AMD Radeon RX 6800S vs NVIDIA GeForce RTX 3070 Ti Mobile Comparison

The AMD Radeon RX 6800S and NVIDIA GeForce RTX 3070 Ti Mobile are both end-of-life mobile GPUs from the same release window, yet they present very different profiles. Benchmark data shows the NVIDIA part wins 9 out of 10 head-to-head tests, but the AMD card holds a narrow lead in one legacy test and offers a dramatically lower power draw. This analysis breaks down where each GPU makes sense based strictly on the measured scores and architectural facts.

Where Each One Wins

The RTX 3070 Ti Mobile is the clear performance leader across nearly every modern workload. In the 3DMark Steel Nomad DX12 test, it scores 2572 versus the RX 6800S’s 1689, a commanding 34.3% advantage. This pattern repeats in compute and general graphics: Geekbench OpenCL shows a 31% lead (106022 vs 73202), and Vulkan testing shows an 18.7% gap (100027 vs 81272). For any DirectX 10, 11, or 12 gaming scenario, the NVIDIA card wins by margins ranging from 9.2% to 26.6%. The PassMark G3D score also favors NVIDIA, 17727 vs 15908, a 10.3% difference. Even in 2D performance (PassMark G2D), the RTX 3070 Ti Mobile edges ahead by 6.5% (783 vs 732).

The AMD Radeon RX 6800S wins exactly one head-to-head test: PassMark DirectX 9, where it scores 200 against NVIDIA’s 196, a slim 2% margin. This is a legacy API test, but it indicates the RDNA 2 architecture retains an edge in older DirectX 9 titles. Beyond that single win, the RX 6800S’s value proposition lies in its efficiency profile. Its TDP is 100 W versus 115 W for the RTX 3070 Ti Mobile, making it the lower-power option. For thin-and-light gaming laptops where thermal headroom is scarce, that 15 W difference can be decisive for sustained performance and chassis design. The data does not show any other workload where AMD takes the lead, so the choice is between raw speed and lower power consumption.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. The RX 6800S uses the Navi 23 chip built on RDNA 2.0 architecture, manufactured on TSMC’s 7 nm process. It packs 11,060 million transistors into a 237 mm² die, yielding a transistor density of 46.7M per mm². The RTX 3070 Ti Mobile uses the GA104 chip with NVIDIA’s Ampere architecture, fabricated on Samsung’s 8 nm process. That chip is substantially larger at 392 mm² and holds 17,400 million transistors, though its density is slightly lower at 44.4M per mm². The larger die and higher transistor count give NVIDIA more raw hardware, but the AMD chip achieves higher density on a more advanced process node.

Core configurations differ sharply. The RX 6800S has 2048 shading units, 128 texture mapping units, and 64 ROPs. It also includes 32 ray tracing cores. The RTX 3070 Ti Mobile has 5888 shading units, 184 TMUs, and 96 ROPs, plus 46 RT cores and 184 tensor cores. NVIDIA’s Ampere architecture includes tensor cores for AI-accelerated workloads, which the AMD chip lacks entirely. Clock speeds tell the opposite story: the RX 6800S runs at a base of 1800 MHz and boosts to 2100 MHz, with a game clock of 1975 MHz. The RTX 3070 Ti Mobile has a much lower base of 915 MHz and a boost of 1410 MHz. Despite lower clocks, the NVIDIA part’s massive core count delivers far higher peak throughput: 16.60 TFLOPS FP32 versus 8.602 TFLOPS for AMD.

Memory subsystems also diverge. Both have 8 GB of GDDR6, but the RX 6800S uses a 128-bit bus yielding 256.0 GB/s bandwidth, while the RTX 3070 Ti Mobile uses a 256-bit bus for 448.0 GB/s. The NVIDIA card also has a wider PCIe interface: PCIe 4.0 x16 versus x8 on the AMD side. Both support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. The RX 6800S has a 2:1 FP16 ratio (17.20 TFLOPS), while the RTX 3070 Ti Mobile runs FP16 at 1:1 (16.60 TFLOPS), meaning NVIDIA’s FP16 and FP32 throughput are identical.

Head-to-Head Benchmarks

The biggest win for the RTX 3070 Ti Mobile comes in the 3DMark Steel Nomad DX12 test, where it beats the RX 6800S by 34.3% (2572 vs 1689). This is a modern DirectX 12 workload, and the gap is substantial enough to translate into noticeably higher frame rates in current titles. Geekbench OpenCL shows a 31% lead for NVIDIA (106022 vs 73202), which is relevant for compute-heavy applications that leverage OpenCL. The Vulkan test gap is smaller but still significant at 18.7% (100027 vs 81272), suggesting the NVIDIA architecture is more efficient in this API as well.

Legacy DirectX tests show a mixed bag. In PassMark DirectX 10, NVIDIA wins by 26.6% (124 vs 91). In DirectX 11, the margin narrows to 9.2% (153 vs 139). DirectX 12 shows an 11% NVIDIA lead (73 vs 65). The only AMD victory is DirectX 9, where it wins by 2% (200 vs 196) — a narrow but real advantage for older game libraries. PassMark G3D, an overall 3D graphics score, favors NVIDIA by 10.3% (17727 vs 15908). Compute performance in PassMark GPU Compute shows NVIDIA ahead by 18.1% (7504 vs 6144). The 2D test is close, with NVIDIA winning 783 vs 732, a 6.5% margin.

