NVIDIA GeForce GTX 460 v2 vs NVIDIA GeForce GTX 660 Comparison

The NVIDIA GeForce GTX 660 and the NVIDIA GeForce GTX 460 v2 are two end-of-life graphics cards from different generations, and the benchmark data paints a clear picture of generational progress. The most direct comparison, the Geekbench OpenCL test, shows the GTX 660 scoring 11,347 against the GTX 460 v2's 8,743, a 29.8% delta in favor of the newer card. This is a substantial margin that aligns with the architectural leap between the two, and it reflects the GTX 660's overall average benchmark score of 9,022 compared to the GTX 460 v2's 8,743. In terms of relative standing, the GTX 660 sits at the 45th percentile of all GPUs, while the GTX 460 v2 sits just one point lower at the 44th percentile, indicating that while both are older parts, the GTX 660 holds a slight edge in the broader performance hierarchy.

Head-to-Head Benchmarks

The only direct head-to-head benchmark available is Geekbench OpenCL, and it is a decisive win for the NVIDIA GeForce GTX 660. The GTX 660 scores 11,347, while the GTX 460 v2 scores 8,743, resulting in a delta of 29.8%. This is not a marginal victory; it is a significant performance gap that suggests the GTX 660 is roughly a third faster in compute workloads that leverage OpenCL. For context, the GTX 660's average benchmark score of 9,022 is 279 points higher than the GTX 460 v2's 8,743, reinforcing that this single-test advantage is representative of its overall superiority.

Looking at the rivals surrounding each card, the GTX 660's nearest competitor is the NVIDIA TITAN V CEO Edition, which averages 9,037, a mere 0.2% above the GTX 660. The GTX 660 also edges out the AMD Radeon 550X by 1.2% and the AMD Radeon Pro WX 5100 by 1.8%. On the other side, the GTX 460 v2's closest rival is the NVIDIA GeForce RTX 3050 A Mobile with a 0% delta, and it trails the AMD Radeon R9 M265X by 1.2% and the AMD Radeon Pro WX 5100 by 1.4%. Notably, the AMD Radeon Pro WX 5100 appears as a rival for both cards, and the GTX 660 beats it by 1.8%, while the GTX 460 v2 loses to it by 1.4%, a swing of 3.2% that further underscores the GTX 660's stronger position. The data shows no benchmark where the GTX 460 v2 wins, making the head-to-head comparison a one-sided affair in favor of the GTX 660.

Where Each One Wins

Based strictly on the benchmark data, the NVIDIA GeForce GTX 660 wins in every measurable category. Its 29.8% lead in OpenCL performance makes it the clear choice for any application that utilizes GPU compute, such as rendering, video encoding, or scientific simulations. The GTX 660 also benefits from a higher average benchmark score (9,022 vs 8,743), which suggests it maintains its advantage across a broader range of workloads, even those not directly tested in the head-to-head. For a user deciding between these two cards today, the GTX 660 is the only one with a winning record in the provided benchmarks.

The GTX 460 v2, conversely, does not win any benchmark in the data. Its average score of 8,743 places it at the 44th percentile, and its nearest rival, the RTX 3050 A Mobile, matches it exactly with a 0% delta. This means that in any scenario where compute performance is the primary concern, the GTX 460 v2 is strictly inferior to the GTX 660. The only potential advantage for the GTX 460 v2 lies outside the benchmark results: it supports a dual-slot cooler and has a shorter length of 210 mm (8.3 inches) compared to the GTX 660's 241 mm (9.5 inches), which could matter in a compact case, but that is a physical attribute, not a performance win.

Architecture Differences

The architectural divide between these two cards is stark and explains the performance gap. The GTX 660 is built on the Kepler architecture using the GK106 chip, manufactured on a 28 nm process at TSMC. This allows for 2,540 million transistors packed into a 221 mm² die, yielding a transistor density of 11.5 million per mm². In contrast, the GTX 460 v2 uses the Fermi 2.0 architecture with the GF114 chip, built on an older 40 nm process. It has 1,950 million transistors on a much larger 332 mm² die, resulting in a density of just 5.9 million per mm². The newer 28 nm node is a fundamental advantage, enabling higher clock speeds and efficiency.

The compute resources differ significantly. The GTX 660 boasts 960 shading units, 80 texture mapping units (TMUs), and 24 raster operation units (ROPs). The GTX 460 v2 has only 336 shading units, 56 TMUs, and 24 ROPs. This nearly threefold difference in shading units is a primary driver of the GTX 660's superior compute throughput, which is reflected in its 1.981 TFLOPS of FP32 performance versus the GTX 460 v2's 1,046.3 GFLOPS. The texture rate also favors the GTX 660 at 82.56 GTexel/s compared to 43.62 GTexel/s, while the pixel rate is 20.64 GPixel/s versus 10.91 GPixel/s. Memory configurations also differ: the GTX 660 has 2 GB of GDDR5, while the GTX 460 v2 has only 1 GB. Although both use a 192-bit bus, the GTX 660's memory runs at 6 Gbps effective, yielding 144.2 GB/s of bandwidth, far above the GTX 460 v2's 4 Gbps effective and 96.19 GB/s.

