AMD Radeon RX 6700M vs AMD Radeon RX 6800S Comparison
AMD Radeon RX 6700M
Radeon RX 6800S
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon RX 6700M vs AMD Radeon RX 6800S
Both the AMD Radeon RX 6700M and the AMD Radeon RX 6800S are RDNA 2.0 mobile parts from the Radeon RX 6000 series, but the benchmark data shows they are tuned for very different personalities. The 6700M leans on a larger die and more memory bandwidth, while the 6800S counters with higher base clocks and efficiency. The result is a split decision: the 6700M dominates modern API workloads, while the 6800S pulls ahead in legacy and compute-oriented tasks.
Head-to-Head Benchmarks
The most decisive win for the AMD Radeon RX 6700M comes in the Geekbench Vulkan test, where it scores 90816 against the 6800S’s 81272. That is an 11.7% advantage, the largest margin in any modern 3D workload. The 3DMark Steel Nomad DX12 test also goes to the 6700M, with a score of 1845 versus 1689, a 9.2% lead. These two results confirm that the 6700M is the stronger choice for current-generation game engines that rely on DX12 and Vulkan.
The 6700M also wins the Geekbench Metal and OpenCL tests, though by smaller margins. In Metal, it scores 91911 against 85256, a 7.8% difference. In OpenCL, the gap narrows to 6.1%, with scores of 77666 and 73202. These wins suggest the 6700M’s additional shading units and texture units translate into consistent compute throughput across different API abstraction layers.
The PassMark DirectX 10 result is nearly a tie, with the 6700M edging out the 6800S by just 1.1% (92 vs 91). The DirectX 12 test is an exact tie at 65 points each, meaning both cards deliver identical performance in that specific rasterization path.
The AMD Radeon RX 6800S fights back in the older DirectX 9 and DirectX 11 tests. In DirectX 9, it scores 200 versus 155, a massive 22.5% advantage for the 6800S. In DirectX 11, it wins 139 to 129, a 7.2% margin. These results are counterintuitive given the 6700M’s hardware advantage, but they indicate the 6800S has better driver optimization or clock behavior for legacy API paths.
The 6800S also dominates the PassMark 2D and 3D tests. In PassMark G2D, it scores 732 against 555, a 24.2% advantage. In PassMark G3D, the 6800S leads with 15908 versus 13536, a 14.9% margin. The PassMark GPU Compute test goes to the 6800S as well, with 6144 against 5191, a 15.5% lead. These are substantial wins in the PassMark suite, which uses a mix of synthetic and real-world rendering tasks.
Overall, the head-to-head tally is narrow: the 6700M wins 6 tests, and the 6800S wins 5. But the distribution of wins is telling. The 6700M takes every modern API test (DX12, Vulkan, Metal, OpenCL), while the 6800S cleans up in legacy DX9, DX11, and PassMark’s proprietary benchmarks. If you are playing current titles, the 6700M is the clear pick. If you are running older software or compute-heavy workloads, the 6800S often comes out ahead.
FAQ
Q: Which GPU has the higher average benchmark score?
A: The AMD Radeon RX 6700M has an average benchmark score of 25633, placing it in the 71st percentile of all GPUs. The AMD Radeon RX 6800S averages 24063, sitting in the 69th percentile. The 6700M leads by about 6.5% in average score.
Q: How does the 6700M compare to its nearest rivals?
A: The 6700M’s nearest rival is the AMD Radeon Pro W5700, which scores 25726, just 0.4% higher. The NVIDIA GeForce RTX 3080 Mobile is also 0.4% behind at 25740. Both are essentially tied with the 6700M in aggregate performance.
Q: What about the 6800S’s competitive position?
A: The 6800S is 0.4% behind the NVIDIA GeForce RTX 2080 SUPER (24170) and 0.7% behind the GTX 780 Ti (24236). Interestingly, the AMD Radeon RX 9070 scores 23877, which is 0.8% lower than the 6800S, so the 6800S holds a small edge over that newer card.
Q: Which card wins in modern DirectX 12 workloads?
A: The 6700M wins the 3DMark Steel Nomad DX12 test with a score of 1845, a 9.2% advantage over the 6800S’s 1689. In PassMark DirectX 12, both score exactly 65, so there is no separation in that specific test.
Q: Is the 6800S better for older DirectX 9 games?
A: Yes, decisively. The 6800S scores 200 in PassMark DirectX 9 versus 155 for the 6700M, a 22.5% lead. This is the largest single-test margin in the entire comparison.
Q: Which card has the higher memory bandwidth?
A: The 6700M has 320.0 GB/s of bandwidth, compared to 256.0 GB/s for the 6800S. This comes from a 160-bit bus with 10 GB of GDDR6 versus a 128-bit bus with 8 GB.
Where Each One Wins
The AMD Radeon RX 6700M is the winner for any workload that targets modern graphics APIs. Its Vulkan lead of 11.7% and DX12 lead of 9.2% are significant enough to matter in current game engines. The 7.8% Metal advantage also makes it the better option for macOS or iOS development environments. If you are playing 2023 or newer AAA titles, the 6700M’s higher shading unit count (2304 vs 2048) and texture units (144 vs 128) give it the raw throughput needed for high-resolution textures and complex shaders.
