AMD Radeon RX 5700 vs AMD Radeon RX 6800S Comparison

# AMD Radeon RX 6800S vs AMD Radeon RX 5700

The AMD Radeon RX 6800S and AMD Radeon RX 5700 represent two different generations of RDNA architecture, with the 6800S being a mobile-focused part from the RX 6000M series and the 5700 being a desktop card from the RX 5000 series. The benchmark data shows a near-split decision: the 6800S wins 6 of 11 head-to-head tests, while the 5700 wins 5, yet the overall average benchmark scores tell a different story. The RX 6800S posts an average score of 24,063 versus 22,170 for the RX 5700, placing the mobile part at the 69th percentile of all GPUs compared to the 5700’s 67th percentile. That 8.5% average score gap is meaningful, but the individual test results reveal a more nuanced picture, with each card dominating in different API workloads.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon RX 6800S has an average benchmark score of 24,063, which is 8.5% higher than the RX 5700’s 22,170. The 6800S also sits at the 69th percentile of all GPUs, versus the 5700’s 67th percentile.

Q: Does the RX 6800S support hardware ray tracing?

A: Yes, the RX 6800S includes 32 dedicated ray tracing cores, while the RX 5700 has no ray tracing cores at all. This is a direct architectural advantage for the 6800S, which supports DirectX 12 Ultimate (12_2) compared to the 5700’s DirectX 12 (12_1).

Q: How do the two cards compare in Vulkan performance?

A: The RX 6800S is significantly faster in Vulkan, scoring 81,272 in Geekbench Vulkan versus 64,166 for the RX 5700. That is a 26.7% lead for the 6800S, which is the largest single benchmark win for either card in the head-to-head results.

Q: Which card has higher memory bandwidth?

A: The RX 5700 has higher memory bandwidth at 448.0 GB/s, thanks to its 256-bit memory bus, compared to the RX 6800S’s 256.0 GB/s over a 128-bit bus. Both cards have 8 GB of GDDR6 memory, but the 5700’s wider bus provides 75% more bandwidth.

Q: What is the power consumption difference?

A: The RX 6800S has a TDP of 100 W, while the RX 5700 has a TDP of 180 W. The 6800S is an integrated-class mobile part (IGP slot width) with no power connectors, whereas the 5700 is a dual-slot desktop card requiring one 6-pin and one 8-pin power connector.

Q: Which card performs better in DirectX 11 workloads?

A: The RX 6800S is substantially ahead in PassMark DirectX 11, scoring 139 versus 101 for the RX 5700, a 37.6% advantage. This is the largest percentage win for the 6800S in any benchmark test.

Architecture Differences

The architectural divide between these two GPUs is generational. The RX 6800S is built on RDNA 2.0, while the RX 5700 uses the original RDNA 1.0. Both are manufactured on TSMC’s 7 nm process, but the 6800S uses the Navi 23 chip with 11,060 million transistors on a 237 mm² die, while the 5700 uses the larger Navi 10 chip with 10,300 million transistors on a 251 mm² die. That means the 6800S achieves higher transistor density at 46.7M per mm² versus 41.0M per mm² for the 5700.

The most significant functional difference is ray tracing support. The RX 6800S includes 32 ray tracing cores, a feature entirely absent from the RX 5700, which has no RT cores listed. This directly enables the 6800S to support DirectX 12 Ultimate (12_2), whereas the 5700 is limited to DirectX 12 (12_1). Both cards support OpenGL 4.6 and Vulkan 1.4, but the 6800S’s newer architecture brings the hardware capabilities needed for modern graphics features.

Clock speeds also favor the 6800S substantially. The 6800S runs at a base clock of 1800 MHz, a game clock of 1975 MHz, and a boost clock of 2100 MHz. The RX 5700 is slower across the board, with a 1465 MHz base, 1625 MHz game, and 1725 MHz boost clock. The 6800S also runs faster memory at 2000 MHz (16 Gbps effective) versus 1750 MHz (14 Gbps effective) for the 5700. The 6800S compensates for its narrower 128-bit memory bus with higher clocks and newer architecture, though it still trails in raw bandwidth. The 5700’s 256-bit bus gives it 448.0 GB/s of bandwidth, while the 6800S manages 256.0 GB/s.

