AMD Radeon RX 590 vs AMD Radeon RX 6800S Comparison

The AMD Radeon RX 590 and AMD Radeon RX 6800S occupy very different corners of the GPU landscape despite sharing a brand name. The RX 590 is an end-of-life desktop part from the Polaris generation, built on a 12nm process and aimed at 1080p gaming in a full-size tower. The RX 6800S is a mobile-first RDNA 2 part designed for slim gaming laptops, with a dramatically lower power target. Benchmark data shows the RX 6800S holds a clear performance lead in every direct comparison, but the RX 590’s desktop pedigree means it still has a place in older systems. The following analysis breaks down where each card wins, loses, and why you might choose one over the other based strictly on the numbers.

FAQ

Q: Which GPU is faster in the 3DMark Steel Nomad DX12 test?

A: The AMD Radeon RX 6800S scores 1689, which is 36% higher than the RX 590’s 1081. The deltaPct of -36 indicates the RX 590 trails by that margin in this head-to-head benchmark.

Q: How do the two cards compare in Geekbench Metal performance?

A: The RX 6800S posts a score of 85256, while the RX 590 scores 48407. This gives the RX 6800S a 43.2% advantage, the largest single-benchmark gap between the two in the data.

Q: What are the architectural generations of these GPUs?

A: The RX 590 uses GCN 4.0 architecture on a 12nm process, part of the Polaris (RX 500) generation. The RX 6800S uses RDNA 2.0 on a 7nm process, belonging to the Navi Mobile (RX 6000M) generation.

Q: Is the RX 6800S a desktop or laptop part?

A: The data lists the RX 6800S as having an IGP slot width, no power connectors, and display outputs described as "Portable Device Dependent." This indicates it is a mobile GPU, whereas the RX 590 is a dual-slot desktop card with a 1x 8-pin power connector.

Q: Which card has more shading units?

A: The RX 590 has 2304 shading units, while the RX 6800S has 2048. Despite having fewer shading units, the RX 6800S achieves higher performance due to its newer architecture and higher clock speeds.

Q: What is the power consumption difference?

A: The RX 590 has a TDP of 175W, while the RX 6800S has a TDP of 100W. The RX 6800S delivers higher performance while drawing 75W less, reflecting the efficiency gain from the 7nm process.

Architecture Differences

The RX 590 is built on GCN 4.0, a mature architecture from the Polaris era. It uses a 12nm process at GlobalFoundries, packing 5,700 million transistors into a 232 mm² die. This yields a transistor density of 24.6M per mm². The RX 6800S, by contrast, uses RDNA 2.0, a fundamental redesign that prioritizes efficiency and higher clocks. It is manufactured on TSMC’s 7nm process, fitting 11,060 million transistors into a slightly larger 237 mm² die, for a density of 46.7M per mm² — nearly double the RX 590’s.

Clock speeds tell a similar story. The RX 590 runs at a base of 1469 MHz and boosts to 1545 MHz. The RX 6800S starts at 1800 MHz base, reaches 1975 MHz in game mode, and boosts up to 2100 MHz. This clock advantage, combined with architectural improvements, allows the RX 6800S to achieve higher pixel and texture rates despite having fewer execution units. The RX 6800S also features 32 ray tracing cores, a capability the RX 590 lacks entirely.

Memory architecture differs significantly. Both cards have 8 GB of VRAM and the same 256.0 GB/s bandwidth, but the RX 590 uses GDDR5 on a 256-bit bus, while the RX 6800S uses GDDR6 on a 128-bit bus. The newer memory type and higher effective speed (16 Gbps vs. 8 Gbps) allow the RX 6800S to match bandwidth with half the bus width, saving power and die space.

API support also diverges. The RX 590 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. The RX 6800S steps up to DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RX 6800S also supports FP16 at a 2:1 rate (17.20 TFLOPS) versus the RX 590’s 1:1 rate (7.119 TFLOPS), which matters for compute workloads that use half precision.

Where Each One Wins

The RX 6800S wins outright in every head-to-head benchmark recorded. In 3DMark Steel Nomad DX12, it leads by 36%. In Geekbench Metal, it leads by 43.2%. Looking at its broader benchmark suite, the RX 6800S also shows strength in compute and legacy tests: it scores 73202 in Geekbench OpenCL, 81272 in Geekbench Vulkan, and 15908 in Passmark G3D. These numbers suggest it is the clear choice for any modern workload, from gaming to GPU-accelerated tasks.

The RX 590’s wins are more situational. Its 70th percentile vs. all GPUs is marginally better than the RX 6800S’s 69th percentile, and its average benchmark score of 24744 is actually higher than the RX 6800S’s 24063. This is because the RX 590’s benchmark pool is smaller and skews toward specific tests. The RX 590 also has a higher average score than the RX 6800S, despite losing the direct head-to-head tests. This suggests that in some undefined legacy or mixed workloads, the RX 590’s higher shading unit count and wider memory bus may still hold up.

For a desktop builder with an existing PCIe 3.0 system, the RX 590 is a drop-in part with standard display outputs (1x HDMI 2.0b, 3x DisplayPort 1.4a). The RX 6800S, being a mobile part with no power connectors and portable-device-dependent outputs, cannot be installed in a desktop. If you need a card for a standard ATX build, the RX 590 is the only option that physically fits.

