NVIDIA GeForce GTX 460 SE vs NVIDIA GeForce GTX 680M Comparison

# NVIDIA GeForce GTX 680M vs NVIDIA GeForce GTX 460 SE

The NVIDIA GeForce GTX 680M is a Kepler-based mobile GPU from 2012, while the NVIDIA GeForce GTX 460 SE is a Fermi-generation desktop card released in 2010. The data shows a clear performance hierarchy: the GTX 680M averages 7,023 across benchmark scores versus 6,389 for the GTX 460 SE, a difference that places the newer mobile part in the 39th percentile of all GPUs compared to the 37th percentile for the older desktop card. The head-to-head benchmark results reinforce this gap, with the GTX 680M winning the only shared test by a substantial 44.5% margin.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 680M scores 7,023 on average, while the NVIDIA GeForce GTX 460 SE scores 6,389. That is a difference of 634 points in favor of the GTX 680M.

Q: What is the only benchmark where both GPUs have results?

A: The only overlapping test is Geekbench OpenCL. The GTX 680M scores 9,230, and the GTX 460 SE scores 6,389. This makes the GTX 680M the winner with a 44.5% delta.

Q: How do these GPUs compare to their nearest rivals?

A: The GTX 680M sits within 1.9% of the NVIDIA GeForce GTX 970 (7,157) and is 1.1% ahead of the GeForce GTX 675M (6,946). The GTX 460 SE is tied exactly with the GeForce GTX 580M (6,389) at 0% delta and is 1% ahead of the AMD Radeon R7 M350 (6,327).

Q: Which GPU has a higher pixel rate?

A: The GTX 680M delivers 21.22 GPixel/s, whereas the GTX 460 SE delivers 7.800 GPixel/s. The GTX 680M achieves nearly three times the pixel throughput.

Q: Do both GPUs support DirectX 12?

A: Yes, both report DirectX 12 (11_0) support. They also both support OpenGL 4.6, but the GTX 680M adds Vulkan 1.2.175 support while the GTX 460 SE has no Vulkan entry.

Q: What is the transistor density difference?

A: The GTX 680M packs 12.0M transistors per mm² on a 28 nm process, while the GTX 460 SE has 5.9M per mm² on 40 nm. The GTX 680M more than doubles the density.

The Verdict

The benchmark data points to a single clear conclusion: the NVIDIA GeForce GTX 680M outperforms the NVIDIA GeForce GTX 460 SE decisively. In the only direct comparison available, the GTX 680M wins by 44.5% in Geekbench OpenCL, and its average score of 7,023 versus 6,389 confirms this is not a fluke. Users seeking the higher-performing part should choose the GTX 680M without hesitation.

However, context matters. The GTX 460 SE is a desktop card with a 150 W TDP, dual-slot cooler, and two 6-pin power connectors, making it a fixed installation. The GTX 680M is an MXM module with a 100 W TDP and no power connectors, designed for laptops. If the use case demands a portable solution, the GTX 680M is the only option that fits, and it happens to be faster. If the use case is a stationary desktop with an existing PCIe 2.0 x16 slot, the GTX 460 SE remains functional but sacrifices roughly 9% average performance. The data shows no scenario where the GTX 460 SE is the better performer.

Head-to-Head Benchmarks

The Geekbench OpenCL result is the single most telling comparison. The GTX 680M scores 9,230, while the GTX 460 SE scores 6,389. The delta of 44.5% is substantial, and it aligns with the underlying architectural advantages. The GTX 680M has 1,344 shading units, 112 texture mapping units, and 32 ROPs, while the GTX 460 SE has 288 shading units, 48 TMUs, and 32 ROPs. The shading unit count alone is over 4.6 times higher, and the texture rate of 84.90 GTexel/s versus 31.20 GTexel/s explains why the OpenCL workload favors the newer card so heavily.

The pixel rate also favors the GTX 680M: 21.22 GPixel/s compared to 7.800 GPixel/s. Even though both have 32 ROPs, the GTX 680M's higher clock speed — 719 MHz base with 758 MHz boost versus no base clock listed for the GTX 460 SE — drives the fillrate advantage. The memory subsystem shows a similar pattern: the GTX 680M has 4 GB of GDDR5 at 115.2 GB/s bandwidth, while the GTX 460 SE has 1 GB of GDDR5 at 108.8 GB/s. The bandwidth gap is 5.9%, but the capacity gap is 300% in favor of the GTX 680M.

The average benchmark scores reinforce the head-to-head result. The GTX 680M also has a Geekbench Metal score of 4,815, which the GTX 460 SE does not have, giving the newer card an additional performance data point. Across all benchmarks, the GTX 680M wins 1 and loses 0, while the GTX 460 SE wins 0 and loses 1.

