AMD Radeon Pro 555X vs NVIDIA GeForce GTX 1650 SUPER Comparison

Head-to-Head Benchmarks

The recorded data shows a decisive performance gap between the AMD Radeon Pro 555X and the NVIDIA GeForce GTX 1650 SUPER, with the NVIDIA part winning both head-to-head tests in the database. The two compute-focused benchmarks, Geekbench OpenCL and Geekbench Vulkan, reveal the scale of the separation.

In Geekbench OpenCL, the AMD Radeon Pro 555X scores 12,628 points, while the NVIDIA GeForce GTX 1650 SUPER scores 43,875 points. That is a 71.2% deficit for the AMD part, meaning the NVIDIA GPU delivers roughly 3.5 times the OpenCL performance. In Geekbench Vulkan, the margin is even wider: the Radeon Pro 555X scores 12,855 points against 50,519 points for the GTX 1650 SUPER, a 74.6% shortfall. The NVIDIA card is nearly 4 times faster in this API.

These are not marginal differences. The GTX 1650 SUPER wins both tests by overwhelming margins, and the database records zero wins for the Radeon Pro 555X in head-to-head comparisons. The average benchmark score tells a similar story: the AMD part averages 13,321 points across all recorded tests, while the NVIDIA part averages 11,047 points. It is important to note that the averages include different test sets for each card, so the head-to-head results are the more reliable comparison.

The Radeon Pro 555X sits in the 54th percentile of all GPUs, slightly above the GTX 1650 SUPER's 50th percentile, but that ranking is based on each card's own benchmark suite. The GTX 1650 SUPER's nearest rivals in the database include the AMD Radeon RX 550 (average score 11,075, delta -0.2%), the NVIDIA RTX PRO 6000D Blackwell Max-Q (11,088, delta -0.4%), and the NVIDIA GeForce MX350 (10,883, delta 1.5%). The Radeon Pro 555X's nearest rivals are the NVIDIA GeForce GTX 480 (13,300, delta 0.2%), AMD FirePro M6100 (13,354, delta -0.2%), AMD Radeon RX 5500M (13,356, delta -0.3%), and AMD Radeon HD 8950M (13,376, delta -0.4%). The delta values are all within 1.5%, indicating that each card is tightly clustered with its peer group.

Where Each One Wins

The GTX 1650 SUPER wins every directly comparable workload in the database. Its OpenCL and Vulkan scores are both far ahead, and it also posts strong results in other benchmarks that the AMD card does not share. In Passmark G3D, the NVIDIA card scores 10,179 points; in Passmark G2D, 749 points; and in Passmark GPU Compute, 4,477 points. DirectX legacy tests show scores of 148 in DirectX 9, 73 in DirectX 11, 50 in DirectX 10, and 45 in DirectX 12. For modern API workloads, 3DMark Steel Nomad DX12 yields 352 points.

The Radeon Pro 555X has no benchmark category where it records a win over the GTX 1650 SUPER. Its strength lies in its profile as a mobile workstation GPU: it is an integrated graphics processor (IGP) with no power connectors, drawing 75 W, and it targets portable devices where the display outputs are device-dependent. The GTX 1650 SUPER, by contrast, is a dual-slot discrete card with a 100 W TDP, one 6-pin power connector, and a recommended 300 W power supply. It is a desktop-oriented card with fixed display outputs: one DVI, one HDMI 2.0, and one DisplayPort 1.4a.

For compute-heavy tasks such as OpenCL and Vulkan, the data points clearly to the NVIDIA card. For legacy DirectX workloads, the GTX 1650 SUPER also leads, though the AMD card does not have comparable Passmark scores recorded, so those comparisons cannot be made directly. The Radeon Pro 555X is the only option for systems that require an integrated GPU with no additional power cabling, but in raw performance terms the NVIDIA card dominates every shared test.

