AMD Radeon RX 9070 vs NVIDIA GeForce RTX 2080 SUPER Comparison

The NVIDIA GeForce RTX 2080 SUPER and AMD Radeon RX 9070 are separated by nearly six years of GPU architecture, yet their average benchmark scores land within 1.2% of each other. The data shows the RX 9070 wins 8 of 10 head-to-head tests, but the RTX 2080 SUPER takes a decisive victory in Vulkan compute and edges out a win in one DirectX 12 test. The RX 9070 is the newer, faster card for most workloads, but the RTX 2080 SUPER still holds specific advantages that matter for certain users.

The Verdict

Choose the AMD Radeon RX 9070 if you want the stronger all-around performer. It wins the most modern and demanding test, 3DMark Steel Nomad DX12, by a massive 70.1% margin (6290 vs 1882). It also dominates in PassMark G3D (25381 vs 19490, a 23.2% lead), PassMark GPU Compute (14737 vs 8290, a 43.7% lead), and PassMark DirectX 11 (281 vs 165, a 41.3% lead). With 16 GB of memory versus 8 GB, it is the card for future-proofing and heavier workloads.

Choose the NVIDIA GeForce RTX 2080 SUPER if your priority is Vulkan performance. The NVIDIA card scores 111284 in Geekbench Vulkan versus the RX 9070’s 58705 — a 89.6% advantage for NVIDIA. It also wins PassMark DirectX 12 by a narrow 1.4% margin (75 vs 74). For users running Vulkan-based applications or games, the RTX 2080 SUPER is the clear pick despite its age.

Both cards sit at the 69th percentile of all GPUs, meaning they appeal to the same performance tier overall. The RX 9070 is the practical choice for new builds; the RTX 2080 SUPER is a niche pick for Vulkan-specific workloads.

Architecture Differences

The RTX 2080 SUPER uses the TU104 chip on NVIDIA’s Turing architecture, built on a 12 nm process at TSMC. It packs 13,600 million transistors on a 545 mm² die, with a transistor density of 25.0M per mm². The RX 9070 uses the Navi 48 chip on AMD’s RDNA 4.0 architecture, built on a 4 nm process at TSMC. It crams 53,900 million transistors into a 357 mm² die, achieving a density of 151.0M per mm² — roughly six times denser than the Turing chip.

The RX 9070 has more raw compute resources: 3584 shading units, 224 texture mapping units, and 128 ROPs, versus 3072 shaders, 192 TMUs, and 64 ROPs on the RTX 2080 SUPER. The AMD card also has more ray tracing cores (56 vs 48) and lacks tensor cores entirely; the NVIDIA card has 384 tensor cores. The RX 9070’s FP32 throughput is 36.13 TFLOPS, more than three times the RTX 2080 SUPER’s 11.15 TFLOPS. The RX 9070 also delivers FP16 at a 1:1 ratio (36.13 TFLOPS), while the RTX 2080 SUPER achieves 22.30 TFLOPS at a 2:1 ratio.

Memory architecture diverges sharply. The RTX 2080 SUPER uses 8 GB of GDDR6 on a 256-bit bus with 495.9 GB/s bandwidth. The RX 9070 doubles capacity to 16 GB of GDDR6 on the same 256-bit bus but pushes bandwidth to 644.6 GB/s. The RX 9070 also supports PCIe 5.0 x16, while the RTX 2080 SUPER is limited to PCIe 3.0 x16.

Where Each One Wins

The RX 9070 wins in every category except Vulkan compute and one legacy DirectX test. Its biggest victories come in modern, compute-heavy workloads: 3DMark Steel Nomad DX12 (70.1% ahead), PassMark GPU Compute (43.7% ahead), and PassMark DirectX 11 (41.3% ahead). It also wins in OpenCL (131539 vs 99226, a 24.6% lead), PassMark G3D (23.2% lead), PassMark DirectX 9 (33.8% lead), PassMark G2D (28.1% lead), and PassMark DirectX 10 (a marginal 0.7% lead).

The RTX 2080 SUPER’s wins are narrow but meaningful. Its 89.6% lead in Geekbench Vulkan is the single largest margin in any test. That suggests NVIDIA’s Turing architecture retains a significant advantage in Vulkan compute pipelines. The 1.4% lead in PassMark DirectX 12 is negligible, but it does show the older card is not entirely obsolete in the latest DirectX API.

For gaming, the RX 9070 is the obvious choice based on the 3DMark Steel Nomad DX12 result and PassMark G3D. For productivity, the RX 9070’s OpenCL and GPU compute leads make it the better option. For Vulkan-specific applications — emulators, certain game engines, or compute frameworks — the RTX 2080 SUPER is unexpectedly superior.

FAQ

Q: Which card has a higher average benchmark score?

A: The RTX 2080 SUPER has an average benchmark score of 24170, while the RX 9070 scores 23877. The NVIDIA card leads by 1.2%, despite losing most individual tests.

Q: How do these cards compare to their nearest rivals?

