NVIDIA GeForce GTX 560 vs NVIDIA GeForce GTX 960 Comparison

The NVIDIA GeForce GTX 960 and the NVIDIA GeForce GTX 560 are separated by nearly four years of GPU architecture evolution, yet both occupy a similar historical performance tier. The data shows a decisive overall victory for the GTX 960, but the comparison is not without nuance. The sole shared benchmark, Geekbench OpenCL, illustrates the generational gap clearly: the GTX 960 scores 18,925 points against the GTX 560’s 9,058, a delta of 108.9% in favor of the newer card. This single result gives the GTX 960 one win in the head-to-head category, while the GTX 560 records zero, making the outcome unambiguous in raw compute throughput.

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and it is a landslide. The GTX 960 produces a score of 18,925, more than double the GTX 560’s 9,058. The delta percentage of 108.9% indicates the GTX 960 delivers over twice the OpenCL compute performance of the older Fermi-based card. This is not a marginal improvement; it is a complete redefinition of the performance class. For context, the GTX 960’s average benchmark score across all tests is 9,273, a figure that sits close to its nearest rival, the NVIDIA GeForce GTX 465, which averages 9,294 (a -0.2% difference). The GTX 560’s average score of 9,058 places it near the NVIDIA TITAN V CEO Edition (9,037, a 0.2% delta) and the NVIDIA GeForce GTX 660 (9,022, a 0.4% delta). While the average scores are similar, the OpenCL result highlights that the GTX 960 excels specifically in compute workloads, whereas the GTX 560’s single benchmark reflects a more modest capability.

The absence of further head-to-head tests, such as 3DMark Steel Nomad or Geekbench Vulkan, means the GTX 960’s additional benchmark results must be viewed in isolation. The GTX 960 posts 162 in 3DMark Steel Nomad DX12, 8,773 in Geekbench Metal, and 9,231 in Geekbench Vulkan. These scores, combined with its OpenCL figure, produce a broader performance profile that the GTX 560 simply cannot match, as the GTX 560 only has the one OpenCL data point. The delta of 108.9% in the shared test is the definitive metric; it shows the GTX 960 is not just faster, but categorically superior in raw floating-point operations.

Architecture Differences

The two cards are built on fundamentally different architectures. The GTX 960 uses the GM206 chip based on Maxwell 2.0, fabricated on a 28 nm process at TSMC. The GTX 560 uses the GF114 chip based on Fermi 2.0, also from TSMC but on an older 40 nm node. This process shrink is significant: the GTX 960 packs 2,940 million transistors into a 228 mm² die, resulting in a transistor density of 12.9 million per mm². The GTX 560, by contrast, contains 1,950 million transistors on a much larger 332 mm² die, yielding just 5.9 million transistors per mm². The newer process allows the GTX 960 to be more efficient in both power and space, a theme echoed in its power specifications.

The memory subsystems diverge as well. The GTX 960 features 2 GB of GDDR5 on a 128-bit bus, delivering 112.2 GB/s of bandwidth. The GTX 560 has 1024 MB of GDDR5 on a wider 256-bit bus, achieving 128.0 GB/s. Despite the narrower bus, the GTX 960’s higher memory clock of 7 Gbps effective (versus 4 Gbps effective on the GTX 560) allows it to approach similar bandwidth with half the bus width. The GTX 960 also doubles the memory capacity, which is critical for modern textures. In terms of compute units, the GTX 960 has 1,024 shading units, 64 texture mapping units, and 32 ROPs. The GTX 560 has 336 shading units, 56 TMUs, and 32 ROPs. The shading unit count is nearly triple on the GTX 960, explaining its massive OpenCL advantage.

Clock speeds further differentiate them. The GTX 960 runs at a base clock of 1127 MHz with a boost of 1178 MHz, while the GTX 560 has no listed base or boost clock in the data, only a memory clock of 1000 MHz (4 Gbps effective). This lack of clock data for the GTX 560 makes direct frequency comparison impossible, but the pixel and texture rates tell the story. The GTX 960 achieves a pixel rate of 37.70 GPixel/s and a texture rate of 75.39 GTexel/s, versus 11.34 GPixel/s and 45.36 GTexel/s for the GTX 560. The FP32 throughput is 2.413 TFLOPS on the GTX 960, compared to 1,088.6 GFLOPS (roughly 1.09 TFLOPS) on the GTX 560, a 2.2x difference that aligns with the OpenCL benchmark.

FAQ

Q: Which card has a higher average benchmark score?

A: The GTX 960 has an average benchmark score of 9,273, while the GTX 560 has 9,058, making the GTX 960 approximately 2.4% higher on average.

Q: Does the GTX 560 support Vulkan?

A: No, the GTX 560 has no Vulkan API support listed in the data, whereas the GTX 960 supports Vulkan 1.4. The GTX 560 is limited to DirectX 12 (11_0) and OpenGL 4.6.

Q: What is the memory capacity difference?

A: The GTX 960 has 2 GB of GDDR5 memory, while the GTX 560 has 1024 MB (1 GB) of GDDR5. Despite the GTX 560’s wider 256-bit bus, the GTX 960 has higher effective memory speed at 7 Gbps versus 4 Gbps.

Q: Which card has a higher transistor density?

A: The GTX 960 has a transistor density of 12.9 million per mm², far exceeding the GTX 560’s 5.9 million per mm², due to the newer 28 nm process versus 40 nm.

