AMD Radeon RX 590 GME vs NVIDIA GeForce RTX 3080 Ti Comparison

The Verdict

The database places the NVIDIA GeForce RTX 3080 Ti and AMD Radeon RX 590 GME in entirely different performance classes. The RTX 3080 Ti wins all three recorded head-to-head benchmarks, with an average benchmark score of 41187 against 32601 for the RX 590 GME. The RTX 3080 Ti sits at the 83rd percentile of all GPUs, while the RX 590 GME sits at the 77th, a gap that understates the actual performance chasm because the modern RTX card is competing at a far higher absolute level.

The RTX 3080 Ti is the clear choice for anyone running DirectX 12 Ultimate workloads, ray tracing, or Vulkan titles. Its 3DMark Steel Nomad DX12 score of 5077 is 407.2% higher than the RX 590 GME's 1001, meaning the AMD card completes the same scene at roughly one-fifth the frame rate. The Geekbench Vulkan result reinforces this: 192697 versus 60508, a 218.5% advantage. The OpenCL gap is even larger at 368.5%, with 170037 versus 36294.

The RX 590 GME is a legacy Polaris part, end-of-life in both product lines. It has no ray tracing cores, no tensor cores, and supports only DirectX 12 (12_0) rather than 12 Ultimate. Its recorded rivals are similarly aged or lower-tier parts: the AMD FirePro S9300 X2 (0.2% ahead), the AMD Radeon RX 7900 GRE (0.4% ahead), the AMD FirePro S10000 (0.7% ahead), and the NVIDIA T600 Mobile (0.8% behind). The RTX 3080 Ti's nearest rivals include the NVIDIA GeForce RTX 5070 (2% behind), the AMD Radeon Pro 5300 (0.8% ahead), and two Tesla M40 variants (1.2% and 1.7% behind). In short, the RTX 3080 Ti belongs to a much faster generation of hardware.

For users who require maximum frame rates in modern APIs, the RTX 3080 Ti is the only defensible selection from this pairing. The RX 590 GME is not competitive in any recorded metric where the two overlap, so its role is limited to older DirectX 11 or DirectX 9 workloads where the newer card's architectural advantages are less decisive. The data does not support any scenario where the RX 590 GME beats the RTX 3080 Ti.

FAQ

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

A: The NVIDIA GeForce RTX 3080 Ti scores 5077, which is 407.2% higher than the AMD Radeon RX 590 GME's 1001. This is the largest recorded head-to-head delta between the two cards.

Q: Does the RX 590 GME support ray tracing?

A: No. The database lists no ray tracing cores for the AMD Radeon RX 590 GME. The RTX 3080 Ti has 80 RT cores. The RTX card also supports DirectX 12 Ultimate (12_2), while the RX 590 GME is limited to DirectX 12 (12_0).

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

A: The RTX 3080 Ti scores 192697 in Geekbench Vulkan, a 218.5% advantage over the RX 590 GME's 60508.

Q: What is the memory configuration difference?

A: The RTX 3080 Ti has 12 GB of GDDR6X on a 384-bit bus with 912.4 GB/s bandwidth. The RX 590 GME has 8 GB of GDDR5 on a 256-bit bus with 256.0 GB/s bandwidth.

Q: Which card has a higher transistor count?

A: The RTX 3080 Ti contains 28,300 million transistors on a 628 mm² die, versus 5,700 million on a 232 mm² die for the RX 590 GME. The RTX card also uses a denser 8 nm Samsung process compared to the 14 nm GlobalFoundries node.

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

A: The RTX 3080 Ti has an average benchmark score of 41187, placing it at the 83rd percentile. The RX 590 GME averages 32601, at the 77th percentile.

Architecture Differences

The RTX 3080 Ti is built on NVIDIA's Ampere architecture using the GA102 chip, fabricated on an 8 nm Samsung process. The RX 590 GME uses AMD's GCN 4.0 architecture with the Polaris 20 chip, on a 14 nm GlobalFoundries node. These are two generations apart in design philosophy: Ampere is a modern GPU focused on compute, ray tracing, and tensor operations, while GCN 4.0 was designed for conventional rasterization.

The transistor scale is stark. The GA102 die contains 28,300 million transistors across 628 mm², with a transistor density of 45.1 million per mm². The Polaris 20 die has 5,700 million transistors over 232 mm², at a density of 24.6 million per mm². The RTX 3080 Ti has roughly five times more transistors, which directly enables its far larger execution resource pool.

The RTX 3080 Ti features 80 RT cores and 320 tensor cores, both absent from the RX 590 GME. The RT cores handle ray tracing workloads, while tensor cores accelerate AI and deep learning tasks such as DLSS. The RX 590 GME has no equivalent hardware, so it cannot offload these workloads from the shader array.

The shading engine differs dramatically. The RTX 3080 Ti has 10,240 shading units, 320 texture mapping units, and 112 ROPs. The RX 590 GME has 2,304 shading units, 144 TMUs, and 32 ROPs. This represents a 4.4x difference in shader count and a 3.5x difference in ROPs. The RTX card's pixel rate of 186.5 GPixel/s versus 45.44 GPixel/s and texture rate of 532.8 GTexel/s versus 204.5 GTexel/s reflect these resource counts.

The memory subsystem is also architecturally distinct. The RTX 3080 Ti uses GDDR6X at 19 Gbps effective, with 12 GB across a 384-bit bus. The RX 590 GME uses GDDR5 at 8 Gbps effective, with 8 GB across a 256-bit bus. The bandwidth difference is 912.4 GB/s versus 256.0 GB/s, a 3.56x advantage for the NVIDIA card.

