AMD Radeon VII vs NVIDIA GeForce RTX 4080 SUPER Comparison

FAQ

Q: Which GPU has the higher average benchmark score in the database?

A: The AMD Radeon VII has an average benchmark score of 66004, while the NVIDIA GeForce RTX 4080 SUPER has an average score of 54209. Despite the NVIDIA card winning every head-to-head test, the Radeon VII holds a higher average across all recorded benchmarks.

Q: How does each card rank against all other GPUs?

A: The Radeon VII sits in the 90th percentile of all GPUs, while the RTX 4080 SUPER sits in the 86th percentile. The Radeon VII's nearest rivals include the NVIDIA Tesla T4 (1.1% below), Tesla P40 (1.4% above), and AMD Radeon Pro WX 9100 (2.8% above). The RTX 4080 SUPER's nearest rivals include the RTX 4080 (0.1% below), Radeon Pro W5700X (1.1% below), and Radeon RX 6750 GRE 12 GB (2.7% below).

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

A: Both cards have 16 GB of memory, but the Radeon VII uses HBM2 on a 4096-bit bus with 1.02 TB/s bandwidth, while the RTX 4080 SUPER uses GDDR6X on a 256-bit bus with 736.3 GB/s bandwidth.

Q: Which card has a higher boost clock?

A: The RTX 4080 SUPER boosts to 2550 MHz, compared to the Radeon VII's boost of 1750 MHz. The base clocks are 2295 MHz for the RTX 4080 SUPER and 1400 MHz for the Radeon VII.

Q: What are the launch MSRP values for these two cards?

A: The AMD Radeon VII has a launch MSRP of 699 USD, and the NVIDIA GeForce RTX 4080 SUPER has a launch MSRP of 999 USD.

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

A: The RTX 4080 SUPER scores 6600 in this test, while the Radeon VII scores 2304. The NVIDIA card wins by 65.1%, a decisive margin in this DirectX 12 workload.

Architecture Differences

The AMD Radeon VII is built on the Vega 20 chip using the GCN 5.1 architecture, fabricated on a 7 nm process at TSMC. It contains 13,230 million transistors on a 331 mm² die, giving a transistor density of 40.0M per mm². The NVIDIA GeForce RTX 4080 SUPER uses the AD103 chip with the Ada Lovelace architecture, also from TSMC but on a 5 nm process. It packs 45,900 million transistors into a 379 mm² die, for a density of 121.1M per mm². The NVIDIA chip crams over three times the transistors into a slightly larger area.

The Radeon VII has 3840 shading units, 240 texture mapping units, and 64 ROPs. The RTX 4080 SUPER has 10240 shading units, 320 TMUs, and 112 ROPs. The NVIDIA card also includes 80 RT cores and 320 tensor cores, while the Radeon VII has none of either, reflecting its pre-ray-tracing GCN design. The API support differs accordingly: the Radeon VII supports DirectX 12 (12_1), while the RTX 4080 SUPER supports DirectX 12 Ultimate (12_2). Vulkan support is 1.3 for the AMD card and 1.4 for the NVIDIA card. Both support OpenGL 4.6.

Memory architecture is a major differentiator. The Radeon VII uses 16 GB of HBM2 on a 4096-bit bus, achieving 1.02 TB/s of bandwidth. The RTX 4080 SUPER uses 16 GB of GDDR6X on a 256-bit bus, achieving 736.3 GB/s. The AMD card's HBM2 provides significantly more raw bandwidth, but the NVIDIA card's architecture compensates with higher clock speeds and more compute units.

The bus interface also differs: the Radeon VII uses PCIe 3.0 x16, while the RTX 4080 SUPER uses PCIe 4.0 x16. The physical design reflects the performance gap, with the Radeon VII being a dual-slot card measuring 280 mm in length, 125 mm in height, and 40 mm in width. The RTX 4080 SUPER is a triple-slot card at 310 mm long, 140 mm tall, and 61 mm wide. Power delivery uses 2x 8-pin connectors on the AMD card and a single 16-pin connector on the NVIDIA card. The Radeon VII has a 295 W TDP with a 600 W suggested PSU, while the RTX 4080 SUPER has a 320 W TDP with a 700 W suggested PSU. Display outputs are similar, with both offering 1x HDMI and 3x DisplayPort, though the Radeon VII has HDMI 2.0b while the RTX 4080 SUPER has HDMI 2.1.

Where Each One Wins

The recorded head-to-head data shows no test where the Radeon VII comes out ahead. The RTX 4080 SUPER wins all three shared benchmarks: 3DMark Steel Nomad DX12, Geekbench OpenCL, and Geekbench Vulkan. However, the Radeon VII's higher average benchmark score, 66004 versus 54209, indicates that the broader benchmark database includes tests where the AMD card performs relatively better, likely compute-oriented workloads where its HBM2 bandwidth and GCN architecture remain effective.

The Radeon VII's 1.02 TB/s memory bandwidth is nearly 40% higher than the RTX 4080 SUPER's 736.3 GB/s. In workloads that are bandwidth-bound, the AMD card's HBM2 stack can keep more data in flight. Its percentile ranking of 90 versus 86 also suggests that in the full database of GPUs, the Radeon VII outperforms a slightly larger share of the field than the RTX 4080 SUPER does, even though the NVIDIA card dominates in the specific tests shared between these two.