Q: Are the power requirements significantly different?

A: Yes, the GTX 960 has a TDP of 120 W with a single 6-pin power connector and a suggested PSU of 300 W. The GTX 560 has a TDP of 150 W, requires two 6-pin connectors, and suggests a 450 W PSU.

Q: Which card has a higher pixel fill rate?

A: The GTX 960 achieves 37.70 GPixel/s, more than three times the GTX 560’s 11.34 GPixel/s, despite both having 32 ROPs.

The Verdict

The data points decisively to the GTX 960 as the superior card for nearly all use cases. In the single shared benchmark, the GTX 960 is 108.9% ahead, meaning it delivers double the OpenCL performance. Its average benchmark score of 9,273 also edges out the GTX 560’s 9,058, though the difference is small (about 2.3%). However, the GTX 960’s advantage extends beyond raw scores. It supports Vulkan 1.4, a feature entirely absent from the GTX 560, and offers double the memory (2 GB versus 1 GB), which is critical for modern game textures. The GTX 960 also has a lower TDP of 120 W against 150 W, requires only one 6-pin power connector versus two, and suggests a 300 W PSU compared to 450 W. These efficiency gains, driven by the 28 nm process and Maxwell 2.0 architecture, mean the GTX 960 achieves higher performance while consuming less power and generating less heat.

The GTX 560’s only advantages are its wider 256-bit memory bus, which gives it slightly higher raw bandwidth (128.0 GB/s versus 112.2 GB/s), and its smaller physical length of 210 mm (8.3 inches) versus 241 mm (9.5 inches) for the GTX 960. The wider bus is a legacy design choice that does not compensate for the older architecture’s lower compute throughput. For anyone choosing between these two, the GTX 960 is the clear pick, provided the additional 31 mm of length fits the chassis. The GTX 560 might be preferable only in a system with strict power supply limits that cannot accommodate a 300 W suggested PSU but can handle 450 W, which is an unlikely scenario given the GTX 560’s higher TDP.

Specification Differences

The two cards diverge on nearly every specification. The GTX 960 uses the GM206 chip (Maxwell 2.0) on a 28 nm process, while the GTX 560 uses GF114 (Fermi 2.0) on 40 nm. Transistor counts are 2,940 million versus 1,950 million, and die sizes are 228 mm² versus 332 mm². The GTX 960 has a base clock of 1127 MHz and boost of 1178 MHz; the GTX 560 has no base or boost clock listed. Memory speed is 7 Gbps effective on the GTX 960 versus 4 Gbps effective on the GTX 560. Memory size is 2 GB versus 1024 MB, with bus widths of 128-bit and 256-bit, respectively. Bandwidth is 112.2 GB/s on the GTX 960 and 128.0 GB/s on the GTX 560. Shading units are 1,024 versus 336, TMUs are 64 versus 56, while ROPs stay identical at 32. Pixel rates are 37.70 GPixel/s versus 11.34 GPixel/s, and texture rates are 75.39 GTexel/s versus 45.36 GTexel/s.

FP32 compute is 2.413 TFLOPS on the GTX 960 versus 1,088.6 GFLOPS on the GTX 560. TDP is 120 W versus 150 W, power connectors are 1x 6-pin versus 2x 6-pin, and suggested PSU is 300 W versus 450 W. The bus interface is PCIe 3.0 x16 on the GTX 960 and PCIe 2.0 x16 on the GTX 560. Display outputs differ: the GTX 960 has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2, while the GTX 560 has 2x DVI and 1x mini-HDMI 1.3a. API support shows the GTX 960 with DirectX 12 (12_1) and Vulkan 1.4, versus DirectX 12 (11_0) and no Vulkan on the GTX 560. Both support OpenGL 4.6. The GTX 960 is 241 mm long, while the GTX 560 is 210 mm. Both are dual-slot cards, both are end-of-life, and both launched at 199 USD MSRP.

Where Each One Wins

The GTX 960 wins in every performance category that matters for modern workloads. Its OpenCL score of 18,925 is 108.9% higher, making it the obvious choice for compute-heavy applications like video encoding, machine learning inference, or any GPGPU task. Its Vulkan 1.4 support means it can run modern APIs that the GTX 560 cannot use at all. The doubled memory capacity of 2 GB versus 1 GB allows it to handle higher-resolution textures without stuttering. The lower power draw (120 W versus 150 W) and single 6-pin connector make it easier to integrate into existing systems, especially those with modest power supplies. The higher pixel rate of 37.70 GPixel/s versus 11.34 GPixel/s indicates better fill-rate performance for high-resolution rendering, even though both have 32 ROPs.

The GTX 560’s wins are narrow and mostly historical. Its 256-bit memory bus provides a nominal bandwidth advantage of 128.0 GB/s over the GTX 960’s 112.2 GB/s, which could theoretically benefit memory-bound tasks that do not scale with compute units, though no benchmark in the data confirms this. Its shorter length of 210 mm versus 241 mm makes it easier to fit in small form-factor cases. It also requires a 450 W PSU suggestion, which is a higher requirement but may align with pre-existing builds that have spare 6-pin connectors. However, these are minor caveats. In every measurable benchmark, the GTX 960 wins outright, and its architectural advantages in process node, transistor density, shading units, and API support make it the only rational choice for any new build or upgrade from the data presented.