AMD Radeon RX 5700 vs NVIDIA GeForce RTX 3080 Ti Mobile Comparison

# The Verdict

The data separates these two mobile GPUs into two distinct performance classes. The NVIDIA GeForce RTX 3080 Ti Mobile leads in 8 of 10 head-to-head benchmark comparisons, while the AMD Radeon RX 5700 wins only the two DirectX 9 and G2D legacy tests. The average benchmark score of the RTX 3080 Ti Mobile is 25,740, placing it at the 71st percentile of all GPUs in the database. The RX 5700 averages 22,170, sitting at the 67th percentile. The NVIDIA card sits 16% higher in average score, a gap that widens dramatically in modern compute-heavy workloads. For raw compute and current-generation API performance, the RTX 3080 Ti Mobile is the clear choice.

# Architecture Differences

The two chips come from different foundries and design philosophies. NVIDIA uses the GA103 die built on Samsung's 8nm process, packing 22,000 million transistors into a 496 mm² die. That yields a transistor density of 44.4 million per mm². AMD counters with the Navi 10 chip on TSMC's 7nm node, fitting 10,300 million transistors into 251 mm², a density of 41.0 million per mm². Both are dense, modern designs, but the NVIDIA die is physically larger and packs more than twice the transistor count.

The GA103 is an Ampere architecture part with 7,424 shading units, 232 texture mapping units, and 96 ROPs. It also carries 58 ray tracing cores and 232 tensor cores. The Navi 10 in the RX 5700 is RDNA 1.0, with 2,304 shading units, 144 TMUs, and 64 ROPs. It has no ray tracing or tensor core hardware. The NVIDIA part is the more fully featured silicon, with dedicated acceleration blocks.

Clocks and memory also differ notably. The RTX 3080 Ti Mobile runs at a base 810 MHz and boosts to 1260 MHz, with memory effective at 2000 MHz or 16 Gbps. The RX 5700 runs a higher base clock at 1465 MHz, boosting to 1725 MHz, with a game clock of 1625 MHz and memory at 1750 MHz. The AMD card ships with 8GB of GDDR6 across a 256-bit bus, for 448.0 GB/s of bandwidth. NVIDIA answers with 16GB of GDDR6 on a 256-bit bus, reaching 512.0 GB/s. The bandwidth advantage is meaningful in 4K texturing and compute workloads.

# Where Each One Wins

The RTX 3080 Ti Mobile dominates modern API results. In the 3DMark Steel Nomad DX12 test, it scores 2,869 versus 1,862 for the RX 5700, a 54.1% win. Geekbench_OpenCL shows a 117,866 to 69,603 result, again a 69.3% margin. Vulkan follows a similar pattern: 107,502 against 64,166, a 67.5% advantage. These are large, consistent wins in compute and modern graphics.

The RX 5700 strikes back in legacy tests. In Passmark DirectX 9, it takes 209 to 201, a 3.8% lead. Passmark G2D shows 877 to 818, a 6.7% advantage. These are the only two wins for AMD.

The NVIDIA card also wins PassMark DX10 (131 vs 87, a 50.6% margin), DX11 (164 vs 101, 62.4%), and DX12 (88 vs 54, 63%). DirectX results show a pattern: the newer the API, the wider the gap. The same appears in PassMark G3D: 19,288 vs 14,314, a 34.7% margin, and GPU Compute: 8,471 vs 6,517, a 30% margin.

For builders targeting the latest APIs, ray tracing, or heavy compute, the RTX 3080 Ti Mobile is the card to pick. For users stuck in DirectX 9 or 2D legacy workloads, the RX 5700 makes more sense. But it is a narrow niche.

FAQ

Q: Which GPU is faster overall?

A: The NVIDIA GeForce RTX 3080 Ti Mobile has a 16% higher average benchmark score (25,740 vs 22,170).

Q: Is the AMD card ever faster?

A: Yes. It wins in PassMark DirectX 9 by 3.8% and PassMark G2D by 6.7%. These are the only two wins out of ten comparisons.

Q: What about API support?

A: The RTX 3080 Ti Mobile supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. The RX 5700 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

Q: Which card has more memory?

A: The RTX 3080 Ti Mobile has 16GB of GDDR6. The RX 5700 uses 8GB of GDDR6.

Q: What is the power connector requirement?

A: The AMD card uses one 6-pin and one 8-pin connector. The NVIDIA card needs no additional power connectors. It is rated at 115W TDP, while the AMD is a 180W part.

Head-to-Head Benchmarks

The NVIDIA GeForce RTX 3080 Ti Mobile wins the heaviest tests by wide margins. In 3DMark Steel Nomad DX12, it reaches 2,869 against 1,862 for the RX 5700. That is 54.1% ahead. The Geekbench OpenCL test shows 117,866 vs 69,603, a 69.3% lead. Vulkan follows the same story: 107,502 vs 64,166, a 67.5% margin.

DirectX 10 and 11 also favor the NVIDIA part. In DX10, the score goes 131 to 87, a 50.6% lead. In DX11, it is 164 to 101, a 62.4% advantage. DX12 repeats: 88 vs 54, a 63% lead.

AMD pulls back in the legacy APIs. PassMark DX9 shows 203 vs 201, a 3.8% win for the RX 5700. PassMark G2D is 877 vs 818, a 6.7% advantage. PassMark G3D goes the other way: 19,288 for the RTX 3080 Ti Mobile vs 14,314 for the RX 5700, a 34.7% NVIDIA win. PassMark GPU Compute: 8,471 vs 6,517, a 30% win for NVIDIA.

The story across all ten workloads: NVIDIA wins the modern API and compute tests, AMD wins the two legacy 2D/DX9 tests.

Specification Differences

| Field | NVIDIA GeForce RTX 3080 Ti Mobile | AMD Radeon RX 5700 |

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

| Chip | GA103 | Navi 10 |

| Architecture | Ampere | RDNA 1.0 |

| Process | 8nm Samsung | 7nm TSMC |

| Transistors | 22,000 million | 10,300 million |

| Die size | 496 mm² | 251 mm² |

| Transistor density | 44.4M / mm² | 41.0M / mm² |

| Base clock | 810 MHz | 1465 MHz |

| Boost clock | 1260 MHz | 1725 MHz |

| Memory | 16GB GDDR6 | 8GB GDDR6 |

| Memory bus | 256-bit | 256-bit |

| Bandwidth | 512.0 GB/s | 448.0 GB/s |

| Shading units | 7424 | 2304 |

| TMUs | 232 | 144 |

| ROPs | 96 | 64 |

| Ray tracing cores | 58 | None |

| Tensor cores | 232 | None |

| FP32 | 18.71 TFLOPS | 7.949 TFLOPS |

| TDP | 115W | 180W |

| Power connectors | None | 1x 6-pin + 1x 8-pin |

| Suggested PSU | None specified | 450W |

| Bus | PCIe 4.0 x16 | PCIe 4.0 x16 |

| Display outputs | Portable Device Dependent | 1x HDMI 2.0b, 3x DisplayPort 1.4a |