AMD Radeon RX 580 vs NVIDIA GeForce GTX 580 Comparison

Head-to-Head to Head-to-Head, Head-to-Head-to-Head, Head-to-Head-to-Head-to-Head-Head-to-Head-to-Head-to-Head-Head-to-Head-Head-Head-Head-to-Head-Head = 0, Head-Head = 1. The database records a single direct comparison: Geekbench OpenCL. AMD Radeon RX 580 misses NVIDIA GeForce GTX 580, 37453 vs 15283. That is a 59.2% deficit for the older NVIDIA card. In absolute terms, the AMD card more than doubles the NVIDIA score. The delta percent of -59.2 indicates the NVIDIA card is the lower performer in this test. No other head-to-head tests are available in the recorded data, so this one result defines the entire direct comparison. The margin is decisive, not marginal. A score of 37453 places the RX 580 far outside the GTX 580's reach in compute-oriented OpenCL workloads. The GTX 580's best recorded score is 15283, which is nearly identical to its nearest rival, the RTX 2060 at 15290, a 0% delta. The RX 580, by contrast, sits at 37453 in the same test, which is why the win count is 0 for A and 1 for B. The benchmark data shows a one-sided matchup in raw compute throughput. The GTX 580's OpenCL score of 15283 is also close to the AMD Radeon 680M at 15270, only 0.1% behind, and the RTX 3050 OEM at 15199, 0.6% ahead. The RX 580's OpenCL score of 37453 has no direct rival listed at that level, but its overall average score of 12928 places it near the RTX 3050 Ti Mobile at 12940, a -0.1% delta, and the GTX 1660 SUPER at 12986, a -0.4% delta. The head-to-head result is the only place where the two cards meet in the same test, and it is a clear sweep for AMD.

Where Each One Wins

The AMD Radeon RX 580 wins the only shared benchmark, Geekbench OpenCL, with a score of 37453 against the NVIDIA GeForce GTX 580's 15283. This indicates the RX 580 is the stronger choice for compute-heavy tasks that leverage OpenCL, such as general-purpose GPU workloads or certain rendering and simulation applications. The GTX 580, on the other hand, has no recorded wins in any head-to-head test, so the data does not support any scenario where it outperforms the RX 580 in a direct comparison. However, the GTX 580 does have a foothold in its own benchmark pool: its Geekbench OpenCL score of 15283 places it at the 58th percentile among all GPUs, and it closely matches the RTX 2060 (15290, 0% delta) and the Radeon 680M (15270, 0.1% delta). That means the GTX 580 is competitive with much newer entry-level and mid-range cards in OpenCL, even though it loses badly to the RX 580. The RX 580's percentile is 53, slightly lower than the GTX 580's 58, but its average benchmark score across all tests is 12928, which is lower than its OpenCL score alone, indicating that its performance varies widely by workload. For DirectX 10, 11, and 12, the RX 580 records PassMark scores of 46, 60, and 43 respectively, while for DirectX 9 it scores 124. The GTX 580 has no recorded PassMark results, so the RX 580 is the only card with data for those older APIs. The RX 580 also has Geekbench Metal and Vulkan scores of 45235 and 45173, respectively, which the GTX 580 cannot match because it lacks those API entries. In practical terms, the RX 580 wins on versatility across multiple modern APIs, while the GTX 580 only shows up in OpenCL, where it is still far behind. The data suggests the RX 580 is the better all-rounder, especially for anyone using Vulkan, Metal, or DirectX 12, all of which the GTX 580 does not have recorded scores for. The GTX 580's only claim is that its single OpenCL score is respectable for its era, but it does not hold up against the RX 580 in any direct test.

