AMD Radeon RX 590 GME vs AMD Radeon RX 6700 XT Comparison

Where Each One Wins

The data splits these two cards into clearly different performance tiers. The AMD Radeon RX 6700 XT wins every head-to-head benchmark recorded in the database, taking all three contests outright. The AMD Radeon RX 590 GME does not win a single comparative test. That is a decisive sweep, but the scale of the victory varies by workload type.

For raw graphics throughput, the RX 6700 XT dominates. In the 3DMark Steel Nomad DX12 test, the RX 6700 XT scores 2435 against 1001 for the RX 590 GME, a 58.9% advantage. This is a modern DirectX 12 workload that stresses the full graphics pipeline, and the newer RDNA 2 architecture handles it far more efficiently. If the task is high-detail gaming at demanding resolutions, the RX 6700 XT is the only serious option here.

For compute-oriented tasks, the gap narrows but remains substantial. In Geekbench OpenCL, the RX 6700 XT scores 44152 versus 36294 for the RX 590 GME, a 17.8% lead. This is a more balanced workload that includes both graphics and general-purpose compute. The RX 590 GME is not embarrassed here, but it still trails clearly. In Geekbench Vulkan, the RX 6700 XT scores 93500 against 60508 for the RX 590 GME, a 35.3% margin. Vulkan is a low-level API that benefits from the newer architecture's scheduling and memory management.

The practical takeaway is straightforward: if the use case is modern gaming with DirectX 12 or Vulkan, the RX 6700 XT is in a different league. If the use case is general compute or legacy workloads, the RX 590 GME is closer, but still behind. The RX 590 GME's only saving grace is its lower power draw and smaller physical footprint, but in raw performance, it loses every recorded contest.

Architecture Differences

The two cards come from different architectural eras entirely. The RX 590 GME uses the Polaris 20 chip, built on the GCN 4.0 architecture, manufactured on a 14 nm process at GlobalFoundries. The RX 6700 XT uses the Navi 22 chip, built on RDNA 2.0, manufactured on a 7 nm process at TSMC. This process shrink is one of the largest generational leaps in the comparison, enabling higher clocks and better efficiency.

The transistor counts tell the story of the architectural shift. The RX 590 GME packs 5,700 million transistors on a 232 mm² die, for a density of 24.6 million transistors per square millimeter. The RX 6700 XT packs 17,200 million transistors on a 335 mm² die, for a density of 51.3 million per square millimeter. The newer chip has three times the transistors on a die that is only 44% larger, which explains why it can offer substantially more compute resources.

Clock speeds also differ massively. The RX 590 GME runs a base clock of 1257 MHz and a boost clock of 1420 MHz. The RX 6700 XT runs a base of 2321 MHz, a game clock of 2424 MHz, and a boost of 2581 MHz. The boost clock alone is nearly double that of the older card. This is not just a small overclock; it is a fundamental change in how the architecture achieves performance.

Feature support also diverges. The RX 6700 XT includes 40 ray tracing cores, which the RX 590 GME lacks entirely. The RX 6700 XT supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the RX 590 GME only supports DirectX 12 (12_0) and Vulkan 1.3. The newer card also supports half-precision FP16 at a 2:1 ratio, doubling its FP16 throughput to 26.43 TFLOPS, whereas the RX 590 GME runs FP16 at 1:1 with its FP32 rate of 6.543 TFLOPS.

The memory subsystems are also architecturally distinct. The RX 590 GME uses GDDR5 on a 256-bit bus with 256.0 GB/s bandwidth. The RX 6700 XT uses GDDR6 on a 192-bit bus with 384.0 GB/s bandwidth. Despite a narrower bus, the newer memory type provides 50% more bandwidth. The RX 6700 XT also has a larger frame buffer at 12 GB versus 8 GB.

Head-to-Head Benchmarks

The three recorded head-to-head benchmarks all favor the RX 6700 XT, but the margins tell different stories about each workload.

The largest gap appears in 3DMark Steel Nomad DX12. The RX 590 GME scores 1001, while the RX 6700 XT scores 2435. That is a 58.9% deficit for the older card. This test is particularly punishing for the RX 590 GME because it uses modern DirectX 12 features that the GCN 4.0 architecture handles poorly. The RX 6700 XT's ray tracing cores and newer shader pipeline give it a massive advantage in this synthetic but representative workload.

Geekbench Vulkan shows a 35.3% gap. The RX 590 GME scores 60508, and the RX 6700 XT scores 93500. Vulkan is a lower-level API that can sometimes help older architectures by reducing driver overhead, but here the raw compute and memory bandwidth advantages of the RX 6700 XT still shine through. The gap is smaller than in Steel Nomad, suggesting that the RX 590 GME's architecture is not completely obsolete for Vulkan compute, but it is still firmly behind.

