NVIDIA GeForce GTX 460 SE vs NVIDIA GeForce GTX 670M Comparison

# NVIDIA GeForce GTX 670M vs NVIDIA GeForce GTX 460 SE

The NVIDIA GeForce GTX 670M and NVIDIA GeForce GTX 460 SE are both end-of-life Fermi-generation parts, but they target very different segments: the 670M is a mobile MXM module from the GeForce 600M family, while the 460 SE is a dual-slot desktop card from the GeForce 400 series. In the single available benchmark — Geekbench OpenCL — the GTX 670M scores 6513, edging out the GTX 460 SE's 6389 by a narrow 1.9% margin. That slim lead, combined with the 670M's substantially lower 75 W TDP versus 150 W, makes the mobile part the more efficient performer, though the desktop card counters with a wider 256-bit memory bus and higher bandwidth.

FAQ

Q: Which GPU wins in the head-to-head benchmark?

A: The NVIDIA GeForce GTX 670M wins the Geekbench OpenCL test with a score of 6513 against the GTX 460 SE's 6389, a delta of 1.9%.

Q: How do their memory configurations compare?

A: The GTX 670M has 1536 MB of GDDR5 on a 192-bit bus delivering 72.00 GB/s, while the GTX 460 SE has 1024 MB on a 256-bit bus with 108.8 GB/s bandwidth.

Q: What are the TDP differences between the two cards?

A: The GTX 670M is rated at 75 W with no power connectors, whereas the GTX 460 SE draws 150 W and requires 2x 6-pin power connectors with a suggested 450 W PSU.

Q: Do both cards support the same DirectX version?

A: Yes, both support DirectX 12 (11_0) and OpenGL 4.6; neither lists Vulkan support.

Q: Which card has more shading units and texture units?

A: The GTX 670M has 336 shading units and 56 TMUs, while the GTX 460 SE has 288 shading units and 48 TMUs.

Q: What is the process node and die size for both?

A: Both are built on TSMC's 40 nm process with 1,950 million transistors on a 332 mm² die, yielding 5.9M transistors per mm².

Architecture Differences

Both GPUs share the same fundamental Fermi architecture lineage, but they represent distinct iterations. The GTX 670M uses the GF114 chip, described as "Fermi 2.0," while the GTX 460 SE uses the original GF104 chip under the plain "Fermi" architecture label. The 670M is part of the GeForce 600M generation, positioned as the successor to the GeForce 500M series, whereas the 460 SE belongs to the GeForce 400 generation, following the GeForce 200 line.

The transistor count is identical at 1,950 million, and the die size is the same 332 mm², which means the transistor density is also equal at 5.9M / mm². However, the internal configuration differs significantly. The GTX 670M packs 336 shading units, 56 texture mapping units, and 24 ROPs, giving it a pixel rate of 8.372 GPixel/s and a texture rate of 33.49 GTexel/s. The GTX 460 SE, by contrast, has 288 shading units, 48 TMUs, and 32 ROPs, resulting in a lower pixel rate of 7.800 GPixel/s and texture rate of 31.20 GTexel/s. The 670M's higher shader and texture counts explain its FP32 throughput of 803.7 GFLOPS versus the 460 SE's 748.8 GFLOPS.

Memory architecture is another key differentiator. The GTX 670M uses a 192-bit bus with 1536 MB of GDDR5 running at 750 MHz (3 Gbps effective), producing 72.00 GB/s. The GTX 460 SE uses a wider 256-bit bus with 1024 MB of GDDR5 at 850 MHz (3.4 Gbps effective), yielding 108.8 GB/s — a 51% bandwidth advantage for the desktop card. The two cards also differ in physical form: the 670M is an MXM Module with a 3.0 bus interface, while the 460 SE is a dual-slot card measuring 210 mm (8.3 inches) with PCIe 2.0 x16 and display outputs of 2x DVI plus 1x mini-HDMI 1.3a.

The 670M's TDP of 75 W is exactly half the 460 SE's 150 W, and the mobile card requires no power connectors while the desktop card needs 2x 6-pin. Release dates also differ: the 670M launched on 2012-03-21, while the 460 SE launched earlier on 2010-11-14.

Head-to-Head Benchmarks

The only benchmark available is Geekbench OpenCL, and the results show a close contest. The GTX 670M scores 6513, while the GTX 460 SE scores 6389. This gives the 670M a 1.9% victory — a narrow but consistent margin.

