NVIDIA GeForce GTX 760

The NVIDIA GeForce GTX 760 is a Kepler-architecture graphics card from the GeForce 700 generation, built on TSMC’s 28 nm process with 3,540 million transistors on a 294 mm² die. It carries 2 GB of GDDR5 memory on a 256-bit bus, delivering 192.3 GB/s of bandwidth, with a base clock of 980 MHz and a boost clock of 1032 MHz. The card’s average benchmark score across Geekbench Metal, OpenCL, and Vulkan tests is 11259, placing it at the 50th percentile of all GPUs, which positions it as a middle-of-the-road performer in the current database landscape. This analysis examines how it stacks against its nearest rivals, who should consider it, its benchmark behavior, and its power requirements.

How It Compares

The GTX 760’s closest competitor is the AMD Radeon Pro WX 3200, which posts an average score of 11228. The delta is a mere 0.3%, meaning the GTX 760 is statistically tied with the WX 3200 in aggregate performance. In practical terms, benchmark results indicate that neither card holds a meaningful advantage across the three Geekbench compute tests; the difference is within run-to-run variance. For users deciding between these two, the data suggests the choice hinges on other factors—such as driver support or display outputs—rather than raw compute throughput, since the scores are nearly indistinguishable.

Next is the AMD FirePro W4300, with an average score of 11187, putting it 0.6% behind the GTX 760. This is still a razor-thin margin, equivalent to a few dozen points in an aggregate score of over 11,000. The GTX 760 edges ahead, but the lead is so small that it would not be perceptible in real-world workloads. The benchmark evidence implies that the FirePro W4300 and GTX 760 occupy the same performance tier, with the NVIDIA card holding a marginal edge that is unlikely to translate into a measurable difference in any specific application.

The NVIDIA GeForce GTX 870M, a mobile part, scores 11173, which is 0.8% lower than the GTX 760. This is a slightly larger gap than the two AMD rivals, but still within the range of what could be considered a statistical tie. The desktop GTX 760 outperforms the mobile GTX 870M by a small margin, which aligns with the expectation that a desktop card with a 170 W TDP can sustain higher clocks than a mobile variant, even if the underlying architecture is similar. Still, the delta is modest, and the two would be interchangeable in many compute scenarios.

Finally, the NVIDIA GeForce GTX 780M, another mobile part, averages 11012, trailing the GTX 760 by 2.2%. This is the largest gap among the listed rivals, yet it remains a single-digit percentage difference. The GTX 760 holds a clearer advantage here, likely due to higher sustained boost clocks and dedicated desktop power delivery. The data shows that the GTX 760 is the faster card by a measurable margin, but the 780M is not far behind, suggesting that Kepler-based mobile GPUs from the same generation were relatively close to their desktop counterparts in compute performance.

Who Should Consider It

Given that the GTX 760 sits at the 50th percentile of all GPUs, it is not a card for high-resolution or maximum-settings gaming in modern titles. Its compute scores—Geekbench Metal at 6288, OpenCL at 13613, and Vulkan at 13876—indicate a card that can handle older or less demanding games at 1080p with medium to high settings, but it will struggle with 1440p or 4K resolutions or with games that require substantial shader and texture throughput. The pixel rate of 24.77 GPixel/s and texture rate of 99.07 GTexel/s are modest by today’s standards, meaning that fill-rate-bound scenarios will expose its limitations.

For users with a 1080p monitor and a preference for esports titles or games from the early-to-mid 2010s, the GTX 760 remains a viable option. Its 2 GB of GDDR5 memory is sufficient for that era’s textures, but modern games with high-resolution texture packs will exceed that capacity, leading to stuttering or reduced detail. The 1152 shading units and 32 ROPs provide a baseline that can produce playable frame rates in lighter titles, but the card is best suited for users who are not chasing the latest AAA releases at high settings.

Conversely, users targeting 1440p or higher, or those who play VR-heavy or compute-intensive applications, should look elsewhere. The GTX 760’s FP32 throughput of 2.378 TFLOPS is far below that of contemporary cards, and its DirectX support is limited to version 12 (11_0), which means it lacks the feature set required for some modern effects. The card is also end-of-life, so driver optimizations for new games are unlikely to be a priority. In summary, the GTX 760 is for budget-conscious 1080p gamers with modest expectations, not for enthusiasts or professionals needing high-end compute.

