AMD Radeon RX 6800S vs NVIDIA GeForce RTX 2080 SUPER Comparison

The NVIDIA GeForce RTX 2080 SUPER and AMD Radeon RX 6800S occupy the same 69th percentile among all GPUs, and their average benchmark scores differ by only 0.4%, with the RTX 2080 SUPER at 24170 and the RX 6800S at 24063. Despite this near-identical overall standing, the head-to-head benchmark data reveals a consistent pattern of victories for the NVIDIA card, which wins all 10 recorded tests. The following analysis breaks down those results, the architectural reasons behind them, and what each card is best suited for.

Head-to-Head Benchmarks

The data shows a clean sweep for the NVIDIA GeForce RTX 2080 SUPER, which wins every single head-to-head benchmark against the AMD Radeon RX 6800S. The margins vary considerably, ranging from a modest 11.4% advantage to a dominant 53.8% lead. The largest single victory comes in the PassMark DirectX 10 test, where the RTX 2080 SUPER scores 140 against the RX 6800S's 91, a 53.8% difference. This is followed closely by the Geekbench Vulkan result, where the NVIDIA card's 111284 score eclipses the AMD card's 81272, a 36.9% margin.

Compute-oriented workloads also heavily favor the NVIDIA card. In Geekbench OpenCL, the RTX 2080 SUPER delivers 99226 points compared to 73202 for the RX 6800S, a 35.6% lead. The PassMark GPU Compute test shows a similar story, with the NVIDIA card scoring 8290 versus 6144 for the AMD card, a 34.9% advantage. These results indicate that the RTX 2080 SUPER has a substantial edge in general-purpose and parallel compute tasks, not just in graphics rendering.

In more modern DirectX 12 workloads, the gap narrows but remains firmly in NVIDIA's favor. The 3DMark Steel Nomad DX12 test shows the RTX 2080 SUPER at 1882 points, 11.4% ahead of the RX 6800S's 1689. The PassMark DirectX 12 test shows a 15.4% lead for NVIDIA (75 vs 65), while the DirectX 11 test shows an 18.7% advantage (165 vs 139). The legacy DirectX 9 test also goes to NVIDIA, with a 13.5% lead (227 vs 200).

The 2D performance delta is also notable. The PassMark G2D test shows the RTX 2080 SUPER scoring 920, which is 25.7% higher than the RX 6800S's 732. This suggests that even outside of 3D rendering, the NVIDIA card holds a clear performance advantage. The overall PassMark G3D score reinforces the trend, with the RTX 2080 SUPER at 19490 and the RX 6800S at 15908, a 22.5% difference.

While the RX 6800S does not secure a single win in these head-to-head tests, its average benchmark score of 24063 is remarkably close to the RTX 2080 SUPER's 24170. This apparent contradiction is explained by the fact that the head-to-head suite includes different tests than the broader average, and the RX 6800S has no equivalent Geekbench Metal result in this comparison. The data confirms that in the specific tests where both cards are measured, the RTX 2080 SUPER is consistently faster, with an average lead of roughly 20-30% across most workloads.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 2080 SUPER has an average benchmark score of 24170, which is 0.4% higher than the AMD Radeon RX 6800S's average score of 24063.

Q: How much faster is the RTX 2080 SUPER in the Geekbench Vulkan test?

A: The RTX 2080 SUPER scores 111284 in Geekbench Vulkan, which is 36.9% higher than the RX 6800S's score of 81272.

Q: Did the RX 6800S win any head-to-head benchmark?

A: No. The RX 6800S lost all 10 head-to-head benchmarks to the RTX 2080 SUPER, with zero wins recorded.

Q: What is the difference in the PassMark G3D scores?

A: The RTX 2080 SUPER scores 19490 in PassMark G3D, while the RX 6800S scores 15908, giving the NVIDIA card a 22.5% advantage.

Q: How do the two cards compare in the 3DMark Steel Nomad DX12 test?

