NVIDIA GeForce GTX 680 vs NVIDIA GeForce GTX 960A Comparison

The NVIDIA GeForce GTX 680 and the NVIDIA GeForce GTX 960A are both end-of-life mobile and desktop graphics solutions, but they target very different performance brackets. The GTX 680 is a high-end desktop part from the Kepler generation with a large die and substantial compute resources, while the GTX 960A is a low-power MXM module from the Maxwell generation, designed for portable systems. The database shows a single head-to-head benchmark, and the results are decisive: the GTX 680 wins the only recorded test by a wide margin. However, the story is more nuanced than raw performance, as the GTX 960A offers a dramatically different power and form-factor profile. The following analysis walks through the recorded data, specification differences, and architectural choices to determine where each card fits.

Where Each One Wins

The recorded data contains exactly one head-to-head benchmark: Geekbench OpenCL. In that test, the NVIDIA GeForce GTX 680 scores 16,906 points, while the NVIDIA GeForce GTX 960A scores 11,998 points. The delta is 40.9 percent in favor of the GTX 680, meaning the older desktop card delivers substantially higher raw compute throughput in this specific workload. If the task is general-purpose GPU compute via OpenCL, the GTX 680 is the clear winner, and there is no benchmark in the database where the GTX 960A comes out ahead.

That said, the GTX 960A has strengths that do not appear in benchmark scores. Its thermal design power is 75 W, compared to 195 W for the GTX 680. This three-fold difference in power draw is a major advantage for the GTX 960A in compact or battery-powered systems. The GTX 960A is an MXM module with no power connectors, while the GTX 680 requires two 6-pin connectors and a 450 W suggested power supply. The GTX 960A also carries no suggested PSU figure, underscoring its low-power nature. For scenarios where power delivery is limited, such as thin laptops or small form-factor industrial PCs, the GTX 960A is the only viable option, despite its lower compute scores.

In terms of overall performance percentile, the GTX 680 sits at the 55th percentile among all GPUs in the database, with an average benchmark score of 14,150. The GTX 960A is at the 51st percentile with an average score of 11,998. While the GTX 680 is ahead, both cards are in the lower-middle range of overall GPU performance, meaning neither is a top-tier part today. The GTX 680’s nearest rivals include the NVIDIA Tesla K10, which scores 14,029 (0.9 percent behind), and the NVIDIA GeForce GTX 1070 Ti, which scores 14,277 (0.9 percent ahead). The GTX 960A’s nearest rivals include the NVIDIA GeForce GTX 1080 at 11,960 (0.3 percent behind) and the AMD Radeon RX 6500 XT at 11,842 (1.3 percent ahead). These proximity figures show that the GTX 680 competes with mid-range desktop parts from later generations, while the GTX 960A sits near entry-level modern cards.

FAQ

Q: Which card has the higher benchmark score in the database?

A: The NVIDIA GeForce GTX 680 scores 16,906 in Geekbench OpenCL, while the NVIDIA GeForce GTX 960A scores 11,998. The GTX 680 leads by 40.9 percent.

Q: What is the power consumption difference?

A: The GTX 680 has a TDP of 195 W, while the GTX 960A has a TDP of 75 W. The GTX 960A uses no power connectors, whereas the GTX 680 requires two 6-pin connectors and a 450 W suggested power supply.

Q: Are both cards from the same architecture generation?

A: No. The GTX 680 uses the Kepler architecture with the GK104 chip, while the GTX 960A uses the Maxwell architecture with the GM107 chip. Both are manufactured on a 28 nm process by TSMC.

Q: Do both cards have the same memory configuration?

A: Both have 2 GB of GDDR5 memory, but the GTX 680 uses a 256-bit bus with 192.3 GB/s bandwidth, while the GTX 960A uses a 128-bit bus with 80.19 GB/s bandwidth.

Q: Which card has more shading units?

A: The GTX 680 has 1,536 shading units, while the GTX 960A has 640. The GTX 680 also has 128 texture mapping units and 32 ROPs, compared to 40 TMUs and 16 ROPs for the GTX 960A.

Q: What are the release dates?

A: The GTX 680 was released on March 21, 2012, while the GTX 960A was released on March 12, 2015. Both are end-of-life products.

Head-to-Head Benchmarks

The only direct comparison in the database is the Geekbench OpenCL test. The GTX 680 records 16,906 points, and the GTX 960A records 11,998 points. The delta is 40.9 percent, which is a substantial margin. To put this in context, consider the GTX 680’s nearest rivals: it is only 0.9 percent behind the GeForce GTX 1070 Ti and 1.2 percent behind the Intel Iris Xe MAX Graphics. Meanwhile, the GTX 960A is only 0.3 percent behind the GeForce GTX 1080 and 2.7 percent ahead of the GeForce GTX 1660. This means that despite the large gap between the two cards in this test, the GTX 960A is actually closer in score to some high-end desktop parts than its raw numbers might suggest, because those rivals also score in the 11,000 to 12,000 range. The GTX 680, on the other hand, sits in a slightly higher performance tier, closer to cards like the GTX 1070 Ti.

The 40.9 percent delta is driven by the GTX 680’s much larger compute configuration. It has more than double the shading units (1,536 vs. 640), triple the texture mapping units (128 vs. 40), and double the ROPs (32 vs. 16). Its FP32 throughput is 3.250 TFLOPS, compared to 1.505 TFLOPS for the GTX 960A, a ratio of 2.16 to 1. The memory bandwidth also differs sharply: 192.3 GB/s versus 80.19 GB/s. These hardware differences translate directly into the benchmark result, with the GTX 680 outperforming the GTX 960A by roughly 41 percent, which is consistent with the overall resource advantage.

