AMD Radeon 860M vs AMD Radeon RX 6800S Comparison

The AMD Radeon 860M and AMD Radeon RX 6800S represent two distinct philosophies in mobile graphics: one is a power-sipping integrated solution built for efficiency, the other a dedicated mobile GPU engineered for raw throughput. The benchmark data shows a decisive performance gap, with the RX 6800S dominating in every shared test, yet the 860M holds its own as a remarkably capable IGP. The head-to-head results are lopsided, but the architectural and practical implications of each chip tell a more nuanced story for different use cases.

Head-to-Head Benchmarks

The two available shared benchmarks paint a clear picture of the performance hierarchy. In Geekbench OpenCL, the AMD Radeon RX 6800S scores 73,202, while the AMD Radeon 860M manages 22,759. This represents a delta of -68.9% for the 860M, meaning the dedicated GPU delivers more than three times the compute performance in this workload. The Vulkan results follow a similar pattern: the RX 6800S posts 81,272 against the 860M's 30,043, a -63% gap. In both cases, the RX 6800S is the undisputed winner, and the margin is substantial enough to classify these as entirely different performance tiers.

Looking at broader benchmark context, the RX 6800S also benefits from a richer data set. Its average benchmark score of 24,063 across multiple tests (including 3DMark Steel Nomad, Passmark DirectX variants, and Geekbench Metal) places it in the 69th percentile of all GPUs. The 860M, with only two benchmark entries, achieves an average score of 26,401 and sits in the 72nd percentile. This percentile ranking is intriguing: despite losing both head-to-head tests by wide margins, the 860M's average score is actually higher than the RX 6800S's. This discrepancy stems from the different test suites—the 860M's two Geekbench results are strong for an integrated part, while the RX 6800S's Passmark DirectX 9 score of 200 and DirectX 10 score of 91 drag down its aggregate.

Examining the nearest rivals for each chip provides additional context. The 860M's average score places it within 1.1% of the NVIDIA RTX A4000 (26,683), 0.6% of the AMD Radeon RX 5700 XT 50th Anniversary (26,553), and essentially tied with the NVIDIA GeForce MX550 (26,421, delta -0.1%) and RTX 5060 (26,331, delta +0.3%). This clustering suggests the 860M's aggregate performance is competitive with a range of discrete and workstation GPUs, even though it loses to the RX 6800S in direct comparison. The RX 6800S, meanwhile, sits within 0.9% of the NVIDIA GeForce GTX 1630 (24,277) and 0.8% of the AMD Radeon RX 9070 (23,877), while being 0.4% behind the RTX 2080 SUPER (24,170) and 0.7% behind the GTX 780 Ti (24,236). The RX 6800S's aggregate performance is thus comparable to older high-end desktop cards, a signal of its sustained capability despite being a mobile part.

Architecture Differences

The two GPUs represent two generations of AMD graphics architecture, with significant differences in manufacturing process and core design. The 860M is built on TSMC's 4 nm process node, while the RX 6800S uses the older 7 nm node. This process advantage allows the 860M to pack its RDNA 3.5 architecture into a much smaller power envelope—15 W TDP versus the RX 6800S's 100 W TDP. The RX 6800S, based on RDNA 2.0, uses the Navi 23 chip with 11,060 million transistors on a 237 mm² die, yielding a transistor density of 46.7M per mm². The 860M's transistor count and die size are listed as unknown, but its 4 nm process and integrated nature suggest a far more compact implementation.

Core configuration differences are stark. The RX 6800S features 2,048 shading units, 128 texture mapping units, and 64 ROPs, while the 860M has 512 shading units, 32 TMUs, and 16 ROPs. This 4:1 ratio in shading units and TMUs, and 4:1 in ROPs, directly explains the RX 6800S's massive performance advantage. Ray tracing cores follow the same pattern: the RX 6800S has 32 RT cores versus the 860M's 8. The RX 6800S also offers higher raw throughput numbers—8.602 TFLOPS FP32 against 3.072 TFLOPS for the 860M—and a significantly higher pixel rate of 134.4 GPixel/s versus 48.00 GPixel/s. Texture rate similarly favors the dedicated card: 268.8 GTexel/s versus 96.00 GTexel/s.

Clock speeds tell a different story. The 860M has a base clock of 600 MHz and a boost clock of 3000 MHz, while the RX 6800S runs at 1800 MHz base and 2100 MHz boost, with a game clock of 1975 MHz. The 860M's higher boost clock partially compensates for its fewer cores, but the sheer core count difference is insurmountable. Memory architecture is another major divergence. The RX 6800S has 8 GB of dedicated GDDR6 memory on a 128-bit bus, delivering 256.0 GB/s of bandwidth. The 860M uses System Shared memory, with bandwidth described as System Dependent. This means the 860M's memory performance varies with the host system's RAM configuration, while the RX 6800S has a fixed, dedicated memory pool.

