NVIDIA GeForce GTX 1070 vs NVIDIA GeForce GTX 950A Comparison

The NVIDIA GeForce GTX 950A and the NVIDIA GeForce GTX 1070 represent two distinct eras of NVIDIA's mobile and desktop GPU strategy, separated by a full process-node generation and a dramatic shift in performance class. The data paints a stark picture: in the sole direct head-to-head benchmark available, the Geekbench OpenCL test, the GTX 1070 scores 44,700 points against the GTX 950A's 10,273 points. This translates to a 77% deficit for the older chip, a margin that underscores the generational leap between the two. While the GTX 950A finds its nearest rivals within a tight 2% band, the GTX 1070's average benchmark score of 9,780 places it in a similar overall percentile (47th vs. 48th), but the peak scores tell a very different story of capability.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and it is decisively one-sided. The GTX 1070 delivers a score of 44,700, while the GTX 950A manages just 10,273. The 77% delta is not merely a numerical gap; it represents a fundamental difference in compute throughput. The GTX 1070's raw FP32 performance of 6.463 TFLOPS dwarfs the GTX 950A's 1,438.7 GFLOPS, a 4.5x advantage that explains the benchmark disparity. This is not a close contest where architectural efficiency tips the scales; it is a brute-force victory for the newer, larger chip.

Looking at the GTX 1070's other benchmark results provides context for its OpenCL score. Its Passmark G3D score of 13,498 and Geekbench Vulkan score of 22,121 indicate strong sustained performance across different API loads. The GTX 950A has no comparable scores in the data, so the OpenCL result is the only direct measure. In terms of nearest rivals, the GTX 950A's 10,273 score sits between the AMD Radeon RX 550X (10,481, a 2% lead for the AMD) and the AMD Radeon R9 M375 (10,070, a 2% lead for the NVIDIA). The GTX 1070's average score of 9,780 is nearly matched by the NVIDIA Tesla M10 (9,724, a 0.6% lead for the 1070), showing that the 1070's average performance across all its tests is lower than its peak OpenCL run, but still competitive with professional-grade compute cards.

The texture and pixel rates further illustrate the gap. The GTX 1070's texture rate of 202.0 GTexel/s is 4.5x higher than the GTX 950A's 44.96 GTexel/s. Similarly, the pixel rate of 107.7 GPixel/s on the 1070 is a 6x improvement over the 950A's 17.98 GPixel/s. For gaming workloads, these numbers suggest the GTX 1070 can handle far higher resolutions and more complex shaders without becoming fill-rate limited. The benchmark data, while limited to one direct test, is consistent with the theoretical specifications: the GTX 1070 is in a different performance tier entirely.

Architecture Differences

The two GPUs are built on fundamentally different architectures from different process nodes. The GTX 950A uses the GM107 chip, based on the Maxwell architecture, fabricated on a 28 nm process at TSMC. The GTX 1070 uses the GP104 chip, based on the Pascal architecture, also from TSMC but on a much denser 16 nm node. This process shrink is critical: the GTX 1070 packs 7,200 million transistors into a 314 mm² die, while the GTX 950A contains only 1,870 million transistors on a 148 mm² die. The transistor density tells the story clearly — 22.9M per mm² for Pascal versus 12.6M per mm² for Maxwell — meaning the GTX 1070 crams nearly twice the logic into the same area.

The core configurations are vastly different. The GTX 950A has 640 shading units, 40 texture mapping units, and 16 ROPs. The GTX 1070 scales this up to 1,920 shading units, 120 TMUs, and 64 ROPs. This is a 3x increase in shading units and a 4x increase in ROPs, which directly enables the higher pixel and texture fill rates seen in the benchmarks. The GTX 1070 also carries a significant clock-speed advantage: a base clock of 1506 MHz and boost of 1683 MHz, compared to the GTX 950A's 993 MHz base and 1124 MHz boost. Higher clocks plus more cores equals the massive FP32 lead.

Memory architecture is another major divergence. The GTX 950A uses 2 GB of DDR3 on a 128-bit bus, yielding just 32.03 GB/s of bandwidth. The GTX 1070 uses 8 GB of GDDR5 on a 256-bit bus, achieving 256.3 GB/s. That is an 8x improvement in memory bandwidth, which is crucial for modern game assets and compute workloads. The effective memory clock also jumps from 2 Gbps to 8 Gbps. The GTX 1070 supports DirectX 12 (12_1), while the GTX 950A only supports DirectX 12 (11_0), indicating a difference in feature-level support for modern rendering techniques. Both support OpenGL 4.6 and Vulkan 1.4, but the Pascal chip is more future-proof on the Microsoft API front.