There are no other head-to-head tests in the database, so the OpenCL result stands as the sole quantitative comparison. The GTX 680 wins that test, giving it a 1-0 record in wins. The GTX 960A has zero recorded wins.

Specification Differences

The two cards differ in nearly every measurable specification except memory size and type. Both have 2 GB of GDDR5 memory, but the bus width and bandwidth are different. The GTX 680 uses a 256-bit bus, yielding 192.3 GB/s, while the GTX 960A uses a 128-bit bus, yielding 80.19 GB/s. The GTX 680’s memory clock is 1502 MHz (6 Gbps effective), while the GTX 960A runs at 1253 MHz (5 Gbps effective).

Clock speeds are close but not identical. The GTX 680 has a base clock of 1006 MHz and a boost clock of 1058 MHz. The GTX 960A has a base clock of 1097 MHz and a boost clock of 1176 MHz. The GTX 960A actually runs at higher frequencies, but its much smaller chip (1,870 million transistors on a 148 mm² die) cannot compensate for the GTX 680’s larger configuration (3,540 million transistors on a 294 mm² die). Transistor density is slightly higher on the GTX 960A at 12.6 million per mm², versus 12.0 million per mm² for the GTX 680, but this is a minor difference.

Power and physical specifications diverge significantly. The GTX 680 is a dual-slot card with a length of 256 mm (10.1 inches), a height of 111 mm (4.4 inches), and a width of 38 mm (1.5 inches). It requires two 6-pin power connectors and a 450 W suggested power supply. The GTX 960A is an MXM module with no power connectors and no suggested PSU. Its dimensions are not recorded in the database. The GTX 680’s display outputs are 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the GTX 960A’s outputs are described as “Portable Device Dependent,” reflecting its mobile design.

The bus interface also differs: the GTX 680 uses PCIe 3.0 x16, while the GTX 960A uses MXM-B (3.0). The GTX 680 has a launch MSRP of 499 USD, which can be stated once. The GTX 960A has no launch MSRP recorded.

Architecture Differences

The GTX 680 is built on the Kepler architecture, using the GK104 chip, and belongs to the GeForce 600 generation. The GTX 960A is built on the Maxwell architecture, using the GM107 chip, and belongs to the GeForce 900A generation. Both are manufactured by TSMC on a 28 nm process, and both use the same foundry.

The transistor counts and die sizes reflect the different design goals. The GTX 680 packs 3,540 million transistors into a 294 mm² die, while the GTX 960A packs 1,870 million transistors into a 148 mm² die. The GTX 960A’s die is less than half the area, and its transistor count is roughly 53 percent of the GTX 680’s. This smaller footprint is what enables the 75 W TDP, as opposed to the GTX 680’s 195 W.

Shading unit counts are starkly different: 1,536 for the GTX 680 versus 640 for the GTX 960A. The GTX 680 also has 128 TMUs and 32 ROPs, while the GTX 960A has 40 TMUs and 16 ROPs. Pixel rate for the GTX 680 is 33.86 GPixel/s, and texture rate is 135.4 GTexel/s. The GTX 960A achieves 18.82 GPixel/s and 47.04 GTexel/s. FP32 compute is 3.250 TFLOPS for the GTX 680 and 1.505 TFLOPS for the GTX 960A. Neither card has RT cores or tensor cores, and neither has FP16 performance listed.

In terms of API support, both cards support DirectX 12 (11_0) and OpenGL 4.6. The Vulkan versions differ slightly: the GTX 680 supports Vulkan 1.2.175, while the GTX 960A supports Vulkan 1.4. This suggests the newer Maxwell architecture has more modern driver-level API support, even though its raw hardware resources are smaller.

The GTX 680’s predecessor is the GeForce 500 series, and its successor is the GeForce 700 series. The GTX 960A’s predecessor is the GeForce 800A series, and it has no successor listed. The GTX 680 was released in March 2012, while the GTX 960A came three years later in March 2015.

The Verdict

The data is unambiguous on raw performance: the NVIDIA GeForce GTX 680 wins the only head-to-head benchmark, and it wins by 40.9 percent. Its larger Kepler chip, with double the shading units, triple the TMUs, and more than double the memory bandwidth, delivers 16,906 points in Geekbench OpenCL versus 11,998 for the GTX 960A. If the use case is compute-heavy workloads, such as OpenCL acceleration, and power and space are not constraints, the GTX 680 is the superior choice. It also has a higher overall percentile rank (55th vs. 51st) and a higher average benchmark score (14,150 vs. 11,998).

However, the GTX 960A is not without merit. Its 75 W TDP is 120 W lower than the GTX 680’s 195 W, and it requires no external power connectors. It is an MXM module, meaning it can fit into portable or compact systems where a dual-slot, 256 mm long desktop card would be impossible. The GTX 960A’s higher base and boost clocks (1097 MHz and 1176 MHz, versus 1006 MHz and 1058 MHz) show that Maxwell is more efficient per clock, but that efficiency does not overcome the GTX 680’s massive resource advantage in the recorded benchmark.

For a user building a desktop system with a standard power supply and space for a full-length card, the GTX 680 is the clear pick based on the recorded data. For a user needing a low-power, mobile-compatible GPU that can still run OpenCL workloads at a moderate level, the GTX 960A is the only option. The GTX 680’s launch MSRP was 499 USD, but that figure is historical and not relevant to current selection. In summary, the GTX 680 wins on compute performance, while the GTX 960A wins on power efficiency and form factor, and the choice depends entirely on which constraint matters more.