NVIDIA GeForce GTX 1070 vs NVIDIA GeForce GTX 960M Comparison

The NVIDIA GeForce GTX 1070 and NVIDIA GeForce GTX 960M occupy different ends of the mobile and desktop spectrum, yet the data places them in adjacent percentiles. The GTX 1070 ranks in the 47th percentile of all GPUs, while the GTX 960M sits at the 46th percentile, but their average benchmark scores—9780 versus 9645—reveal a narrow overall margin of about 1.4%. This proximity is deceptive, as the head-to-head benchmark data shows a massive performance gulf in compute workloads, where the GTX 1070 leads by over 300% in OpenCL. The GTX 960M is a Maxwell-based mobile chip designed for portability, while the GTX 1070 is a Pascal-based desktop card with far greater resources. The sections below dissect their architectural differences, benchmark wins, and specification gaps to clarify which device suits which workload, based strictly on the provided facts.

The Verdict

From the data, the NVIDIA GeForce GTX 1070 is the decisive choice for any workload involving general-purpose compute or graphics acceleration that leverages OpenCL or Vulkan. In the head-to-head benchmarks, it wins both tests outright: a 304.7% higher score in Geekbench OpenCL and a 168.3% higher score in Geekbench Vulkan. The GTX 960M, meanwhile, records no wins in any head-to-head comparison. For users prioritizing raw performance in compute-heavy tasks, the GTX 1070 is the only rational pick, as its average benchmark score of 9780 slightly exceeds the GTX 960M's 9645, despite the latter's lower power envelope.

However, the GTX 960M is not without merit for specific use cases. Its 75 W TDP and MXM Module form factor indicate a design for thin-and-light laptops, whereas the GTX 1070 requires 150 W, a dual-slot cooler, and a 1x 8-pin power connector. If the priority is portability and battery life, the GTX 960M's lower power draw makes it the feasible option for mobile systems. But strictly from benchmark performance, the GTX 1070 dominates: it scores 44,700 in Geekbench OpenCL against the GTX 960M's 11,045, and 22,121 in Geekbench Vulkan against 8,245. The verdict is clear: pick the GTX 1070 for maximum compute throughput, and pick the GTX 960M only when power constraints and form factor are non-negotiable.

Architecture Differences

The two GPUs stem from different architectural generations. The GTX 1070 is built on the Pascal architecture, fabricated on a 16 nm process at TSMC, while the GTX 960M uses the older Maxwell architecture on a 28 nm process, also from TSMC. This process shrink is significant: the GTX 1070 packs 7,200 million transistors into a 314 mm² die, yielding a transistor density of 22.9 million per square millimeter. In contrast, the GTX 960M houses 1,870 million transistors on a 148 mm² die, with a density of 12.6 million per square millimeter. The Pascal chip (GP104) thus has nearly four times the transistor count and more than double the die area, explaining its higher resource allocation.

Core counts amplify the difference. The GTX 1070 features 1920 shading units, 120 texture mapping units (TMUs), and 64 raster output units (ROPs). The GTX 960M, using the GM107 chip, has only 640 shading units, 40 TMUs, and 16 ROPs—one-third the shading units and a quarter of the ROPs. Clock speeds also favor the desktop card: the GTX 1070 runs at a base of 1506 MHz and boosts to 1683 MHz, while the GTX 960M operates at 1097 MHz base and 1176 MHz boost. Memory configurations differ starkly: the GTX 1070 uses 8 GB of GDDR5 on a 256-bit bus, achieving 256.3 GB/s bandwidth, whereas the GTX 960M has 4 GB of GDDR5 on a 128-bit bus, delivering only 80.19 GB/s. The GTX 1070's memory clock is 2002 MHz (8 Gbps effective), compared to the GTX 960M's 1253 MHz (5 Gbps effective).

API support also diverges. The GTX 1070 supports DirectX 12 (12_1), while the GTX 960M is limited to DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4, but the GTX 1070's higher feature level in DirectX 12 is a notable architectural advantage. The GTX 960M has no listed FP16 performance, while the GTX 1070 achieves 101.0 GFLOPS (1:64 ratio). These architectural gaps—transistor count, process node, core counts, memory bus, and API level—collectively drive the benchmark disparity.

Where Each One Wins

Based on the head-to-head data, the GTX 1070 wins in every measured category. It leads by 304.7% in Geekbench OpenCL (44,700 versus 11,045) and by 168.3% in Geekbench Vulkan (22,121 versus 8,245). These are compute-oriented workloads, suggesting the GTX 1070 is the superior choice for GPU-accelerated applications such as rendering, machine learning inference, or scientific simulation that leverage OpenCL or Vulkan compute. The GTX 1070's broader benchmark suite also shows strength in DirectX titles: it scores 82 in Passmark DirectX 10, 100 in DirectX 11, 48 in DirectX 12, and 197 in DirectX 9, alongside a Passmark G3D score of 13,498 and a GPU Compute score of 6,102. The GTX 960M lacks these specific benchmark entries, so direct comparison is limited to the two Geekbench tests.

