AMD Radeon RX 6800S vs NVIDIA GeForce RTX 2080 Comparison

Head-to-Head Benchmarks

The benchmark data presents a remarkably one-sided comparison. Across the ten head-to-head tests, the NVIDIA GeForce RTX 2080 claims victory in every single instance, with the AMD Radeon RX 6800S failing to secure a single win. The most decisive margin appears in the Passmark DirectX 10 test, where the RTX 2080 leads by 33.1 percent, scoring 136 against the RX 6800S’s 91. This is not a narrow gap; it represents a substantial performance chasm in legacy API workloads.

The 3DMark Steel Nomad DX12 test, often a strong indicator of modern gaming performance, shows the RTX 2080 ahead by a narrower 3.6 percent, with scores of 1752 versus 1689. This close result suggests that in contemporary DirectX 12 titles, the two GPUs are more evenly matched than the overall win count implies. However, the margin still favors NVIDIA, and the pattern of RTX 2080 superiority holds across every other workload category.

Compute performance reveals a similar story. In Geekbench OpenCL, the RTX 2080 scores 91313, which is 19.8 percent higher than the RX 6800S’s 73202. The Vulkan compute gap is even wider, with NVIDIA leading by 24.6 percent (107797 versus 81272). These are not trivial differences; they indicate that the RTX 2080 maintains a commanding advantage in general-purpose GPU compute tasks, likely benefiting from its larger shading unit count and higher FP32 throughput.

The Passmark suite further reinforces NVIDIA’s dominance. In DirectX 11, the RTX 2080 leads by 12 percent (158 versus 139). DirectX 12 shows a 9.7 percent gap (72 versus 65), while DirectX 9 sees a 10.3 percent difference (223 versus 200). The 2D graphics test, Passmark G2D, shows a 19.3 percent NVIDIA advantage (907 versus 732), which is notable for a test that often favors memory bandwidth and rasterization efficiency.

The G3D test, a broader measure of 3D rendering performance, has the RTX 2080 ahead by 15 percent, scoring 18720 against 15908. GPU compute in Passmark shows a 22 percent gap (7872 versus 6144), consistent with the Geekbench OpenCL results. The overall average benchmark score tells a similar tale: the RTX 2080 averages 22895, while the RX 6800S averages 24063. Interestingly, the RX 6800S actually has a higher average score, but this is skewed by its strong Geekbench Metal result of 85256, a test that the RTX 2080 does not appear in the head-to-head list for. Excluding that outlier, the RTX 2080’s consistency across the common tests gives it the edge in direct competition.

Where Each One Wins

The data shows no test where the AMD Radeon RX 6800S emerges victorious, so the use-case split is heavily tilted toward NVIDIA. The RTX 2080 wins across every benchmark category, but the magnitude of its advantage varies significantly by workload type, which provides some nuance for specific use cases.

For legacy DirectX 9 and DirectX 10 applications, the RTX 2080 is overwhelmingly superior. The 33.1 percent lead in DirectX 10 and the 10.3 percent lead in DirectX 9 suggest that older games and applications that rely on these APIs will run noticeably better on the NVIDIA card. This is likely due to driver optimization maturity and the Turing architecture’s handling of legacy rendering paths.

In modern DirectX 12 workloads, the competition tightens considerably. The 3DMark Steel Nomad DX12 test shows only a 3.6 percent gap, and the Passmark DirectX 12 test shows a 9.7 percent gap. This indicates that in current-generation games utilizing DirectX 12, the RX 6800S is nearly competitive, and the performance difference may be imperceptible in real-world gameplay. The RX 6800S’s RDNA 2.0 architecture, with its 32 ray tracing cores and DirectX 12 Ultimate support, appears well-suited to modern rendering techniques.

Compute-heavy tasks, such as those measured by Geekbench OpenCL, Vulkan, and Passmark GPU Compute, clearly favor the RTX 2080. With leads ranging from 19.8 to 24.6 percent, the NVIDIA card is the better choice for GPU-accelerated workloads like rendering, scientific computation, or machine learning inference. The RTX 2080’s 368 tensor cores, which the RX 6800S lacks entirely, likely contribute to this advantage in certain compute tasks.

The Passmark G2D test, which measures 2D graphics performance, shows a 19.3 percent NVIDIA lead. While this test is less relevant for gaming, it can matter for desktop productivity, video playback, and multi-monitor setups. The RTX 2080’s higher memory bandwidth (448.0 GB/s versus 256.0 GB/s) may play a role here, as 2D operations often rely on memory throughput.

The Verdict

Based strictly on the benchmark data, the NVIDIA GeForce RTX 2080 is the superior performer in this head-to-head comparison. It wins all ten tests, with margins ranging from a narrow 3.6 percent in 3DMark Steel Nomad DX12 to a decisive 33.1 percent in Passmark DirectX 10. Users seeking maximum performance across a broad range of workloads—gaming, compute, and legacy applications—should choose the RTX 2080 without hesitation.