FAQ

Q: Which GPU is faster in the Geekbench OpenCL benchmark?

A: The NVIDIA GeForce GTX 1070 is significantly faster, scoring 44,700 versus the GTX 950A's 10,273, a 77% difference in favor of the GTX 1070.

Q: How do the memory configurations compare?

A: The GTX 950A has 2 GB of DDR3 on a 128-bit bus with 32.03 GB/s bandwidth. The GTX 1070 has 8 GB of GDDR5 on a 256-bit bus with 256.3 GB/s bandwidth, an 8x increase in memory bandwidth.

Q: What architectural generation does each GPU belong to?

A: The GTX 950A is based on the Maxwell architecture (GM107 chip), while the GTX 1070 is based on the Pascal architecture (GP104 chip).

Q: Are there differences in API support?

A: Yes. The GTX 1070 supports DirectX 12 (12_1), whereas the GTX 950A only supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4.

Q: What is the transistor count and process node for each?

A: The GTX 950A has 1,870 million transistors on a 28 nm process with a 148 mm² die. The GTX 1070 has 7,200 million transistors on a 16 nm process with a 314 mm² die.

Q: Which card has a higher pixel fill rate?

A: The GTX 1070 has a pixel rate of 107.7 GPixel/s, which is 6x higher than the GTX 950A's 17.98 GPixel/s.

Specification Differences

| Specification | NVIDIA GeForce GTX 950A | NVIDIA GeForce GTX 1070 |

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

| Chip | GM107 | GP104 |

| Architecture | Maxwell | Pascal |

| Process Node | 28 nm | 16 nm |

| Transistors | 1,870 million | 7,200 million |

| Die Size | 148 mm² | 314 mm² |

| Transistor Density | 12.6M / mm² | 22.9M / mm² |

| Base Clock | 993 MHz | 1506 MHz |

| Boost Clock | 1124 MHz | 1683 MHz |

| Memory Clock | 2 Gbps effective | 8 Gbps effective |

| Memory Size | 2 GB | 8 GB |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus Width | 128 bit | 256 bit |

| Memory Bandwidth | 32.03 GB/s | 256.3 GB/s |

| Shading Units | 640 | 1920 |

| TMUs | 40 | 120 |

| ROPs | 16 | 64 |

| Pixel Rate | 17.98 GPixel/s | 107.7 GPixel/s |

| Texture Rate | 44.96 GTexel/s | 202.0 GTexel/s |

| FP32 Performance | 1,438.7 GFLOPS | 6.463 TFLOPS |

| TDP | 75 W | 150 W |

| Slot Width | MXM Module | Dual-slot |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | Not specified | 450 W |

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

| Display Outputs | Portable Device Dependent | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a |

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

The Verdict

The data is unambiguous. The NVIDIA GeForce GTX 1070 outperforms the GTX 950A in every measurable way, from raw compute to memory bandwidth. The 77% lead in the only direct benchmark is compounded by a 4.5x advantage in FP32 throughput, a 6x advantage in pixel rate, and an 8x advantage in memory bandwidth. If the workload requires any modern gaming, high-resolution rendering, or compute tasks, the GTX 1070 is the only viable choice between the two.

The GTX 950A, however, is not without a niche. Its 75 W TDP and lack of external power connectors make it suitable for low-power or portable MXM-based systems, where a dedicated card is required but space and thermal headroom are minimal. Its nearest rivals in the data are all within a 2% performance band, indicating it is a serviceable entry-level part. For a user constrained to a legacy MXM slot with a 75 W budget, the GTX 950A is a functional if unspectacular option.

For anyone building or upgrading a desktop system with a standard PCIe slot, the choice is clear. The GTX 1070's 150 W TDP is reasonable for the performance class, and its 8 GB of GDDR5 memory ensures it remains relevant for larger textures and datasets. The GTX 1070's benchmark scores — including a Passmark G3D score of 13,498 and a Geekbench Vulkan score of 22,121 — demonstrate broad capability across APIs. The GTX 950A has no such breadth in the data. The verdict is straightforward: the GTX 1070 is the performance pick, and the GTX 950A serves only the most constrained, low-power legacy environments.