AMD Radeon RX 9070 vs NVIDIA GeForce RTX 2080 Comparison

The AMD Radeon RX 9070 is the decisive winner in this comparison, taking 9 of 10 head-to-head benchmarks and delivering an average benchmark score of 23,877 versus 22,895 for the NVIDIA GeForce RTX 2080. The RX 9070’s lead is not merely consistent but often dramatic, with the largest margin reaching 259% in a modern DirectX 12 test. The RTX 2080 manages only a single victory, albeit a significant one in Vulkan compute, but the overall data paints a clear picture of generational dominance for AMD’s RDNA 4.0 architecture.

Head-to-Head Benchmarks

The most striking result is in 3DMark Steel Nomad (DX12), where the RX 9070 scores 6,290 against the RTX 2080’s 1,752. This represents a 259% advantage, meaning the AMD card delivers over three and a half times the performance in this specific workload. This is not a marginal improvement; it is a complete overhaul in raw rendering capability under DirectX 12. The delta is so large that it dwarfs every other comparison in the dataset.

Beyond that headline number, the RX 9070 shows consistent strength across legacy and compute-oriented tests. In Passmark GPU Compute, it scores 14,737 versus 7,872, a 87.2% lead. This near-doubling of compute throughput aligns with the card’s raw specifications, which show a FP32 rating of 36.13 TFLOPS against the RTX 2080’s 10.07 TFLOPS. The result is a clear indicator that the RX 9070 is built for heavy parallel workloads, not just gaming. Similarly, in Geekbench OpenCL, the RX 9070 posts 131,539 points, a 44.1% improvement over the RTX 2080’s 91,313, reinforcing the compute advantage.

The gap narrows considerably in older DirectX API tests, but the RX 9070 still wins. In Passmark DirectX 11, the score is 281 versus 158, a 77.8% lead. For Passmark DirectX 9, the RX 9070 wins 343 to 223, a 53.8% margin. Even in Passmark DirectX 10, the RX 9070 edges ahead 141 to 136, a modest 3.7% gain. The pattern suggests that while the RX 9070 excels in modern and compute-heavy scenarios, it also handles legacy APIs without regression, though the relative advantage shrinks as the workload becomes older and less parallel.

The 2D performance also favors the new card. Passmark G2D shows the RX 9070 at 1,280 versus 907, a 41.1% lead. This is relevant for desktop compositing and non-3D tasks. The Passmark DirectX 12 result is the closest of all, with the RX 9070 scoring 74 and the RTX 2080 scoring 72, a mere 2.8% difference. This near-parity in a pure DX12 feature test is surprising given the massive gap in the 3DMark Steel Nomad DX12 test, suggesting that the latter stresses raw throughput while the former may be limited by other factors.

The single loss for the RX 9070 comes in Geekbench Vulkan, where the RTX 2080 scores 107,797 against 58,705, a 45.5% deficit for the AMD card. This is a substantial reversal and indicates that NVIDIA’s Turing architecture retains a significant advantage in Vulkan compute workloads. The RTX 2080’s FP16 performance of 20.14 TFLOPS (2:1 ratio) versus the RX 9070’s FP16 of 36.13 TFLOPS (1:1) does not explain this, as the AMD card has higher raw numbers. The result suggests driver or architectural scheduling differences favor NVIDIA in this specific API. The Passmark G3D score, often used as a general gaming proxy, still goes to the RX 9070 at 25,381 versus 18,720, a 35.6% win.

The Verdict

The data is unambiguous: the AMD Radeon RX 9070 is the superior graphics card for the vast majority of workloads. It wins 9 out of 10 benchmarks, including all DirectX variants, OpenCL, and compute tests. The average benchmark score of 23,877 places it in the 69th percentile of all GPUs, while the RTX 2080 sits at the 68th percentile with an average of 22,895. The RX 9070’s nearest rivals include the NVIDIA GeForce RTX 2080 SUPER (avg score 24,170, -1.2% delta) and the AMD Radeon RX 6800S (avg score 24,063, -0.8% delta), showing it is competitive with a slightly higher-tier card from the previous generation.

For gamers targeting modern DirectX 12 titles, the RX 9070 is the obvious choice. The 259% lead in 3DMark Steel Nomad is a strong predictor of future-proofing, as that test is designed to stress current and next-generation rendering techniques. The RTX 2080’s single Vulkan win is notable but narrow in scope; it does not compensate for the losses elsewhere. If a user’s primary application relies heavily on Vulkan compute, the RTX 2080 holds a specific advantage, but this is a niche scenario.

The RTX 2080 is end-of-life, while the RX 9070 is active in production. This status difference, combined with the benchmark dominance, makes the RX 9070 the rational purchase for any new build. The RTX 2080’s only saving grace is its Vulkan performance, but the RX 9070’s wins in OpenCL and compute suggest that the AMD card is more versatile overall. The verdict is that the RX 9070 is the stronger card, and the RTX 2080 should only be considered for users with workloads that are exclusively Vulkan-based.

Where Each One Wins

The AMD Radeon RX 9070 wins in every DirectX scenario, from legacy DX9 to modern DX12, and in all compute tests except Vulkan. Its 87.2% lead in Passmark GPU Compute and 44.1% lead in Geekbench OpenCL make it the clear choice for GPU-accelerated rendering, machine learning inference, and scientific computing that leverages OpenCL. The 35.6% win in Passmark G3D also makes it the better gaming card for the vast majority of titles that use DirectX. The 41.1% lead in G2D indicates better desktop responsiveness and 2D application performance.

