NVIDIA GeForce GTX 460 vs NVIDIA GeForce GTX 680M Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct benchmark comparison between the NVIDIA GeForce GTX 460 and the NVIDIA GeForce GTX 680M: the Geekbench OpenCL test. The GTX 680M wins this matchup decisively, posting a score of 9230 against the GTX 460's 7925. That represents a 14.1% advantage for the newer mobile part, a meaningful gap in compute throughput that reflects the generational shift between the two architectures.

Looking at the broader database context, the GTX 460's OpenCL score of 7925 places it in the 41st percentile of all GPUs. Its nearest rivals in the database are tightly clustered: the NVIDIA Quadro K4100M scores 7906 (just 0.2% behind), the NVIDIA Quadro P5000 scores 8039 (1.4% ahead), the NVIDIA GeForce GTX 880M scores 8040 (1.4% ahead), and the NVIDIA GeForce GTX 650 Ti scores 8053 (1.6% ahead). This clustering indicates that the GTX 460 sits in a very competitive performance band, where even a small clock or memory advantage can shift rankings. The GTX 460 is essentially at parity with these adjacent parts, trailing the fastest of them by less than two percentage points.

The GTX 680M, by contrast, records an OpenCL score of 9230, but its average benchmark score across all recorded tests is 7023, which places it in the 39th percentile of all GPUs. This discrepancy between the OpenCL score and the average is worth noting: the GTX 680M also has a Geekbench Metal score of 4815 in the database, which drags its average down. Its nearest rivals on average score are the NVIDIA T600 at 7035 (0.2% ahead), the AMD Radeon R5 M240 at 6975 (0.7% behind), the NVIDIA GeForce GTX 675M at 6946 (1.1% behind), and the NVIDIA GeForce GTX 970 at 7157 (1.9% ahead). In terms of average performance, the GTX 680M is effectively interchangeable with these parts, all within a two-percentage-point band.

The head-to-head OpenCL result is the only direct measurement where both cards appear, and it strongly favors the GTX 680M. A 14.1% win in compute workload is substantial, not a marginal edge. The GTX 460's raw score, while respectable for its era, simply cannot match the throughput of the Kepler-based mobile chip. The data shows one clear victory for the GTX 680M, and zero for the GTX 460, in their shared test.

Architecture Differences

The two cards represent distinct generations of NVIDIA design. The GTX 460 uses the GF104 chip built on the Fermi architecture, fabricated on a 40 nm process at TSMC. The GTX 680M uses the GK104 chip on the Kepler architecture, fabricated on a 28 nm process, also at TSMC. This process shrink is the single biggest architectural differentiator: the newer chip packs far more transistors into a smaller area.

The GTX 460 contains 1,950 million transistors on a die size of 332 mm², yielding a transistor density of 5.9 million per square millimeter. The GTX 680M contains 3,540 million transistors on a die size of 294 mm², yielding a density of 12.0 million per square millimeter. In short, the GTX 680M has roughly 81% more transistors in a smaller physical package, more than doubling the transistor density. This is the physical foundation for its higher performance.

The compute resources differ sharply. The GTX 460 has 336 shading units, 56 texture mapping units, and 24 raster output units. The GTX 680M has 1,344 shading units, 112 texture mapping units, and 32 raster output units. That is four times the shading units, exactly double the texture units, and a 33% increase in ROPs. These are not incremental changes; they represent a fundamentally wider execution engine.

Clock behavior also differs. The GTX 460 lists only a memory clock of 900 MHz (3.6 Gbps effective), with no base or boost clock recorded. The GTX 680M has a base clock of 719 MHz and a boost clock of 758 MHz, with the same memory clock of 900 MHz (3.6 Gbps effective). The GTX 680M's boost capability is a feature the Fermi card lacks entirely in the database.

Memory configurations are very different. The GTX 460 has 768 MB of GDDR5 on a 192-bit bus, delivering 86.40 GB/s of bandwidth. The GTX 680M has 4 GB of GDDR5 on a 256-bit bus, delivering 115.2 GB/s. The GTX 680M offers more than five times the capacity and a 33% bandwidth increase. For workloads that scale with memory size or bandwidth, this is a decisive advantage.

The GTX 460 has a TDP of 160 W, while the GTX 680M is rated at 100 W. The mobile part delivers higher performance at lower power, a direct consequence of the 28 nm process and Kepler's efficiency improvements. The GTX 460 is a dual-slot desktop card requiring two 6-pin power connectors and a 450 W suggested PSU. The GTX 680M is an MXM module with no power connectors, designed for portable devices. The bus interfaces also differ: PCIe 2.0 x16 for the GTX 460 versus MXM-B (3.0) for the GTX 680M.

API support is mostly aligned: both support DirectX 12 (11_0) and OpenGL 4.6. However, the GTX 680M additionally supports Vulkan 1.2.175, while the GTX 460 has no Vulkan support recorded. For modern applications leveraging Vulkan, the GTX 680M is the only viable option of the two.

Where Each One Wins

The direct benchmark data gives the GTX 680M a clean win in OpenCL compute, with a 14.1% lead. This makes it the stronger choice for general-purpose GPU compute tasks, including OpenCL-based rendering, physics simulation, and data-parallel workloads. Its 4 GB memory capacity and higher bandwidth also favor it for larger datasets and higher-resolution textures.

