AMD Radeon RX 9070 GRE vs NVIDIA GeForce RTX 3080 Ti Comparison

Head-to-Head Benchmarks

The recorded data shows a split decision between these two cards. In the 3DMark Steel Nomad DX12 test, the AMD Radeon RX 9070 GRE posts a score of 5424, while the NVIDIA GeForce RTX 3080 Ti trails at 5077. That gives AMD a 6.8% advantage in this modern DirectX 12 workload, a meaningful margin that points to the newer architecture handling current rendering paths more efficiently.

The picture flips dramatically in the Geekbench OpenCL compute test. Here, the RTX 3080 Ti scores 170037 against the RX 9070 GRE's 109309, a 35.7% deficit for AMD. This is a substantial gap in raw compute throughput, and it highlights how differently these two cards are built under the hood. The NVIDIA card's massive shading unit count and tensor core array give it a clear edge in this type of workload.

Looking at the broader database context, the RX 9070 GRE holds an 87th percentile ranking among all GPUs, while the RTX 3080 Ti sits at the 83rd percentile. The average benchmark score for the RX 9070 GRE is 57367, which places it within 1.8% of the AMD Radeon RX 6950 XT, its closest rival in the database. The RTX 3080 Ti averages 41187, sitting just 2% ahead of the NVIDIA GeForce RTX 5070 and 0.8% ahead of the AMD Radeon Pro 5300. These adjacent rankings show that despite the RTX 3080 Ti's compute advantage in one test, its overall database position is lower than the newer AMD card.

The 3DMark result deserves extra attention because it uses a modern DirectX 12 renderer with ray-traced features. A 6.8% win there suggests the RX 9070 GRE has better efficiency in contemporary game workloads, even though its raw compute numbers look weaker. The Geekbench OpenCL result, by contrast, measures pure parallel computation across all cores, which plays directly into the strengths of the RTX 3080 Ti's 10240 shading units.

One win each is the headline, but the margins matter more than the count. AMD's win is modest at 6.8%; NVIDIA's is overwhelming at 35.7%. Anyone choosing between these cards needs to weigh which workload type matches their usage patterns, since neither card dominates across the board.

Architecture Differences

The RX 9070 GRE is built on TSMC's 4 nm process with the Navi 48 chip, while the RTX 3080 Ti uses Samsung's 8 nm process with the GA102 die. This process gap explains a great deal about efficiency and transistor density. The AMD chip packs 53,900 million transistors into a 357 mm² die, yielding a density of 151.0M transistors per mm². The NVIDIA chip uses 28,300 million transistors across a much larger 628 mm² die, resulting in just 45.1M transistors per mm². The smaller, denser AMD die is a direct result of the newer manufacturing node.

Clock speeds reflect this difference. The RX 9070 GRE runs a base clock of 1420 MHz and boosts to 2790 MHz, with a game clock rated at 2220 MHz. The RTX 3080 Ti operates at a 1365 MHz base and 1665 MHz boost. The AMD card's higher boost clock, combined with its RDNA 4.0 architecture, allows it to achieve comparable FP32 throughput of 34.28 TFLOPS versus the RTX 3080 Ti's 34.10 TFLOPS, despite far fewer shading units. AMD does this with 3072 shading units; NVIDIA needs 10240 to reach essentially the same number. Both cards deliver FP16 at a 1:1 ratio with FP32.

Memory configurations differ in type and bus width but match in capacity. Both cards have 12 GB of VRAM. The RX 9070 GRE uses GDDR6 on a 192 bit bus, delivering 432.0 GB/s of bandwidth. The RTX 3080 Ti uses GDDR6X on a 384 bit bus, delivering 912.4 GB/s, more than double the AMD card's bandwidth. This is a significant architectural difference that affects texture-heavy and high-resolution workloads.

The RTX 3080 Ti carries 80 RT cores and 320 tensor cores, while the RX 9070 GRE has 48 RT cores and no tensor core equivalent listed. NVIDIA's tensor cores support its proprietary DLSS features, while AMD's RDNA 4.0 relies on its own upscaling approach. The AMD architecture supports PCIe 5.0 x16, while the RTX 3080 Ti uses PCIe 4.0 x16. Display output also differs: the RX 9070 GRE offers 1x HDMI 2.1b and 3x DisplayPort 2.1a, while the RTX 3080 Ti provides 1x HDMI 2.1 and 3x DisplayPort 1.4a.

Power requirements tell a clear efficiency story. The RX 9070 GRE has a TDP of 220 W with a suggested 550 W PSU, using dual 8-pin connectors. The RTX 3080 Ti draws 350 W with a suggested 750 W PSU, using a single 12-pin connector. The AMD card delivers comparable FP32 performance at nearly two-thirds the power draw.

The RTX 3080 Ti also includes a full suite of PassMark legacy DirectX tests, indicating its driver stack has been validated across older APIs. The RX 9070 GRE does not have these results recorded in the database. Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Where Each One Wins

The RX 9070 GRE wins in modern gaming workloads. Its 6.8% advantage in 3DMark Steel Nomad DX12 shows that RDNA 4.0 handles current generation rendering techniques, including ray tracing and mesh shaders, more efficiently than Ampere. The lower TDP of 220 W versus 350 W also makes it the better choice for smaller form factor builds or systems with older power supplies, since the suggested PSU drops from 750 W to 550 W. The PCIe 5.0 x16 interface provides additional headroom for future platform compatibility, and the DisplayPort 2.1a outputs support higher bandwidth displays than the RTX 3080 Ti's DisplayPort 1.4a.