The AMD Radeon RX 6800S is the winner for legacy software and compute-heavy tasks. Its 22.5% DirectX 9 advantage is enormous, making it the better choice for older game libraries or emulation scenarios. The 15.5% lead in PassMark GPU Compute and 14.9% lead in PassMark G3D suggest the 6800S has stronger performance in mixed compute and older rasterization pipelines. The 24.2% advantage in PassMark G2D also indicates better 2D acceleration, which is relevant for desktop composition or CAD-style applications.
The 6800S also wins on efficiency. It has a 100 W TDP versus 135 W for the 6700M, meaning it delivers competitive performance with 35 W less power draw. For thin-and-light laptops, that lower power budget is a practical advantage.
Specification Differences
The most fundamental difference is the GPU die. The 6700M uses the Navi 22 chip, while the 6800S uses Navi 23. This leads to a major transistor and die size gap: the Navi 22 has 17,200 million transistors on a 335 mm² die, while the Navi 23 has 11,060 million on 237 mm². The 6700M’s transistor density is 51.3M / mm², slightly higher than the 6800S’s 46.7M / mm².
Memory configuration is another clear differentiator. The 6700M has 10 GB of GDDR6 on a 160-bit bus, yielding 320.0 GB/s of bandwidth. The 6800S has 8 GB on a 128-bit bus, yielding 256.0 GB/s. Both run memory at 2000 MHz with 16 Gbps effective speed.
Clock speeds tell a different story. The 6800S has a higher base clock at 1800 MHz versus 1489 MHz, but the 6700M has a higher boost clock at 2400 MHz versus 2100 MHz. The game clock also favors the 6700M at 2300 MHz versus 1975 MHz. This suggests the 6700M is designed for sustained high-frequency operation, while the 6800S’s higher base clock helps with idle and low-load scenarios.
The compute units differ as well. The 6700M has 2304 shading units, 144 TMUs, and 64 ROPs, plus 36 ray tracing cores. The 6800S has 2048 shading units, 128 TMUs, and 64 ROPs, with 32 ray tracing cores. The pixel rate is 153.6 GPixel/s for the 6700M versus 134.4 for the 6800S, and the texture rate is 345.6 GTexel/s versus 268.8.
Finally, the bus interface differs: the 6700M uses PCIe 4.0 x16, while the 6800S uses PCIe 4.0 x8. Both are IGP form factors with no power connectors, and both support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4.
Architecture Differences
Both GPUs are built on RDNA 2.0 using a 7 nm process from TSMC, but they represent different design philosophies within that architecture. The 6700M is a larger die with more transistors and higher power budget, clearly aimed at performance-first gaming laptops. The 6800S is a smaller, more efficient die that prioritizes power consumption.
The FP32 compute throughput highlights this: the 6700M delivers 11.06 TFLOPS, while the 6800S delivers 8.602 TFLOPS. In FP16, the 6700M hits 22.12 TFLOPS (2:1) versus 17.20 TFLOPS for the 6800S. This 22-28% compute advantage for the 6700M directly explains its wins in Vulkan and Metal, which scale well with raw shader throughput.
The 6800S compensates with a higher base clock. At 1800 MHz, it starts higher than the 6700M’s 1489 MHz, which helps in bursty, low-duration workloads that don’t require sustained boost. The lower boost clock of 2100 MHz means it peaks lower, but the higher base clock suggests better consistency at lower power states.
Both cards have the same ROP count (64) and the same memory type (GDDR6), but the 6700M’s larger bus width (160-bit vs 128-bit) gives it a 25% bandwidth advantage. This is critical for high-resolution textures and 4K workloads, where memory bandwidth often becomes the bottleneck.
Ray tracing cores are present on both, but the 6700M has 36 versus 32 on the 6800S. Neither card has tensor cores, so all AI and DLSS-style workloads are handled by the shader units.
The 6800S also has a lower transistor density (46.7M / mm²) than the 6700M (51.3M / mm²), which suggests the 6800S is a more relaxed design that prioritizes thermal headroom and battery life over raw density.
The Verdict
The data tells a clear story: the AMD Radeon RX 6700M is the superior GPU for modern gaming and high-performance compute. It wins every modern API benchmark by substantial margins, has 28% more FP32 throughput, 25% more memory bandwidth, and 35% more power budget. If your priority is playing current AAA titles at high settings, the 6700M’s 9.2% DX12 lead and 11.7% Vulkan lead are decisive.
The AMD Radeon RX 6800S is the better pick for efficiency-conscious buyers and users with legacy software needs. Its 100 W TDP is 35 W lower than the 6700M, making it suitable for thinner laptops. The 22.5% DirectX 9 advantage and 15.5% compute lead mean it will handle older games and compute tasks faster, despite being behind in modern APIs.
For a gamer buying a laptop in 2024, the 6700M is the rational choice. Its 71st percentile ranking versus the 6800S’s 69th percentile confirms it as the faster card overall. But for a professional who runs legacy CAD or OpenGL applications, the 6800S’s wins in PassMark G2D and DirectX 9 make it a more versatile tool.
The tie in PassMark DirectX 12 (65 each) is a reminder that these cards are more similar than different in some paths. However, the 6700M’s wins in every modern API test are weighted toward real-world current usage. If you value future-proofing and maximum performance in new games, take the 6700M. If you value power efficiency and backward compatibility, take the 6800S. There is no wrong answer—just different priorities.