Contextualizing these scores against the nearest rivals provides additional insight. The GTX 670M sits at the 38th percentile of all GPUs, with its closest competitors being the NVIDIA GeForce GT 555M (6493, 0.3% behind), the NVIDIA Quadro M5000M (6481, 0.5% behind), and the AMD Radeon Vega 10 Mobile (6476, 0.6% behind). It also trails the Intel UHD Graphics P750 (6554) by 0.6%. The GTX 460 SE sits at the 37th percentile, with the NVIDIA GeForce GTX 580M matching it exactly at 6389 (0% delta), the NVIDIA RTX PRO 5000 72 GB Blackwell at 6407 (0.3% ahead), the AMD Radeon Pro WX 4100 at 6330 (0.9% behind), and the AMD Radeon R7 M350 at 6327 (1% behind).

The 1.9% delta between the two cards is small enough that real-world differences would be marginal in most workloads. However, the 670M's higher shading unit count (336 vs 288) and higher FP32 throughput (803.7 vs 748.8 GFLOPS) align with its benchmark lead. The 460 SE's higher memory bandwidth (108.8 vs 72.00 GB/s) does not translate into an OpenCL win, suggesting that compute performance in this test is more sensitive to shader throughput than memory bandwidth.

Specification Differences

| Specification | GTX 670M | GTX 460 SE |

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

| Architecture | Fermi 2.0 | Fermi |

| Chip | GF114 | GF104 |

| Generation | GeForce 600M | GeForce 400 |

| Memory Size | 1536 MB | 1024 MB |

| Memory Bus Width | 192 bit | 256 bit |

| Memory Clock | 750 MHz (3 Gbps) | 850 MHz (3.4 Gbps) |

| Memory Bandwidth | 72.00 GB/s | 108.8 GB/s |

| Shading Units | 336 | 288 |

| TMUs | 56 | 48 |

| ROPs | 24 | 32 |

| Pixel Rate | 8.372 GPixel/s | 7.800 GPixel/s |

| Texture Rate | 33.49 GTexel/s | 31.20 GTexel/s |

| FP32 | 803.7 GFLOPS | 748.8 GFLOPS |

| TDP | 75 W | 150 W |

| Slot Width | MXM Module | Dual-slot |

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

| Suggested PSU | None | 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 specified | 210 mm (8.3 inches) |

| Release Date | 2012-03-21 | 2010-11-14 |

| Launch MSRP | None | 160 USD |

| Predecessor | GeForce 500M | GeForce 200 |

| Successor | GeForce 700M | GeForce 500 |

The two cards share identical transistor counts (1,950 million), die size (332 mm²), process node (40 nm), foundry (TSMC), and API support (DirectX 12 (11_0), OpenGL 4.6). Both are end-of-life products.

Where Each One Wins

The GTX 670M wins in compute-oriented workloads based on the Geekbench OpenCL result. Its advantage comes from having 48 more shading units, 8 more TMUs, and higher FP32 throughput (803.7 vs 748.8 GFLOPS). The 670M also dominates on power efficiency: at 75 W, it delivers its benchmark score at half the TDP of the 460 SE's 150 W. For mobile or compact systems using MXM modules, the 670M is the only viable choice of the two, and its lack of power connectors makes installation simpler.

The GTX 460 SE wins in memory-bandwidth-sensitive scenarios, though no benchmark in the data confirms this directly. Its 256-bit bus and 108.8 GB/s bandwidth — over 51% higher than the 670M's 72.00 GB/s — could favor high-resolution textures or memory-heavy workloads, but the OpenCL score does not reflect such an advantage. The 460 SE also has more ROPs (32 vs 24), which could benefit pixel-heavy rendering, and its desktop form factor with dual DVI and mini-HDMI outputs makes it suitable for multi-monitor setups. Its 1024 MB memory is smaller, but the wider bus partially compensates.

The 460 SE's launch MSRP was 160 USD, while the 670M has no listed launch price. The desktop card's longer length (210 mm) and dual-slot design mean it requires more case space, whereas the MXM module is designed for portable devices.

The Verdict

The data points to the GTX 670M as the better overall performer. It wins the only available benchmark by 1.9%, achieves a higher percentile ranking (38th vs 37th), and does so with a 75 W TDP that is half the 460 SE's 150 W draw. For anyone prioritizing compute performance per watt, the 670M is the clear choice.

However, the verdict changes based on form factor and workload. If the requirement is a desktop GPU with high memory bandwidth for texture-heavy applications, the GTX 460 SE's 108.8 GB/s and 256-bit bus offer a structural advantage that the 670M cannot match, even if the benchmark does not reflect it. The 460 SE also provides more ROPs and dedicated display outputs, making it more flexible for traditional desktop use with multiple monitors.

For mobile users or those integrating an MXM module, the GTX 670M is the only option between the two. For desktop builders with a 450 W PSU and space for a dual-slot card, the GTX 460 SE remains a functional choice, though its benchmark deficit and double the power draw make it harder to recommend on pure performance grounds. The 1.9% benchmark gap is small, but the efficiency gap is enormous — the 670M delivers a higher score at half the power, which is the decisive factor in this comparison.