NVIDIA GeForce GTX 1070 vs NVIDIA GeForce GTX 760 Comparison

The NVIDIA GeForce GTX 1070 and the NVIDIA GeForce GTX 760 represent two distinct eras of NVIDIA's GPU design, separated by three generations of architecture. The benchmark data shows a decisive generational leap, with the GTX 1070 winning all three head-to-head tests by substantial margins, yet the GTX 760 still holds a place in the historical record as a capable 1080p card from its time. This analysis compares their raw performance, architectural foundations, and the specific use cases where each card's strengths are most relevant.

Head-to-Head Benchmarks

The performance gap between these two cards is stark, as demonstrated by the three common benchmark tests. In Geekbench Metal, the GTX 1070 scores 18,801 points, a massive 315.6% higher than the GTX 760's 4,524 points. This is not a marginal improvement; it represents a multi-generational shift in compute capability, particularly relevant for applications leveraging Apple's Metal API. The GTX 1070 is essentially in a different performance class.

Similarly, in Geekbench OpenCL, the GTX 1070 posts a score of 44,700, which is 295.6% ahead of the GTX 760's 11,299. This test highlights the raw compute throughput of the GP104 chip, and the delta shows that workloads heavily reliant on general-purpose GPU computing will see dramatic speedups on the newer card. For tasks like video encoding or scientific simulations, the GTX 1070 provides a transformative upgrade.

The closest contest comes in Geekbench Vulkan, where the GTX 1070 scores 22,121 against the GTX 760's 12,551. While still a decisive 76.2% win for the GTX 1070, this smaller margin suggests that the older Kepler architecture handles modern low-level graphics APIs relatively better than its compute performance would imply. However, a 76.2% lead is still a crushing defeat, meaning even in the GTX 760's most favorable benchmark, the GTX 1070 remains far superior.

The overall average benchmark scores confirm this hierarchy. The GTX 1070 averages 9,780 points across all its tested benchmarks, placing it in the 47th percentile of all GPUs. The GTX 760, with an average of 9,458 points, sits in the 46th percentile. While the percentile ranks are nearly identical, the average score is misleading due to the different test suites; the head-to-head data is the definitive comparison. The GTX 1070's nearest rivals are the AMD FirePro W5000 (deltaPct -0.2), NVIDIA Quadro M2000M (deltaPct -0.5), and NVIDIA Tesla M10 (deltaPct 0.6), placing it in a tight cluster of mid-range professional cards. The GTX 760's nearest rivals include the NVIDIA GeForce GTX TITAN BLACK (deltaPct -0.2) and AMD Radeon R7 M380 (deltaPct 1.6), showing it competes with older high-end parts, but not with the GTX 1070's generation.

Architecture Differences

The architectural chasm between these two cards is the root cause of their performance disparity. The GTX 1070 is built on the Pascal architecture, specifically the GP104 chip, fabricated on a 16 nm process at TSMC. In contrast, the GTX 760 uses the Kepler architecture, with the GK104 chip, manufactured on a 28 nm process. This node shrink is fundamental: it allows the GTX 1070 to pack 7,200 million transistors into a 314 mm² die, achieving a transistor density of 22.9 million per square millimeter. The GTX 760, by comparison, contains only 3,540 million transistors on a slightly smaller 294 mm² die, yielding a density of just 12.0 million per square millimeter. This nearly doubling of transistor density is the primary enabler of the GTX 1070's performance.

The core configurations diverge significantly as well. The GTX 1070 features 1,920 shading units, 120 texture mapping units (TMUs), and 64 raster output units (ROPs). The GTX 760 is equipped with 1,152 shading units, 96 TMUs, and only 32 ROPs. This means the GTX 1070 has 66.7% more shading units and double the ROPs, directly translating to higher pixel and texture throughput. The pixel rate for the GTX 1070 is 107.7 GPixel/s versus 24.77 GPixel/s for the GTX 760, and the texture rate is 202.0 GTexel/s versus 99.07 GTexel/s.

Clock speeds and memory also tell the story of progress. The GTX 1070 runs at a base clock of 1506 MHz with a boost of 1683 MHz, while the GTX 760 operates at 980 MHz base and 1032 MHz boost. This higher clock speed, combined with the architectural efficiency, yields an FP32 performance of 6.463 TFLOPS for the GTX 1070, compared to just 2.378 TFLOPS for the GTX 760. The GTX 1070 also supports FP16 compute at 101.0 GFLOPS (1:64), a feature absent from the GTX 760. Memory configurations differ as well: the GTX 1070 offers 8 GB of GDDR5 on a 256-bit bus with 256.3 GB/s bandwidth, while the GTX 760 has 2 GB of GDDR5 on the same 256-bit bus but with a lower 192.3 GB/s bandwidth. The GTX 1070's memory clock is 2002 MHz (8 Gbps effective) versus 1502 MHz (6 Gbps effective) on the GTX 760.

Feature support also reflects their different generations. The GTX 1070 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 760 supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6. Display outputs differ: the GTX 1070 has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a, whereas the GTX 760 offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The GTX 1070 is also more power-efficient despite its higher performance, with a TDP of 150 W versus 170 W for the GTX 760, and it requires a single 8-pin power connector compared to the GTX 760's two 6-pin connectors.

Where Each One Wins

Based on the benchmark data, the GTX 1070 wins in every measurable category. There are no tests where the GTX 760 outperforms it, as the winsA count is 3 and winsB is 0. However, the nature of the wins suggests different levels of dominance. The GTX 1070's largest victory is in Metal (315.6% delta), followed closely by OpenCL (295.6% delta), indicating that its compute and general-purpose processing capabilities are its strongest assets. The Vulkan test, while still a 76.2% win, shows a relatively smaller gap, suggesting that the GTX 760 is less outdated in gaming-oriented, low-level API workloads.