A: The RTX 2080 SUPER is 0.4% ahead of the AMD Radeon RX 6800S and 1.1% behind the AMD Radeon RX 6600 XT. The RX 9070 is 1.1% ahead of the NVIDIA GeForce RTX 3080 Mobile and 0.8% behind the RX 6800S.

Q: Which card has more ray tracing cores?

A: The RX 9070 has 56 ray tracing cores, while the RTX 2080 SUPER has 48. The AMD card has a 16.7% advantage in ray tracing core count.

Q: What is the memory capacity difference?

A: The RX 9070 has 16 GB of GDDR6 memory, exactly double the 8 GB on the RTX 2080 SUPER. The RX 9070 also has higher bandwidth at 644.6 GB/s versus 495.9 GB/s.

Q: Which card is more power-efficient?

A: The RX 9070 has a lower TDP of 220 W compared to the RTX 2080 SUPER’s 250 W, and it requires a 550 W suggested PSU versus 600 W for the NVIDIA card, despite delivering significantly higher performance.

Q: Does the RTX 2080 SUPER support newer PCIe standards?

A: No. The RTX 2080 SUPER uses PCIe 3.0 x16, while the RX 9070 uses PCIe 5.0 x16. The AMD card supports a much newer bus interface.

Head-to-Head Benchmarks

The largest single margin in the entire comparison is Geekbench Vulkan, where the RTX 2080 SUPER scores 111284 against the RX 9070’s 58705 — a 89.6% lead for NVIDIA. This is the clearest indicator that Turing’s Vulkan implementation is far ahead of RDNA 4.0 in this specific test.

The 3DMark Steel Nomad DX12 result is the reverse story. The RX 9070 scores 6290 versus 1882 for the RTX 2080 SUPER, a 70.1% margin. This is the most modern test in the set and the best predictor of current DX12 gaming performance. The RX 9070’s lead here is enormous and decisive.

PassMark GPU Compute shows the RX 9070’s raw throughput advantage: 14737 versus 8290, a 43.7% lead. This aligns with the FP32 specification gap, where the AMD card delivers 36.13 TFLOPS versus 11.15 TFLOPS for NVIDIA.

In PassMark DirectX 11, the RX 9070 scores 281 versus 165, a 41.3% lead. This is a legacy API, but it still matters for older games. The RX 9070 also wins PassMark DirectX 9 by 33.8% (343 vs 227) and PassMark G2D by 28.1% (1280 vs 920).

OpenCL results favor the RX 9070 by 24.6% (131539 vs 99226). PassMark G3D gives the RX 9070 a 23.2% win (25381 vs 19490). The closest test is PassMark DirectX 10, where the RX 9070 wins by just 0.7% (141 vs 140). The only other NVIDIA win is PassMark DirectX 12, by 1.4% (75 vs 74) — a statistical tie.

The overall tally is 8 wins for the RX 9070 and 2 for the RTX 2080 SUPER. But the NVIDIA wins are not trivial: the Vulkan margin is the largest of any test, and the DirectX 12 win, however small, shows the older architecture still has life in the newest API.

Specification Differences

The two cards differ in nearly every measurable specification. The RX 9070 uses a 4 nm process versus 12 nm for the RTX 2080 SUPER. The RX 9070 has 53,900 million transistors on a 357 mm² die; the RTX 2080 SUPER has 13,600 million on a 545 mm² die. Transistor density is 151.0M per mm² versus 25.0M per mm².

Clock speeds favor the RX 9070 at the top end: boost clock of 2520 MHz versus 1815 MHz for the RTX 2080 SUPER. The base clocks are closer (1330 MHz vs 1650 MHz, with NVIDIA actually higher), but the AMD card has a game clock of 2070 MHz. Memory clocks differ too: the RX 9070 runs at 2518 MHz (20.1 Gbps effective) versus 1937 MHz (15.5 Gbps effective) for NVIDIA.

Compute resources: 3584 shading units, 224 TMUs, and 128 ROPs on the RX 9070; 3072 shading units, 192 TMUs, and 64 ROPs on the RTX 2080 SUPER. Ray tracing cores: 56 vs 48. Tensor cores: 384 on NVIDIA, none listed on AMD. Pixel rate is 322.6 GPixel/s for AMD versus 116.2 GPixel/s for NVIDIA. Texture rate is 564.5 GTexel/s versus 348.5 GTexel/s.

Power and connectivity: The RX 9070 has a 220 W TDP and requires a 550 W PSU; the RTX 2080 SUPER has a 250 W TDP and requires a 600 W PSU. Both are dual-slot, but the RX 9070 uses 2x 8-pin connectors while the RTX 2080 SUPER uses 1x 6-pin + 1x 8-pin. The RX 9070 uses PCIe 5.0 x16; the RTX 2080 SUPER uses PCIe 3.0 x16. Display outputs: the RX 9070 has 1x HDMI 2.1b and 3x DisplayPort 2.1a; the RTX 2080 SUPER has 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C.

The RTX 2080 SUPER is end-of-life with a launch MSRP of 699 USD. The RX 9070 is active with a launch MSRP of 549 USD. The RX 9070 also has a smaller physical footprint of 357 mm² versus 545 mm² for the NVIDIA card, though its dimensions are not listed.