AMD Radeon RX 6800S

AMD Radeon RX 6800S is a mobile graphics solution built on the RDNA 2.0 architecture, utilizing the Navi 23 chip fabricated on a 7 nm process at TSMC. This end-of-life product, released in early 2022, is positioned within the Radeon RX 6000 series and targets portable devices, as indicated by its IGP slot width and portable-device-dependent display outputs. The GPU integrates 11,060 million transistors on a 237 mm² die, resulting in a transistor density of 46.7M per mm², and is configured with 2048 shading units, 128 texture mapping units, and 64 render output units.

Power and Cooling

The RX 6800S carries a thermal design power (TDP) of 100 W, a figure that underscores its efficiency-focused design for mobile form factors. This power envelope is notably modest for the performance tier it occupies, allowing for thinner and lighter laptop implementations. The data shows no dedicated power connectors are required, meaning the card draws all its power through the motherboard slot, which is consistent with its integrated graphics processor (IGP) classification. Consequently, there is no suggested PSU rating provided in the specifications, as the power delivery is handled entirely by the host system's design rather than an end-user upgradeable component.

Cooling solutions for this GPU are inherently dependent on the portable device manufacturer, with the slot width listed as IGP. The 7 nm process node from TSMC is a key enabler of the 100 W TDP, as it allows for higher clock speeds—1800 MHz base, 1975 MHz game, and 2100 MHz boost—without escalating thermal output. The memory operates at 2000 MHz, translating to 16 Gbps effective, across a 128-bit bus, which yields a bandwidth of 256.0 GB/s. This combination of a modest TDP and modern process node suggests that a capable air cooler, as designed by the laptop vendor, is sufficient to maintain sustained performance, though the exact thermal solution remains device-specific.

How It Compares

The benchmark data positions the RX 6800S in a tight cluster of competing GPUs, with its average benchmark score of 23964 placing it at the 67th percentile of all GPUs. The nearest rivals are separated by margins of less than 1.5%, indicating that performance differences are minimal in aggregate.

Against the AMD Radeon Pro W5700, the RX 6800S trails by a negligible 0.7% in average score (24131 vs. 23964). This places the two products on nearly identical footing, with the Pro W5700's workstation-oriented drivers likely accounting for its slight edge in professional applications, though gaming performance would be expected to be indistinguishable.

The NVIDIA GeForce GTX 1630 holds a 1.1% advantage over the RX 6800S, with an average score of 24233. This margin is within typical run-to-run variance, meaning in real-world usage, the two GPUs should deliver comparable frame rates, though architectural differences in feature support may differentiate them in specific workloads.

The RX 6800S actually leads the NVIDIA GeForce RTX 2080 by 1.3%, with scores of 23964 and 23664 respectively. This is a notable result, as the RTX 2080 is a desktop-class GPU from a previous generation, yet the mobile RDNA 2 part manages to edge ahead in aggregate benchmark performance, likely due to newer architectural efficiencies.

Similarly, the RX 6800S outperforms the NVIDIA GeForce RTX 3080 Mobile by 1.4%, scoring 23964 against 23628. This margin is modest but consistent, suggesting that the AMD part delivers competitive raw performance despite the RTX 3080 Mobile's higher tier positioning in NVIDIA's lineup, though the latter may excel in ray-traced scenarios.

Ray Tracing and Feature Set

The RX 6800S integrates 32 ray accelerators, AMD's implementation of dedicated ray tracing hardware within the RDNA 2.0 architecture. This enables hardware-accelerated ray tracing in supported titles, a feature that was a key addition to the Radeon RX 6000 series. The GPU also supports DirectX 12 Ultimate (12_2), which mandates ray tracing and variable rate shading support, alongside OpenGL 4.6 and Vulkan 1.4 for broader API compatibility.