The API support differs as well. The RTX 3080 Ti supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RX 590 GME supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. The RTX card also uses PCIe 4.0 x16, while the RX card uses PCIe 3.0 x16.

The RTX 3080 Ti is the successor to the GeForce 20 series and predecessor to the GeForce 40 series. The RX 590 GME is part of the Polaris (RX 500) generation, with Arctic Islands as its predecessor and Vega as its successor. Both are listed as end-of-life products.

Specification Differences

The table below lists only the fields where the two cards differ in the recorded data.

| Specification | NVIDIA GeForce RTX 3080 Ti | AMD Radeon RX 590 GME |

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

| Architecture | Ampere | GCN 4.0 |

| Chip | GA102 | Polaris 20 |

| Process node | 8 nm (Samsung) | 14 nm (GlobalFoundries) |

| Transistors | 28,300 million | 5,700 million |

| Die size | 628 mm² | 232 mm² |

| Transistor density | 45.1M / mm² | 24.6M / mm² |

| Base clock | 1365 MHz | 1257 MHz |

| Boost clock | 1665 MHz | 1420 MHz |

| Memory clock | 19 Gbps effective | 8 Gbps effective |

| Memory size | 12 GB | 8 GB |

| Memory type | GDDR6X | GDDR5 |

| Memory bus | 384 bit | 256 bit |

| Memory bandwidth | 912.4 GB/s | 256.0 GB/s |

| Shading units | 10240 | 2304 |

| TMUs | 320 | 144 |

| ROPs | 112 | 32 |

| RT cores | 80 | None |

| Tensor cores | 320 | None |

| Pixel rate | 186.5 GPixel/s | 45.44 GPixel/s |

| Texture rate | 532.8 GTexel/s | 204.5 GTexel/s |

| FP32 | 34.10 TFLOPS | 6.543 TFLOPS |

| FP16 | 34.10 TFLOPS (1:1) | 6.543 TFLOPS (1:1) |

| TDP | 350 W | 175 W |

| Power connectors | 1x 12-pin | 1x 8-pin |

| Suggested PSU | 750 W | 450 W |

| Bus interface | PCIe 4.0 x16 | PCIe 3.0 x16 |

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

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

| Vulkan | 1.4 | 1.3 |

| Length | 285 mm (11.2 inches) | 241 mm (9.5 inches) |

| Height | 112 mm (4.4 inches) | Not recorded |

| Width | 40 mm (1.6 inches) | Not recorded |

| Release date | 2021-05-30 | 2020-03-08 |

| Predecessor | GeForce 20 | Arctic Islands |

| Successor | GeForce 40 | Vega |

| Launch MSRP | 1,199 USD | Not recorded |

Head-to-Head Benchmarks

The recorded head-to-head data contains three benchmarks, and the RTX 3080 Ti wins all three. The largest margin is in 3DMark Steel Nomad DX12, where the RTX 3080 Ti scores 5077 against 1001 for the RX 590 GME. That is a 407.2% delta, meaning the AMD card achieves less than 20% of the NVIDIA card's score. In practical terms, a game that runs at 60 FPS on the RTX 3080 Ti would run at roughly 12 FPS on the RX 590 GME, assuming linear scaling of this workload.

The Geekbench OpenCL benchmark shows a 368.5% advantage for the RTX 3080 Ti, with scores of 170037 versus 36294. This is a compute-oriented test that stresses raw floating-point throughput. The RTX card's 34.10 TFLOPS FP32 rate versus 6.543 TFLOPS for the RX 590 GME is a 5.2x difference, which aligns with the observed OpenCL margin. The RX 590 GME's nearest rival in this metric, the AMD FirePro S9300 X2, is only 0.2% ahead of it in average score, confirming that the Polaris card is near the bottom of its own performance tier.

The Geekbench Vulkan benchmark shows a 218.5% delta, with the RTX 3080 Ti scoring 192697 and the RX 590 GME scoring 60508. Vulkan is a low-level API that benefits from the RTX card's larger shader array and faster memory bandwidth. The RX 590 GME's 256.0 GB/s bandwidth is a significant bottleneck for modern Vulkan workloads, while the RTX 3080 Ti's 912.4 GB/s provides ample headroom.

The remaining benchmarks in the database are only recorded for the RTX 3080 Ti. These include PassMark DirectX 9 (274), DirectX 10 (184), DirectX 11 (223), DirectX 12 (110), G2D (1091), G3D (26896), and GPU Compute (15282). No corresponding scores exist for the RX 590 GME in these tests, so a direct comparison is impossible. However, the three head-to-head results are consistent: the RTX 3080 Ti is between 2.2x and 5.1x faster depending on the workload.

The average benchmark score gap, 41187 versus 32601, is 26.3% in favor of the RTX 3080 Ti, but this understates the head-to-head margins because the average includes tests where the RX 590 GME has no recorded score. The percentile rankings, 83 versus 77, place both cards above the median of all GPUs, but the RTX 3080 Ti sits in the upper quartile while the RX 590 GME is only slightly above the midpoint.

The RTX 3080 Ti's nearest rivals in the database are the NVIDIA GeForce RTX 5070 (2% behind), the AMD Radeon Pro 5300 (0.8% ahead), and two Tesla M40 variants (1.2% and 1.7% behind). The RX 590 GME's nearest rivals are all within 1% of its average score: the AMD FirePro S9300 X2 (0.2% ahead), the AMD Radeon RX 7900 GRE (0.4% ahead), the AMD FirePro S10000 (0.7% ahead), and the NVIDIA T600 Mobile (0.8% behind). These rival clusters confirm that the two cards occupy entirely different performance strata.