Power and connectivity also reflect the generational shift. The GTX 660 has a TDP of 140 W and requires a single 6-pin power connector with a 300 W suggested PSU, while the GTX 460 v2 has a higher TDP of 160 W, needs two 6-pin connectors, and a 450 W suggested PSU. The GTX 660 also supports PCIe 3.0 x16, whereas the GTX 460 v2 is limited to PCIe 2.0 x16. API support is similar for DirectX (12 (11_0)) and OpenGL (4.6), but the GTX 660 adds Vulkan 1.2.175 support, which the GTX 460 v2 lacks entirely.

FAQ

Q: Which card is faster in OpenCL benchmarks?

A: The NVIDIA GeForce GTX 660 is 29.8% faster, scoring 11,347 versus the GTX 460 v2's 8,743.

Q: Does the GTX 460 v2 win any benchmarks?

A: No. The data shows zero benchmark wins for the GTX 460 v2; the GTX 660 wins the only head-to-head test.

Q: What is the memory bandwidth difference?

A: The GTX 660 has 144.2 GB/s of bandwidth, while the GTX 460 v2 has 96.19 GB/s, a difference of 48 GB/s.

Q: Which card has more shading units?

A: The GTX 660 has 960 shading units, compared to 336 on the GTX 460 v2.

Q: Are there any API differences?

A: Yes. The GTX 660 supports Vulkan 1.2.175, while the GTX 460 v2 has no Vulkan support listed. Both support DirectX 12 (11_0) and OpenGL 4.6.

Q: Which card requires a more powerful PSU?

A: The GTX 460 v2 requires a 450 W suggested PSU, while the GTX 660 only needs 300 W.

Specification Differences

| Specification | NVIDIA GeForce GTX 660 | NVIDIA GeForce GTX 460 v2 |

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

| Architecture | Kepler | Fermi 2.0 |

| Process Node | 28 nm | 40 nm |

| Transistors | 2,540 million | 1,950 million |

| Die Size | 221 mm² | 332 mm² |

| Transistor Density | 11.5M / mm² | 5.9M / mm² |

| Base Clock | 980 MHz | N/A |

| Boost Clock | 1032 MHz | N/A |

| Memory Clock | 6 Gbps effective | 4 Gbps effective |

| Memory Size | 2 GB | 1024 MB |

| Memory Bandwidth | 144.2 GB/s | 96.19 GB/s |

| Shading Units | 960 | 336 |

| TMUs | 80 | 56 |

| Pixel Rate | 20.64 GPixel/s | 10.91 GPixel/s |

| Texture Rate | 82.56 GTexel/s | 43.62 GTexel/s |

| FP32 | 1.981 TFLOPS | 1,046.3 GFLOPS |

| TDP | 140 W | 160 W |

| Power Connectors | 1x 6-pin | 2x 6-pin |

| Suggested PSU | 300 W | 450 W |

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

| Display Outputs | 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 | 2x DVI, 1x mini-HDMI 1.3a |

| Vulkan | 1.2.175 | None |

| Length | 241 mm (9.5 inches) | 210 mm (8.3 inches) |

| Release Date | 2012-09-05 | 2011-09-23 |

| Launch MSRP | 229 USD | 199 USD |

The Verdict

The data is unambiguous: the NVIDIA GeForce GTX 660 is the superior card in every benchmarked metric. Its 29.8% lead in OpenCL performance, coupled with a higher average score (9,022 vs 8,743), makes it the only rational choice for anyone prioritizing compute performance. The GTX 660 also offers double the memory (2 GB vs 1 GB), higher bandwidth, and a more efficient 28 nm process, all while consuming less power (140 W vs 160 W) and requiring a smaller PSU. The only area where the GTX 460 v2 could be argued as preferable is physical size, being shorter at 210 mm versus 241 mm, which might fit in tighter cases, but that is a niche concern.

For a user with an older system, the GTX 660 is the clear upgrade path. It delivers a generational leap in architecture, feature set, and raw performance. The GTX 460 v2, while having a lower launch MSRP (199 USD vs 229 USD), offers no performance advantage in the data. If you must choose between these two specific cards, the GTX 660 is the only one that wins benchmarks, and it should be selected without hesitation. The GTX 460 v2 remains a functional card, but it is outclassed in every measurable way by its newer counterpart.