AMD Radeon RX 590 GME vs AMD Radeon RX 6800 Comparison

The AMD Radeon RX 590 GME and AMD Radeon RX 6800 represent two distinct generations of AMD graphics technology, separated by a fundamental architectural shift. The RX 590 GME is built on the older Polaris 20 chip using GCN 4.0 architecture on a 14 nm process, while the RX 6800 utilizes the Navi 21 chip with RDNA 2.0 architecture on a 7 nm process. Benchmark data shows a split decision: the RX 590 GME wins in one of three head-to-head tests, while the RX 6800 takes the other two, including a dominant victory in modern DirectX 12 workloads.

Where Each One Wins

The RX 590 GME demonstrates its strength in OpenCL compute workloads. In the Geekbench OpenCL test, it scores 36,294 against the RX 6800's 24,508, a 48.1% advantage. This is a substantial margin that suggests the older GCN architecture retains an edge in certain general-purpose compute tasks. The RX 590 GME also holds a higher average benchmark score of 32,601 compared to the RX 6800's 30,095, despite being the less powerful card by most conventional metrics. This places the RX 590 GME in the 77th percentile of all GPUs, slightly ahead of the RX 6800's 75th percentile.

The RX 6800 wins decisively in the other two head-to-head tests. In 3DMark Steel Nomad DX12, it scores 3,188 versus the RX 590 GME's 1,001, a 68.6% lead. This is a massive gap that reflects the architectural superiority of RDNA 2.0 in modern gaming APIs. The RX 6800 also wins the Geekbench Vulkan test with 115,107 points against 60,508, a 47.4% advantage. Vulkan is another modern graphics API, and the RX 6800's dominance here further confirms that the newer architecture is far better suited to contemporary rendering workloads.

The use-case split is clear from the data. The RX 590 GME wins in OpenCL compute, making it potentially relevant for specific compute-oriented applications that rely on that API. The RX 6800 wins in DirectX 12 and Vulkan, which are the primary APIs for modern PC gaming. For gaming workloads, the RX 6800 is overwhelmingly superior. For compute workloads that favor OpenCL, the RX 590 GME surprisingly outperforms its newer counterpart.

Architecture Differences

The two cards differ at nearly every architectural level. The RX 590 GME uses the Polaris 20 chip with GCN 4.0 architecture, manufactured on a 14 nm process by GlobalFoundries. It contains 5,700 million transistors on a 232 mm² die, giving it a transistor density of 24.6 million per square millimeter. The RX 6800 uses the Navi 21 chip with RDNA 2.0 architecture, manufactured on a 7 nm process by TSMC. It contains 26,800 million transistors on a 520 mm² die, with a transistor density of 51.5 million per square millimeter.

The core configurations reflect the generational leap. The RX 590 GME has 2,304 shading units, 144 texture mapping units, and 32 render output units. The RX 6800 has 3,840 shading units, 240 TMUs, and 96 ROPs. The RX 6800 also includes 60 ray tracing cores, a feature entirely absent from the RX 590 GME. Clock speeds are also significantly different: the RX 590 GME has a base clock of 1257 MHz and a boost clock of 1420 MHz, while the RX 6800 has a base clock of 1700 MHz, a game clock of 1815 MHz, and a boost clock of 2105 MHz.

Memory subsystems differ substantially. The RX 590 GME has 8 GB of GDDR5 memory on a 256-bit bus, delivering 256.0 GB/s of bandwidth at 2000 MHz (8 Gbps effective). The RX 6800 has 16 GB of GDDR6 memory on the same 256-bit bus, but delivers 512.0 GB/s of bandwidth at 2000 MHz (16 Gbps effective). This double bandwidth is critical for high-resolution gaming and texture-heavy workloads. The RX 6800 also supports PCIe 4.0 x16, while the RX 590 GME is limited to PCIe 3.0 x16.

The RX 6800 supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the RX 590 GME supports DirectX 12 (12_0) and Vulkan 1.3. Both support OpenGL 4.6. The RX 6800 has a higher TDP of 250 W with 2x 8-pin power connectors and a suggested 600 W PSU, while the RX 590 GME has a 175 W TDP with a single 8-pin connector and a suggested 450 W PSU.

Head-to-Head Benchmarks

The 3DMark Steel Nomad DX12 test produces the largest performance gap between the two cards. The RX 6800 scores 3,188, which is 68.6% higher than the RX 590 GME's 1,001. This test is a modern DirectX 12 workload, and the RX 6800's RDNA 2.0 architecture with its 60 ray tracing cores and 3,840 shading units delivers a crushing victory. The RX 590 GME's GCN 4.0 architecture, with its 2,304 shading units and no ray tracing support, simply cannot compete in this modern rendering environment.

The Geekbench Vulkan test shows a similar but slightly smaller margin. The RX 6800 scores 115,107 versus the RX 590 GME's 60,508, a 47.4% advantage. Vulkan is another modern low-level API, and the RX 6800's superior geometry processing, texture rate, and pixel rate contribute to this win. The RX 6800 has a texture rate of 505.2 GTexel/s and a pixel rate of 202.1 GPixel/s, compared to the RX 590 GME's 204.5 GTexel/s and 45.44 GPixel/s.

The Geekbench OpenCL test is the one notable victory for the RX 590 GME. It scores 36,294 against the RX 6800's 24,508, a 48.1% advantage. This is counterintuitive given the RX 6800's higher compute throughput in theory. The RX 6800 has an FP32 rating of 16.17 TFLOPS, while the RX 590 GME has 6.543 TFLOPS. Yet in this specific OpenCL workload, the older card wins decisively. The reasons may lie in driver optimizations or workload characteristics that favor GCN's compute scheduling, but the data is clear: the RX 590 GME outperforms the RX 6800 in OpenCL by a wide margin.

