AMD Radeon RX 6800S vs NVIDIA GeForce RTX 3090 Comparison

Head-to-Head Benchmarks

The recorded data shows a decisive overall victory for the NVIDIA GeForce RTX 3090, which wins 9 of the 10 shared benchmark tests. The lone exception is Geekbench Vulkan, where the AMD Radeon RX 6800S takes a significant lead. In that test, the AMD card scores 81,272 against the NVIDIA card's 53,927, a 33.6% advantage. This is a notable outlier, as the AMD card's Vulkan performance is substantially higher despite its lower raw compute specifications.

Outside of Vulkan, the RTX 3090 dominates across every other workload. The largest margin comes in 3DMark Steel Nomad DX12, where the RTX 3090 scores 5,118 versus 1,689 for the RX 6800S, a 203% advantage. That is more than triple the AMD card's score and represents the single biggest gap in the comparison. In Geekbench OpenCL, the RTX 3090 posts 172,758 points against 73,202, a 136% lead, reinforcing its strength in general-purpose compute workloads.

The Passmark suite tells a similar story. In Passmark G3D, the RTX 3090 scores 26,645 versus 15,908, a 67.5% lead. Passmark GPU Compute shows an even larger disparity: 15,356 versus 6,144, a 149.9% difference. The RTX 3090 also wins all DirectX versions tested: DirectX 10 (182 vs 91, 100% ahead), DirectX 11 (220 vs 139, 58.3% ahead), DirectX 12 (110 vs 65, 69.2% ahead), and DirectX 9 (268 vs 200, 34% ahead). Even in Passmark G2D, a 2D graphics test, the RTX 3090 leads 1,063 to 732, a 45.2% margin.

The average benchmark score for the RTX 3090 across all recorded tests is 27,565, while the RX 6800S averages 24,063. This places the RTX 3090 in the 73rd percentile of all GPUs in the database, compared to the 69th percentile for the RX 6800S. The RTX 3090's nearest rivals include the NVIDIA GeForce RTX 4070 Mobile (27,435, 0.5% behind) and the AMD Radeon RX 6700 XT (27,425, 0.5% behind), while the RX 6800S sits close to the NVIDIA GeForce RTX 2080 SUPER (24,170, 0.4% ahead) and the AMD Radeon RX 9070 (23,877, 0.8% behind). These proximity scores indicate that the RTX 3090 competes in a higher performance tier, while the RX 6800S is more aligned with mid-range desktop cards from an earlier generation.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. The RTX 3090 uses the GA102 chip built on NVIDIA's Ampere architecture, manufactured on an 8 nm process at Samsung. The RX 6800S uses the Navi 23 chip based on AMD's RDNA 2.0 architecture, built on a 7 nm process at TSMC. The process node difference is relatively small, but the chip scales are not. The GA102 die measures 628 mm² and contains 28,300 million transistors, while the Navi 23 die is 237 mm² with 11,060 million transistors. Transistor density is similar (45.1M per mm² for NVIDIA, 46.7M per mm² for AMD), but the RTX 3090's much larger die gives it far more resources.

The RTX 3090 ships with 10,496 shading units, 328 TMUs, and 112 ROPs, alongside 82 RT cores and 328 tensor cores. The RX 6800S has 2,048 shading units, 128 TMUs, and 64 ROPs, with 32 RT cores and no tensor cores. This is a roughly 5:1 ratio in shading units and a 2.6:1 ratio in RT cores. The FP32 compute figures reflect this: the RTX 3090 delivers 35.58 TFLOPS, while the RX 6800S delivers 8.602 TFLOPS. In FP16, the RTX 3090 again hits 35.58 TFLOPS (1:1 ratio), while the RX 6800S reaches 17.20 TFLOPS (2:1 ratio), a smaller gap but still a clear NVIDIA lead.

Memory architecture is another major differentiator. The RTX 3090 has 24 GB of GDDR6X on a 384-bit bus, yielding 936.2 GB/s of bandwidth. The RX 6800S has 8 GB of GDDR6 on a 128-bit bus, yielding 256.0 GB/s. That is a 3.66x bandwidth advantage for the RTX 3090, which directly explains its dominance in compute-heavy and high-resolution workloads. The RX 6800S compensates with higher clocks: its base clock is 1800 MHz and boost is 2100 MHz, while the RTX 3090 runs at 1395 MHz base and 1695 MHz boost. The AMD card's game clock is listed at 1975 MHz. However, the massive difference in core count and memory bandwidth overwhelms the clock speed advantage.

Power and physical design also diverge sharply. The RTX 3090 has a 350 W TDP, requires a triple-slot cooler, uses a single 12-pin power connector, and needs a 750 W suggested PSU. It measures 336 mm in length, 140 mm in height, and 61 mm in width. The RX 6800S is rated at 100 W TDP, is listed as IGP (integrated graphics processor) slot width, requires no power connectors, and has no listed PSU recommendation or dimensions, being portable-device dependent. The RX 6800S also uses a PCIe 4.0 x8 interface, while the RTX 3090 uses PCIe 4.0 x16. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs differ: the RTX 3090 has 1x HDMI 2.1 and 3x DisplayPort 1.4a, while the RX 6800S is listed as portable device dependent.

