NVIDIA GeForce GTX 560 Ti vs NVIDIA GeForce GTX 870M Comparison

The NVIDIA GeForce GTX 560 Ti and the NVIDIA GeForce GTX 870M represent two distinct eras of mobile and desktop graphics, separated by three years of architectural evolution. The most direct comparison available in the data is the Geekbench OpenCL benchmark, where the GTX 870M scores 12,630 against the GTX 560 Ti’s 10,690. This translates to a 15.4% advantage for the notebook part, a significant margin that flips the expected performance hierarchy based on their respective positions in the overall GPU landscape. The GTX 560 Ti sits at the 49th percentile of all GPUs, while the GTX 870M sits at the 48th percentile, yet the newer mobile chip clearly outpaces the older desktop card in this compute-oriented test.

Head-to-Head Benchmarks

The only shared benchmark between these two GPUs is Geekbench OpenCL, and the results are unambiguous. The GTX 870M delivers a score of 12,630, while the GTX 560 Ti achieves 10,690. The deltaPct of -15.4% indicates that the GTX 560 Ti trails the GTX 870M by that margin, making the GTX 870M the decisive winner in this single head-to-head contest. This is not a marginal victory; a 15.4% gap in raw compute performance is substantial enough to affect real-world workloads that leverage OpenCL acceleration.

When placed in the context of their nearest rivals, the picture becomes more nuanced. The GTX 560 Ti’s score of 10,690 places it just 0.6% ahead of the AMD Radeon RX 6600S (10,629) and a mere 1% ahead of the AMD Radeon R9 M275X (10,582). However, it trails the NVIDIA Quadro K2200 (10,761) by 0.7% and the AMD Radeon Pro 450 (10,804) by 1.1%. This clustering suggests that the GTX 560 Ti is firmly in the middle of a tight performance band, where a few percentage points separate competitors. The GTX 870M, with its 12,630 score, sits in a different league among its peers: it is 1.1% ahead of the NVIDIA Quadro 6000 (9,846) and 0.5% behind the AMD Radeon Pro 5300M (10,013), though those deltas are misleading because the raw score gap is over 2,600 points. The GTX 870M’s nearest rivals all score near 10,000, making its 12,630 result a standout outlier—it outperforms its closest competitor, the NVIDIA Quadro K5100M (10,043), by roughly 25.7% based on the raw scores.

The benchmark data indicates that the GTX 870M is not just faster than the GTX 560 Ti; it is significantly faster. The 15.4% delta is consistent across the board, and the GTX 870M’s position relative to its own rivals suggests that this is a generational leap rather than a minor refresh. The GTX 560 Ti, despite being a desktop card with a higher thermal envelope, cannot match the architectural efficiency of the newer mobile chip in this test. For any workload that relies on OpenCL, the GTX 870M is the clear choice.

Architecture Differences

The architectural gap between these two GPUs is stark. The GTX 560 Ti is built on the GF114 chip using the Fermi 2.0 architecture, manufactured on a 40 nm process at TSMC. The GTX 870M, in contrast, uses the GK104 chip with the Kepler architecture, built on a more advanced 28 nm process, also at TSMC. This process shrink is fundamental: the GTX 870M packs 3,540 million transistors onto a 294 mm² die, resulting in a transistor density of 12.0M per mm². The GTX 560 Ti, with 1,950 million transistors on a 332 mm² die, achieves only 5.9M transistors per mm². The Kepler design is more than twice as dense, enabling far more compute resources in a smaller physical footprint.

The compute resources tell a similar story. The GTX 870M features 1,344 shading units, 112 texture mapping units, and 24 ROPs. The GTX 560 Ti has just 384 shading units, 64 TMUs, and 32 ROPs. While the GTX 560 Ti has more ROPs, the GTX 870M’s massive advantage in shading units—3.5 times more—drives its superior OpenCL performance. The pixel rate for the GTX 870M is 27.08 GPixel/s, compared to 13.17 GPixel/s for the GTX 560 Ti, and the texture rate is 108.3 GTexel/s versus 52.67 GTexel/s. The FP32 compute output is 2.599 TFLOPS for the GTX 870M, more than double the GTX 560 Ti’s 1,263.4 GFLOPS.

Memory configurations also diverge significantly. The GTX 870M comes with 3 GB of GDDR5 memory on a 192-bit bus, yielding 120.0 GB/s of bandwidth. The GTX 560 Ti has only 1 GB of GDDR5 on a 256-bit bus, delivering 128.3 GB/s. Despite the narrower bus, the GTX 560 Ti actually has slightly higher memory bandwidth, but the GTX 870M compensates with triple the capacity, which is critical for modern textures and larger datasets. Memory clocks differ as well: the GTX 870M runs at 1250 MHz (5 Gbps effective), while the GTX 560 Ti runs at 1002 MHz (4 Gbps effective).

Feature support also separates them. The GTX 560 Ti supports DirectX 12 (11_0) and OpenGL 4.6, but has no Vulkan support listed. The GTX 870M also supports DirectX 12 (11_0) and OpenGL 4.6, but adds Vulkan 1.2.175, making it more future-proof for modern APIs. The GTX 870M also has explicit base and boost clocks of 941 MHz and 967 MHz, respectively, while the GTX 560 Ti has no base or boost clock listed, only its memory clock.