Specification Differences

| Specification | GTX 680M | GTX 460 SE |

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

| Chip | GK104 | GF104 |

| Process Node | 28 nm | 40 nm |

| Transistors | 3,540 million | 1,950 million |

| Die Size | 294 mm² | 332 mm² |

| Transistor Density | 12.0M / mm² | 5.9M / mm² |

| Base Clock | 719 MHz | Not listed |

| Boost Clock | 758 MHz | Not listed |

| Memory Clock | 900 MHz (3.6 Gbps effective) | 850 MHz (3.4 Gbps effective) |

| Memory Size | 4 GB | 1024 MB |

| Memory Bandwidth | 115.2 GB/s | 108.8 GB/s |

| Shading Units | 1,344 | 288 |

| TMUs | 112 | 48 |

| ROPs | 32 | 32 |

| Pixel Rate | 21.22 GPixel/s | 7.800 GPixel/s |

| Texture Rate | 84.90 GTexel/s | 31.20 GTexel/s |

| FP32 | 2.038 TFLOPS | 748.8 GFLOPS |

| TDP | 100 W | 150 W |

| Slot Width | MXM Module | Dual-slot |

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

| Suggested PSU | Not listed | 450 W |

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

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

| Length | Not listed | 210 mm (8.3 inches) |

| Vulkan Support | 1.2.175 | None |

| Release Date | 2012-06-03 | 2010-11-14 |

| Generation | GeForce 600M | GeForce 400 |

| Predecessor | GeForce 500M | GeForce 200 |

| Successor | GeForce 700M | GeForce 500 |

Architecture Differences

The two GPUs represent different architectural generations from NVIDIA. The GTX 680M uses the Kepler architecture with the GK104 chip, built on a 28 nm TSMC process. The GTX 460 SE uses the older Fermi architecture with the GF104 chip, built on a 40 nm TSMC process. This process shrink is significant: the GTX 680M fits 3,540 million transistors into a 294 mm² die, achieving a density of 12.0M transistors per mm². The GTX 460 SE has 1,950 million transistors on a larger 332 mm² die, yielding only 5.9M per mm². The Kepler part is denser and smaller while carrying nearly double the transistor count.

The compute capabilities diverge sharply. The GTX 680M delivers 2.038 TFLOPS of FP32 performance, while the GTX 460 SE manages 748.8 GFLOPS. This 2.7x gap in raw floating-point throughput is the core of the benchmark difference. The GTX 680M's 1,344 shading units operate at a base clock of 719 MHz with a boost to 758 MHz, while the GTX 460 SE's 288 shading units have no listed base or boost clock. The texture unit count also differs: 112 versus 48, contributing to the 84.90 GTexel/s versus 31.20 GTexel/s texture fill rates.

Memory architecture shows equal bus widths — both use a 256-bit interface — but the GTX 680M runs its GDDR5 at 900 MHz (3.6 Gbps effective) versus 850 MHz (3.4 Gbps effective) for the GTX 460 SE. The resulting bandwidth is 115.2 GB/s versus 108.8 GB/s. The GTX 680M also has 4 GB of memory versus 1 GB, which is a fourfold capacity advantage. Both support DirectX 12 (11_0) and OpenGL 4.6, but only the GTX 680M supports Vulkan 1.2.175. The GTX 460 SE lacks Vulkan entirely.

The power envelope is inverted relative to performance. The GTX 680M consumes 100 W as an MXM module, while the GTX 460 SE draws 150 W as a dual-slot desktop card. The GTX 460 SE requires two 6-pin power connectors and a suggested 450 W power supply, whereas the GTX 680M has no power connectors because it relies on the laptop's MXM-B (3.0) interface. The GTX 460 SE has a fixed length of 210 mm, while the GTX 680M's dimensions are not listed.

Where Each One Wins

The GTX 680M wins every measurable performance category in the data. It has higher pixel rate (21.22 GPixel/s versus 7.800 GPixel/s), higher texture rate (84.90 GTexel/s versus 31.20 GTexel/s), higher FP32 throughput (2.038 TFLOPS versus 748.8 GFLOPS), and more memory bandwidth (115.2 GB/s versus 108.8 GB/s). It also has more memory capacity (4 GB versus 1 GB) and more shading units (1,344 versus 288). In the only shared benchmark, Geekbench OpenCL, it wins by 44.5%.

The GTX 460 SE wins in areas that are not about raw speed. It has a longer physical footprint at 210 mm, which is not an advantage but a form factor choice for desktop cases. It offers fixed display outputs — 2x DVI and 1x mini-HDMI 1.3a — which may be more convenient than the GTX 680M's portable-device-dependent outputs. It also has a lower transistor count (1,950 million versus 3,540 million), which is a point of efficiency in terms of manufacturing cost, though not performance. The GTX 460 SE's launch MSRP was 160 USD, while the GTX 680M has no listed launch MSRP.

For use cases, the GTX 680M is the clear pick for any performance-sensitive workload. Its higher memory capacity and bandwidth make it suitable for texture-heavy applications or larger datasets, while its superior compute rates benefit OpenCL tasks. The GTX 460 SE, with its lower power connectors and desktop form factor, fits only in a traditional tower PC with a PCIe 2.0 x16 slot and a 450 W power supply. The data shows no benchmark where the GTX 460 SE outperforms the GTX 680M, so the older card's only niche is as a legacy desktop component where the MXM form factor of the GTX 680M is not compatible.