NVIDIA GeForce GTX 580 vs NVIDIA GeForce GTX 965M Comparison

The NVIDIA GeForce GTX 580 and the NVIDIA GeForce GTX 965M represent two distinct eras of GPU design, separated by four years of architectural evolution. The data shows a clear, albeit narrow, victory for the older desktop card in raw compute benchmarks, while the newer mobile chip counters with superior feature support and efficiency metrics. This analysis examines their head-to-head performance, architectural philosophies, and the specific trade-offs each brings to the table.

Head-to-Head Benchmarks

The only direct benchmark comparison available is the Geekbench OpenCL test, and it yields a decisive result for the GTX 580. The desktop card scores 15,283 points, while the GTX 965M trails with 14,509 points. This translates to a 5.3% advantage for the GTX 580, a meaningful gap that places the two GPUs in adjacent performance tiers. The GTX 580’s win is consistent with its position in the broader database, where it holds a 58th percentile ranking among all GPUs, compared to the GTX 965M’s 56th percentile.

Contextualizing these scores against their nearest rivals clarifies their standing. The GTX 580’s score of 15,283 places it within 0.1% of the NVIDIA GeForce RTX 2060 (15,290) and 0.6% of the RTX 3050 OEM (15,199). This is remarkable for a GPU released in 2010; it effectively matches modern mainstream discrete cards in OpenCL compute throughput. The GTX 965M, with its 14,509 points, is similarly competitive against its peers, sitting just 0.1% ahead of the AMD Radeon RX Vega 11 (14,385) and 0.2% ahead of the NVIDIA GeForce GTX TITAN (14,373). The delta between the two cards is larger than the gaps to their respective rivals, underscoring that while both are mid-pack performers, the GTX 580 holds a slight but consistent edge.

It is importantly the GTX 965M also has a Vulkan benchmark score of 14,299, which is not available for the GTX 580. This indicates the mobile GPU’s modern API support, but the lack of a comparable score prevents a direct cross-API comparison. In the single shared test, the GTX 580 wins, and the data does not support any other conclusion.

Architecture Differences

The architectural divide between these two GPUs is substantial, reflecting a generational shift in NVIDIA’s design philosophy. The GTX 580 is built on the GF110 chip, using the Fermi 2.0 architecture fabricated on a 40 nm process at TSMC. In contrast, the GTX 965M uses the GM204 chip with the Maxwell 2.0 architecture on a 28 nm node, also from TSMC. This process shrink allowed NVIDIA to pack significantly more transistors: the GTX 965M contains 5,200 million transistors compared to the GTX 580’s 3,000 million. However, the GTX 580’s die is physically larger at 520 mm² versus 398 mm², yielding a transistor density of 5.8 million per square millimeter for the Fermi chip versus 13.1 million per square millimeter for Maxwell.

Core configuration tells a story of divergent priorities. The GTX 580 employs 512 shading units, 64 texture mapping units (TMUs), and 48 raster operation units (ROPs). The GTX 965M doubles the shading units to 1,024 but cuts ROPs down to 32, while keeping TMUs at 64. This explains the compute gap: despite having twice the shaders, the GTX 965M’s lower ROP count and narrower memory bus limit its fill-rate performance. The GTX 580 achieves a pixel rate of 24.70 GPixel/s and a texture rate of 49.41 GTexel/s, while the GTX 965M posts 30.40 GPixel/s and 60.80 GTexel/s respectively. Interestingly, the Maxwell chip wins on both fill-rate metrics, yet still loses the OpenCL test, indicating that the Fermi architecture’s raw FP32 throughput—1.581 TFLOPS versus 1.946 TFLOPS for Maxwell—is not the sole determinant of compute performance.

Memory subsystems differ radically. The GTX 580 uses 1,536 MB of GDDR5 on a 384-bit bus, delivering 192.4 GB/s of bandwidth. The GTX 965M has 2 GB of GDDR5 but is constrained to a 128-bit interface, resulting in just 80.19 GB/s. This 58% reduction in bandwidth is a critical weakness for the mobile card. Clock speeds also favor the newer chip: the GTX 965M runs at a 924 MHz base and 950 MHz boost, whereas the GTX 580’s memory clock is 1,002 MHz (4 Gbps effective), but its core clocks are not listed in the data. The GTX 965M’s memory operates at 1,253 MHz (5 Gbps effective).

