 3DMark DX12 test by 35.4% and the PassMark G3D test by 6.2%. It also wins every PassMark DirectX gaming test except DirectX 10. For gamers targeting DirectX 11 or DirectX 12 titles — which covers the vast majority of modern games — the AMD card is the stronger choice based on benchmark evidence.

The NVIDIA GeForce RTX 2080 SUPER has one overwhelming advantage: OpenCL compute, where it leads by 83.5%. It also wins the Vulkan test by 11.9%, which could matter for Vulkan-based games. However, its overall average benchmark score is 7.1% lower (24170 vs 26011), and it sits in the 69th percentile versus the AMD card's 71st percentile.

The AMD card also offers more memory (12 GB vs 8 GB) and higher FP32 throughput (13.31 TFLOPS vs 11.15 TFLOPS), alongside a more modern 7 nm process and PCIe 4.0 interface. The NVIDIA card counters with a wider memory bus (256-bit vs 192-bit) and higher bandwidth (495.9 GB/s vs 432.0 GB/s), plus 384 tensor cores that the AMD card lacks.

For most gaming scenarios, the AMD Radeon RX 6750 XT is the data-backed pick. Its wins span more benchmark categories, its margins in gaming tests are consistent, and its modern architecture delivers higher raw throughput. The NVIDIA card is only preferable for users whose workloads specifically depend on OpenCL compute performance or Vulkan API efficiency, where its advantages are substantial. Given the AMD card's lower launch MSRP of 549 USD versus 699 USD for the NVIDIA card, the benchmark data supports the AMD card as the superior option for general use.