The Verdict

The data unequivocally favors the GTX 870M for raw compute performance. Its OpenCL score is 15.4% higher than the GTX 560 Ti’s, and it dominates in shading units, texture rate, and pixel rate. The GTX 870M is the faster GPU by every measurable compute metric in the fact pack, despite being a mobile part with a 100 W TDP versus the GTX 560 Ti’s 170 W. The GTX 560 Ti’s only advantages are memory bandwidth (128.3 GB/s vs 120.0 GB/s) and ROP count (32 vs 24), but these do not compensate for the 3.5x shading unit deficit.

For a PC builder, the choice is straightforward if performance is the priority: the GTX 870M is the better option. It delivers more compute throughput, more VRAM, and better API support in a lower-power package. The GTX 560 Ti might appeal to someone with a legacy desktop system that requires a dual-slot card with 2x 6-pin power connectors, but the benchmark data shows it will be slower in any OpenCL-heavy task. The GTX 870M’s MXM form factor is a different physical requirement, but for those who can accommodate it, the performance is superior.

FAQ

Q: Which GPU has a higher Geekbench OpenCL score?

A: The NVIDIA GeForce GTX 870M scores 12,630, while the NVIDIA GeForce GTX 560 Ti scores 10,690, giving the GTX 870M a 15.4% advantage.

Q: How does the GTX 560 Ti compare to its nearest rivals?

A: The GTX 560 Ti scores 0.6% higher than the AMD Radeon RX 6600S, 1% higher than the AMD Radeon R9 M275X, but 0.7% lower than the NVIDIA Quadro K2200 and 1.1% lower than the AMD Radeon Pro 450.

Q: What are the memory specifications for each GPU?

A: The GTX 560 Ti has 1024 MB of GDDR5 on a 256-bit bus with 128.3 GB/s bandwidth. The GTX 870M has 3 GB of GDDR5 on a 192-bit bus with 120.0 GB/s bandwidth.

Q: Which GPU has more shading units?

A: The GTX 870M has 1,344 shading units, compared to the GTX 560 Ti’s 384 shading units, a 3.5x difference.

Q: Does either GPU support Vulkan?

A: The GTX 870M supports Vulkan 1.2.175, while the GTX 560 Ti has no Vulkan support listed in the data.

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

A: The GTX 870M’s GK104 chip has a transistor density of 12.0M per mm², while the GTX 560 Ti’s GF114 chip has a density of 5.9M per mm².

Where Each One Wins

The GTX 870M wins in virtually every performance category where data exists. Its OpenCL benchmark victory is decisive, and its architectural advantages in shading units, texture units, and pixel rate all point to superior compute and rasterization performance. The GTX 870M also wins on memory capacity, offering 3 GB versus 1 GB, which is essential for modern games and applications with large texture pools. It also supports Vulkan, which the GTX 560 Ti lacks, making it more relevant for current-generation software.

The GTX 560 Ti does have a few narrow wins. Its memory bandwidth of 128.3 GB/s is 6.9% higher than the GTX 870M’s 120.0 GB/s, which could help in bandwidth-sensitive scenarios. It also has 32 ROPs versus 24, potentially improving fill-rate-bound workloads. However, these advantages are minor in the context of the overall benchmark gap. The GTX 560 Ti is also a desktop card with a dual-slot form factor and 2x 6-pin power connectors, which might be easier to integrate into a traditional PC build than the GTX 870M’s MXM module.

For practical use, the GTX 870M is the clear winner for anyone running OpenCL-accelerated applications, modern games, or Vulkan-based titles. The GTX 560 Ti would only be preferable in a legacy desktop system where the MXM form factor is incompatible and where the slightly higher memory bandwidth might be leveraged in specific niche workloads. The data does not support choosing the GTX 560 Ti for general performance.

Specification Differences

| Specification | NVIDIA GeForce GTX 560 Ti | NVIDIA GeForce GTX 870M |

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

| Chip | GF114 | GK104 |

| Architecture | Fermi 2.0 | Kepler |

| Generation | GeForce 500 | GeForce 800M |

| Process Node | 40 nm | 28 nm |

| Transistors | 1,950 million | 3,540 million |

| Die Size | 332 mm² | 294 mm² |

| Transistor Density | 5.9M / mm² | 12.0M / mm² |

| Memory | 1024 MB GDDR5 | 3 GB GDDR5 |

| Memory Bus Width | 256 bit | 192 bit |

| Memory Bandwidth | 128.3 GB/s | 120.0 GB/s |

| Shading Units | 384 | 1,344 |

| TMUs | 64 | 112 |

| ROPs | 32 | 24 |

| Pixel Rate | 13.17 GPixel/s | 27.08 GPixel/s |

| Texture Rate | 52.67 GTexel/s | 108.3 GTexel/s |

| FP32 Performance | 1,263.4 GFLOPS | 2.599 TFLOPS |

| TDP | 170 W | 100 W |

| Slot Width | Dual-slot | MXM Module |

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

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

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

| Vulkan Support | None | 1.2.175 |

| Release Date | 2011-01-24 | 2014-03-11 |

| Launch MSRP | 249 USD | None |