NVIDIA GeForce 945M vs NVIDIA GeForce GTX 560 Comparison

FAQ

Q: Which GPU has the higher average benchmark score in the database?

A: The NVIDIA GeForce GTX 560 records an average benchmark score of 9058, while the NVIDIA GeForce 945M records 8099. The GTX 560 leads by 11.8% in the head-to-head Geekbench OpenCL test.

Q: How do the two cards rank against all other GPUs?

A: The GTX 560 sits at the 45th percentile of all GPUs, while the GeForce 945M sits at the 42nd percentile. Both are near the middle of the overall distribution, but the GTX 560 places slightly higher.

Q: What are the nearest performance rivals for each card?

A: The GTX 560 is within 1.7% of the AMD Radeon 890M (9210, 1.7% faster), the NVIDIA TITAN V CEO Edition (9037, 0.2% slower), the NVIDIA GeForce GTX 660 (9022, 0.4% slower), and the AMD Radeon 550X (8918, 1.6% slower). The GeForce 945M is within 0.4% of the NVIDIA GeForce GTX 950M (8135, 0.4% faster), the AMD Radeon R9 M360 (8129, 0.4% faster), the NVIDIA GRID K2 (8080, 0.2% slower), and the NVIDIA GeForce GTX 650 Ti Boost (8067, 0.4% slower).

Q: Which card has the higher memory bandwidth?

A: The GTX 560 delivers 128.0 GB/s over a 256-bit bus with GDDR5 memory. The GeForce 945M delivers 28.80 GB/s over a 128-bit bus with DDR3 memory. The GTX 560 has over four times the bandwidth.

Q: What is the transistor density difference between the two chips?

A: The GTX 560 uses a 40 nm process and has a transistor density of 5.9M per mm². The GeForce 945M uses a 28 nm process and has a transistor density of 12.6M per mm², more than double the density.

Q: Which card consumes more power?

A: The GTX 560 has a TDP of 150 W and requires two 6-pin power connectors. The GeForce 945M has a TDP of 75 W and requires no external power connectors.

Architecture Differences

The two GPUs come from different NVIDIA architectures separated by several generations. The GTX 560 is built on the Fermi 2.0 architecture using the GF114 chip, while the GeForce 945M uses the Maxwell architecture with the GM107 chip. This architectural shift is visible in nearly every design decision.

The manufacturing process differs significantly. The GTX 560 uses a 40 nm process at TSMC, while the GeForce 945M uses a denser 28 nm process at TSMC. The die size reflects this: the GF114 measures 332 mm², while the GM107 measures 148 mm². Interestingly, the transistor counts are similar, 1,950 million for the GTX 560 versus 1,870 million for the GeForce 945M, but the Maxwell chip packs those transistors into less than half the area. The density figures tell the story: 5.9M transistors per mm² for Fermi versus 12.6M per mm² for Maxwell.

The compute units are organized differently as well. The GTX 560 has 336 shading units, 56 texture mapping units, and 32 ROPs. The GeForce 945M has 640 shading units, 40 TMUs, and only 16 ROPs. Maxwell uses a wider, shallower design with many more shaders but fewer fixed-function units. The GTX 560 also has a much larger memory bus at 256-bit versus 128-bit, and uses GDDR5 memory while the 945M uses DDR3. Memory clock figures also differ, with the GTX 560 running at 1000 MHz (4 Gbps effective) and the 945M at 900 MHz (1800 Mbps effective).

The output capabilities reflect their intended platforms. The GTX 560 is a dual-slot desktop card with 2x DVI and 1x mini-HDMI 1.3a outputs, a PCIe 2.0 x16 interface, and a 210 mm length. The GeForce 945M is an MXM module for laptops, with no fixed display outputs and a bus interface of MXM-B (3.0). The 945M does support Vulkan 1.4 in addition to DirectX 12 (11_0) and OpenGL 4.6, while the GTX 560 lists DirectX 12 (11_0) and OpenGL 4.6 with no Vulkan entry.

Where Each One Wins

The GTX 560 wins the only recorded benchmark, the Geekbench OpenCL test, by 11.8%. Given its 128.0 GB/s memory bandwidth, 256-bit bus, and higher pixel rate of 11.34 GPixel/s, the data suggests it is better suited for workloads that are sensitive to memory throughput. Its texture rate of 45.36 GTexel/s is also slightly higher than the 945M's 40.80 GTexel/s, giving it an edge in texture-heavy rendering.

The GeForce 945M wins on raw compute throughput in one respect: its FP32 rating of 1,305.6 GFLOPS exceeds the GTX 560's 1,088.6 GFLOPS. The 945M also has a higher pixel rate at 16.32 GPixel/s, despite having half the ROPs, thanks to its higher clock speeds of 928 MHz base and 1020 MHz boost. The GTX 560 has no base or boost clock listed. However, these theoretical advantages do not translate into a higher benchmark score in the recorded data.