FAQ

Q: Which GPU is faster in OpenCL benchmarks?

A: The NVIDIA GeForce GTX 580 scores 15,283 in Geekbench OpenCL, which is 5.3% higher than the GTX 965M’s 14,509.

Q: How does the GTX 580 compare to modern GPUs?

A: Its OpenCL score of 15,283 is within 0.1% of the RTX 2060 (15,290) and 0.6% of the RTX 3050 OEM (15,199), indicating it remains competitive with much newer discrete cards.

Q: What are the key memory differences?

A: The GTX 580 has 1,536 MB on a 384-bit bus with 192.4 GB/s bandwidth, while the GTX 965M offers 2 GB on a 128-bit bus with 80.19 GB/s bandwidth.

Q: Does the GTX 965M support newer graphics APIs?

A: Yes, the GTX 965M supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 580 only reaches DirectX 12 (11_0) and has no Vulkan support listed.

Q: Which card has higher shading unit count?

A: The GTX 965M has 1,024 shading units, exactly double the GTX 580’s 512.

Q: What is the transistor density difference?

A: The GTX 965M’s 28 nm process yields 13.1M transistors per mm², compared to the GTX 580’s 5.8M per mm² on 40 nm.

The Verdict

The data presents a nuanced picture. For raw compute performance in OpenCL, the GTX 580 is the clear winner, besting the GTX 965M by 5.3% and achieving a higher percentile ranking (58th versus 56th). Its massive 384-bit memory bus and 48 ROPs provide substantial bandwidth advantages that the newer card cannot overcome, despite its higher shader count and clock speeds. Users prioritizing compute tasks or legacy DirectX 11 workloads would find the GTX 580 data more compelling.

However, the GTX 965M counters with architectural modernity. It supports DirectX 12 (12_1) and Vulkan 1.4, whereas the GTX 580 is limited to DirectX 12 (11_0) with no Vulkan support. This makes the GTX 965M the more future-proof option for applications leveraging these newer APIs. Its higher FP32 throughput (1.946 TFLOPS) and superior fill rates (30.40 GPixel/s, 60.80 GTexel/s) also indicate strengths in certain rendering scenarios. The GTX 965M’s mobile form factor (MXM Module) and lack of external power connectors also suggest a fundamentally different use case, targeting laptops where the GTX 580’s 244 W TDP and dual-slot design are impractical.

The choice is clear: the GTX 580 wins on benchmark scores and memory bandwidth, while the GTX 965M wins on API support and architectural efficiency. Neither card is a universal victor, and the selection depends entirely on whether the workload favors raw OpenCL compute or modern feature support.

Specification Differences

| Specification | NVIDIA GeForce GTX 580 | NVIDIA GeForce GTX 965M |

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

| Chip | GF110 | GM204 |

| Architecture | Fermi 2.0 | Maxwell 2.0 |

| Process Node | 40 nm | 28 nm |

| Transistors | 3,000 million | 5,200 million |

| Die Size | 520 mm² | 398 mm² |

| Transistor Density | 5.8M / mm² | 13.1M / mm² |

| Core Clock (Base) | Not listed | 924 MHz |

| Core Clock (Boost) | Not listed | 950 MHz |

| Memory Clock | 1002 MHz (4 Gbps effective) | 1253 MHz (5 Gbps effective) |

| Memory Size | 1536 MB | 2 GB |

| Memory Bus Width | 384 bit | 128 bit |

| Memory Bandwidth | 192.4 GB/s | 80.19 GB/s |

| Shading Units | 512 | 1024 |

| ROPs | 48 | 32 |

| Pixel Rate | 24.70 GPixel/s | 30.40 GPixel/s |

| Texture Rate | 49.41 GTexel/s | 60.80 GTexel/s |

| FP32 Performance | 1.581 TFLOPS | 1.946 TFLOPS |

| TDP | 244 W | Not listed |

| Slot Width | Dual-slot | MXM Module |

| Power Connectors | 1x 6-pin + 1x 8-pin | None |

| Bus Interface | PCIe 2.0 x16 | MXM-B (3.0) |

| Display Outputs | 2x DVI, 1x mini-HDMI 1.3a | Portable Device Dependent |

| DirectX Support | 12 (11_0) | 12 (12_1) |

| Vulkan Support | Not listed | 1.4 |

| Release Date | 2010-11-08 | 2015-01-08 |

| Launch MSRP | 499 USD | Not listed |