The RX 6800 also shows a strong result in the Passmark G3D test with a score of 22,067, though no comparable RX 590 GME score exists in the data for direct comparison. The RX 6800's Passmark GPU Compute score of 10,864 further indicates its compute capabilities in DirectX-based workloads. The RX 590 GME's average benchmark score of 32,601 is 8.3% higher than the RX 6800's 30,095, driven largely by the OpenCL result.

The Verdict

The data supports a clear split recommendation based on workload. For gaming, the RX 6800 is the overwhelming choice. Its 68.6% lead in 3DMark Steel Nomad DX12 and 47.4% lead in Geekbench Vulkan demonstrate that it is vastly superior in modern graphics APIs. The RX 6800 also offers 16 GB of GDDR6 memory with 512.0 GB/s bandwidth, double the capacity and bandwidth of the RX 590 GME. Its 60 ray tracing cores enable hardware-accelerated ray tracing, which the RX 590 GME cannot do at all. The RX 6800's DirectX 12 Ultimate support further future-proofs it for upcoming titles.

For OpenCL compute workloads, the RX 590 GME is the better performer according to the data. Its 48.1% lead in Geekbench OpenCL is substantial and unexpected given the architectural gap. The RX 590 GME also has a higher average benchmark score and sits in a higher percentile of all GPUs. However, this advantage is likely narrow in scope, as the RX 6800's Passmark GPU Compute score of 10,864 indicates strong compute performance in other APIs. The RX 590 GME's lower TDP of 175 W and single 8-pin power connector make it easier to integrate into systems with less robust power supplies.

The RX 6800's nearest rival is the NVIDIA GeForce RTX 3070 Ti, which scores 29,945 on average, just 0.5% behind. The RX 590 GME's nearest rival is the AMD FirePro S9300 X2, which scores 32,540, a 0.2% difference. These comparisons place both cards in competitive positions within their respective performance tiers. The RX 6800 was released on 2020-10-27 with a launch MSRP of 579 USD, while the RX 590 GME was released on 2020-03-08 with no listed launch MSRP.

FAQ

Q: Which card wins in DirectX 12 gaming performance?

A: The RX 6800 wins decisively. In the 3DMark Steel Nomad DX12 test, it scores 3,188 versus the RX 590 GME's 1,001, a 68.6% advantage.

Q: Is the RX 590 GME better at any benchmark?

A: Yes, the RX 590 GME wins the Geekbench OpenCL test with a score of 36,294, which is 48.1% higher than the RX 6800's 24,508.

Q: How do the memory configurations differ?

A: The RX 590 GME has 8 GB of GDDR5 with 256.0 GB/s bandwidth, while the RX 6800 has 16 GB of GDDR6 with 512.0 GB/s bandwidth. Both use a 256-bit bus.

Q: Does the RX 6800 support ray tracing?

A: Yes, the RX 6800 includes 60 ray tracing cores. The RX 590 GME has no ray tracing cores.

Q: What are the power requirements for each card?

A: The RX 590 GME has a 175 W TDP with a single 8-pin power connector and a suggested 450 W PSU. The RX 6800 has a 250 W TDP with dual 8-pin connectors and a suggested 600 W PSU.

Q: How do the average benchmark scores compare?

A: The RX 590 GME has an average benchmark score of 32,601, which is 8.3% higher than the RX 6800's 30,095. The RX 590 GME sits in the 77th percentile of all GPUs, while the RX 6800 sits in the 75th.

Specification Differences

| Specification | AMD Radeon RX 590 GME | AMD Radeon RX 6800 |

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

| Chip | Polaris 20 | Navi 21 |

| Architecture | GCN 4.0 | RDNA 2.0 |

| Process Node | 14 nm | 7 nm |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 5,700 million | 26,800 million |

| Die Size | 232 mm² | 520 mm² |

| Transistor Density | 24.6M / mm² | 51.5M / mm² |

| Base Clock | 1257 MHz | 1700 MHz |

| Boost Clock | 1420 MHz | 2105 MHz |

| Game Clock | Not specified | 1815 MHz |

| Memory Size | 8 GB | 16 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bandwidth | 256.0 GB/s | 512.0 GB/s |

| Memory Speed | 2000 MHz (8 Gbps effective) | 2000 MHz (16 Gbps effective) |

| Shading Units | 2304 | 3840 |

| TMUs | 144 | 240 |

| ROPs | 32 | 96 |

| Ray Tracing Cores | Not specified | 60 |

| Pixel Rate | 45.44 GPixel/s | 202.1 GPixel/s |

| Texture Rate | 204.5 GTexel/s | 505.2 GTexel/s |

| FP32 | 6.543 TFLOPS | 16.17 TFLOPS |

| FP16 | 6.543 TFLOPS (1:1) | 32.33 TFLOPS (2:1) |

| TDP | 175 W | 250 W |

| Power Connectors | 1x 8-pin | 2x 8-pin |

| Suggested PSU | 450 W | 600 W |

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

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

| Vulkan Support | 1.3 | 1.4 |

| Display Outputs | 1x HDMI 2.0b, 3x DisplayPort 1.4a | 1x HDMI 2.1, 2x DisplayPort 1.4a, 1x USB Type-C |

| Dimensions | 241 mm (9.5 inches) | 267 mm (10.5 inches) x 120 mm (4.7 inches) x 40 mm (1.6 inches) |