NVIDIA GeForce GTX 460 vs NVIDIA GeForce GTX 950M Comparison

The NVIDIA GeForce GTX 950M and the NVIDIA GeForce GTX 460 represent two very different eras of mobile and desktop GPU design. The data shows a clear performance hierarchy, with the newer Maxwell-based mobile part taking a decisive lead in the only shared benchmark. However, the older Fermi card was a high-power desktop part with its own architectural strengths, making this comparison more nuanced than a simple generational sweep.

Head-to-Head Benchmarks

The single direct comparison available in the data is the Geekbench OpenCL test, and the result is lopsided. The GTX 950M scores 9,745 points, while the GTX 460 manages 7,925 points. This translates to a 23% advantage for the GTX 950M, a substantial margin that underscores the efficiency and architectural maturity of Maxwell. This is not a marginal victory; it is a commanding lead that places the mobile chip firmly ahead in raw compute throughput as measured by this workload.

Interpreting this score through the lens of their respective peer groups adds context. The GTX 950M’s average benchmark score of 8,135 places it at the 42nd percentile of all GPUs. Its nearest rivals are remarkably close: the AMD Radeon R9 M360 is essentially tied at 8,129 (a 0.1% delta), and the NVIDIA GeForce 945M trails by just 0.4% with a score of 8,099. The GTX 950M also edges out the NVIDIA GRID K2 by 0.7%. This indicates that while the GTX 950M is not a high-end part, it sits at the top of its immediate competitive cluster. Its 23% win over the GTX 460 is far larger than any delta seen among its closest rivals, making the victory over the older card particularly emphatic.

The GTX 460, by contrast, has an average benchmark score of 7,925, placing it at the 41st percentile. Its nearest rivals tell a different story. It is slightly ahead of the NVIDIA Quadro K4100M by 0.2%, but it loses ground to more modern parts: the NVIDIA Quadro P5000 is 1.4% faster, and the NVIDIA GeForce GTX 880M is also 1.4% faster. The NVIDIA GeForce GTX 650 Ti leads it by 1.6%. The GTX 460 is thus positioned at the lower end of its performance tier, and the 23% deficit to the GTX 950M is the largest gap in this entire comparative landscape. The data is unambiguous: in the one test they share, the GTX 950M wins decisively.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process technologies. The GTX 950M is based on the GM107 chip, utilizing NVIDIA’s Maxwell architecture. It is fabricated on a 28 nm process at TSMC, with a die size of 148 mm². The GTX 460, in contrast, uses the GF104 chip and the older Fermi architecture, built on a much larger 332 mm² die using a 40 nm process. The transistor counts are surprisingly similar — 1,870 million for the GTX 950M and 1,950 million for the GTX 460 — but the density differs dramatically. The GTX 950M packs 12.6 million transistors per mm², while the GTX 460 manages only 5.9 million per mm². This density advantage is a direct result of the more advanced 28 nm node and is a primary reason for the newer card’s superior performance-per-watt.

Compute resources also diverge sharply. The GTX 950M has 640 shading units, 40 texture mapping units (TMUs), and 16 raster output units (ROPs). The GTX 460 has fewer shading units at 336, but more TMUs (56) and more ROPs (24). This is a classic example of architectural philosophy: Fermi was designed with a wider memory interface and more textured throughput, while Maxwell focuses on higher shader efficiency and lower power. The clock speeds reflect this, with the GTX 950M running at a base of 993 MHz and a boost of 1124 MHz, whereas the GTX 460 has no listed base or boost clocks, relying on its memory subsystem for performance.

The memory subsystems are entirely different as well. The GTX 950M uses 4 GB of DDR3 memory on a 128-bit bus, yielding a bandwidth of 28.80 GB/s. The GTX 460 uses 768 MB of GDDR5 memory on a 192-bit bus, delivering a much higher 86.40 GB/s of bandwidth. This is a significant divergence: the older card has nearly triple the memory bandwidth, which was critical for Fermi’s design. However, the GTX 950M compensates with a much higher pixel rate (17.98 GPixel/s vs 9.450 GPixel/s) and texture rate (44.96 GTexel/s vs 37.80 GTexel/s), showing that Maxwell’s compute units are far more efficient at processing. The FP32 throughput also favors the newer card: 1,438.7 GFLOPS vs 907.2 GFLOPS for the GTX 460.

FAQ

Q: Which GPU wins in the Geekbench OpenCL benchmark?

A: The NVIDIA GeForce GTX 950M wins decisively, scoring 9,745 points compared to the GTX 460’s 7,925 points, a 23% advantage.