The GTX 460, despite losing the head-to-head, remains competitive in its own context. Its OpenCL score of 7925 is within 1.6% of the GTX 650 Ti and within 1.4% of the GTX 880M. For a desktop card from the GeForce 400 generation, it holds its own against significantly newer parts in the database. It has no recorded Vulkan support, but for DirectX 11-era titles and OpenGL applications, it remains serviceable.

The GTX 680M's average benchmark score of 7023 is actually lower than the GTX 460's single OpenCL score of 7925, but this is because the GTX 680M's Metal score of 4815 pulls its average down. In pure OpenCL, the GTX 680M is clearly ahead. The GTX 460 has no Metal score recorded, so cross-API comparisons are limited to OpenCL.

For gaming, the GTX 680M's higher pixel rate (21.22 GPixel/s versus 9.45 GPixel/s) and texture rate (84.90 GTexel/s versus 37.80 GTexel/s) indicate substantially faster rasterization and texturing. The GTX 680M also has a higher FP32 throughput at 2.038 TFLOPS versus 907.2 GFLOPS, more than doubling the single-precision compute ceiling. The GTX 460's advantage, if any, lies in its desktop form factor with dedicated display outputs (2x DVI, 1x mini-HDMI 1.3a), whereas the GTX 680M's outputs are portable-device dependent.

The GTX 680M wins on raw performance, memory capacity, power efficiency, and modern API support. The GTX 460 wins on form factor flexibility for desktop systems and has a higher percentile ranking (41st versus 39th), though that ranking is based on different benchmark sets.

The Verdict

The data is unambiguous: the NVIDIA GeForce GTX 680M is the faster GPU in the shared OpenCL test, winning by 14.1% with a score of 9230 against 7925. It also offers four times the shading units, double the texture units, more than double the FP32 throughput, and over five times the memory capacity. For any workload that can utilize these resources, the GTX 680M is the correct choice.

The GTX 680M is also more power-efficient, consuming 100 W versus 160 W, and it supports Vulkan, which the GTX 460 does not. Its MXM form factor makes it suitable for portable devices, while the GTX 460 is a dual-slot desktop card requiring a 450 W PSU. If the system can accommodate an MXM module, the GTX 680M is the superior part in nearly every measurable way.

The GTX 460, however, is not without merit. Its OpenCL score places it in the 41st percentile, slightly higher than the GTX 680M's 39th percentile average. But that percentile is based on a single test, while the GTX 680M's average includes a weaker Metal result. In the only direct comparison available, the GTX 680M wins clearly.

Buyers or system builders choosing between these two should select the GTX 680M for compute performance, memory capacity, and efficiency. The GTX 460 is a reasonable fallback for legacy desktop builds where MXM modules are not an option, but the data shows no performance category where it beats the GTX 680M. The verdict is a decisive win for the GTX 680M.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The NVIDIA GeForce GTX 680M scores 9230, which is 14.1% higher than the NVIDIA GeForce GTX 460's 7925.

Q: How much more memory does the GTX 680M have?

A: The GTX 680M has 4 GB of GDDR5 memory, while the GTX 460 has 768 MB, a difference of more than five times.

Q: Do both GPUs support the same APIs?

A: Both support DirectX 12 (11_0) and OpenGL 4.6, but only the GTX 680M supports Vulkan 1.2.175.

Q: What is the power consumption difference?

A: The GTX 460 has a TDP of 160 W, while the GTX 680M has a TDP of 100 W, making the GTX 680M more power-efficient.

Q: Which GPU has more shading units?

A: The GTX 680M has 1,344 shading units, exactly four times the GTX 460's 336 shading units.

Q: What form factor does each GPU use?

A: The GTX 460 is a dual-slot desktop card with two 6-pin power connectors, while the GTX 680M is an MXM module with no power connectors, designed for portable devices.

Specification Differences

| Field | NVIDIA GeForce GTX 460 | NVIDIA GeForce GTX 680M |

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

| Chip | GF104 | GK104 |

| Architecture | Fermi | Kepler |

| Generation | GeForce 400 | GeForce 600M |

| Process Node | 40 nm | 28 nm |

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

| Die Size | 332 mm² | 294 mm² |

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

| Base Clock | Not recorded | 719 MHz |

| Boost Clock | Not recorded | 758 MHz |

| Memory Size | 768 MB | 4 GB |

| Memory Bus Width | 192 bit | 256 bit |

| Memory Bandwidth | 86.40 GB/s | 115.2 GB/s |

| Shading Units | 336 | 1,344 |

| Texture Mapping Units | 56 | 112 |

| Raster Output Units | 24 | 32 |

| Pixel Rate | 9.450 GPixel/s | 21.22 GPixel/s |

| Texture Rate | 37.80 GTexel/s | 84.90 GTexel/s |

| FP32 | 907.2 GFLOPS | 2.038 TFLOPS |

| TDP | 160 W | 100 W |

| Slot Width | Dual-slot | MXM Module |

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

| Suggested PSU | 450 W | Not recorded |

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

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

| Vulkan Support | Not recorded | 1.2.175 |

| Release Date | 2010-07-11 | 2012-06-03 |

| Predecessor | GeForce 200 | GeForce 500M |

| Successor | GeForce 500 | GeForce 700M |

| Launch MSRP | 199 USD | Not recorded |