AMD Radeon Pro W5700 vs NVIDIA GeForce RTX 2080 Comparison

Head-to-Head Benchmarks

The benchmark data shows a clear, though not universal, performance hierarchy between the AMD Radeon Pro W5700 and the NVIDIA GeForce RTX 2080. Out of nine common tests, the RTX 2080 claims eight victories, while the W5700 manages a single, narrow win. The average benchmark score reinforces this split: the AMD card sits at 25,726, while the NVIDIA card records 22,895, a difference that places them in different performance tiers despite the W5700's higher percentile ranking among all GPUs (71st percentile versus 68th).

The largest single margin comes in the Geekbench Vulkan test, where the RTX 2080 scores 107,797 against the W5700's 70,706, a 34.4% advantage. This is a substantial gap that indicates the NVIDIA architecture handles Vulkan workloads with significantly greater efficiency. The DirectX 10 and DirectX 11 results follow a similar pattern, with the RTX 2080 leading by 35.3% (136 versus 88) and 34.2% (158 versus 104), respectively. These are not marginal differences; they represent a full tier of separation in legacy API performance.

The DirectX 12 test shows the RTX 2080 ahead by 25% (72 versus 54), while the OpenCL benchmark gives NVIDIA an 18.3% edge (91,313 versus 74,613). The Passmark G3D test, which often reflects overall gaming and graphics rendering capability, shows the RTX 2080 at 18,720 versus 14,520, a 22.4% lead. Even in GPU compute, where workstation cards often excel, the RTX 2080 wins by 17.5% (7,872 versus 6,495).

The AMD card's sole victory comes in the DirectX 9 test, where it scores 225 against 223, a 0.9% margin that is effectively a statistical tie. The Passmark G2D test is similarly close, with NVIDIA winning by only 0.9% (907 versus 899). These near-dead heats suggest that in older, simpler rendering paths, the two cards are functionally equivalent. The data does not support a scenario where the W5700 outperforms the RTX 2080 in anything beyond a negligible margin in one legacy API.

FAQ

Q: Which card has the higher average benchmark score?

A: The AMD Radeon Pro W5700 has a higher average benchmark score of 25,726, compared to the NVIDIA GeForce RTX 2080's 22,895. However, this aggregate figure is somewhat misleading, as the head-to-head tests show the RTX 2080 winning eight of nine comparisons.

Q: How do the cards compare in Geekbench Vulkan performance?

A: The NVIDIA GeForce RTX 2080 is significantly faster in Geekbench Vulkan, scoring 107,797 versus the AMD Radeon Pro W5700's 70,706. This represents a 34.4% advantage for NVIDIA.

Q: Is there any test where the AMD card wins?

A: Yes, the AMD Radeon Pro W5700 wins the Passmark DirectX 9 test with a score of 225, narrowly beating the NVIDIA GeForce RTX 2080's 223, a 0.9% difference.

Q: What is the performance gap in DirectX 12?

A: The NVIDIA GeForce RTX 2080 leads in Passmark DirectX 12 with a score of 72, while the AMD Radeon Pro W5700 scores 54. This gives NVIDIA a 25% advantage.

Q: How do the cards compare in GPU compute workloads?

A: The NVIDIA GeForce RTX 2080 scores 7,872 in Passmark GPU Compute, while the AMD Radeon Pro W5700 scores 6,495, giving NVIDIA a 17.5% lead.

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

A: The AMD Radeon Pro W5700 ranks in the 71st percentile, while the NVIDIA GeForce RTX 2080 ranks in the 68th percentile, despite the RTX 2080 winning most head-to-head tests.

Architecture Differences

The two cards are built on fundamentally different architectures. The AMD Radeon Pro W5700 uses the Navi 10 chip based on RDNA 1.0, while the NVIDIA GeForce RTX 2080 uses the TU104 chip based on Turing. This architectural divergence explains many of the observed performance differences.

The manufacturing process is a key differentiator. The W5700 is fabricated on a 7 nm process at TSMC, while the RTX 2080 uses a 12 nm process, also at TSMC. The smaller node allows AMD to pack 10,300 million transistors into a 251 mm² die, resulting in a transistor density of 41.0M per mm². In contrast, the RTX 2080 houses 13,600 million transistors on a much larger 545 mm² die, yielding a lower density of 25.0M per mm². This means the AMD chip is more compact and efficient in terms of area, but the NVIDIA chip has more raw transistors to work with.

The feature sets diverge significantly. The RTX 2080 includes 46 RT cores and 368 tensor cores, hardware dedicated to ray tracing and AI acceleration, respectively. The W5700 has no such dedicated units. This makes the NVIDIA card architecturally equipped for workloads that leverage these accelerators, while the AMD card relies entirely on its general-purpose shading units.