A: The RTX 2080 SUPER scores 1882 points, which is 11.4% higher than the RX 6800S's score of 1689.

Q: Do both cards support the same DirectX version?

A: Yes, both the NVIDIA GeForce RTX 2080 SUPER and the AMD Radeon RX 6800S support DirectX 12 Ultimate (12_2), as well as OpenGL 4.6 and Vulkan 1.4.

Architecture Differences

The two GPUs are built on fundamentally different architectures and manufacturing processes. The NVIDIA GeForce RTX 2080 SUPER uses the TU104 chip based on the Turing architecture, fabricated on TSMC's 12 nm process. This is a large chip, with a die size of 545 mm² and 13,600 million transistors, resulting in a transistor density of 25.0M per mm². In contrast, the AMD Radeon RX 6800S uses the Navi 23 chip based on the RDNA 2.0 architecture, built on TSMC's more advanced 7 nm process. The Navi 23 die is significantly smaller at 237 mm² and contains 11,060 million transistors, giving it a higher transistor density of 46.7M per mm².

The compute resources differ substantially in configuration. The RTX 2080 SUPER has 3072 shading units, 192 texture mapping units (TMUs), and 64 render output units (ROPs). It also includes 48 dedicated ray tracing cores and 384 tensor cores, which are specific to NVIDIA's Turing architecture. The RX 6800S has fewer shading units at 2048, 128 TMUs, and the same 64 ROPs. It includes 32 ray accelerators (listed as rtCores) but has no tensor cores, as those are not part of AMD's RDNA 2.0 design.

The memory subsystems also reflect different design philosophies. The RTX 2080 SUPER uses a 256-bit memory bus with 8 GB of GDDR6 memory, achieving a bandwidth of 495.9 GB/s. The RX 6800S also has 8 GB of GDDR6 memory, but on a narrower 128-bit bus, resulting in a bandwidth of 256.0 GB/s — roughly half of the NVIDIA card's bandwidth. This is a significant architectural difference that likely contributes to the performance gap in memory-intensive workloads.

Clock speeds and power characteristics show the intended use cases. The RTX 2080 SUPER has a base clock of 1650 MHz and a boost clock of 1815 MHz, with a TDP of 250 W. The RX 6800S has a higher base clock of 1800 MHz and a boost clock of 2100 MHz, but a much lower TDP of 100 W. The RX 6800S is classified as an IGP (integrated graphics processor) with no power connectors, indicating it is designed for mobile platforms, while the RTX 2080 SUPER is a dual-slot discrete card requiring a 1x 6-pin and 1x 8-pin power connector and a 600 W suggested PSU.

Specification Differences

The two cards differ in nearly every core specification field. The process node is a clear differentiator: the RTX 2080 SUPER uses 12 nm, while the RX 6800S uses 7 nm. The chip design also differs, with TU104 versus Navi 23. The RTX 2080 SUPER has a larger die (545 mm² vs 237 mm²) and more transistors (13,600 million vs 11,060 million), but the RX 6800S has a higher transistor density (46.7M / mm² vs 25.0M / mm²).

Clock speeds favor the RX 6800S in raw terms: it has a base clock of 1800 MHz and a boost of 2100 MHz, compared to the RTX 2080 SUPER's 1650 MHz base and 1815 MHz boost. The RX 6800S also has a defined game clock of 1975 MHz, which the RTX 2080 SUPER lacks. However, the RTX 2080 SUPER counters with a much wider memory bus (256 bit vs 128 bit) and higher memory bandwidth (495.9 GB/s vs 256.0 GB/s). The memory clock is also higher on the NVIDIA card at 1937 MHz (15.5 Gbps effective) versus 2000 MHz (16 Gbps effective) on the AMD card.

