NVIDIA GeForce GTX 460M vs NVIDIA GeForce GTX 970M Comparison

The NVIDIA GeForce GTX 970M and NVIDIA GeForce GTX 460M represent two distinct eras of mobile graphics, separated by four generations of architecture. The data shows a decisive overall performance gap, but the specific numbers reveal the magnitude of that divide and highlight where each part retains relevance. This analysis draws exclusively on benchmark results and architectural specifications to compare the two.

Head-to-Head Benchmarks

The only direct benchmark comparison available is the Geekbench OpenCL test, which measures general-purpose compute performance. The results are stark: the GTX 970M scores 18,946 points, while the GTX 460M scores 4,282 points. This gives the GTX 970M a 342.5% advantage, meaning it delivers more than four times the compute throughput of the older chip. In practical terms, this translates to dramatically faster performance in any workload that leverages OpenCL, from video encoding to physics simulations.

Looking at the broader benchmark suite, the GTX 970M's average benchmark score of 4,628 points places it in the 27th percentile of all GPUs. Its nearest rivals include the NVIDIA Quadro M3000M (average score 4,621, deltaPct 0.2%) and the AMD Radeon R5 M320 (average score 4,657, deltaPct -0.6%). This indicates the GTX 970M sits in a competitive mid-range tier, trading blows with professional mobile workstations and entry-level desktop parts from the same era. The GTX 460M, by contrast, has an average benchmark score of 4,282 points, placing it in the 25th percentile. Its nearest rivals include the AMD FirePro W2100 (average score 4,295, deltaPct -0.3%) and the AMD Radeon Vega 3 (average score 4,268, deltaPct 0.3%).

The GTX 970M's individual benchmark results show a profile consistent with a balanced mobile GPU. It achieves a Passmark G3D score of 5,704, which is its strongest DirectX-era result, while its Passmark G2D score of 381 indicates modest 2D performance. The GPU compute score of 2,289 further confirms its compute-oriented strengths. The GTX 460M, however, has no corresponding individual benchmark scores in the data beyond the single Geekbench OpenCL result, making direct comparisons in other tests impossible. The available data nonetheless suggests that the GTX 970M is not merely faster but categorically superior in raw compute output, a conclusion reinforced by the architectural differences detailed below.

Architecture Differences

The two GPUs are built on fundamentally different architectures, which explains the performance gap. The GTX 970M uses the GM204 chip based on Maxwell 2.0 architecture, fabricated on a 28 nm process at TSMC. It contains 5,200 million transistors on a die size of 398 mm², yielding a transistor density of 13.1 million transistors per square millimeter. The GTX 460M, in contrast, uses the GF106 chip based on the older Fermi architecture, also fabricated at TSMC but on a larger 40 nm process. It contains 1,170 million transistors on a 238 mm² die, with a transistor density of 4.9 million per square millimeter.

These process and density differences have direct consequences for computing resources. The GTX 970M packs 1,280 shading units, 80 texture mapping units (TMUs), and 48 raster output units (ROPs). The GTX 460M has only 192 shading units, 32 TMUs, and 24 ROPs. This represents a 6.7x increase in shading units and a 2.5x increase in TMUs for the newer part, while ROPs double. The pixel rate of the GTX 970M is 49.82 GPixel/s versus 5.400 GPixel/s for the GTX 460M, and texture rates are 83.04 GTexel/s versus 21.60 GTexel/s. Floating-point performance tells a similar story: the GTX 970M delivers 2.657 TFLOPS of FP32 compute, while the GTX 460M manages 518.4 GFLOPS.

Memory configurations also diverge significantly. The GTX 970M ships with 6 GB of GDDR5 memory on a 192-bit bus, running at 1,253 MHz (5 Gbps effective), yielding a bandwidth of 120.3 GB/s. The GTX 460M has 1,536 MB (1.5 GB) of GDDR5 on the same 192-bit bus width, but at a lower 625 MHz (2.5 Gbps effective), resulting in 60.00 GB/s of bandwidth. The GTX 970M thus offers four times the memory capacity and exactly double the bandwidth. In terms of API support, the GTX 970M supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 460M supports DirectX 12 (11_0) and OpenGL 4.6, but has no Vulkan support listed. Both use MXM Module slot width and have no power connectors, with display outputs described as "Portable Device Dependent."

The bus interface differs as well: the GTX 970M uses MXM-B (3.0), while the GTX 460M uses PCIe 2.0 x16. The GTX 460M has a listed TDP of 50 W, while the GTX 970M's TDP is not specified in the data. The GTX 970M belongs to the GeForce 900M generation, succeeding the GeForce 800M, while the GTX 460M is part of the GeForce 400M generation, succeeding the GeForce 300M. Release dates are 2014-10-06 for the GTX 970M and 2010-09-02 for the GTX 460M, confirming the four-year gap. Both are marked as end-of-life production status.

