AMD Radeon RX 6600S vs NVIDIA GeForce GTX 560 Ti Comparison

Head-to-Head Benchmarks

The benchmark data presents a decisive picture: the AMD Radeon RX 6600S wins the sole head-to-head comparison by an overwhelming margin. In Geekbench OpenCL, the RX 6600S scores 66,435, while the NVIDIA GeForce GTX 560 Ti manages 10,690. That is a delta of -83.9% for the GTX 560 Ti, meaning the RX 6600S is roughly six times faster in this compute-oriented test. The gap is so wide that it dwarfs the differences seen among the nearest rivals for either card. For context, the GTX 560 Ti’s closest competitor, the NVIDIA Quadro K2200, scores 10,761 — just 0.7% higher — while the RX 6600S’s nearest rival, the AMD Radeon R9 M275X, scores 10,582, a mere 0.4% lower. Neither card’s typical competitive window comes anywhere close to the chasm separating these two in OpenCL compute.

The wins tally reflects this: the AMD part claims 1 win, while the NVIDIA part claims none. The GTX 560 Ti’s average benchmark score is 10,690, which is nearly identical to its OpenCL result since that is its only recorded benchmark. The RX 6600S, by contrast, has a broader benchmark suite. Its PassMark G3D score of 12,649 and its PassMark DirectX 9 score of 176 illustrate strengths in legacy rasterization workloads, while its PassMark DirectX 12 score of 56 and DirectX 10 score of 81 show more modest results in those specific tests. The RX 6600S also posts a PassMark GPU Compute score of 4,879, which is substantial relative to its G3D score, indicating strong compute throughput beyond just graphics.

The percentile rankings are identical — both sit at the 49th percentile among all GPUs — which is somewhat paradoxical given the massive OpenCL delta. This suggests that the percentile is driven by the average of all benchmarks, and the GTX 560 Ti’s single OpenCL score is its only data point, pulling its average to 10,690. The RX 6600S’s average of 10,629 is slightly lower, yet its individual OpenCL score is far higher. The takeaway is that in the one test where both are evaluated identically — Geekbench OpenCL — the RX 6600S is categorically superior, but in aggregate, the two cards land in the same performance tier relative to the entire GPU landscape. That is a nuance worth noting: average scores can obscure extreme single-test disparities.

FAQ

Q: Which card wins the only direct benchmark comparison?

A: The AMD Radeon RX 6600S wins the Geekbench OpenCL test with a score of 66,435 versus the NVIDIA GeForce GTX 560 Ti’s 10,690, a delta of -83.9% in AMD’s favor.

Q: Do the two cards have similar overall performance rankings?

A: Yes. Both sit at the 49th percentile among all GPUs, and their average benchmark scores are close — 10,690 for the GTX 560 Ti and 10,629 for the RX 6600S — despite the huge OpenCL gap.

Q: How does the RX 6600S perform across different DirectX versions?

A: The RX 6600S scores 176 in PassMark DirectX 9, 105 in DirectX 11, 81 in DirectX 10, and 56 in DirectX 12, showing the strongest legacy DirectX 9 result and the weakest DirectX 12 result in its suite.

Q: What is the GTX 560 Ti’s only benchmark result?

A: The GTX 560 Ti has a single recorded benchmark: Geekbench OpenCL with a score of 10,690, which also serves as its average benchmark score.

Q: Is the RX 6600S’s compute performance consistent with its graphics performance?

A: Not entirely. Its PassMark GPU Compute score is 4,879, while its PassMark G3D score is 12,649 — compute is substantial but lower than the graphics score, whereas its OpenCL score of 66,435 is far above both.

Q: Which card has more wins in head-to-head comparisons?

A: The AMD Radeon RX 6600S has 1 win, and the NVIDIA GeForce GTX 560 Ti has 0 wins.

Architecture Differences

The architectural gap between these two GPUs is generational and profound. The GTX 560 Ti is built on Fermi 2.0 using the GF114 chip, fabricated on a 40 nm process at TSMC, with 1,950 million transistors on a 332 mm² die. That yields a transistor density of 5.9 million per mm². The RX 6600S uses the Navi 23 chip with RDNA 2.0 architecture, also fabricated at TSMC but on a 7 nm node, packing 11,060 million transistors into a smaller 237 mm² die, achieving a far higher density of 46.7 million per mm². The process node difference — 40 nm versus 7 nm — explains much of the efficiency and density disparity.

Core counts diverge sharply. The GTX 560 Ti has 384 shading units, 64 texture mapping units, and 32 ROPs. The RX 6600S has 1,792 shading units, 112 TMUs, and 64 ROPs. The RX 6600S also introduces 28 ray tracing cores, a feature entirely absent from the Fermi-based GTX 560 Ti. Memory configurations differ as well: the GTX 560 Ti uses 1,024 MB of GDDR5 on a 256-bit bus, yielding 128.3 GB/s of bandwidth. The RX 6600S uses 4 GB of GDDR6 on a 128-bit bus, but achieves 224.0 GB/s due to faster memory clocks — 1,750 MHz or 14 Gbps effective versus the GTX 560 Ti’s 1,002 MHz or 4 Gbps effective.