The 945M is the clear choice for portable systems. It is an MXM module with no power connectors, a 75 W TDP, and no fixed dimensions, while the GTX 560 is a 210 mm dual-slot card requiring a 450 W suggested PSU and two 6-pin connectors. The 945M also has double the memory capacity at 2 GB versus 1024 MB, which can matter for larger working sets even if the bandwidth is far lower.

Specification Differences

| Field | GTX 560 | GeForce 945M |

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

| Architecture | Fermi 2.0 | Maxwell |

| Chip | GF114 | GM107 |

| Generation | GeForce 500 | GeForce 900M |

| Process node | 40 nm | 28 nm |

| Die size | 332 mm² | 148 mm² |

| Transistor density | 5.9M / mm² | 12.6M / mm² |

| Base clock | None listed | 928 MHz |

| Boost clock | None listed | 1020 MHz |

| Memory clock | 1000 MHz, 4 Gbps effective | 900 MHz, 1800 Mbps effective |

| Memory size | 1024 MB | 2 GB |

| Memory type | GDDR5 | DDR3 |

| Memory bus | 256 bit | 128 bit |

| Memory bandwidth | 128.0 GB/s | 28.80 GB/s |

| Shading units | 336 | 640 |

| TMUs | 56 | 40 |

| ROPs | 32 | 16 |

| Pixel rate | 11.34 GPixel/s | 16.32 GPixel/s |

| Texture rate | 45.36 GTexel/s | 40.80 GTexel/s |

| FP32 | 1,088.6 GFLOPS | 1,305.6 GFLOPS |

| TDP | 150 W | 75 W |

| Slot width | Dual-slot | MXM Module |

| Power connectors | 2x 6-pin | None |

| Suggested PSU | 450 W | None listed |

| Bus interface | PCIe 2.0 x16 | MXM-B (3.0) |

| Display outputs | 2x DVI, 1x mini-HDMI 1.3a | Portable Device Dependent |

| Vulkan | None listed | 1.4 |

| Release date | 2011-05-16 | 2015-10-26 |

| Predecessor | GeForce 400 | GeForce 800M |

| Successor | GeForce 600 | GeForce 10 Mobile |

| Launch MSRP | 199 USD | None listed |

Head-to-Head Benchmarks

The database contains one head-to-head benchmark between these two GPUs: Geekbench OpenCL. The GTX 560 scores 9058, and the GeForce 945M scores 8099. The GTX 560 wins by 11.8%. This is a substantial margin in a compute-oriented workload, especially considering the 945M has nearly 20% higher FP32 peak throughput on paper.

The result is consistent with the memory system differences. The GTX 560 has a 256-bit bus and GDDR5 memory producing 128.0 GB/s of bandwidth, while the 945M has a 128-bit bus with DDR3 producing 28.80 GB/s. OpenCL workloads often scale with memory bandwidth, and the GTX 560's 4.4x bandwidth advantage likely explains why it outperforms despite the 945M's higher shader count and clock speeds.

The nearest-rival data reinforces the gap. The GTX 560 lands between the AMD Radeon 890M (9210) and the NVIDIA TITAN V CEO Edition (9037), with the TITAN V CEO Edition only 0.2% behind and the GTX 660 only 0.4% behind. The GeForce 945M sits in a lower performance cluster, with the GTX 950M 0.4% ahead and the GRID K2 0.2% behind. The two cards occupy different performance tiers despite both being mid-range in their respective generations.

The pixel and texture rates also favor the GTX 560 on texture work (45.36 GTexel/s versus 40.80 GTexel/s) but favor the 945M on pixel fill (16.32 GPixel/s versus 11.34 GPixel/s). Since the single recorded benchmark is compute-based rather than rasterization-based, the texture rate advantage appears more relevant to the outcome.

The Verdict

The recorded data points to the GTX 560 as the stronger performer. Its average benchmark score of 9058 beats the GeForce 945M's 8099 by 11.8%, and it sits at the 45th percentile compared to the 945M's 42nd percentile. For any workload captured by Geekbench OpenCL, the GTX 560 is the better choice. Its launch MSRP was 199 USD, and it carries a 150 W TDP with two 6-pin connectors.

The GeForce 945M is not without merit, but its advantages are confined to specific areas. It doubles the memory capacity (2 GB versus 1024 MB), has a higher FP32 rating (1,305.6 GFLOPS versus 1,088.6 GFLOPS), higher pixel rate (16.32 GPixel/s versus 11.34 GPixel/s), and consumes half the power (75 W versus 150 W). It also supports Vulkan 1.4, which the GTX 560 does not list, and it requires no external power connectors.

Buyers who need a desktop card for compute tasks should select the GTX 560 based on the benchmark data. Buyers who need a mobile MXM module with lower power draw and greater memory capacity should select the GeForce 945M, accepting the lower benchmark score. The 945M is also the more modern design, released in 2015 versus 2011, but the GTX 560 remains the faster card in the one test the database records. The choice comes down to platform and power constraints, not raw performance.