AMD Radeon R9 M290X vs AMD Radeon RX 6800S Comparison

The AMD Radeon RX 6800S and the AMD Radeon R9 M290X represent two distinct eras of mobile graphics, separated by a generational gap in architecture and process technology. Benchmark data shows the RX 6800S decisively outperforming the R9 M290X, yet both cards hold a near-identical standing in the overall performance percentile, at 69 and 68 respectively. This creates an interesting dynamic where the newer card is a clear upgrade, but the older card remains a surprisingly relevant baseline for its time.

The Verdict

The data is unambiguous: the AMD Radeon RX 6800S is the superior performer in every measurable category. In the two available head-to-head benchmark comparisons, the RX 6800S wins both, with an average win rate of 100% over the R9 M290X. The magnitude of this victory is substantial, with the RX 6800S delivering a 247.6% higher score in Geekbench Metal and a 232.3% higher score in Geekbench OpenCL. For any user comparing these two for a modern workload, the RX 6800S is the only logical choice based on raw compute capability.

However, the R9 M290X is not without its own story. Its average benchmark score of 23276 places it within 0.1% of the AMD Radeon RX 6600M and 0.1% of the AMD Radeon Pro Vega 16, and it even edges out the NVIDIA P106-100 by 0.1%. This suggests that despite its age, it still operates in a performance class comparable to much more recent entry-level solutions. For legacy applications or systems where the R9 M290X is already installed, the data shows it remains a functional, if not competitive, option. The recommendation is straightforward: if choosing between the two, select the RX 6800S without hesitation; if you already own the R9 M290X, it remains a viable part for older systems, but it will not keep pace with modern titles or compute tasks.

Architecture Differences

The architectural divide between these two GPUs is vast, explaining the performance chasm. The RX 6800S is built on the RDNA 2.0 architecture, utilizing the Navi 23 chip, while the R9 M290X uses the older GCN 1.0 architecture with the Neptune chip. This is a fundamental shift in design philosophy, with RDNA 2.0 being optimized for modern gaming and compute efficiency.

The manufacturing process highlights the generational leap. The RX 6800S is fabricated on a 7 nm process at TSMC, whereas the R9 M290X uses a 28 nm process. This allows the RX 6800S to pack 11,060 million transistors into a 237 mm² die, achieving a transistor density of 46.7 million per mm². In contrast, the R9 M290X contains only 2,800 million transistors on a 212 mm² die, with a density of 13.2 million per mm². This density difference is a primary driver of the RX 6800S's superior performance and efficiency.

Core configurations also differ significantly. The RX 6800S features 2048 shading units, 128 texture mapping units, and 64 ROPs. The R9 M290X has 1280 shading units, 80 TMUs, and 32 ROPs. The RX 6800S also introduces 32 ray tracing cores, a feature entirely absent from the R9 M290X, as it lacks any RT core count. Clock speeds reinforce this advantage, with the RX 6800S boosting to 2100 MHz compared to the R9 M290X's 900 MHz boost.

Memory subsystems are another point of divergence. The RX 6800S is equipped with 8 GB of GDDR6 memory on a 128-bit bus, delivering 256.0 GB/s of bandwidth. The R9 M290X uses 4 GB of GDDR5 memory on a wider 256-bit bus, but its bandwidth is capped at 153.6 GB/s. These architectural differences—from the node to the core count to the memory type—collectively explain why the newer card is so much faster.

Head-to-Head Benchmarks

The head-to-head data provides a clear quantitative picture of the RX 6800S's dominance. In the Geekbench Metal test, the RX 6800S scores 85,256 against the R9 M290X's 24,524, a delta of 247.6% in favor of the newer card. This test is a strong indicator of general compute and graphics performance, and the margin is enormous.

The Geekbench OpenCL test tells a similar story. The RX 6800S achieves a score of 73,202, while the R9 M290X manages 22,028. This represents a 232.3% advantage for the RX 6800S. The consistency of these results across different OpenCL and Metal APIs indicates that the performance gap is not workload-specific but rather a reflection of the fundamental architectural superiority of the RDNA 2.0 part.

Looking at the broader benchmark suite for the RX 6800S, it posts an average benchmark score of 24,063, placing it within 0.4% of the NVIDIA GeForce RTX 2080 SUPER and 0.8% of the AMD Radeon RX 9070. This contextualizes its performance: it is not just fast relative to the old R9 M290X, but it holds its own against modern desktop-class GPUs. The R9 M290X, with an average score of 23,276, sits near the RX 6800S in overall percentile but achieves this with far lower peak scores, suggesting its average is propped up by older, less demanding tests where it remains competitive.

FAQ

Q: Is the AMD Radeon RX 6800S always faster than the R9 M290X?

A: Yes. In the two head-to-head benchmarks available, the RX 6800S wins both, with a 247.6% lead in Geekbench Metal and a 232.3% lead in Geekbench OpenCL.

Q: How does the R9 M290X compare to other GPUs of its era?

A: The R9 M290X has an average benchmark score of 23,276, putting it within 0.1% of the AMD Radeon RX 6600M and 0.1% of the AMD Radeon Pro Vega 16. It is also 0.1% ahead of the NVIDIA P106-100.

Q: What are the key differences in memory capacity and bandwidth?

A: The RX 6800S has 8 GB of GDDR6 memory with 256.0 GB/s bandwidth on a 128-bit bus. The R9 M290X has 4 GB of GDDR5 memory with 153.6 GB/s bandwidth on a 256-bit bus.

Q: Does the R9 M290X support ray tracing?

A: No. The R9 M290X has no ray tracing cores. The RX 6800S includes 32 ray tracing cores.

Q: What is the transistor count difference between the two?

A: The RX 6800S contains 11,060 million transistors, while the R9 M290X contains 2,800 million transistors. This is due to the RX 6800S's 7 nm process versus the R9 M290X's 28 nm process.

Q: How does the RX 6800S perform relative to its closest rivals?

A: The RX 6800S has an average benchmark score of 24,063, which is 0.4% lower than the NVIDIA GeForce RTX 2080 SUPER and 0.8% higher than the AMD Radeon RX 9070.

Where Each One Wins

The RX 6800S wins in every modern workload category that requires raw compute throughput. Its FP32 performance of 8.602 TFLOPS dwarfs the R9 M290X's 2.304 TFLOPS, making it ideal for complex 3D rendering, high-resolution gaming, and GPU-accelerated compute tasks. The presence of 32 ray tracing cores means it can handle hardware-accelerated ray tracing, a feature the R9 M290X cannot offer at all. Its higher pixel rate of 134.4 GPixel/s and texture rate of 268.8 GTexel/s also make it superior for high-fidelity visual effects and texture-heavy scenes.

The R9 M290X's wins are more about context than direct performance. Its 68th percentile standing shows it is not a slouch, and its proximity to the RX 6600M in average score means it can still handle older DirectX 11-era games or light compute workloads. It supports DirectX 12 (11_1) and Vulkan 1.2.170, so it is not entirely obsolete, but its 28.80 GPixel/s pixel rate and 72.00 GTexel/s texture rate are bottlenecks for modern games. Where the R9 M290X might be preferable is in legacy system compatibility, specifically for users with MXM module slots, as opposed to the RX 6800S's IGP form factor. For any user prioritizing current-generation gaming, features like ray tracing, or high-bandwidth memory, the RX 6800S is the definitive winner. The R9 M290X remains a serviceable fallback for dated applications, but it offers no benchmark-driven advantage over the RX 6800S.