Geekbench OpenCL shows the smallest gap, at 17.8%. The RX 590 GME scores 36294, and the RX 6700 XT scores 44152. OpenCL is a general-purpose compute workload that does not rely heavily on graphics-specific features. Here, the RX 590 GME's 2304 shading units and 144 texture mapping units are closer to the RX 6700 XT's 2560 shading units and 160 TMUs. The newer card still wins, but the margin is less than a third of the Steel Nomad gap. This suggests that for compute-heavy tasks that do not use ray tracing or modern DX12 features, the RX 590 GME is a more competitive, though still inferior, option.

The aggregate benchmark scores reflect this hierarchy. The RX 590 GME has an average benchmark score of 32601, while the RX 6700 XT has an average of 27425. However, this aggregate is skewed by the fact that the RX 6700 XT has many more recorded benchmarks, including several low-scoring Passmark tests. The head-to-head data is the more reliable indicator, and it shows the RX 6700 XT winning every contest by at least 17.8%.

Specification Differences

The following specifications differ between the two cards, based on the recorded data:

| Specification | AMD Radeon RX 590 GME | AMD Radeon RX 6700 XT |

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

| Chip | Polaris 20 | Navi 22 |

| Architecture | GCN 4.0 | RDNA 2.0 |

| Generation | Polaris (RX 500) | Navi II (RX 6000) |

| Process Node | 14 nm | 7 nm |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 5,700 million | 17,200 million |

| Die Size | 232 mm² | 335 mm² |

| Transistor Density | 24.6M / mm² | 51.3M / mm² |

| Base Clock | 1257 MHz | 2321 MHz |

| Boost Clock | 1420 MHz | 2581 MHz |

| Game Clock | None recorded | 2424 MHz |

| Memory Type | GDDR5 | GDDR6 |

| Memory Size | 8 GB | 12 GB |

| Memory Bus Width | 256 bit | 192 bit |

| Memory Bandwidth | 256.0 GB/s | 384.0 GB/s |

| Effective Memory Speed | 8 Gbps | 16 Gbps |

| Shading Units | 2304 | 2560 |

| Texture Mapping Units | 144 | 160 |

| Raster Operation Units | 32 | 64 |

| Ray Tracing Cores | None | 40 |

| Pixel Rate | 45.44 GPixel/s | 165.2 GPixel/s |

| Texture Rate | 204.5 GTexel/s | 413.0 GTexel/s |

| FP32 Performance | 6.543 TFLOPS | 13.21 TFLOPS |

| FP16 Performance | 6.543 TFLOPS (1:1) | 26.43 TFLOPS (2:1) |

| TDP | 175 W | 230 W |

| Power Connectors | 1x 8-pin | 1x 6-pin + 1x 8-pin |

| Suggested PSU | 450 W | 550 W |

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

| Display Outputs | 1x HDMI 2.0b, 3x DisplayPort 1.4a | 1x HDMI 2.1, 3x DisplayPort 1.4a |

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

| Vulkan Support | 1.3 | 1.4 |

| Dimensions | 241 mm (9.5 inches) long | 267 mm (10.5 inches) long, 110 mm (4.3 inches) high, 40 mm (1.6 inches) wide |

| Release Date | 2020-03-08 | 2021-03-02 |

| Predecessor | Arctic Islands | Navi |

| Successor | Vega | Navi III |

| Launch MSRP | None recorded | 479 USD |

FAQ

Q: Which card is faster in 3DMark Steel Nomad DX12?

A: The AMD Radeon RX 6700 XT scores 2435, which is 58.9% higher than the RX 590 GME's score of 1001.

Q: How much memory bandwidth does each card have?

A: The RX 590 GME has 256.0 GB/s of bandwidth from GDDR5 on a 256-bit bus. The RX 6700 XT has 384.0 GB/s from GDDR6 on a 192-bit bus.

Q: Does the RX 590 GME support ray tracing?

A: No. The RX 590 GME has no ray tracing cores. The RX 6700 XT includes 40 ray tracing cores.

Q: What is the difference in FP32 compute performance?

A: The RX 6700 XT delivers 13.21 TFLOPS of FP32, exactly double the 6.543 TFLOPS of the RX 590 GME.

Q: Which card has a higher boost clock?

A: The RX 6700 XT has a boost clock of 2581 MHz, compared to 1420 MHz for the RX 590 GME.

Q: What is the average benchmark score for each card?

A: The RX 590 GME has an average benchmark score of 32601, while the RX 6700 XT has an average of 27425. However, in direct head-to-head tests, the RX 6700 XT wins all three recorded benchmarks.