Specification Differences

| Specification | AMD Radeon RX 590 | AMD Radeon RX 6800S |

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

| Architecture | GCN 4.0 | RDNA 2.0 |

| Process Node | 12 nm | 7 nm |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 5,700 million | 11,060 million |

| Die Size | 232 mm² | 237 mm² |

| Transistor Density | 24.6M / mm² | 46.7M / mm² |

| Base Clock | 1469 MHz | 1800 MHz |

| Boost Clock | 1545 MHz | 2100 MHz |

| Game Clock | None | 1975 MHz |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus Width | 256 bit | 128 bit |

| Memory Speed | 8 Gbps effective | 16 Gbps effective |

| Shading Units | 2304 | 2048 |

| TMUs | 144 | 128 |

| ROPs | 32 | 64 |

| Ray Tracing Cores | None | 32 |

| Pixel Rate | 49.44 GPixel/s | 134.4 GPixel/s |

| Texture Rate | 222.5 GTexel/s | 268.8 GTexel/s |

| FP32 | 7.119 TFLOPS | 8.602 TFLOPS |

| FP16 | 7.119 TFLOPS (1:1) | 17.20 TFLOPS (2:1) |

| TDP | 175 W | 100 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 8-pin | None |

| Suggested PSU | 450 W | None |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

| Display Outputs | 1x HDMI 2.0b, 3x DisplayPort 1.4a | Portable Device Dependent |

| DirectX Support | 12 (12_0) | 12 Ultimate (12_2) |

| Vulkan Support | 1.3 | 1.4 |

| Dimensions | 241 mm (9.5 inches) | Not specified |

| Release Date | 2018-11-14 | 2022-01-03 |

| Launch MSRP | 279 USD | None |

Head-to-Head Benchmarks

The 3DMark Steel Nomad DX12 test is a modern DirectX 12 workload that stresses raw rasterization and compute. The RX 6800S scores 1689, while the RX 590 scores 1081. The deltaPct of -36 means the RX 590 is 36% slower. This is a decisive win for the RX 6800S, and it aligns with the architectural advantages: RDNA 2’s higher clock speeds (2100 MHz boost vs. 1545 MHz) and double the ROPs (64 vs. 32) give it a massive pixel throughput edge (134.4 GPixel/s vs. 49.44 GPixel/s).

Geekbench Metal shows an even wider gap. The RX 6800S scores 85256, compared to the RX 590’s 48407. The deltaPct here is -43.2%, indicating the RX 590 is nearly half as fast. Metal is Apple’s graphics API, and this test often favors newer architectures with better geometry processing and memory compression. The RX 6800S’s higher FP32 throughput (8.602 TFLOPS vs. 7.119 TFLOPS) and faster memory (16 Gbps effective vs. 8 Gbps) contribute to this lead.

The RX 590’s only statistical advantage appears in its average benchmark score of 24744, which edges out the RX 6800S’s 24063. This is an artifact of the benchmark selection: the RX 590 has only two recorded tests (both of which it loses), while the RX 6800S has eleven, including several Passmark legacy tests where the RX 6800S scores relatively low (e.g., 65 in Passmark DirectX 12, 91 in DirectX 10). These low scores drag down its average but do not reflect real-world modern performance.

The Verdict

Pick the AMD Radeon RX 6800S if you are comparing these as raw performance entities. The data is unambiguous: it wins both head-to-head benchmarks by margins of 36% and 43.2%, has nearly double the FP16 throughput, supports ray tracing, and does all of this at 100W TDP versus the RX 590’s 175W. Its 69th percentile vs. all GPUs places it alongside the NVIDIA GeForce RTX 2080 SUPER (deltaPct -0.4) and above the AMD Radeon RX 9070 (deltaPct 0.8) in its nearest rival group. For any modern gaming or compute workload, the RX 6800S is the superior silicon.

Pick the AMD Radeon RX 590 if you are constrained by physical compatibility. The RX 6800S is an IGP mobile part with no power connectors and portable-device-dependent outputs — it cannot be installed in a desktop. The RX 590 is a 241 mm dual-slot card with a standard PCIe 3.0 x16 interface, a 1x 8-pin connector, and a 450W suggested PSU. It also has a launch MSRP of 279 USD, which you can reference as historical context. Its 70th percentile vs. all GPUs is marginally higher than the RX 6800S’s 69th, and its average benchmark score (24744) slightly exceeds the RX 6800S’s (24063), suggesting that in legacy or mixed workloads, the RX 590’s wider 256-bit memory bus and higher shading unit count still provide some residual value.

The verdict is simple: if you are shopping for a laptop or a system that can accept a mobile-class part, the RX 6800S wins on every measurable benchmark. If you need a desktop card today, the RX 590 is the only one of the two that physically works, and its performance — while generations behind — remains competitive with cards like the Intel Arc A350M (deltaPct 0.4) and the AMD Radeon RX 6600 XT (deltaPct 1.2) in its nearest rival list. Choose accordingly based on your platform, not just the scoreboard.