For a modern user, the GTX 1070 is the clear choice for any application that benefits from high compute throughput, large memory capacity (8 GB vs 2 GB), and modern API support. This includes contemporary gaming at high settings, video editing, 3D rendering, and any GPU-accelerated compute task. The GTX 760, on the other hand, could still serve in a legacy system running older DirectX 11 games or as a basic display adapter, but its 2 GB memory buffer is a severe limitation for modern titles, and its lack of support for full DirectX 12 (11_0 vs 12_1) means it cannot fully utilize newer graphics features.

The GTX 1070's higher TDP efficiency (150 W vs 170 W) also makes it a better fit for systems with modest power supplies, despite both cards suggesting a 450 W PSU. The GTX 1070 is also physically longer at 267 mm (10.5 inches) versus 241 mm (9.5 inches) for the GTX 760, which is a consideration for small form factor cases, though the GTX 760 has unspecified height and width dimensions.

FAQ

Q: Is the GTX 1070 worth upgrading to from the GTX 760?

A: The benchmark data indicates a massive generational leap. The GTX 1070 is 315.6% faster in Metal, 295.6% faster in OpenCL, and 76.2% faster in Vulkan. This translates to a transformative improvement in any GPU-intensive workload.

Q: Which card has better memory specifications?

A: The GTX 1070 has a clear advantage with 8 GB of GDDR5 memory versus the GTX 760's 2 GB. Both use a 256-bit bus, but the GTX 1070 achieves 256.3 GB/s bandwidth compared to 192.3 GB/s on the GTX 760, due to its higher effective memory clock of 8 Gbps versus 6 Gbps.

Q: How do their power requirements compare?

A: The GTX 1070 is more power-efficient, with a TDP of 150 W, while the GTX 760 has a TDP of 170 W. The GTX 1070 uses a single 8-pin power connector, whereas the GTX 760 requires two 6-pin connectors. Both suggest a 450 W power supply.

Q: Do both cards support modern APIs?

A: They support different versions. The GTX 1070 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 760 supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Q: What is the physical size difference?

A: The GTX 1070 is longer, measuring 267 mm (10.5 inches) in length, compared to the GTX 760's 241 mm (9.5 inches). The GTX 1070 also has specified dimensions of 112 mm height and 40 mm width, while the GTX 760 only lists its length.

Q: Which card has a higher average benchmark score?

A: The GTX 1070 has a higher average benchmark score of 9,780 points, compared to the GTX 760's 9,458 points. However, this difference is small and the head-to-head tests show a much larger performance gap, as the average scores are drawn from different test suites.

The Verdict

The data is unequivocal: the NVIDIA GeForce GTX 1070 is the superior card in every benchmark test where they are directly compared. It offers more than double the performance in compute-heavy workloads like Metal and OpenCL, and a 76.2% advantage even in the Vulkan test where the GTX 760 is relatively strongest. For any user building a system for modern gaming, content creation, or GPU computing, the GTX 1070 is the only rational choice between these two.

The GTX 760, while historically significant, is a product of a different era. Its 2 GB memory capacity and lower compute throughput make it unsuitable for contemporary demanding applications. Its only potential role is in a legacy system where its smaller physical footprint (241 mm vs 267 mm) might be an advantage in a compact case, but this is a niche scenario. The GTX 1070's superior performance, higher memory capacity, and better power efficiency (150 W vs 170 W) make it a comprehensive winner.

Specification Differences

The following table lists only the fields where the two cards differ, based on the fact pack data:

| Specification | NVIDIA GeForce GTX 1070 | NVIDIA GeForce GTX 760 |

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

| Architecture | Pascal | Kepler |

| Chip | GP104 | GK104 |

| Process Node | 16 nm | 28 nm |

| Transistors | 7,200 million | 3,540 million |

| Die Size | 314 mm² | 294 mm² |

| Transistor Density | 22.9M / mm² | 12.0M / mm² |

| Base Clock | 1506 MHz | 980 MHz |

| Boost Clock | 1683 MHz | 1032 MHz |

| Memory Clock | 2002 MHz (8 Gbps effective) | 1502 MHz (6 Gbps effective) |

| Memory Size | 8 GB | 2 GB |

| Memory Bandwidth | 256.3 GB/s | 192.3 GB/s |

| Shading Units | 1920 | 1152 |

| TMUs | 120 | 96 |

| ROPs | 64 | 32 |

| Pixel Rate | 107.7 GPixel/s | 24.77 GPixel/s |

| Texture Rate | 202.0 GTexel/s | 99.07 GTexel/s |

| FP32 | 6.463 TFLOPS | 2.378 TFLOPS |

| FP16 | 101.0 GFLOPS (1:64) | null |

| TDP | 150 W | 170 W |

| Power Connectors | 1x 8-pin | 2x 6-pin |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a | 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 |

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

| Vulkan Support | 1.4 | 1.2.175 |

| Length | 267 mm (10.5 inches) | 241 mm (9.5 inches) |

| Height | 112 mm (4.4 inches) | null |

| Width | 40 mm (1.6 inches) | null |

| Release Date | 2016-06-09 | 2013-06-24 |

| Predecessor | GeForce 900 | GeForce 600 |

| Successor | GeForce 20 | GeForce 900 |

| Launch MSRP | 379 USD | 249 USD |