The RTX 3080 Ti wins decisively in compute-heavy tasks. Its 35.7% OpenCL advantage makes it the stronger pick for GPU compute workloads, rendering tasks, or any application that scales with raw shading unit count. The 912.4 GB/s memory bandwidth is more than double the AMD card's 432.0 GB/s, which matters for large dataset operations and texture-heavy scenarios. The 320 tensor cores provide dedicated hardware for AI-accelerated workloads and DLSS upscaling, features that the RX 9070 GRE cannot match with its architecture.

For users who play a mix of older DirectX titles, the RTX 3080 Ti has recorded PassMark results across DirectX 9, 10, 11, and 12, plus a G2D score of 1091 and G3D score of 26896. The RX 9070 GRE has no such legacy test data in the database, which suggests its driver maturity across older APIs has not been validated to the same degree. Users running legacy games may prefer the card with documented compatibility across those API generations.

The production status matters for long-term planning. The RX 9070 GRE is listed as Active, while the RTX 3080 Ti is End-of-life with a successor in the GeForce 40 series. New stock and driver support will continue for the AMD card, while the NVIDIA card is already past its prime in the product cycle.

FAQ

Q: Which card is faster in DirectX 12 gaming benchmarks?

A: The AMD Radeon RX 9070 GRE scores 5424 in 3DMark Steel Nomad DX12, which is 6.8% higher than the RTX 3080 Ti's 5077.

Q: Why does the RTX 3080 Ti have such a large lead in OpenCL compute?

A: The RTX 3080 Ti scores 170037 in Geekbench OpenCL versus 109309 for the RX 9070 GRE, a 35.7% advantage. This comes from its 10240 shading units and 320 tensor cores, compared to 3072 shading units for the AMD card.

Q: Do both cards have the same amount of memory?

A: Yes, both have 12 GB. However, the RTX 3080 Ti uses GDDR6X on a 384 bit bus with 912.4 GB/s bandwidth, while the RX 9070 GRE uses GDDR6 on a 192 bit bus with 432.0 GB/s.

Q: Which card requires a more powerful power supply?

A: The RTX 3080 Ti has a 350 W TDP and suggests a 750 W PSU. The RX 9070 GRE has a 220 W TDP and suggests a 550 W PSU.

Q: How do these cards rank among all GPUs in the database?

A: The RX 9070 GRE is at the 87th percentile with an average benchmark score of 57367. The RTX 3080 Ti is at the 83rd percentile with an average score of 41187.

Q: Which card is still in production?

A: The RX 9070 GRE is listed as Active production. The RTX 3080 Ti is End-of-life and has been succeeded by the GeForce 40 series.

The Verdict

The data supports a clear split recommendation. Choose the AMD Radeon RX 9070 GRE if your priority is modern gaming performance, lower power draw, and future platform compatibility. It wins the DirectX 12 benchmark by 6.8%, uses 130 W less power, requires a smaller PSU, and sits higher in the overall database percentile ranking at 87 versus 83. Its Active production status and PCIe 5.0 interface make it the more future-proof option.

Choose the NVIDIA GeForce RTX 3080 Ti if your workload is compute-intensive. Its 35.7% OpenCL lead is the largest margin recorded in this comparison, and its 912.4 GB/s memory bandwidth and 320 tensor cores provide capabilities the AMD card lacks. The PassMark legacy DirectX results also give it documented compatibility across older APIs. Be aware that it requires a 750 W PSU, draws 350 W, and is End-of-life, but for raw compute throughput, the data favors NVIDIA decisively.

The average benchmark scores suggest the RX 9070 GRE is the more balanced card overall, with a 57367 average versus 41187 for the RTX 3080 Ti. The nearest rival data reinforces this: the RX 9070 GRE sits within 1.8% of the RX 6950 XT, a well-regarded high-end card, while the RTX 3080 Ti's closest competition includes the RTX 5070 and mid-range workstation cards. If you need one card for everything, the AMD option has the better all-around profile. If your specific application rewards massive compute parallelism, the NVIDIA option remains a potent choice despite its age.

Specification Differences

| Specification | AMD Radeon RX 9070 GRE | NVIDIA GeForce RTX 3080 Ti |

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

| Architecture | RDNA 4.0 | Ampere |

| Process node | 4 nm | 8 nm |

| Foundry | TSMC | Samsung |

| Transistors | 53,900 million | 28,300 million |

| Die size | 357 mm² | 628 mm² |

| Transistor density | 151.0M / mm² | 45.1M / mm² |

| Base clock | 1420 MHz | 1365 MHz |

| Boost clock | 2790 MHz | 1665 MHz |

| Game clock | 2220 MHz | Not specified |

| Memory type | GDDR6 | GDDR6X |

| Memory bus | 192 bit | 384 bit |

| Memory bandwidth | 432.0 GB/s | 912.4 GB/s |

| Shading units | 3072 | 10240 |

| TMUs | 192 | 320 |

| ROPs | 96 | 112 |

| RT cores | 48 | 80 |

| Tensor cores | Not specified | 320 |

| Pixel rate | 267.8 GPixel/s | 186.5 GPixel/s |

| Texture rate | 535.7 GTexel/s | 532.8 GTexel/s |

| FP32 | 34.28 TFLOPS | 34.10 TFLOPS |

| TDP | 220 W | 350 W |

| Power connectors | 2x 8-pin | 1x 12-pin |

| Suggested PSU | 550 W | 750 W |

| Bus interface | PCIe 5.0 x16 | PCIe 4.0 x16 |

| Display outputs | 1x HDMI 2.1b, 3x DisplayPort 2.1a | 1x HDMI 2.1, 3x DisplayPort 1.4a |

| Production status | Active | End-of-life |

| Release date | 2025-05-07 | 2021-05-30 |

| Launch MSRP | 549 USD | 1,199 USD |

| Dimensions | Not specified | 285 mm / 11.2 inches length, 112 mm / 4.4 inches height, 40 mm / 1.6 inches width |