Average benchmark scores from the broader database tell a similar story. The RX 6800S averages 24063 across all tests, which is 0.4% behind the NVIDIA GeForce RTX 2080 SUPER (24170) and 0.7% behind the GTX 780 Ti (24236). It sits 0.8% ahead of the AMD Radeon RX 9070 (23877). The RTX 3070 Ti Mobile averages 23518, which is 0.5% behind the RTX 3080 Mobile (23628) and 0.9% ahead of the AMD Radeon AI PRO R9700 (23315). Both GPUs land at the 69th percentile against all GPUs, indicating they occupy a similar tier in the overall performance distribution, even though the head-to-head results favor NVIDIA consistently.

FAQ

Q: Which GPU is faster in DirectX 12 gaming?

A: The RTX 3070 Ti Mobile wins decisively. In the 3DMark Steel Nomad DX12 test, it scores 2572 versus the RX 6800S’s 1689, a 34.3% advantage. PassMark DirectX 12 also favors NVIDIA, 73 vs 65, an 11% lead.

Q: Are there any tests where the AMD RX 6800S wins?

A: Yes, exactly one. The RX 6800S wins PassMark DirectX 9 with a score of 200 versus 196 for the RTX 3070 Ti Mobile, a 2% margin. It also has a lower TDP of 100 W compared to 115 W, which is an efficiency win.

Q: How do these GPUs compare in compute performance?

A: NVIDIA leads across the board. Geekbench OpenCL shows 106022 vs 73202 (31% lead), and PassMark GPU Compute shows 7504 vs 6144 (18.1% lead). The RTX 3070 Ti Mobile also has 184 tensor cores, which the RX 6800S lacks entirely.

Q: What is the memory bandwidth difference?

A: The RTX 3070 Ti Mobile has 448.0 GB/s bandwidth on a 256-bit bus, while the RX 6800S has 256.0 GB/s on a 128-bit bus. Both have 8 GB of GDDR6 memory.

Q: Which GPU has higher clock speeds?

A: The RX 6800S boosts to 2100 MHz with a base of 1800 MHz and a game clock of 1975 MHz. The RTX 3070 Ti Mobile boosts to only 1410 MHz with a base of 915 MHz. Despite this, NVIDIA wins on raw throughput due to its larger core count.

Q: How do these cards compare to their nearest rivals?

A: The RX 6800S averages 24063, within 0.7% of the GTX 780 Ti (24236) and 0.4% of the RTX 2080 SUPER (24170). The RTX 3070 Ti Mobile averages 23518, within 0.5% of the RTX 3080 Mobile (23628) and 0.9% ahead of the AMD Radeon AI PRO R9700 (23315).

The Verdict

The data points to a straightforward conclusion: pick the RTX 3070 Ti Mobile if raw performance is the priority. It wins 9 of 10 head-to-head tests, with margins from 6.5% to 34.3%. The largest gap is in modern DX12 workloads, which is the best indicator of future gaming performance. Its 16.60 TFLOPS FP32 throughput is nearly double the RX 6800S’s 8.602 TFLOPS, and the wider 256-bit memory bus provides 448.0 GB/s bandwidth versus 256.0 GB/s. The tensor cores are a bonus for any AI-adjacent workload.

Pick the RX 6800S only if power draw is the critical constraint. Its 100 W TDP is 15 W lower than the RTX 3070 Ti Mobile’s 115 W, which matters in thin chassis where sustained boost clocks are limited by thermals. The AMD card also wins the DirectX 9 test, which could matter for older game libraries. Its higher clock speeds (2100 MHz boost vs 1410 MHz) and denser 7 nm process make it a more efficient design, but the benchmark results show that efficiency does not translate into a performance win anywhere except the legacy DX9 test. Both GPUs sit at the 69th percentile overall, so they are peers in the broader GPU landscape, but the head-to-head data is unambiguous about which one is faster.

Specification Differences

| Specification | AMD Radeon RX 6800S | NVIDIA GeForce RTX 3070 Ti Mobile |

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

| Architecture | RDNA 2.0 | Ampere |

| Process Node | 7 nm (TSMC) | 8 nm (Samsung) |

| Transistors | 11,060 million | 17,400 million |

| Die Size | 237 mm² | 392 mm² |

| Transistor Density | 46.7M / mm² | 44.4M / mm² |

| Base Clock | 1800 MHz | 915 MHz |

| Boost Clock | 2100 MHz | 1410 MHz |

| Game Clock | 1975 MHz | — |

| Memory Clock | 2000 MHz / 16 Gbps effective | 1750 MHz / 14 Gbps effective |

| Memory Bus Width | 128 bit | 256 bit |

| Memory Bandwidth | 256.0 GB/s | 448.0 GB/s |

| Shading Units | 2048 | 5888 |

| TMUs | 128 | 184 |

| ROPs | 64 | 96 |

| Ray Tracing Cores | 32 | 46 |

| Tensor Cores | — | 184 |

| Pixel Rate | 134.4 GPixel/s | 135.4 GPixel/s |

| Texture Rate | 268.8 GTexel/s | 259.4 GTexel/s |

| FP32 Performance | 8.602 TFLOPS | 16.60 TFLOPS |

| FP16 Performance | 17.20 TFLOPS (2:1) | 16.60 TFLOPS (1:1) |

| TDP | 100 W | 115 W |

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

| Predecessor | Polaris Mobile | GeForce 20 Mobile |