The power envelope is radically different. The 6800S is designed as an integrated mobile part with a 100 W TDP, no external power connectors, and a slot width listed as IGP. The 5700 is a traditional dual-slot desktop card rated at 180 W, requiring a 6-pin and 8-pin power connector, with a suggested PSU of 450 W. The 6800S achieves its performance at nearly half the power draw, which is the defining characteristic of its mobile design.

The Verdict

The data shows a clear split based on use case. The AMD Radeon RX 6800S is the better choice for modern API workloads and efficiency. It wins decisively in Vulkan (26.7% ahead), DirectX 11 (37.6% ahead), and DirectX 12 (20.4% ahead), and it offers ray tracing hardware that the RX 5700 simply does not have. Its 8.5% higher average benchmark score and superior percentile ranking (69th versus 67th) confirm its overall performance advantage. The 100 W TDP makes it suitable for thin-and-light gaming laptops where power draw is critical.

The AMD Radeon RX 5700, despite being an older desktop card, still holds its own in specific scenarios. It wins in DirectX 9, Metal, and GPU compute tests, and its 448.0 GB/s memory bandwidth is 75% higher than the 6800S. For users running legacy DirectX 9 titles or compute-heavy workloads, the 5700’s wider memory bus and higher compute scores (6,517 versus 6,144 in PassMark GPU Compute) give it an edge. It also wins in 3DMark Steel Nomad DX12, a modern test, by 9.3%.

The verdict depends on priorities. If you want maximum performance in current and upcoming APIs, ray tracing support, and drastically lower power consumption, the RX 6800S is the data-backed pick. If you need raw memory bandwidth, better compute performance, or are primarily playing older DirectX 9 games, the RX 5700 remains competitive despite its age. The 5700’s launch MSRP was 349 USD, but this is not a price comparison — it is a capability comparison.

Specification Differences

The two cards differ across nearly every specification category. The RX 6800S uses the Navi 23 chip with RDNA 2.0 architecture and 11,060 million transistors on a 237 mm² die, while the RX 5700 uses Navi 10 with RDNA 1.0 and 10,300 million transistors on a larger 251 mm² die. Transistor density favors the 6800S at 46.7M per mm² versus 41.0M per mm².

Clock speeds are uniformly higher on the 6800S: 1800 MHz base versus 1465 MHz, 1975 MHz game versus 1625 MHz, and 2100 MHz boost versus 1725 MHz. Memory clocks follow the same pattern, with the 6800S at 2000 MHz (16 Gbps effective) versus 1750 MHz (14 Gbps effective) for the 5700. However, the 5700 has a 256-bit memory bus providing 448.0 GB/s bandwidth, while the 6800S has a 128-bit bus providing 256.0 GB/s.

The 5700 has more shading units (2304 versus 2048) and more texture mapping units (144 versus 128), but both have 64 ROPs. The 6800S adds 32 ray tracing cores, which the 5700 lacks entirely. Pixel rate favors the 6800S at 134.4 GPixel/s versus 110.4 GPixel/s, and texture rate is also higher at 268.8 GTexel/s versus 248.4 GTexel/s. FP32 compute is 8.602 TFLOPS for the 6800S versus 7.949 TFLOPS for the 5700, with FP16 at 17.20 TFLOPS versus 15.90 TFLOPS (both at 2:1 ratio).

