NVIDIA GeForce GTX 560 vs NVIDIA GeForce GTX 960M Comparison

The NVIDIA GeForce GTX 960M and the NVIDIA GeForce GTX 560 are two end-of-life graphics cards from different eras of NVIDIA's mobile and desktop lineups. The GTX 960M is a Maxwell-based mobile part from the GeForce 900M generation, while the GTX 560 is a Fermi 2.0-based desktop card from the GeForce 500 series. Despite their age, benchmark data provides a clear picture of how they compare, with the GTX 960M holding a decisive edge in the one common test available.

FAQ

Q: Which GPU is faster in the Geekbench OpenCL benchmark?

A: The NVIDIA GeForce GTX 960M is faster, scoring 11,045 points compared to the GTX 560's 9,058 points. This represents a 21.9% performance advantage for the GTX 960M.

Q: How do these cards rank against all other GPUs?

A: The GTX 960M sits at the 46th percentile of all GPUs, while the GTX 560 sits at the 45th percentile. Their average benchmark scores are 9,645 and 9,058, respectively, placing them in a similar tier overall, but with the 960M consistently ahead.

Q: What are the closest rivals to the GTX 960M in performance?

A: The GTX 960M's nearest rivals include the NVIDIA Quadro K5000 (average score 9,637, a 0.1% difference), the AMD Radeon Pro WX 2100 (9,653, -0.1% difference), and the NVIDIA Quadro P4000 (9,665, -0.2% difference). It trails the NVIDIA Tesla C2070 (9,716) by 0.7%.

Q: What are the closest rivals to the GTX 560 in performance?

A: The GTX 560's nearest rivals include the NVIDIA TITAN V CEO Edition (average score 9,037, a 0.2% difference) and the NVIDIA GeForce GTX 660 (9,022, a 0.4% difference). It is 1.6% ahead of the AMD Radeon 550X (8,918) and 1.7% behind the AMD Radeon 890M (9,210).

Q: Do both cards support the same modern APIs?

A: Both support DirectX 12 (11_0) and OpenGL 4.6. However, the GTX 960M supports Vulkan 1.4, while the GTX 560 does not have a listed Vulkan version.

Q: What are the memory specifications for each card?

A: The GTX 960M has 4 GB of GDDR5 memory on a 128-bit bus, providing 80.19 GB/s of bandwidth. The GTX 560 has 1 GB of GDDR5 memory on a 256-bit bus, providing 128.0 GB/s of bandwidth.

The Verdict

The data points to a clear winner for general compute performance: the NVIDIA GeForce GTX 960M. Its 21.9% lead in the Geekbench OpenCL test is substantial and reflects its architectural advantages. For a mobile GPU, it outpaces an older desktop card with a much larger die and higher power draw, which speaks to the efficiency gains of the Maxwell architecture.

The GTX 560 is not without its merits, however. It offers significantly more memory bandwidth (128.0 GB/s vs. 80.19 GB/s) and a wider memory bus (256-bit vs. 128-bit), which could matter in memory-bandwidth-bound scenarios, though the benchmark data does not show this translating into a compute win. It also has a higher pixel rate (11.34 GPixel/s vs. 18.82 GPixel/s is actually lower, but the 560 has more ROPs) and more texture mapping units (56 vs. 40).

For a user looking at raw compute performance based on the available data, the GTX 960M is the superior choice. Its higher shading unit count (640 vs. 336), higher clock speeds, and newer architecture deliver a decisive win in the OpenCL test. The GTX 560's only advantage lies in its memory subsystem, but that does not appear to compensate for its older and less efficient core design in this benchmark. The GTX 960M is the card to pick for general-purpose compute tasks.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and it is a decisive victory for the GTX 960M. The GTX 960M scores 11,045, while the GTX 560 scores 9,058. This is a 21.9% difference in favor of the newer mobile card.

This result is notable because it shows the GTX 960M overcoming several apparent hardware disadvantages. The GTX 560 has a wider 256-bit memory bus and higher total memory bandwidth (128.0 GB/s vs. 80.19 GB/s). It also has more ROPs (32 vs. 16) and more TMUs (56 vs. 40). Yet, the GTX 960M's advantage in shading units (640 vs. 336) and its much higher clock speeds (1097 MHz base / 1176 MHz boost vs. no listed base/boost clocks) are enough to secure a significant win.

The GTX 960M's texture rate is also slightly higher at 47.04 GTexel/s compared to the GTX 560's 45.36 GTexel/s, despite having fewer TMUs. This is a direct result of its higher clock speeds. The single benchmark result strongly suggests that the Maxwell architecture is far more efficient per clock and per transistor than Fermi 2.0, allowing the 960M to deliver more compute throughput despite a smaller die (148 mm² vs. 332 mm²) and lower transistor count (1,870 million vs. 1,950 million).

Specification Differences

The two cards differ in nearly every core specification, reflecting their different target markets and release generations.