Architecture Differences

The two cards come from different architectural eras. The NVIDIA GeForce GTX 580 uses the GF110 chip, based on the Fermi 2.0 architecture, part of the GeForce 500 generation. It is built on a 40 nm process at TSMC, with 3,000 million transistors on a 520 mm² die, giving a transistor density of 5.8 million per mm². The AMD Radeon RX 580 uses the Polaris 20 chip, based on GCN 4.0, part of the Polaris (RX 500) generation. It is built on a 14 nm process at GlobalFoundries, with 5,700 million transistors on a 232 mm² die, giving a transistor density of 24.6 million per mm². The RX 580 has nearly double the transistor count on less than half the die area, a direct result of the newer 14 nm node. The GTX 580 has 512 shading units, 64 texture mapping units, and 48 ROPs. The RX 580 has 2304 shading units, 144 TMUs, and 32 ROPs. The RX 580 has far more shading units and TMUs, but fewer ROPs. Clock speeds also differ significantly: the GTX 580 has no base or boost clock listed, only a memory clock of 1002 MHz (4 Gbps effective), while the RX 580 has a base clock of 1257 MHz and a boost clock of 1340 MHz, with a memory clock of 2000 MHz (8 Gbps effective). The RX 580's memory configuration is 8 GB GDDR5 on a 256-bit bus, yielding 256.0 GB/s bandwidth. The GTX 580 has 1536 MB GDDR5 on a 384-bit bus, yielding 192.4 GB/s bandwidth. Despite the narrower bus, the RX 580 achieves higher bandwidth due to faster memory clocks. Pixel and texture rates tell a similar story: the GTX 580 outputs 24.70 GPixel/s and 49.41 GTexel/s, while the RX 580 outputs 42.88 GPixel/s and 193.0 GTexel/s. The RX 580 also has a much higher FP32 throughput at 6.175 TFLOPS, compared to the GTX 580's 1.581 TFLOPS. The RX 580 even lists FP16 at 6.175 TFLOPS (1:1), while the GTX 580 has no FP16 entry. Power draw is lower on the RX 580 at 185 W, versus the GTX 580's 244 W, and the RX 580 uses a single 8-pin power connector instead of the GTX 580's 1x 6-pin plus 1x 8-pin. The RX 580 also has a lower suggested PSU of 450 W, compared to 550 W for the GTX 580. Bus interface differs: the GTX 580 uses PCIe 2.0 x16, while the RX 580 uses PCIe 3.0 x16. Display outputs are also different: the GTX 580 offers 2x DVI and 1x mini-HDMI 1.3a, while the RX 580 offers 1x HDMI 2.0b and 3x DisplayPort 1.4a. API support is a major gap: the GTX 580 supports DirectX 12 (11_0) and OpenGL 4.6, but has no Vulkan entry, while the RX 580 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. The RX 580 is physically shorter at 241 mm (9.5 inches) versus the GTX 580's 267 mm (10.5 inches), and it is also lighter on power requirements. Both are dual-slot cards. The GTX 580 was released on 2010-11-08, while the RX 580 came later on 2017-04-17. Both are end-of-life products, with the GTX 580 preceded by GeForce 400 and succeeded by GeForce 600, and the RX 580 preceded by Arctic Islands and succeeded by Vega.

The Verdict

The data points to one clear choice for most users: the AMD Radeon RX 580. It wins the only direct head-to-head test, Geekbench OpenCL, by a margin of 59.2% over the NVIDIA GeForce GTX 580. The RX 580 also offers higher memory capacity (8 GB vs 1536 MB), higher bandwidth (256.0 GB/s vs 192.4 GB/s), more shading units (2304 vs 512), and a much higher FP32 throughput (6.175 TFLOPS vs 1.581 TFLOPS). It consumes less power (185 W vs 244 W), requires a smaller PSU (450 W vs 550 W), and supports newer APIs including Vulkan 1.3 and DirectX 12 (12_0), whereas the GTX 580 only reaches DirectX 12 (11_0) and lacks Vulkan. The RX 580 also has recorded scores in Metal, Vulkan, and multiple DirectX versions, giving it a broader software ecosystem. The GTX 580's only advantage in the recorded data is its higher percentile ranking (58th vs 53rd) and its OpenCL score of 15283, which is close to modern cards like the RTX 2060, but that score is still 59.2% lower than the RX 580's OpenCL score. The GTX 580's launch MSRP was 499 USD, while the RX 580's launch MSRP was 229 USD. For anyone choosing between these two based on the database, the RX 580 is the superior performer in every measured category. The GTX 580 should only be considered if a user specifically needs a Fermi-era card for legacy compatibility, but the data provides no performance reason to pick it. The RX 580 is faster, cooler, more efficient, and more feature-rich. The verdict is unambiguous: the AMD Radeon RX 580 is the better GPU, and the NVIDIA GeForce GTX 580 is outclassed in this matchup.