The NVIDIA GeForce RTX 2080 wins exclusively in Geekbench Vulkan, with a 45.5% advantage. This makes it the superior option for software that is heavily optimized for Vulkan compute, such as certain emulators, CAD applications, or specialized rendering engines that prefer this API. The RTX 2080’s 46 RT cores and 368 tensor cores are present, but the benchmark data does not show a distinct win in ray tracing or tensor workloads specifically. The single Vulkan win is the only data point in its favor.

For a user deciding between the two, the choice hinges on the software ecosystem. If the primary applications are DirectX-based games or OpenCL compute, the RX 9070 is overwhelmingly better. If a user has a specific Vulkan compute workload that is critical, the RTX 2080 offers a performance edge that cannot be ignored. However, the breadth of the RX 9070’s wins across 9 benchmarks provides a much wider safety net for general-purpose use.

FAQ

Q: Which card has a higher average benchmark score?

A: The AMD Radeon RX 9070 has an average benchmark score of 23,877, while the NVIDIA GeForce RTX 2080 has an average of 22,895.

Q: What is the largest performance difference between the two cards?

A: The largest difference is in 3DMark Steel Nomad (DX12), where the RX 9070 scores 6,290 and the RTX 2080 scores 1,752, giving the AMD card a 259% lead.

Q: Does the NVIDIA GeForce RTX 2080 win any benchmarks?

A: Yes, the RTX 2080 wins the Geekbench Vulkan test with a score of 107,797 versus the RX 9070’s 58,705, a 45.5% advantage.

Q: How do the cards compare in compute performance?

A: The RX 9070 leads in Passmark GPU Compute with 14,737 points versus 7,872, a 87.2% difference, and in Geekbench OpenCL with 131,539 versus 91,313, a 44.1% difference.

Q: What are the production statuses of the two cards?

A: The AMD Radeon RX 9070 is listed as "Active" in production, while the NVIDIA GeForce RTX 2080 is listed as "End-of-life".

Q: Which card has a higher percentile ranking among all GPUs?

A: The RX 9070 is in the 69th percentile of all GPUs, while the RTX 2080 is in the 68th percentile.

Architecture Differences

The two cards represent fundamentally different architectural eras. The AMD Radeon RX 9070 uses the Navi 48 chip built on RDNA 4.0 architecture, manufactured on a 4 nm process at TSMC. This is a modern, dense design with 53,900 million transistors packed into a 357 mm² die, yielding a transistor density of 151.0M / mm². The NVIDIA GeForce RTX 2080 uses the TU104 chip based on the older Turing architecture, built on a 12 nm process, also at TSMC. It contains 13,600 million transistors on a much larger 545 mm² die, resulting in a transistor density of just 25.0M / mm².

The RX 9070 features 3,584 shading units, 224 TMUs, and 128 ROPs, alongside 56 ray tracing cores. The RTX 2080 has 2,944 shading units, 184 TMUs, and 64 ROPs, with 46 ray tracing cores and 368 tensor cores. The AMD card does not list tensor cores, while the NVIDIA card includes them for AI acceleration. The RX 9070’s FP32 performance is 36.13 TFLOPS, and its FP16 is also 36.13 TFLOPS (1:1 ratio). The RTX 2080’s FP32 is 10.07 TFLOPS, but its FP16 is 20.14 TFLOPS (2:1 ratio).

The memory subsystems also differ. The RX 9070 uses 16 GB of GDDR6 on a 256-bit bus with 644.6 GB/s bandwidth. The RTX 2080 uses 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth. The RX 9070 has a higher pixel rate of 322.6 GPixel/s versus 109.4 GPixel/s for the RTX 2080, and a texture rate of 564.5 GTexel/s versus 314.6 GTexel/s. The RX 9070 supports PCIe 5.0 x16, while the RTX 2080 uses PCIe 3.0 x16.

Specification Differences

The most obvious difference is memory capacity: the RX 9070 has 16 GB, the RTX 2080 has 8 GB. The RX 9070 also leads in memory bandwidth at 644.6 GB/s versus 448.0 GB/s. The clock speeds differ, with the RX 9070 having a base clock of 1330 MHz and boost of 2520 MHz, while the RTX 2080 has a base of 1515 MHz and boost of 1710 MHz. The RX 9070 also has a listed game clock of 2070 MHz, which the RTX 2080 lacks.

The power connectors are different: the RX 9070 requires 2x 8-pin, while the RTX 2080 uses 1x 6-pin + 1x 8-pin. Both have a TDP of 220 W and 215 W respectively, and both suggest a 550 W PSU. The display outputs also differ: the RX 9070 has 1x HDMI 2.1b and 3x DisplayPort 2.1a, while the RTX 2080 has 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C. The RTX 2080 has physical dimensions listed as 267 mm in length, 116 mm in height, and 35 mm in width; the RX 9070’s dimensions are not provided. The release dates are far apart: the RX 9070 launched on 2025-03-05, while the RTX 2080 launched on 2018-09-19. The launch MSRP for the RX 9070 is 549 USD, and for the RTX 2080 it is 699 USD.