• Process Node: The GTX 960M uses a 28 nm process, while the GTX 560 uses a larger 40 nm process.
• Transistor Density: The GTX 960M packs 12.6M transistors per mm², while the GTX 560 has a density of 5.9M per mm².
• Die Size: The GTX 960M has a 148 mm² die, while the GTX 560 has a much larger 332 mm² die.
• Shading Units: The GTX 960M has 640 shading units, compared to 336 on the GTX 560.
• Texture Mapping Units (TMUs): The GTX 560 has more TMUs, with 56, versus 40 on the GTX 960M.
• Render Output Units (ROPs): The GTX 560 has 32 ROPs, double the 16 found on the GTX 960M.
• Base Clock: The GTX 960M has a base clock of 1097 MHz; the GTX 560 has no listed base clock.
• Boost Clock: The GTX 960M has a boost clock of 1176 MHz; the GTX 560 has no listed boost clock.
• Memory Size: The GTX 960M has 4 GB of memory, while the GTX 560 has 1 GB.
• Memory Bus Width: The GTX 560 has a 256-bit bus, while the GTX 960M has a 128-bit bus.
• Memory Bandwidth: The GTX 560 provides 128.0 GB/s of bandwidth, exceeding the GTX 960M's 80.19 GB/s.
• Pixel Rate: The GTX 960M has a pixel rate of 18.82 GPixel/s, higher than the GTX 560's 11.34 GPixel/s.
• Texture Rate: The GTX 960M's texture rate is 47.04 GTexel/s, slightly higher than the GTX 560's 45.36 GTexel/s.
• FP32 Performance: The GTX 960M delivers 1.505 TFLOPS, compared to 1,088.6 GFLOPS (1.09 TFLOPS) for the GTX 560.
• TDP: The GTX 960M has a 75 W TDP, while the GTX 560 has a 150 W TDP.
• Slot Width: The GTX 960M is an MXM Module, while the GTX 560 is a Dual-slot card.
• Power Connectors: The GTX 960M has none, while the GTX 560 requires 2x 6-pin connectors.
• Suggested PSU: The GTX 960M has no suggested PSU, but the GTX 560 recommends a 450 W power supply.
• Bus Interface: The GTX 960M uses MXM-B (3.0), while the GTX 560 uses PCIe 2.0 x16.
• Display Outputs: The GTX 960M's outputs are listed as "Portable Device Dependent," while the GTX 560 has 2x DVI and 1x mini-HDMI 1.3a.
• Card Length: The GTX 560 is 210 mm (8.3 inches) long; the GTX 960M has no listed length.
• Release Date: The GTX 960M was released on 2015-03-12, while the GTX 560 was released on 2011-05-16.
• Launch MSRP: The GTX 560 had a launch MSRP of 199 USD. The GTX 960M has no listed launch MSRP.

Architecture Differences

The GTX 960M is built on the Maxwell architecture (chip GM107), while the GTX 560 is based on the older Fermi 2.0 architecture (chip GF114). This is the most fundamental difference between the two.

Maxwell was designed for higher performance-per-watt and improved compute efficiency over its predecessor. This is evident in the data: the GTX 960M achieves a higher FP32 performance (1.505 TFLOPS) than the GTX 560 (1,088.6 GFLOPS) while consuming half the power (75 W vs. 150 W). The Maxwell architecture's efficiency is also reflected in its transistor density of 12.6M / mm², which is more than double the Fermi 2.0's 5.9M / mm², even though the GTX 560 has slightly more total transistors (1,950 million vs. 1,870 million).

The GTX 960M's core config leans heavily on shader throughput, with 640 shading units, while the GTX 560's Fermi 2.0 design relies more on its memory subsystem and ROP count. The GTX 960M supports Vulkan 1.4, whereas the GTX 560 has no Vulkan support listed, indicating a more modern API feature set. Both cards are built by TSMC and support DirectX 12 (11_0) and OpenGL 4.6, but the underlying microarchitecture differences clearly favor the newer Maxwell part in compute workloads.

Where Each One Wins

NVIDIA GeForce GTX 960M:

• Compute Performance: The GTX 960M wins the only head-to-head benchmark, the Geekbench OpenCL test, by a 21.9% margin.
• Shader Throughput: With 640 shading units and a high boost clock of 1176 MHz, it has a significant theoretical advantage in parallel compute tasks.
• Efficiency: It delivers higher FP32 performance (1.505 TFLOPS) and a higher pixel rate (18.82 GPixel/s) while using a lower 75 W TDP.
• Memory Capacity: It has 4 GB of VRAM, which is four times the GTX 560's 1 GB, and is more suitable for modern workloads that require larger memory footprints.
• Modern API Support: It supports Vulkan 1.4, a feature absent from the GTX 560.

NVIDIA GeForce GTX 560:

• Memory Bandwidth: The GTX 560 provides 128.0 GB/s of bandwidth over a 256-bit bus, which is significantly higher than the GTX 960M's 80.19 GB/s. This could be an advantage in bandwidth-sensitive tasks, though it did not lead to a win in the OpenCL test.
• Texture and ROP Count: It has more TMUs (56 vs. 40) and double the ROPs (32 vs. 16) of the GTX 960M, which could offer benefits in specific rasterization or texture-heavy workloads.
• Physical Form Factor: As a dual-slot desktop card with standard display outputs (2x DVI, 1x mini-HDMI), it is easier to integrate into a traditional desktop system than the MXM module format of the GTX 960M.