FAQ

Q: Which GPU wins the only head-to-head benchmark recorded?

A: The AMD Radeon RX 580 wins Geekbench OpenCL with a score of 37453, against the NVIDIA GeForce GTX 580's 15283, a delta of -59.2% for the NVIDIA card.

Q: How does the GTX 580's OpenCL score compare to its nearest rivals?

A: The GTX 580's OpenCL score of 15283 is 0% off the NVIDIA GeForce RTX 2060's score of 15290, 0.1% ahead of the AMD Radeon 680M at 15270, and 0.6% ahead of the RTX 3050 OEM at 15199.

Q: What is the memory capacity difference between the two cards?

A: The GTX 580 has 1536 MB of GDDR5, while the RX 580 has 8 GB of GDDR5, with bandwidths of 192.4 GB/s and 256.0 GB/s respectively.

Q: Does the GTX 580 support Vulkan?

A: No, the GTX 580 has no Vulkan entry in its API list, while the RX 580 supports Vulkan 1.3.

Q: What are the power consumption figures for each card?

A: The GTX 580 has a TDP of 244 W and a suggested PSU of 550 W, while the RX 580 has a TDP of 185 W and a suggested PSU of 450 W.

Q: What is the transistor density difference?

A: The GTX 580 has a transistor density of 5.8 million per mm², while the RX 580 has a density of 24.6 million per mm², on a 40 nm and 14 nm process respectively.

Specification Differences

| Field | NVIDIA GeForce GTX 580 | AMD Radeon RX 580 |

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

| Chip | GF110 | Polaris 20 |

| Architecture | Fermi 2.0 | GCN 4.0 |

| Process Node | 40 nm | 14 nm |

| Foundry | TSMC | GlobalFoundries |

| Transistors | 3,000 million | 5,700 million |

| Die Size | 520 mm² | 232 mm² |

| Transistor Density | 5.8M / mm² | 24.6M / mm² |

| Base Clock | Not listed | 1257 MHz |

| Boost Clock | Not listed | 1340 MHz |

| Memory Clock | 1002 MHz (4 Gbps effective) | 2000 MHz (8 Gbps effective) |

| Memory Size | 1536 MB | 8 GB |

| Memory Bus | 384 bit | 256 bit |

| Memory Bandwidth | 192.4 GB/s | 256.0 GB/s |

| Shading Units | 512 | 2304 |

| TMUs | 64 | 144 |

| ROPs | 48 | 32 |

| Pixel Rate | 24.70 GPixel/s | 42.88 GPixel/s |

| Texture Rate | 49.41 GTexel/s | 193.0 GTexel/s |

| FP32 | 1.581 TFLOPS | 6.175 TFLOPS |

| FP16 | Not listed | 6.175 TFLOPS (1:1) |

| TDP | 244 W | 185 W |

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

| Suggested PSU | 550 W | 450 W |

| Bus Interface | PCIe 2.0 x16 | PCIe 3.0 x16 |

| Display Outputs | 2x DVI, 1x mini-HDMI 1.3a | 1x HDMI 2.0b, 3x DisplayPort 1.4a |

| DirectX | 12 (11_0) | 12 (12_0) |

| Vulkan | Not listed | 1.3 |

| Length | 267 mm (10.5 inches) | 241 mm (9.5 inches) |

| Release Date | 2010-11-08 | 2017-04-17 |

| Predecessor | GeForce 400 | Arctic Islands |

| Successor |