Where Each One Wins

The RX 6800S is the clear choice for compute-intensive and high-resolution gaming workloads. Its 8 GB of dedicated GDDR6 memory and 256.0 GB/s bandwidth provide a stable foundation for texture-heavy scenes and larger framebuffers, while its 4x shading unit count delivers the raw geometry and pixel throughput needed for modern titles at high settings. The 3DMark Steel Nomad DX12 score of 1,689 indicates solid DirectX 12 Ultimate performance, and the Geekbench Metal score of 85,256 suggests strong cross-platform compute capabilities. For users who prioritize frame rates and visual fidelity over power consumption, the RX 6800S is the obvious winner.

The 860M, by contrast, excels in efficiency-constrained scenarios. Its 15 W TDP makes it suitable for thin-and-light laptops where battery life and thermals are paramount. The integrated design means no separate VRAM allocation is needed; the GPU shares system memory, which simplifies system design but introduces performance variability. The 860M's benchmark scores, while lower in absolute terms, are still respectable for an IGP: a Geekbench OpenCL score of 22,759 and Vulkan score of 30,043 place it in the 72nd percentile of all GPUs, ahead of several discrete mobile parts. For light gaming, content consumption, and productivity tasks that don't require massive compute throughput, the 860M offers a compelling balance of performance and efficiency.

The architectural split also matters. The 860M's RDNA 3.5 architecture, despite fewer cores, benefits from the newer 4 nm process and potentially better instruction-level efficiency per clock. Its FP16 performance is listed as 3.072 TFLOPS (1:1), indicating no special half-precision acceleration, whereas the RX 6800S offers 17.20 TFLOPS FP16 (2:1), meaning it can process twice as many half-precision operations as FP32. This makes the RX 6800S better suited for AI inference and certain compute workloads that leverage FP16 operations. Both chips support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature sets are identical.

FAQ

Q: Which GPU has higher raw compute performance?

A: The AMD Radeon RX 6800S, with 8.602 TFLOPS FP32 versus the 860M's 3.072 TFLOPS. The RX 6800S also leads in pixel rate (134.4 GPixel/s vs 48.00 GPixel/s) and texture rate (268.8 GTexel/s vs 96.00 GTexel/s).

Q: How do their power requirements compare?

A: The 860M has a 15 W TDP, while the RX 6800S has a 100 W TDP. This makes the 860M significantly more power-efficient for portable devices.

Q: Do both GPUs support the same modern APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither has dedicated tensor cores.

Q: What memory configurations do they use?

A: The RX 6800S has 8 GB of dedicated GDDR6 memory on a 128-bit bus with 256.0 GB/s bandwidth. The 860M uses System Shared memory, with bandwidth dependent on the host system.

Q: Which GPU has a better aggregate benchmark score?

A: The 860M has a higher average benchmark score of 26,401 compared to the RX 6800S's 24,063. However, this is due to different test suites; in direct head-to-head comparisons, the RX 6800S wins both tests by wide margins.

Q: What is the production status of each GPU?

A: The 860M is listed as Active, while the RX 6800S is End-of-life. The 860M was released on 2025-02-28, and the RX 6800S on 2022-01-03.

Specification Differences

| Specification | AMD Radeon 860M | AMD Radeon RX 6800S |

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

| Architecture | RDNA 3.5 | RDNA 2.0 |

| Process Node | 4 nm | 7 nm |

| Transistors | Unknown | 11,060 million |

| Die Size | Unknown | 237 mm² |

| Transistor Density | N/A | 46.7M / mm² |

| Base Clock | 600 MHz | 1800 MHz |

| Boost Clock | 3000 MHz | 2100 MHz |

| Game Clock | N/A | 1975 MHz |

| Memory Size | System Shared | 8 GB |

| Memory Type | System Shared | GDDR6 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 256.0 GB/s |

| Shading Units | 512 | 2048 |

| TMUs | 32 | 128 |

| ROPs | 16 | 64 |

| RT Cores | 8 | 32 |

| Pixel Rate | 48.00 GPixel/s | 134.4 GPixel/s |

| Texture Rate | 96.00 GTexel/s | 268.8 GTexel/s |

| FP32 Performance | 3.072 TFLOPS | 8.602 TFLOPS |

| FP16 Performance | 3.072 TFLOPS (1:1) | 17.20 TFLOPS (2:1) |

| TDP | 15 W | 100 W |

| Production Status | Active | End-of-life |

| Release Date | 2025-02-28 | 2022-01-03 |

| Predecessor | Navi II IGP | Polaris Mobile |