The RTX 4080 SUPER wins in every shared test category. Its 3DMark Steel Nomad score of 6600 is nearly three times the Radeon VII's 2304. In Geekbench Vulkan, it scores 260075 versus 91788, and in Geekbench OpenCL, 219065 versus 91947. The NVIDIA card's advantage comes from its higher shading unit count, higher clocks, and modern Ada Lovelace feature set including RT and tensor cores.

Specification Differences

| Specification | AMD Radeon VII | NVIDIA GeForce RTX 4080 SUPER |

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

| Architecture | GCN 5.1 | Ada Lovelace |

| Process node | 7 nm | 5 nm |

| Transistors | 13,230 million | 45,900 million |

| Die size | 331 mm² | 379 mm² |

| Transistor density | 40.0M / mm² | 121.1M / mm² |

| Base clock | 1400 MHz | 2295 MHz |

| Boost clock | 1750 MHz | 2550 MHz |

| Memory type | HBM2 | GDDR6X |

| Memory bus width | 4096 bit | 256 bit |

| Memory bandwidth | 1.02 TB/s | 736.3 GB/s |

| Shading units | 3840 | 10240 |

| TMUs | 240 | 320 |

| ROPs | 64 | 112 |

| RT cores | None | 80 |

| Tensor cores | None | 320 |

| FP32 | 13.44 TFLOPS | 52.22 TFLOPS |

| FP16 | 26.88 TFLOPS (2:1) | 52.22 TFLOPS (1:1) |

| Pixel rate | 112.0 GPixel/s | 285.6 GPixel/s |

| Texture rate | 420.0 GTexel/s | 816.0 GTexel/s |

| TDP | 295 W | 320 W |

| Slot width | Dual-slot | Triple-slot |

| Power connectors | 2x 8-pin | 1x 16-pin |

| Suggested PSU | 600 W | 700 W |

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

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

| Vulkan support | 1.3 | 1.4 |

| Dimensions | 280 x 125 x 40 mm | 310 x 140 x 61 mm |

| Release date | 2019-02-06 | 2024-01-30 |

| Launch MSRP | 699 USD | 999 USD |

Head-to-Head Benchmarks

The shared benchmark suite consists of three tests, and the RTX 4080 SUPER wins all of them by wide margins. In 3DMark Steel Nomad DX12, the RTX 4080 SUPER scores 6600 against the Radeon VII's 2304, a delta of 65.1%. This is the largest percentage gap of the three tests and reflects the NVIDIA card's superior rasterization throughput, with 52.22 TFLOPS FP32 versus 13.44 TFLOPS.

In Geekbench Vulkan, the RTX 4080 SUPER scores 260075 against 91788 for the Radeon VII, a 64.7% advantage. The Vulkan test exercises the graphics pipeline heavily, and the NVIDIA card's 10240 shading units at a 2550 MHz boost clock overwhelm the Radeon VII's 3840 units at 1750 MHz.

In Geekbench OpenCL, the RTX 4080 SUPER scores 219065 against 91947, a 58% advantage. This is the smallest margin of the three, but still a dominant result. OpenCL compute can benefit from memory bandwidth, and the Radeon VII's 1.02 TB/s HBM2 does narrow the gap somewhat, but not enough to overcome the NVIDIA card's raw compute advantage.

The Radeon VII's best relative showing is in Geekbench OpenCL, where its 58% deficit is smaller than the 64.7% and 65.1% deficits in the other two tests. This aligns with the memory bandwidth story: HBM2 helps in compute-heavy OpenCL workloads, but the NVIDIA card's architecture is simply too far ahead in every other metric.

The Verdict

The data is unambiguous in the head-to-head tests: the NVIDIA GeForce RTX 4080 SUPER wins every shared benchmark, with deltas ranging from 58% to 65.1%. In 3DMark Steel Nomad DX12, the NVIDIA card nearly triples the Radeon VII's score. In Geekbench Vulkan and OpenCL, it more than doubles it. For any user choosing between these two for modern DirectX 12 gaming or Vulkan workloads, the RTX 4080 SUPER is the clear pick based on the recorded results.

The Radeon VII does hold one statistical advantage: its average benchmark score across the database is 66004 versus 54209 for the RTX 4080 SUPER, and its percentile ranking is 90 versus 86. This suggests that in the broader GPU landscape, the Radeon VII performs well in a wider set of legacy or compute-oriented tests, likely due to its 1.02 TB/s HBM2 bandwidth. The RTX 4080 SUPER also has a lower average score than its nearest rival, the RTX 4080, by 0.1%, and sits below the Radeon Pro W5700X by 1.1% and the RX 6750 GRE 12 GB by 2.7% in the database's aggregate rankings.

For users prioritizing modern gaming performance, ray tracing capability, and DirectX 12 Ultimate features, the RTX 4080 SUPER is the only choice between these two. It offers RT cores, tensor cores, and a 5 nm process with 121.1M transistors per mm². For users working with legacy compute workloads that benefit from massive memory bandwidth, the Radeon VII's 1.02 TB/s HBM2 and 90th percentile ranking make it a relevant option in the database, despite its end-of-life status and 2019 release date. The RTX 4080 SUPER, released in 2024, represents a later generation with a higher launch MSRP of 999 USD versus 699 USD for the Radeon VII, and the benchmark data shows that price difference is reflected in performance across every shared test.