Q: How does the GTX 950M compare to its nearest rivals?

A: The GTX 950M is essentially tied with the AMD Radeon R9 M360 (0.1% delta) and slightly ahead of the NVIDIA GeForce 945M (0.4% delta) and NVIDIA GRID K2 (0.7% delta). It sits at the 42nd percentile of all GPUs.

Q: Does the GTX 460 have any advantages in memory bandwidth?

A: Yes. The GTX 460 has 86.40 GB/s of bandwidth thanks to its 192-bit GDDR5 memory, compared to just 28.80 GB/s for the GTX 950M’s 128-bit DDR3 memory. However, this does not translate to a benchmark win.

Q: What are the process node differences between the two cards?

A: The GTX 950M is built on a 28 nm process, while the GTX 460 uses a 40 nm process. This leads to a transistor density of 12.6M / mm² for the newer card versus 5.9M / mm² for the older one.

Q: Which card has a higher TDP?

A: The GTX 460 has a much higher TDP of 160 W, while the GTX 950M is rated at only 75 W. The GTX 460 also requires two 6-pin power connectors and a 450 W suggested PSU, whereas the GTX 950M has no power connectors.

Q: What is the GTX 460’s closest rival according to the data?

A: The GTX 460’s nearest rival is the NVIDIA Quadro K4100M, which has an average score of 7,906, a 0.2% delta. It also trails the NVIDIA GeForce GTX 880M and Quadro P5000 by 1.4%.

Specification Differences

The following table highlights the key fields where the two GPUs differ:

| Specification | NVIDIA GeForce GTX 950M | NVIDIA GeForce GTX 460 |

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

| Architecture | Maxwell | Fermi |

| Process Node | 28 nm | 40 nm |

| Die Size | 148 mm² | 332 mm² |

| Transistor Density | 12.6M / mm² | 5.9M / mm² |

| Base Clock | 993 MHz | N/A |

| Boost Clock | 1124 MHz | N/A |

| Memory Size | 4 GB | 768 MB |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus Width | 128 bit | 192 bit |

| Memory Bandwidth | 28.80 GB/s | 86.40 GB/s |

| Shading Units | 640 | 336 |

| TMUs | 40 | 56 |

| ROPs | 16 | 24 |

| Pixel Rate | 17.98 GPixel/s | 9.450 GPixel/s |

| Texture Rate | 44.96 GTexel/s | 37.80 GTexel/s |

| FP32 Performance | 1,438.7 GFLOPS | 907.2 GFLOPS |

| TDP | 75 W | 160 W |

| Slot Width | IGP | Dual-slot |

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

| Suggested PSU | N/A | 450 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 2.0 x16 |

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

| Vulkan Support | 1.4 | N/A |

| Length | N/A | 210 mm (8.3 inches) |

| Launch MSRP | N/A | 199 USD |

The Verdict

The data points to a clear winner for general compute performance: the NVIDIA GeForce GTX 950M. Its 23% lead in the Geekbench OpenCL test is decisive and is reinforced by superior FP32 throughput (1,438.7 GFLOPS vs 907.2 GFLOPS), a higher pixel rate, and a higher texture rate. The newer Maxwell architecture, built on a more efficient 28 nm process, delivers more performance while consuming half the power — 75 W versus 160 W — and requiring no external power connectors. For any user prioritizing raw compute results in this benchmark, the GTX 950M is the obvious choice.

However, the GTX 460 is not without its merits. Its memory subsystem is vastly superior, offering 86.40 GB/s of bandwidth compared to the GTX 950M’s 28.80 GB/s. This suggests that in workloads heavily dependent on memory throughput, the older card might have an edge, though this is not reflected in the shared benchmark. The GTX 460 also has more TMUs (56 vs 40) and ROPs (24 vs 16), which could theoretically benefit certain texture-heavy or fill-rate-limited scenarios. Its 199 USD launch MSRP also indicates it was a mainstream desktop product, whereas the GTX 950M is an integrated mobile part.

The verdict depends on the use case. For the benchmark data presented, the GTX 950M is the superior performer, and its low power draw makes it ideal for portable devices. The GTX 460, with its high bandwidth and dual-slot design, was built for a different era of desktop gaming and remains competitive only in specific memory-intensive tasks. Users seeking the highest OpenCL score from this pair should choose the GTX 950M; those with a legacy desktop system prioritizing memory bandwidth might still consider the GTX 460, but the data strongly favors the newer mobile chip.