The Verdict

The data unequivocally favors the GTX 970M in every measurable category. The 342.5% lead in Geekbench OpenCL is not a marginal improvement but a generational leap. The GTX 970M's higher shading unit count, transistor density, memory bandwidth, and API support all point to a part designed for modern workloads, while the GTX 460M is a product of an earlier era. The 27th percentile ranking for the GTX 970M versus the 25th percentile for the GTX 460M may seem close, but this is misleading; the percentile reflects all GPUs ever benchmarked, and the GTX 970M's average score of 4,628 versus 4,282 for the GTX 460M represents a meaningful shift in capability.

For users choosing between these two, the GTX 970M is the clear choice for any task involving compute, modern DirectX 12 titles, or Vulkan applications. Its 6 GB memory capacity (versus 1.5 GB) also makes it far more suitable for texture-heavy workloads or larger datasets. The GTX 460M, however, still supports DirectX 12 (11_0) and OpenGL 4.6, meaning it can run some modern software, albeit with significantly lower performance. The GTX 460M's lower TDP of 50 W could be an advantage in power-constrained systems, but the data does not provide enough comparative power figures to draw a definitive conclusion. Strictly from benchmark results and architectural specifications, the GTX 970M wins decisively.

FAQ

Q: How much faster is the GTX 970M than the GTX 460M in compute performance?

A: In the Geekbench OpenCL test, the GTX 970M scores 18,946, which is 342.5% higher than the GTX 460M's score of 4,282.

Q: What are the memory capacities of these two GPUs?

A: The GTX 970M has 6 GB of GDDR5 memory, while the GTX 460M has 1,536 MB (1.5 GB) of GDDR5 memory. Both use a 192-bit bus width.

Q: Do both GPUs support DirectX 12?

A: Yes, but with different feature levels. The GTX 970M supports DirectX 12 (12_1), while the GTX 460M supports DirectX 12 (11_0).

Q: Which GPU has higher texture fill rate?

A: The GTX 970M has a texture rate of 83.04 GTexel/s, compared to 21.60 GTexel/s for the GTX 460M.

Q: What is the transistor count difference?

A: The GTX 970M contains 5,200 million transistors on a 398 mm² die, while the GTX 460M contains 1,170 million transistors on a 238 mm² die.

Q: Do both GPUs support Vulkan?

A: The GTX 970M supports Vulkan 1.4, but the GTX 460M has no Vulkan support listed in the data.

Where Each One Wins

The GTX 970M wins in every category where direct comparison data exists. Its Geekbench OpenCL score is more than four times higher, and its architectural specifications are superior across the board: more shading units (1,280 vs. 192), higher pixel rate (49.82 GPixel/s vs. 5.400 GPixel/s), higher texture rate (83.04 GTexel/s vs. 21.60 GTexel/s), and double the memory bandwidth (120.3 GB/s vs. 60.00 GB/s). The GTX 970M also supports a newer DirectX feature level (12_1 vs. 11_0) and Vulkan 1.4, which the GTX 460M lacks entirely. The GTX 970M's 6 GB memory capacity is four times the GTX 460M's 1.5 GB, making it more viable for modern game textures or larger compute datasets.

The GTX 460M's only potential advantage lies in its lower TDP of 50 W, which could make it suitable for older or more power-sensitive laptops. However, the data does not provide a comparable TDP figure for the GTX 970M, so this cannot be quantified as a win. The GTX 460M also has a smaller die size (238 mm² vs. 398 mm²) and fewer transistors, which historically implies lower manufacturing cost, but pricing is not addressed in this data. For any performance-oriented task, the GTX 970M is the unambiguous winner.

Specification Differences

| Specification | NVIDIA GeForce GTX 970M | NVIDIA GeForce GTX 460M |

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

| Architecture | Maxwell 2.0 | Fermi |

| Generation | GeForce 900M | GeForce 400M |

| Process Node | 28 nm | 40 nm |

| Foundry | TSMC | TSMC |

| Transistors | 5,200 million | 1,170 million |

| Die Size | 398 mm² | 238 mm² |

| Transistor Density | 13.1M / mm² | 4.9M / mm² |

| Memory Size | 6 GB | 1536 MB |

| Memory Type | GDDR5 | GDDR5 |

| Memory Clock | 1253 MHz (5 Gbps effective) | 625 MHz (2.5 Gbps effective) |

| Memory Bandwidth | 120.3 GB/s | 60.00 GB/s |

| Shading Units | 1,280 | 192 |

| TMUs | 80 | 32 |

| ROPs | 48 | 24 |

| Pixel Rate | 49.82 GPixel/s | 5.400 GPixel/s |

| Texture Rate | 83.04 GTexel/s | 21.60 GTexel/s |

| FP32 Performance | 2.657 TFLOPS | 518.4 GFLOPS |

| TDP | Not specified | 50 W |

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

| DirectX Support | 12 (12_1) | 12 (11_0) |

| Vulkan Support | 1.4 | None |

| Release Date | 2014-10-06 | 2010-09-02 |

| Predecessor | GeForce 800M | GeForce 300M |

| Successor | GeForce 10 Mobile | GeForce 500M |