The absence of tensor cores in the specification sheet indicates that AMD relies on its shader units for compute tasks rather than dedicated AI acceleration hardware, unlike NVIDIA's Tensor Core approach. Despite this, the GPU achieves a stated FP32 performance of 8.602 TFLOPS and FP16 performance of 17.20 TFLOPS (at a 2:1 ratio), providing substantial raw compute throughput for general-purpose workloads. The feature set is rounded out by PCIe 4.0 x8 interface support, which offers sufficient bandwidth for the 8 GB GDDR6 memory pool, and the pixel rate of 134.4 GPixel/s with a texture rate of 268.8 GTexel/s.

FAQ

Q: What is the thermal design power of the RX 6800S?

A: The RX 6800S has a TDP of 100 W, which is a relatively low figure for its performance class, enabling its use in slim portable devices.

Q: Does the RX 6800S require an external power connector?

A: No, the specifications list power connectors as "None," meaning the GPU draws all necessary power through the motherboard interface, consistent with its IGP slot width.

Q: How does the RX 6800S compare to the NVIDIA GeForce RTX 3080 Mobile in average benchmark score?

A: The RX 6800S scores 23964, which is 1.4% higher than the RTX 3080 Mobile's average score of 23628, indicating a slight aggregate performance lead for the AMD part.

Q: What is the memory configuration of the RX 6800S?

A: It features 8 GB of GDDR6 memory on a 128-bit bus, with a memory clock of 2000 MHz (16 Gbps effective), providing a bandwidth of 256.0 GB/s.

Q: Which API versions are supported by the RX 6800S?

A: The GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, ensuring compatibility with modern graphics APIs.

Q: What is the production status of the RX 6800S?

A: The production status is listed as "End-of-life," and it was released on 2022-01-03, marking it as a discontinued product in the Radeon RX 6000 series.

Benchmark Performance

The aggregate benchmark data places the RX 6800S at the 67th percentile of all GPUs, with an average score of 23964 across multiple tests. This percentile ranking indicates that it outperforms roughly two-thirds of all GPUs in the database, a strong showing for a mobile part. The individual benchmark results reveal a varied profile: in 3DMark Steel Nomad (DX12), it scores 1689, while Geekbench scores are 85256 in Metal, 73202 in OpenCL, and 80181 in Vulkan, showcasing strong compute performance across different APIs. PassMark scores are notably lower, with G3D at 15908 and GPU Compute at 6144, but DirectX 9, 10, 11, and 12 scores of 200, 91, 139, and 65 respectively indicate that legacy API performance is less of a focus.

Comparing the RX 6800S to its nearest rivals, the data shows a remarkably tight grouping. The AMD Radeon Pro W5700 leads by 0.7% (24131 vs. 23964), while the NVIDIA GeForce GTX 1630 leads by 1.1% (24233 vs. 23964). Conversely, the RX 6800S leads the NVIDIA GeForce RTX 2080 by 1.3% (23964 vs. 23664) and the NVIDIA GeForce RTX 3080 Mobile by 1.4% (23964 vs. 23628). These margins, all under 1.5%, suggest that the RX 6800S delivers essentially equivalent performance to these four rivals in aggregate terms, with differences that would be imperceptible in most gaming scenarios.

The 1.4% delta over the RTX 3080 Mobile is particularly noteworthy given the latter's higher tier in NVIDIA's mobile lineup, but the RX 6800S compensates with its RDNA 2 architecture and higher boost clock of 2100 MHz. The FP32 throughput of 8.602 TFLOPS supports this competitive positioning, as does the 256.0 GB/s memory bandwidth, which, while not exceptional, is sufficient for the 128-bit bus. In rasterized workloads, the RX 6800S should perform on par with these rivals, though in ray-traced titles, the 32 ray accelerators may not match the dedicated RT cores of NVIDIA parts, potentially shifting the balance in specific games. Overall, the benchmark data indicates that the RX 6800S is a well-rounded mobile GPU that trades blows with its closest competitors, offering a balanced mix of compute and graphics performance within a 100 W power envelope.