The API support also differs. The RTX 2080 supports DirectX 12 Ultimate (12_2), while the W5700 supports DirectX 12 (12_1). Both cards support OpenGL 4.6 and Vulkan 1.4, so the gap is specific to the newest DirectX feature level. The W5700's PCIe 4.0 x16 interface offers twice the bandwidth of the RTX 2080's PCIe 3.0 x16, though the benchmark data does not show this translating into a performance advantage in the recorded tests.

Specification Differences

The two cards share several specifications, including 8 GB of GDDR6 memory on a 256-bit bus with 448.0 GB/s of bandwidth, a dual-slot design, 1x 6-pin plus 1x 8-pin power connectors, a 550 W suggested PSU, and identical 267 mm lengths. The differences are where the cards diverge in capability.

The AMD Radeon Pro W5700 has 2,304 shading units, 144 TMUs, and 64 ROPs. The NVIDIA GeForce RTX 2080 has 2,944 shading units, 184 TMUs, and 64 ROPs. The RTX 2080 also adds 46 RT cores and 368 tensor cores, which the W5700 lacks entirely.

Clock speeds differ notably. The W5700 has a base clock of 1400 MHz and a boost clock of 1880 MHz. The RTX 2080 has a higher base clock of 1515 MHz but a lower boost clock of 1710 MHz. Despite the lower boost clock, the RTX 2080 achieves higher theoretical throughput: 10.07 TFLOPS FP32 versus 8.663 TFLOPS for the W5700. FP16 performance follows the same pattern, with the RTX 2080 at 20.14 TFLOPS (2:1) and the W5700 at 17.33 TFLOPS (2:1).

Pixel and texture rates also favor NVIDIA. The RTX 2080 delivers 109.4 GPixel/s and 314.6 GTexel/s, while the W5700 manages 120.3 GPixel/s and 270.7 GTexel/s. The AMD card actually wins on pixel rate, but NVIDIA's higher texture rate reflects its greater TMU count.

Power and physical dimensions differ slightly. The RTX 2080 has a TDP of 215 W, while the W5700 is rated at 205 W. The W5700 is shorter in height at 111 mm versus 116 mm, and the RTX 2080 has a listed width of 35 mm. The display outputs differ: the W5700 offers 5x mini-DisplayPort 1.4a plus 1x USB Type-C, while the RTX 2080 has 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C. The RTX 2080 launched on 2018-09-19 with a launch MSRP of 699 USD, while the W5700 arrived on 2019-11-18 with a launch MSRP of 799 USD.

Where Each One Wins

The NVIDIA GeForce RTX 2080 is the clear winner for most graphics-intensive workloads. Its 34.4% lead in Geekbench Vulkan and 35.3% lead in DirectX 10 suggest strong performance across both modern and legacy APIs. The 25% DirectX 12 advantage and 22.4% Passmark G3D lead indicate that for gaming, real-time rendering, and general 3D acceleration, the RTX 2080 is the superior choice. The 17.5% GPU compute lead also means that for compute-heavy tasks, the NVIDIA card holds the edge, despite its lower average benchmark score.

The RTX 2080's dedicated RT cores and tensor cores, absent from the W5700, make it the only card in this comparison capable of hardware-accelerated ray tracing and AI-based features. For any workload that leverages these units, the NVIDIA card wins by default. Its higher shading unit count (2,944 versus 2,304) and higher FP32 throughput (10.07 TFLOPS versus 8.663 TFLOPS) translate directly into the benchmark advantages observed.

The AMD Radeon Pro W5700 wins in a very narrow set of circumstances. Its sole benchmark victory is a 0.9% margin in DirectX 9, which is within the margin of error and unlikely to be perceptible in real-world use. Its higher pixel rate of 120.3 GPixel/s versus 109.4 GPixel/s suggests a theoretical advantage in fill-rate-bound scenarios, but the recorded tests do not confirm this advantage in practice. The W5700 also has a higher average benchmark score (25,726 versus 22,895) and a higher percentile ranking (71st versus 68th), but these figures are driven by the inclusion of benchmarks not shared in the head-to-head comparison.

For professional or workstation use, the W5700's feature set as a Radeon Pro product might imply advantages in certified applications, but the benchmark data does not show this. The RTX 2080 wins in every modern API test, every compute test, and every DirectX test except for the near-tie in DirectX 9. The only scenario where the W5700 appears preferable is if a user specifically requires its 5x mini-DisplayPort output configuration or its PCIe 4.0 interface, both of which are specification advantages that do not translate into measured performance wins.