The compute unit counts are higher on the RTX 2080 SUPER: 3072 shading units vs 2048, 192 TMUs vs 128, and 48 RT cores vs 32. Both have 64 ROPs. The RTX 2080 SUPER also has 384 tensor cores, which the RX 6800S does not have at all. This translates to higher theoretical performance for the NVIDIA card, with an FP32 rating of 11.15 TFLOPS versus 8.602 TFLOPS, and an FP16 rating of 22.30 TFLOPS versus 17.20 TFLOPS.

The physical and power specifications are starkly different. The RTX 2080 SUPER is a dual-slot card measuring 267 mm in length, 116 mm in height, and 35 mm in width, with a TDP of 250 W. The RX 6800S has no listed dimensions, is an IGP, and has a TDP of 100 W. The bus interface differs as well: the RTX 2080 SUPER uses PCIe 3.0 x16, while the RX 6800S uses PCIe 4.0 x8. The RTX 2080 SUPER has a launch MSRP of 699 USD, while the RX 6800S has no listed launch MSRP.

The Verdict

Based strictly on benchmark data, the NVIDIA GeForce RTX 2080 SUPER is the faster card. It wins all 10 head-to-head benchmarks, with margins ranging from 11.4% to 53.8%. The most significant wins are in compute (Geekbench OpenCL at 35.6%, PassMark GPU Compute at 34.9%) and API-specific tests (Geekbench Vulkan at 36.9%, PassMark DirectX 10 at 53.8%). The RTX 2080 SUPER also holds a 22.5% lead in the overall PassMark G3D score and a 25.7% lead in the G2D test. Its average benchmark score of 24170 is higher than the RX 6800S's 24063, and it achieves this with a higher FP32 throughput (11.15 TFLOPS vs 8.602 TFLOPS) and more than double the memory bandwidth (495.9 GB/s vs 256.0 GB/s).

The AMD Radeon RX 6800S, however, is not without merit. Its 7 nm process node and lower TDP of 100 W make it a far more power-efficient design, and its higher clock speeds (2100 MHz boost vs 1815 MHz boost) show architectural efficiency. The RX 6800S also supports PCIe 4.0, while the RTX 2080 SUPER is limited to PCIe 3.0. But in terms of raw performance, the data does not support the RX 6800S as a competitor to the RTX 2080 SUPER in the tested workloads. The choice is clear for users prioritizing maximum frame rates and compute performance: the RTX 2080 SUPER is the superior option.

Where Each One Wins

The NVIDIA GeForce RTX 2080 SUPER wins in every benchmark category where both cards were tested. This includes DirectX 10, 11, and 12 tests, where it leads by 53.8%, 18.7%, and 15.4% respectively. It also wins in legacy DirectX 9 by 13.5%. For users running older API titles, the RTX 2080 SUPER is particularly dominant. In modern 3DMark Steel Nomad DX12, it holds an 11.4% lead. The NVIDIA card is also the clear choice for compute-heavy applications, as evidenced by its 35.6% lead in Geekbench OpenCL, 36.9% lead in Geekbench Vulkan, and 34.9% lead in PassMark GPU Compute.

The RTX 2080 SUPER also wins in 2D performance, with a 25.7% lead in PassMark G2D. This suggests it may offer snappier desktop responsiveness and faster 2D graphics acceleration, though the practical impact is likely minor compared to the 3D gains. The card's tensor cores, which the RX 6800S lacks entirely, provide additional compute capability for AI-accelerated workloads, though no specific benchmark in this data directly tests that feature.

The AMD Radeon RX 6800S does not win any benchmark, but its strengths lie in its power profile and physical form factor. With a TDP of 100 W versus 250 W for the RTX 2080 SUPER, the RX 6800S is designed for mobile or low-power systems. Its IGP classification and lack of power connectors indicate it is meant for portable devices where power draw is a primary concern. The RX 6800S's higher transistor density (46.7M / mm²) and smaller die size (237 mm²) also show that it is a more modern, efficient design. For a user building a compact, power-conscious system, the RX 6800S offers competitive average benchmark scores (24063 vs 24170) while consuming far less power, even if it cannot match the RTX 2080 SUPER's peak performance in any tested workload.