NVIDIA GeForce GTX 570 vs NVIDIA GeForce GTX 960A Comparison

# NVIDIA GeForce GTX 570 vs NVIDIA GeForce GTX 960A

The database places the NVIDIA GeForce GTX 570 and the NVIDIA GeForce GTX 960A in adjacent performance tiers, separated by only a few percentage points in average benchmark scores. The GTX 570, built on the Fermi 2.0 architecture with the GF110 chip, is the older and larger card. The GTX 960A, using the Maxwell architecture and GM107 chip, represents a later generation with a much smaller die and lower power envelope. Their average OpenCL scores are close enough that the choice between them depends on workload specifics and platform constraints.

FAQ - Frequently Asked Questions

Q: What is the average benchmark score difference between the two cards?

A: The recorded data shows the GTX 570 averaging 13,515 points, while the GTX 960A averages 11,998 points. This places the GTX 570 ahead by roughly 12.6% in the geekbench_opencl test.

Q: Which card has the higher transistor density?

A: The GTX 960A is far more dense. It packs 12.6 million transistors per square millimeter, versus 5.8 million per square millimeter for the GTX 570.

Q: Are both cards the same physical size?

A: No. The GTX 570 is a dual-slot card measuring 10.5 inches in length. The GTX 960A is an MXM module, with no length specified in the database.

Q: What is the power consumption difference?

A: The GTX 570 has a 219 W thermal design power and uses two 6-pin power connectors. The GTX 960A has a 75 W TDP and uses no external power connectors.

Q: Which card has the higher memory clock?

A: The GTX 960A memory operates at 1253 MHz, which is higher than the 950 MHz of the GTX 570.

Q: What is the production status of these cards?

A: Both are listed as end-of-life products. The GTX 570 was released in December 2010, and the GTX 960A in March 2015.

Architecture Differences

The two cards belong to different NVIDIA design eras. The GTX 570 uses the Fermi 2.0 architecture on a 40 nm process at TSMC, with a large 520 mm² die hosting 3,000 million transistors. In contrast, the GTX 960A sits on a much smaller 28 nm process at TSMC, with a die size of 148 mm² and 1,870 million transistors. The density advantage is substantial: the Maxwell based GTX 960A reaches 12.6 million transistors per mm², which is more than double the 5.8 million per mm² of the Fermi die.

Feature differences follow from these design choices. The GTX 570 has 480 shading units, 60 texture mapping units, and 40 render output units. The GTX 960A scales these down to 640 shading units, 40 texture mapping units, and 16 ROPs. Both cards ship with a 320-bit memory bus on the GTX 570 versus a 128-bit bus on the GTX 960A. Memory type differs too: the older card uses GDDR5 (1280 MB), while the newer one uses 2 GB of GDDR5.

The compute feature set also differs. The GTX 570 supports DirectX 12 (11_0) and OpenGL 4.6, but no Vulkan API. The GTX 960A supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. The output configuration also differs: the GTX 570 provides two DVI and one mini-HDMI 1.3a output, whereas the GTX 960A outputs are device-dependent portable.

The Verdict

From the raw data, the GTX 570 is the faster card in the OpenCL workload recorded. The benchmark score of 13,515 for the GTX 570 is 12.6% higher than the 11,998 of the GTX 960A. The data shows the GTX 570 wins the direct head-to-head comparison. The GTX 960A, however, is a more efficient, cooler card with a smaller die, a much higher transistor density, and a newer architecture. The choice should be based on platform and power constraints: if the maximum performance in this particular compute benchmark is the priority and a dual-slot card with external power is acceptable, the GTX 570 is the data-backed pick. If the system is a mobile or small form factor, and lower power draw is important, the GTX 960A is the more suitable model.

Specification Differences

The database records distinct specifications for these two cards in several key areas.

| Field | NVIDIA GeForce GTX 570 | NVIDIA GeForce GTX 960A |

|---|---|

| Architecture | Fermi 2.0 | Maxwell |

| Process Node | 40 nm | 28 nm |

| Transistors | 3,000 million | 1,870 million |

| Die Size | 520 mm² | 148 mm² |

| Transistor Density | 5.8M / mm² | 12.6M / mm² |

| Memory Size | 1280 MB | 2 GB |

| Memory Type | GDDR5 | GDDR5 |

| Memory Bus Width | 320 bit | 128 bit |

| Memory Clock | 950 MHz | 1253 MHz |

| Shading Units | 480 | 640 |

| Texture Mapping Units | 40 | 36 |

| ROPs | 40 | 16 |

| Pixel Rate | 21.96 GPixel/s | 18.82 GPixel/s |

| Texture Rate | 43.92 GTexel/s | 47.04 GTexel/s |

| FP32 | 1,405.4 GFLOPS | 1,505 TFLOPS |

| TDP | 219 W | 75 W |

| Power Connectors | 2x 6-pin | None |

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

| DirectX | 12 (11_11) | 12 (11_0) |

| OpenGL | 4.6 | 4.6 |

| Vulkan | Not listed | 1.4 |

Head-to-Head Benchmarks

The geekbench_opencl result is the only direct comparison in the database. The GTX 570 scored 13,515, while the GTX 960A scored 11,998. That is a 12.6% lead for the GTX 570. This is the single recorded head-to-head win for the older Fermi card.

Looking at the nearest rivals, the relative position of the GTX 570 is tight. It is only 0.2% behind the AMD Radeon HD 7770M, 0.2% behind the AMD Radeon RX 9070 XT, and 0.3% ahead of the NVIDIA P106-090. The GTX 960A is 1.3% behind the AMD Radeon RX 6500 XT, 2.7% behind the NVIDIA GTX 1660, and 3.2% behind the AMD Radeon RX 7800 XT.

Where Each One Wins

NVIDIA GeForce GTX 570 wins when:

• The workload is heavily compute-bound and every point in the OpenCL score matters. The 12.6% lead in the head-to-head benchmark is the largest margin in the dataset.
• The system has enough physical space and power delivery for a dual-slot card. The GTX 570 requires two slot widths and two 6-pin power connectors.
• Legacy PCIe 2.0 platform compatibility is acceptable. The card uses a PCIe 2.0 x16 interface.

NVIDIA GeForce GTX 960A wins when:

• Power draw and thermal envelope are the primary constraints. The GTX 960A draws 75 W TDP versus 219 W for the GTX 570, a difference of 144 W.
• The chassis is limited to a compact MXM module. The GTX 960A is an MXM-B (3.0) card, while the GTX 570 is a dual-slot card with no MXM form factor.
• The system requires a more modern API stack. The GTX 960A supports Vulkan 1.4 and DirectX 12 (11_11), while the GTX 570 stops at DirectX 12 (11_11) and OpenGL 4.6 without Vulkan.
• The memory density or bandwidth is more important than raw compute. The GTX 960A has more memory (2 GB vs 1280 MB) and a higher memory clock (1253 MHz vs 950 MHz).