Clock behavior also separates them. The GTX 560 Ti has no base or boost clock listed, while the RX 6600S runs a base of 1,700 MHz, a game clock of 1,881 MHz, and a boost of 2,000 MHz. The pixel and texture rates reflect the architectural leaps: the RX 6600S delivers 128.0 GPixel/s and 224.0 GTexel/s, versus the GTX 560 Ti’s 13.17 GPixel/s and 52.67 GTexel/s. FP32 compute is 7.168 TFLOPS for the RX 6600S versus 1,263.4 GFLOPS (1.263 TFLOPS) for the GTX 560 Ti, and the RX 6600S adds FP16 throughput of 14.34 TFLOPS with a 2:1 ratio, which the GTX 560 Ti lacks entirely.

API support marks another major divide. The GTX 560 Ti supports DirectX 12 (11_0), OpenGL 4.6, and no Vulkan. The RX 6600S supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RX 6600S is PCIe 4.0 x8, while the GTX 560 Ti uses PCIe 2.0 x16. Power and physical design are starkly different: the GTX 560 Ti is a dual-slot card with 2x 6-pin power connectors and a 170 W TDP, requiring a 450 W suggested PSU. The RX 6600S is an IGP (integrated graphics processor) with no power connectors and an 80 W TDP, with no suggested PSU listed. The GTX 560 Ti has a 229 mm length and outputs 2x DVI plus 1x mini-HDMI 1.3a, while the RX 6600S’s display outputs are listed as portable device dependent.

The Verdict

The data supports a clear split decision based on workload and context. For any compute-heavy task measured by Geekbench OpenCL, the AMD Radeon RX 6600S is the only viable choice — its 66,435 score is more than six times the GTX 560 Ti’s 10,690. The RX 6600S also offers superior raw specifications across the board: 1,792 shading units versus 384, 7.168 TFLOPS versus 1,263.4 GFLOPS, 224.0 GB/s bandwidth versus 128.3 GB/s, and 4 GB of GDDR6 versus 1 GB of GDDR5. Its 7 nm process, ray tracing cores, and modern API support (DirectX 12 Ultimate, Vulkan 1.4) make it the forward-looking part, despite both being end-of-life products.

However, the aggregate benchmark picture complicates a blanket recommendation. Both GPUs sit at the 49th percentile, and their average scores are nearly identical — 10,690 for the GTX 560 Ti and 10,629 for the RX 6600S. The RX 6600S’s PassMark results are mixed: strong in G3D (12,649) and DirectX 9 (176) but weak in DirectX 12 (56) and DirectX 10 (81). The GTX 560 Ti has no PassMark data, so its only recorded performance is the OpenCL score. If the benchmark database reflects real-world variety, the RX 6600S may not universally dominate in every legacy workload.

Who should pick which? If the priority is modern compute, ray tracing capability, higher memory capacity, and vastly better efficiency (80 W TDP versus 170 W, no power connectors versus 2x 6-pin), the RX 6600S is the data-backed choice. If the use case involves legacy DirectX 9 or 11 workloads where the RX 6600S’s PassMark scores dip (81 for DX10, 56 for DX12), or if the system requires a dual-slot PCIe 2.0 card with 2x DVI output, the GTX 560 Ti might still have a niche — but its single OpenCL score offers no evidence of superiority in any tested metric. The verdict from the numbers: the RX 6600S wins the head-to-head decisively, while the GTX 560 Ti remains a historical artifact with a single benchmark to its name.

Specification Differences

| Field | NVIDIA GeForce GTX 560 Ti | AMD Radeon RX 6600S |

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

| Architecture | Fermi 2.0 | RDNA 2.0 |

| Chip | GF114 | Navi 23 |

| Process Node | 40 nm | 7 nm |

| Transistors | 1,950 million | 11,060 million |

| Die Size | 332 mm² | 237 mm² |

| Transistor Density | 5.9M / mm² | 46.7M / mm² |

| Memory Size | 1024 MB | 4 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus Width | 256 bit | 128 bit |

| Memory Bandwidth | 128.3 GB/s | 224.0 GB/s |

| Memory Clock | 1002 MHz / 4 Gbps effective | 1750 MHz / 14 Gbps effective |

| Shading Units | 384 | 1792 |

| TMUs | 64 | 112 |

| ROPs | 32 | 64 |

| RT Cores | null | 28 |

| Pixel Rate | 13.17 GPixel/s | 128.0 GPixel/s |

| Texture Rate | 52.67 GTexel/s | 224.0 GTexel/s |

| FP32 | 1,263.4 GFLOPS | 7.168 TFLOPS |

| FP16 | null | 14.34 TFLOPS (2:1) |

| TDP | 170 W | 80 W |

| Slot Width | Dual-slot | IGP |

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

| Suggested PSU | 450 W | null |

| Bus Interface | PCIe 2.0 x16 | PCIe 4.0 x8 |

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

| DirectX Support | 12 (11_0) | 12 Ultimate (12_2) |

| Vulkan Support | null | 1.4 |

| Base Clock | null | 1700 MHz |

| Boost Clock | null | 2000 MHz |

| Game Clock | null | 1881 MHz |

| Release Date | 2011-01-24 | 2022-01-03 |

| Launch MSRP | 249 USD | null |

| Length | 229 mm / 9 inches | null |