The GTX 960M wins nowhere in the head-to-head data, but it holds a niche based on power and form factor. With a 75 W TDP and no power connectors, it can fit into portable devices, whereas the GTX 1070's 150 W TDP and dual-slot design require a desktop chassis or a large laptop with adequate cooling. The GTX 960M's MXM-B (3.0) interface and "Portable Device Dependent" display outputs indicate it is designed for upgradeable laptops, while the GTX 1070 uses PCIe 3.0 x16 and offers multiple display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a). For users who must run on battery power or within a slim chassis, the GTX 960M is the only viable option, but it sacrifices compute performance dramatically.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The GTX 1070 has an average benchmark score of 9780, while the GTX 960M scores 9645, a difference of 135 points (about 1.4%). Both GPUs rank in the 47th and 46th percentiles of all GPUs, respectively.

Q: How much faster is the GTX 1070 in OpenCL?

A: In Geekbench OpenCL, the GTX 1070 scores 44,700 against the GTX 960M's 11,045, representing a 304.7% advantage—more than four times the performance.

Q: What is the memory bandwidth difference?

A: The GTX 1070 offers 256.3 GB/s bandwidth on a 256-bit bus with 8 GB GDDR5, whereas the GTX 960M provides 80.19 GB/s on a 128-bit bus with 4 GB GDDR5. The GTX 1070's bandwidth is roughly 3.2 times higher.

Q: Do both GPUs support the same DirectX version?

A: No. The GTX 1070 supports DirectX 12 (12_1), while the GTX 960M supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4.

Q: Which GPU has more shading units?

A: The GTX 1070 has 1920 shading units, compared to the GTX 960M's 640 shading units. This is a threefold difference that directly impacts pixel and texture processing rates.

Q: Is the GTX 960M suitable for desktop use?

A: The data indicates the GTX 960M is an MXM Module with no power connectors and a 75 W TDP, designed for portable devices. It lacks the GTX 1070's PCIe 3.0 x16 interface and desktop display outputs, so it is not a drop-in desktop card.

Head-to-Head Benchmarks

The only two shared benchmarks between the GTX 1070 and GTX 960M are Geekbench OpenCL and Geekbench Vulkan, and the GTX 1070 wins both by wide margins. In Geekbench OpenCL, the GTX 1070 achieves a score of 44,700, while the GTX 960M manages only 11,045. The delta is 304.7%, meaning the GTX 1070 is over four times faster in this compute-heavy test. This is the largest single gap in the head-to-head data, reflecting the combination of higher shading unit count (1920 versus 640), faster memory bandwidth (256.3 GB/s versus 80.19 GB/s), and higher boost clocks (1683 MHz versus 1176 MHz).

In Geekbench Vulkan, the GTX 1070 scores 22,121 against the GTX 960M's 8,245, a 168.3% advantage. While this is a smaller relative gap than OpenCL, it still represents a doubling of performance. Vulkan is a low-overhead API that scales with raw compute resources, and the GTX 1070's superior FP32 throughput (6.463 TFLOPS versus 1.505 TFLOPS) explains this outcome. The GTX 1070's pixel rate of 107.7 GPixel/s and texture rate of 202.0 GTexel/s dwarf the GTX 960M's 18.82 GPixel/s and 47.04 GTexel/s, respectively, confirming that the GTX 1070 is designed for high-resolution rendering and compute tasks.

The GTX 1070's additional benchmark scores—Passmark G3D at 13,498 and GPU Compute at 6,102—are not matched by the GTX 960M in the data, but they reinforce its overall performance profile. Notably, the GTX 1070's nearest rivals include the AMD FirePro W5000 (delta -0.2%) and NVIDIA Quadro M2000M (delta -0.5%), while the GTX 960M's nearest rivals are the NVIDIA Quadro K5000 (delta 0.1%) and AMD Radeon Pro WX 2100 (delta -0.1%). This shows both GPUs sit in a crowded mid-range field, but the GTX 1070's head-to-head wins are decisive.

Specification Differences

The specification sheets for the two GPUs differ in nearly every measurable field. Process node: the GTX 1070 uses 16 nm, the GTX 960M uses 28 nm. Transistor count: 7,200 million versus 1,870 million. Die size: 314 mm² versus 148 mm². Transistor density: 22.9M / mm² versus 12.6M / mm². Base clock: 1506 MHz versus 1097 MHz. Boost clock: 1683 MHz versus 1176 MHz. Memory clock: 2002 MHz (8 Gbps effective) versus 1253 MHz (5 Gbps effective). Memory size: 8 GB versus 4 GB. Memory type is GDDR5 for both, but bus width is 256-bit versus 128-bit, and bandwidth is 256.3 GB/s versus 80.19 GB/s.

Shading units: 1920 versus 640. TMUs: 120 versus 40. ROPs: 64 versus 16. Pixel rate: 107.7 GPixel/s versus 18.82 GPixel/s. Texture rate: 202.0 GTexel/s versus 47.04 GTexel/s. FP32 performance: 6.463 TFLOPS versus 1.505 TFLOPS. FP16: 101.0 GFLOPS (1:64) versus null. TDP: 150 W versus 75 W. Slot width: dual-slot versus MXM Module. Power connectors: 1x 8-pin versus none. Suggested PSU: 450 W versus null. Bus interface: PCIe 3.0 x16 versus MXM-B (3.0). Display outputs: 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a versus portable device dependent. DirectX: 12 (12_1) versus 12 (11_0). The GTX 1070 has a launch MSRP of 379 USD; the GTX 960M has no recorded launch MSRP. The GTX 1070's dimensions are 267 mm x 112 mm x 40 mm, while the GTX 960M lists no dimensions. Release dates differ: the GTX 1070 launched on 2016-06-09, the GTX 960M on 2015-03-12. Both are end-of-life products.