Benchmark Performance

The GTX 760’s average benchmark score of 11259 is derived from three distinct Geekbench tests, each measuring a different compute API. The OpenCL score of 13613 and Vulkan score of 13876 are close, indicating that the card performs similarly across these two APIs, while the Metal score of 6288 is substantially lower—less than half of the other two. This discrepancy is notable: Metal is Apple’s API, and the GTX 760’s drivers for that platform may be less optimized, or the test itself may stress different hardware paths. For users relying on Metal, the card’s performance is significantly weaker than its OpenCL or Vulkan numbers suggest.

Relative to its nearest rivals, the GTX 760’s 0.3% lead over the Radeon Pro WX 3200 is negligible, but its 2.2% advantage over the GTX 780M is more tangible. In a multi-core compute workload, a 2.2% delta means the GTX 760 completes the task approximately 2% faster, which could translate to a few seconds in a long render. The 0.6% and 0.8% leads over the FirePro W4300 and GTX 870M, respectively, are similarly minor, indicating that all four of these cards are clustered tightly in performance.

Breaking down the scores further, the GTX 760’s Vulkan result of 13876 is the highest of its three benchmarks, suggesting that the card’s compute units are well-utilized under Vulkan’s explicit control model. The OpenCL score of 13613 is only 1.9% lower, showing consistent performance across these two APIs. The Metal score, however, is 54% lower than the Vulkan score, which is a massive gap that dwarfs any differences between the GTX 760 and its rivals. This means that for users on macOS or iOS environments, the GTX 760’s standing would be much worse than its average score implies, potentially placing it below many of its listed rivals in Metal-specific tests.

The 50th percentile ranking reinforces the picture of a mid-range card. Half of all GPUs in the database are faster, and half are slower. When compared to the nearest rivals, the GTX 760 is at the center of a tight cluster—the spread from the fastest (GTX 760 at 11259) to the slowest (GTX 780M at 11012) is only 2.2%, so all five cards (including the WX 3200, W4300, and 870M) are functionally equivalent in aggregate compute. The data indicates that the GTX 760’s position is defined more by its API-specific quirks than by its overall speed relative to peers.

Power and Cooling

The GTX 760 has a thermal design power (TDP) of 170 W, which is a moderate figure for a desktop card from its era. The suggested power supply is 450 W, meaning that most standard desktop systems with a quality 450 W PSU can accommodate it, though users with older or lower-wattage units should verify their headroom. The card requires two 6-pin power connectors, which is a common configuration for mid-range GPUs of that generation; these connectors must be present on the PSU’s cables or via adapters.

Cooling is handled by a dual-slot design, which is typical for a 170 W card. The card’s length is 241 mm (9.5 inches), so it will fit in most mid-tower cases, but users with compact cases should measure clearance. The dual-slot cooler is designed to dissipate the 170 W of heat through a combination of heatsink and fans, and while no specific noise or temperature figures are provided, the data shows that the card’s power draw is within the range that a dual-slot air cooler can manage without excessive noise. The 28 nm process node helps keep power in check, but 170 W is still substantial enough that proper case airflow is recommended.

The memory operates at 1502 MHz (6 Gbps effective), which contributes to the 192.3 GB/s bandwidth. This bandwidth is sufficient for the 2 GB frame buffer, but it is not exceptionally high by modern standards. The power delivery system, with two 6-pin connectors, provides more headroom than a single 6-pin, which is appropriate for the 170 W TDP. Users should ensure their PSU has the correct connectors before installation, as adapters from Molex to 6-pin may not be reliable under sustained load. The PCIe 3.0 x16 bus interface is backward-compatible with older slots, but a PCIe 2.0 slot may slightly bottleneck the card, though the impact is likely minimal given the card’s performance tier.

FAQ

Q: What is the GTX 760’s average benchmark score?

A: The average benchmark score is 11259, based on Geekbench Metal (6288), OpenCL (13613), and Vulkan (13876) tests.

Q: How does the GTX 760 compare to the AMD Radeon Pro WX 3200?

A: The GTX 760 is 0.3% ahead of the Radeon Pro WX 3200 in average score, which is a negligible difference in practical terms.

Q: What is the GTX 760’s TDP and recommended power supply?

A: The TDP is 170 W, and the suggested PSU is 450 W. It requires two 6-pin power connectors.

Q: Does the GTX 760 support modern graphics APIs?

A: It supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175, but the DirectX version is limited to feature level 11_0.

Q: What is the memory configuration of the GTX 760?

A: It has 2 GB of GDDR5 memory on a 256-bit bus, with a bandwidth of 192.3 GB/s and an effective memory clock of 6 Gbps.

Q: Is the GTX 760 still in production?

A: No, it is end-of-life, with a release date of June 24, 2013, and its successor is the GeForce 900 series.