AMD Radeon Pro 5600M vs NVIDIA GeForce GTX 1070 Ti Comparison

FAQ

Q: Which GPU wins more benchmark tests?

A: The NVIDIA GeForce GTX 1070 Ti wins 9 of the 10 head-to-head benchmark comparisons. The AMD Radeon Pro 5600M wins only one test, Geekbench Metal.

Q: How big is the AMD Radeon Pro 5600M's win in Geekbench Metal?

A: The AMD card scores 55030 versus 22180 for the GTX 1070 Ti, a massive 148.1% advantage. This is the largest margin recorded in any head-to-head test between the two.

Q: What is the average benchmark score difference?

A: The AMD Radeon Pro 5600M has an average benchmark score of 16351, while the NVIDIA GeForce GTX 1070 Ti averages 14277. The AMD card sits about 14.5% higher on average, despite losing most individual tests.

Q: How do their percentile rankings compare?

A: The AMD Radeon Pro 5600M ranks in the 59th percentile among all GPUs, while the NVIDIA GeForce GTX 1070 Ti ranks in the 56th percentile. The AMD card is positioned slightly higher in the overall distribution.

Q: Which GPU has better Vulkan performance?

A: The NVIDIA GeForce GTX 1070 Ti scores 61006 in Geekbench Vulkan versus 46425 for the AMD Radeon Pro 5600M, a 23.9% advantage for NVIDIA.

Q: What are the power requirements for each card?

A: The AMD Radeon Pro 5600M is rated at 50 W TDP and uses no power connectors, while the NVIDIA GeForce GTX 1070 Ti is rated at 180 W TDP and requires a single 8-pin power connector with a suggested 450 W power supply.

Architecture Differences

The AMD Radeon Pro 5600M uses the Navi 12 chip built on TSMC's 7 nm process with RDNA 1.0 architecture. The NVIDIA GeForce GTX 1070 Ti uses the GP104 chip on TSMC's 16 nm process with Pascal architecture. This node advantage gives AMD a significant manufacturing lead, but the design philosophies diverge sharply.

The AMD card packs 2560 shading units, 160 texture mapping units, and 64 raster operation units. It uses 8 GB of HBM2 memory on a 2048-bit bus, delivering 394.2 GB/s of bandwidth. The NVIDIA card has 2432 shading units, 152 TMUs, and 64 ROPs, paired with 8 GB of GDDR5 memory on a 256-bit bus, offering 256.3 GB/s of bandwidth. The AMD memory subsystem has over 50% more bandwidth, which shows in its Metal test dominance.

Clock speeds tell a different story. The AMD Radeon Pro 5600M runs at an 822 MHz base and 1144 MHz boost, while the GTX 1070 Ti runs at 1607 MHz base and 1683 MHz boost. The NVIDIA card's much higher clocks push its pixel rate to 107.7 GPixel/s and texture rate to 255.8 GTexel/s, versus 73.22 GPixel/s and 183.0 GTexel/s for AMD. In raw FP32 compute, NVIDIA leads with 8.186 TFLOPS against 5.857 TFLOPS.

The FP16 situation is reversed. The AMD card delivers 11.71 TFLOPS with a 2:1 ratio, while the NVIDIA card manages only 127.9 GFLOPS with a 1:64 ratio. This makes the Radeon Pro 5600M far more capable in workloads that use half-precision arithmetic.

The NVIDIA chip is built on a much larger scale: 7,200 million transistors on a 314 mm² die with a density of 22.9M per mm². The AMD die size and transistor count are not recorded in the database. The GTX 1070 Ti is a dual-slot card measuring 267 mm in length, 112 mm in height, and 40 mm in width, while the Radeon Pro 5600M is an integrated graphics processor with no separate slot width.

Both support DirectX 12 (12_1) and OpenGL 4.6. The AMD card supports Vulkan 1.3, while the NVIDIA card supports Vulkan 1.4. The NVIDIA card uses PCIe 3.0 x16, while the AMD card uses PCIe 4.0 x16, offering double the interconnect bandwidth. Display outputs are portable device dependent for AMD, while NVIDIA provides 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a.

Head-to-Head Benchmarks

The single AMD victory is decisive. In Geekbench Metal, the Radeon Pro 5600M scores 55030 versus 22180 for the GTX 1070 Ti, a 148.1% lead. This reflects the AMD card's HBM2 memory bandwidth and RDNA architecture being highly optimized for Apple's Metal API, which is the primary workload for this mobile Mac GPU.

Every other benchmark favors NVIDIA, often by substantial margins. The largest NVIDIA wins come in DirectX 11 and DirectX 9. In Passmark DirectX 11, the GTX 1070 Ti scores 107 versus 59, a 44.9% advantage. In Passmark DirectX 9, it scores 207 versus 118, a 43% lead. These legacy API tests highlight the NVIDIA card's raw rasterization throughput.

Compute workloads also favor NVIDIA. In Passmark GPU Compute, the GTX 1070 Ti scores 7165 versus 4031, a 43.7% margin. The Geekbench OpenCL test is closer: NVIDIA scores 49219 versus 47783, a narrow 2.9% win. This suggests the AMD card is competitive in general-purpose compute but loses ground in more specialized tests.

The Vulkan gap is significant. NVIDIA scores 61006 versus 46425, a 23.9% advantage. This matters for modern cross-platform games and applications. The DirectX 12 test shows a smaller but still clear NVIDIA lead: 51 versus 39, a 23.5% margin.