The Verdict

The data is unambiguous: the NVIDIA GeForce RTX 3090 is the far more powerful GPU in almost every measurable way. It wins 9 of 10 head-to-head tests, with margins ranging from 34% to 203%. Its average benchmark score of 27,565 is 14.5% higher than the RX 6800S's 24,063. For anyone prioritizing raw performance, particularly in DirectX workloads, compute tasks, or high-bandwidth memory scenarios, the RTX 3090 is the clear choice from the recorded data. Its 24 GB VRAM and 936.2 GB/s bandwidth are in a different class entirely from the RX 6800S's 8 GB and 256.0 GB/s.

The RX 6800S does have one meaningful advantage: Geekbench Vulkan, where it leads by 33.6%. This suggests that in Vulkan-optimized applications, the AMD card can outperform its larger rival despite far lower compute specifications. Additionally, the RX 6800S is far more power-efficient in the recorded data, with a 100 W TDP versus 350 W, and it requires no external power connectors, making it suitable for thin portable devices. The RTX 3090, by contrast, needs a triple-slot cooler and a 750 W PSU.

Who should pick which? Strictly from the data, the RTX 3090 is for users who need maximum performance in DX12, OpenCL, or compute-heavy workloads, and who have the physical space and power delivery to accommodate a 336 mm triple-slot card. The RX 6800S is for users who prioritize Vulkan performance and need a low-power, compact solution that fits in a portable device without dedicated power connectors. The RX 6800S sits at the 69th percentile of all GPUs, close to the RTX 2080 SUPER, which indicates it is a capable mid-range option, but it is not in the same performance tier as the RTX 3090.

Specification Differences

| Specification | NVIDIA GeForce RTX 3090 | AMD Radeon RX 6800S |

|----------------|------------------------|---------------------|

| Chip | GA102 | Navi 23 |

| Architecture | Ampere | RDNA 2.0 |

| Process Node | 8 nm | 7 nm |

| Foundry | Samsung | TSMC |

| Transistors | 28,300 million | 11,060 million |

| Die Size | 628 mm² | 237 mm² |

| Base Clock | 1395 MHz | 1800 MHz |

| Boost Clock | 1695 MHz | 2100 MHz |

| Game Clock | null | 1975 MHz |

| Memory Clock | 1219 MHz (19.5 Gbps effective) | 2000 MHz (16 Gbps effective) |

| Memory Size | 24 GB | 8 GB |

| Memory Type | GDDR6X | GDDR6 |

| Memory Bus | 384 bit | 128 bit |

| Memory Bandwidth | 936.2 GB/s | 256.0 GB/s |

| Shading Units | 10496 | 2048 |

| TMUs | 328 | 128 |

| ROPs | 112 | 64 |

| RT Cores | 82 | 32 |

| Tensor Cores | 328 | null |

| Pixel Rate | 189.8 GPixel/s | 134.4 GPixel/s |

| Texture Rate | 556.0 GTexel/s | 268.8 GTexel/s |

| FP32 | 35.58 TFLOPS | 8.602 TFLOPS |

| FP16 | 35.58 TFLOPS (1:1) | 17.20 TFLOPS (2:1) |

| TDP | 350 W | 100 W |

| Slot Width | Triple-slot | IGP |

| Power Connectors | 1x 12-pin | None |

| Suggested PSU | 750 W | null |

| Bus Interface | PCIe 4.0 x16 | PCIe 4.0 x8 |

| Display Outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a | Portable Device Dependent |

| Dimensions | 336 mm x 140 mm x 61 mm | null |

| Release Date | 2020-08-31 | 2022-01-03 |

| Launch MSRP | 1,499 USD | null |

FAQ

Q: Which GPU wins more head-to-head benchmarks?

A: The NVIDIA GeForce RTX 3090 wins 9 out of 10 shared tests, with the AMD Radeon RX 6800S winning only Geekbench Vulkan.

Q: How large is the performance gap in 3DMark Steel Nomad DX12?

A: The RTX 3090 scores 5,118 versus 1,689 for the RX 6800S, a 203% advantage, the largest margin in any shared test.

Q: What is the memory bandwidth difference?

A: The RTX 3090 has 936.2 GB/s of bandwidth on a 384-bit bus with 24 GB GDDR6X, while the RX 6800S has 256.0 GB/s on a 128-bit bus with 8 GB GDDR6.

Q: Does the AMD card have any advantage in compute workloads?

A: No, in Geekbench OpenCL the RTX 3090 scores 172,758 versus 73,202, a 136% lead, and in Passmark GPU Compute it scores 15,356 versus 6,144, a 149.9% lead.

Q: What is the power consumption difference?

A: The RTX 3090 has a 350 W TDP and requires a 750 W suggested PSU, while the RX 6800S has a 100 W TDP and requires no power connectors.

Q: How do their average benchmark scores compare to rival GPUs?

A: The RTX 3090 averages 27,565, placing it near the RTX 4070 Mobile (27,435, 0.5% behind) and RX 6700 XT (27,425, 0.5% behind). The RX 6800S averages 24,063, close to the RTX 2080 SUPER (24,170, 0.4% ahead) and RX 9070 (23,877, 0.8% behind).