However, the data does not tell a completely one-sided story. The RX 6800S posts a higher average benchmark score (24063 versus 22895) when including the Geekbench Metal test, where it achieves 85256. This suggests that in Apple-centric ecosystems or Metal-accelerated applications, the AMD card may actually outperform the NVIDIA card. Additionally, the RX 6800S’s slim 3.6 percent deficit in modern DirectX 12 gaming means that for players focused solely on current titles, the two cards are functionally equivalent in many scenarios.

The power consumption figures are also worth noting, though they are not directly benchmarked. The RX 6800S has a TDP of 100 W, while the RTX 2080 draws 215 W. The AMD card achieves 80-90 percent of the RTX 2080’s performance in several tests while consuming less than half the power, which could be a decisive factor for laptop users or those with strict thermal budgets. The RX 6800S is an integrated graphics package (IGP) with no power connectors, designed for portable devices, whereas the RTX 2080 is a dual-slot desktop card requiring a 550 W power supply.

For desktop users with power to spare and a need for maximum performance, the RTX 2080 is the clear choice. For mobile users or those prioritizing efficiency, the RX 6800S offers competitive modern gaming performance with dramatically lower power draw. The RTX 2080’s launch MSRP was 699 USD, which provides a reference point for its original market positioning.

FAQ

Q: Which GPU wins more benchmark tests?

A: The NVIDIA GeForce RTX 2080 wins all 10 head-to-head tests listed in the data. The AMD Radeon RX 6800S does not win a single test.

Q: What is the biggest performance gap between the two cards?

A: The largest margin is in the Passmark DirectX 10 test, where the RTX 2080 leads by 33.1 percent, scoring 136 versus the RX 6800S’s 91.

Q: How close are the two cards in modern DirectX 12 gaming?

A: In the 3DMark Steel Nomad DX12 test, the RTX 2080 leads by only 3.6 percent (1752 versus 1689), indicating near-parity in modern DirectX 12 workloads.

Q: Does the AMD card have any advantage in average benchmark scores?

A: Yes. The RX 6800S has a higher average benchmark score of 24063 compared to the RTX 2080’s 22895, largely due to its strong Geekbench Metal score of 85256.

Q: How do the power requirements compare?

A: The RX 6800S has a TDP of 100 W and requires no power connectors, while the RTX 2080 has a TDP of 215 W and needs a 550 W power supply with one 6-pin and one 8-pin connector.

Q: What is the memory bandwidth difference?

A: The RTX 2080 has 448.0 GB/s of memory bandwidth, while the RX 6800S has 256.0 GB/s, a difference of 192.0 GB/s.

Architecture Differences

The two GPUs represent fundamentally different design philosophies and manufacturing generations. The AMD Radeon RX 6800S is built on the RDNA 2.0 architecture using a 7 nm process at TSMC, with a die size of 237 mm² and 11,060 million transistors. This results in a transistor density of 46.7 million per square millimeter. In contrast, the NVIDIA GeForce RTX 2080 uses the Turing architecture on a 12 nm process, also at TSMC, but with a much larger die of 545 mm² and 13,600 million transistors, yielding a lower density of 25.0 million per square millimeter. The RX 6800S’s smaller, denser process node explains its significantly lower power consumption.

The compute configurations differ substantially. The RTX 2080 packs 2944 shading units, 184 texture mapping units, and 64 ROPs. The RX 6800S has fewer shading units at 2048, fewer TMUs at 128, but the same 64 ROPs. This gives the RTX 2080 a higher texture rate of 314.6 GTexel/s versus 268.8 GTexel/s, though the RX 6800S has a higher pixel rate of 134.4 GPixel/s compared to 109.4 GPixel/s. In raw FP32 throughput, the RTX 2080 achieves 10.07 TFLOPS against the RX 6800S’s 8.602 TFLOPS.

Ray tracing and tensor capabilities are another major differentiator. The RTX 2080 includes 46 ray tracing cores and 368 tensor cores, while the RX 6800S has 32 ray tracing cores and no tensor cores. Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The memory subsystems differ in bus width: the RX 6800S uses a 128-bit bus with 8 GB of GDDR6 at 16 Gbps effective, while the RTX 2080 uses a 256-bit bus with 8 GB of GDDR6 at 14 Gbps effective. This results in the RTX 2080’s higher bandwidth of 448.0 GB/s versus 256.0 GB/s.

The interface and form factor also diverge. The RX 6800S uses PCIe 4.0 x8 and is an integrated graphics package (IGP) with no power connectors, designed for portable devices with display outputs dependent on the host device. The RTX 2080 uses PCIe 3.0 x16, is a dual-slot card measuring 267 mm in length, and includes 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C outputs. The RTX 2080 requires one 6-pin and one 8-pin power connector with a suggested 550 W PSU. Clock speeds show the RX 6800S running higher at 1800 MHz base and 2100 MHz boost, versus the RTX 2080’s 1515 MHz base and 1710 MHz boost, though the NVIDIA card’s larger memory bus compensates in bandwidth-sensitive tasks.