In the overall gaming proxy, Passmark G3D, the GTX 1070 Ti scores 14674 versus 9279, a 36.8% lead. The Passmark DirectX 10 test shows NVIDIA ahead by 31.1% (90 versus 62). Even the 2D test favors NVIDIA: 876 versus 679, a 22.5% margin.

The average benchmark score tells a different story. The Radeon Pro 5600M averages 16351 across all recorded tests, placing it near the AMD Radeon RX 5700 XT (16361, only 0.1% behind) and above the NVIDIA RTX PRO 6000 Blackwell (16408, 0.3% behind). The GTX 1070 Ti averages 14277, sitting near the Intel Iris XE MAX (14315, 0.3% behind) and AMD Radeon Vega 11 (14352, 0.5% behind). The AMD card's high score is driven by its exceptional Metal result, which pulls the average up despite losing nine individual comparisons.

The Verdict

The data draws a clear split. If the workload is Metal-based, the AMD Radeon Pro 5600M is the overwhelming choice. Its 148.1% advantage in Geekbench Metal is not a marginal edge but a categorical one, and its average benchmark score of 16351 sits 14.5% above the NVIDIA card's 14277.

If the workload is anything else, the NVIDIA GeForce GTX 1070 Ti wins. It leads in every DirectX test, both Vulkan and OpenCL, GPU compute, and 2D performance. The margins range from 2.9% in OpenCL to 44.9% in DirectX 11. For gaming, DirectX 12, or general compute, the GTX 1070 Ti is the stronger part, evidenced by its 36.8% lead in Passmark G3D.

The architecture explains the divide. The Radeon Pro 5600M is a 50 W integrated GPU with HBM2 memory and RDNA 1.0, designed for Mac portables. The GTX 1070 Ti is a 180 W dual-slot discrete card with GDDR5 and Pascal, built for desktop performance. The power draw difference, 50 W versus 180 W, reflects fundamentally different design targets.

The NVIDIA card also holds the edge in legacy API support, with higher DirectX 9 and 10 scores, and in Vulkan, where its 1.4 support versus AMD's 1.3 gives it a specification advantage. The AMD card counters with PCIe 4.0 x16 versus PCIe 3.0 x16, which helps in bandwidth-sensitive tasks but does not translate to benchmark wins outside Metal.

Specification Differences

| Specification | AMD Radeon Pro 5600M | NVIDIA GeForce GTX 1070 Ti |

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

| Chip | Navi 12 | GP104 |

| Architecture | RDNA 1.0 | Pascal |

| Process Node | 7 nm | 16 nm |

| Transistors | Not recorded | 7,200 million |

| Die Size | Not recorded | 314 mm² |

| Transistor Density | Not recorded | 22.9M / mm² |

| Base Clock | 822 MHz | 1607 MHz |

| Boost Clock | 1144 MHz | 1683 MHz |

| Memory Type | HBM2 | GDDR5 |

| Memory Bus Width | 2048 bit | 256 bit |

| Memory Bandwidth | 394.2 GB/s | 256.3 GB/s |

| Memory Clock | 770 MHz (1540 Mbps effective) | 2002 MHz (8 Gbps effective) |

| Shading Units | 2560 | 2432 |

| TMUs | 160 | 152 |

| ROPs | 64 | 64 |

| Pixel Rate | 73.22 GPixel/s | 107.7 GPixel/s |

| Texture Rate | 183.0 GTexel/s | 255.8 GTexel/s |

| FP32 | 5.857 TFLOPS | 8.186 TFLOPS |

| FP16 | 11.71 TFLOPS (2:1) | 127.9 GFLOPS (1:64) |

| TDP | 50 W | 180 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | Not recorded | 450 W |

| Bus Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |

| Vulkan | 1.3 | 1.4 |

| Display Outputs | Portable Device Dependent | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a |

| Dimensions | Not recorded | 267 mm x 112 mm x 40 mm |

| Release Date | 2020-06-14 | 2017-11-01 |

| Launch MSRP | Not recorded | 399 USD |

Where Each One Wins

AMD Radeon Pro 5600M wins in Metal environments. Its 55030 Geekbench Metal score is 148.1% higher than NVIDIA's, making it the clear choice for macOS applications that leverage the Metal API. This is a mobile GPU with a 50 W TDP, no power connectors, and an integrated form factor, so it suits thin-and-light Mac portables where power efficiency and compact design matter. Its FP16 performance, 11.71 TFLOPS versus 127.9 GFLOPS, also gives it a distinct edge in half-precision compute workloads. The higher average benchmark score, 16351 versus 14277, reflects its stronger overall position in the database, despite losing most individual tests.

NVIDIA GeForce GTX 1070 Ti wins in gaming and general compute. It leads in every DirectX test, from DirectX 9 (207 versus 118) through DirectX 12 (51 versus 39), and in Passmark G3D (14674 versus 9279). Its Vulkan score is 61006 versus 46425, a 23.9% margin, and its OpenCL score is 49219 versus 47783. The GPU compute score of 7165 versus 4031 shows a 43.7% advantage in compute-heavy tasks. The GTX 1070 Ti also wins in 2D performance (876 versus 679) and legacy DirectX 10 (90 versus 62).

The NVIDIA card's higher clocks, 1683 MHz boost versus 1144 MHz, drive its superior pixel and texture rates. Its 180 W TDP and dual-slot design indicate a desktop-oriented card that can sustain high performance. For users running DirectX titles, Vulkan applications, or OpenCL workloads, the GTX 1070 Ti is the stronger choice across the board. The choice between these two GPUs depends entirely on the operating system and application ecosystem. Metal users should pick AMD; everyone else should pick NVIDIA.