Power and physical design diverge completely. The 6800S is a 100 W integrated mobile part with no power connectors and IGP slot width, while the 5700 is a 180 W dual-slot desktop card with 1x 6-pin + 1x 8-pin connectors and a 450 W suggested PSU. The 5700 has specific dimensions of 268 mm length, 111 mm height, and 36 mm width, while the 6800S lists no dimensions as it is portable-device dependent. The 5700 uses PCIe 4.0 x16, while the 6800S uses PCIe 4.0 x8. Display outputs also differ: the 5700 has 1x HDMI 2.0b and 3x DisplayPort 1.4a, while the 6800S is portable-device dependent. The 6800S supports DirectX 12 Ultimate (12_2), while the 5700 supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The 6800S was released on 2022-01-03, while the 5700 was released on 2019-07-06, and both are end-of-life.

Head-to-Head Benchmarks

The benchmark results show a clear pattern: the RX 6800S dominates in modern API tests, while the RX 5700 wins in legacy and compute workloads. The 6800S’s biggest win is in PassMark DirectX 11, where it scores 139 versus 101 for the 5700, a 37.6% advantage. That is a massive lead and indicates the newer architecture handles DirectX 11 far more efficiently. In Geekbench Vulkan, the 6800S posts 81,272 against 64,166 for the 5700, a 26.7% win that highlights the RDNA 2.0 improvements in low-level API performance. PassMark DirectX 12 also favors the 6800S, with a score of 65 versus 54, a 20.4% lead.

The 6800S also wins in PassMark G3D (15,908 versus 14,314, an 11.1% advantage), Geekbench OpenCL (73,202 versus 69,603, a 5.2% win), and PassMark DirectX 10 (91 versus 87, a 4.6% edge). These results show consistent superiority across the board in synthetic gaming workloads.

The RX 5700 fights back in five tests. Its largest win is in PassMark G2D, scoring 877 versus 732 for the 6800S, a 16.5% advantage in 2D graphics performance. In 3DMark Steel Nomad DX12, the 5700 scores 1,862 against 1,689, a 9.3% win. PassMark GPU Compute favors the 5700 at 6,517 versus 6,144, a 5.7% lead. The 5700 wins narrowly in Geekbench Metal (86,079 versus 85,256, a 1% edge) and in PassMark DirectX 9 (209 versus 200, a 4.3% advantage).

The 3DMark Steel Nomad result is particularly interesting because it is a modern DX12 test, yet the older 5700 wins there while losing the PassMark DX12 test. This suggests the 5700’s higher memory bandwidth (448.0 GB/s) helps in certain memory-intensive workloads, even though the 6800S has higher clocks and newer architecture.

Where Each One Wins

The RX 6800S is the clear winner for modern gaming and general GPU compute through modern APIs. It takes 6 of 11 head-to-head tests, and its wins are concentrated in the APIs that matter most for current games: DirectX 11, DirectX 12, and Vulkan. The 37.6% DirectX 11 win is the single largest margin in the entire comparison, and the 26.7% Vulkan win shows that RDNA 2.0’s architectural improvements translate directly to real-world performance. The 6800S also has ray tracing cores and DirectX 12 Ultimate support, making it the only one of the two that can handle ray-traced games. Its 8.5% higher average benchmark score and 2-percentile advantage in the global GPU ranking confirm that it is the faster card overall. For anyone building a mobile gaming system or prioritizing efficiency, the 6800S’s 100 W TDP versus the 5700’s 180 W TDP is a decisive factor.

The RX 5700 wins in specific scenarios that favor its wider memory bus and higher compute throughput. Its 448.0 GB/s bandwidth is 75% higher than the 6800S, which explains its 9.3% win in 3DMark Steel Nomad and its 5.7% win in PassMark GPU Compute. The 16.5% win in PassMark G2D suggests the 5700 handles 2D workloads substantially better. The 5700 also wins in DirectX 9 (4.3% ahead) and Geekbench Metal (1% ahead), making it the better choice for legacy game compatibility and Apple Metal environments. The 5700’s compute advantage (6,517 versus 6,144) is notable for users running compute-heavy applications that rely on OpenCL or other general-purpose GPU workloads. For a desktop user with a 450 W PSU who plays older titles or runs compute tasks, the RX 5700 remains a viable option despite being from an older generation.