Architecture Differences

The two GPUs come from different architectural generations and foundries. The AMD Radeon Pro 555X uses the Polaris 21 chip built on GCN 4.0 architecture, manufactured on a 14 nm process at GlobalFoundries. The NVIDIA GeForce GTX 1650 SUPER uses the TU116 chip built on Turing architecture, manufactured on a 12 nm process at TSMC. The transistor counts differ substantially: the AMD chip contains 3,000 million transistors on a 123 mm² die, for a density of 24.4 million transistors per mm². The NVIDIA chip contains 6,600 million transistors on a 284 mm² die, for a density of 23.2 million transistors per mm². The NVIDIA die is more than twice the physical size and holds more than twice the transistor count, though the density is slightly lower.

The compute resources are also unevenly distributed. The Radeon Pro 555X has 768 shading units, 48 texture mapping units, and 16 raster operations pipelines. The GTX 1650 SUPER has 1,280 shading units, 80 TMUs, and 32 ROPs. That means the NVIDIA card has 66.7% more shading units, 66.7% more TMUs, and twice the ROPs. The pixel rate for the AMD card is 14.51 GPixel/s, while the NVIDIA card achieves 55.20 GPixel/s. Texture rates are 43.54 GTexel/s for AMD and 138.0 GTexel/s for NVIDIA. FP32 compute is 1,393.2 GFLOPS for the AMD card versus 4.416 TFLOPS for the NVIDIA card. FP16 rates show a different ratio: the AMD card does 1,393.2 GFLOPS at a 1:1 ratio, while the NVIDIA card does 8.832 TFLOPS at a 2:1 ratio, indicating that the Turing architecture processes half-precision at twice the single-precision rate.

Memory configurations share the same 4 GB capacity and 128 bit bus width, but the type and speed differ. The AMD card uses GDDR5 at 1470 MHz with 5.9 Gbps effective, producing 94.08 GB/s of bandwidth. The NVIDIA card uses GDDR6 at 1500 MHz with 12 Gbps effective, producing 192.0 GB/s, which is more than double the AMD bandwidth. Clock behavior also differs: the Radeon Pro 555X has no listed base or boost clocks, while the GTX 1650 SUPER runs at 1530 MHz base and 1725 MHz boost.

API support is similar but not identical. Both cards support DirectX 12 and OpenGL 4.6. The AMD card supports DirectX 12 (12_0) and Vulkan 1.3; the NVIDIA card supports DirectX 12 (12_1) and Vulkan 1.4. Neither card has dedicated ray tracing or tensor cores, so those fields are null for both. The bus interface differs as well: the AMD card uses PCIe 3.0 x8, while the NVIDIA card uses PCIe 3.0 x16. The release dates are about 16 months apart, with the AMD card launching on 2018-07-15 and the NVIDIA card on 2019-11-21. Both are end-of-life. The NVIDIA card has a listed predecessor (GeForce 10) and successor (GeForce 20), while the AMD card has neither.

FAQ

Q: Which GPU is faster in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 1650 SUPER scores 43,875 points versus 12,628 points for the AMD Radeon Pro 555X, a 71.2% lead.

Q: How do the two compare in Geekbench Vulkan?

A: The GTX 1650 SUPER scores 50,519 points, which is 74.6% ahead of the Radeon Pro 555X's 12,855 points.

Q: Do both cards support the same DirectX version?

A: No. The Radeon Pro 555X supports DirectX 12 (12_0), while the GTX 1650 SUPER supports DirectX 12 (12_1). Both support OpenGL 4.6, but Vulkan support differs, with the NVIDIA card at 1.4 and the AMD card at 1.3.

Q: What are the memory bandwidth figures?

A: The AMD card has 94.08 GB/s from GDDR5 memory, while the NVIDIA card has 192.0 GB/s from GDDR6, both on a 128 bit bus with 4 GB capacity.

Q: Which card has higher FP32 compute?

A: The GTX 1650 SUPER delivers 4.416 TFLOPS, compared to 1,393.2 GFLOPS for the Radeon Pro 555X.

Q: What is the power requirement difference?

A: The Radeon Pro 555X is an integrated GPU with a 75 W TDP and no power connectors, while the GTX 1650 SUPER is a dual-slot card with a 100 W TDP, one 6-pin connector, and a 300 W suggested power supply.

Specification Differences

| Specification | AMD Radeon Pro 555X | NVIDIA GeForce GTX 1650 SUPER |

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

| Architecture | GCN 4.0 | Turing |

| Process Node | 14 nm | 12 nm |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 3,000 million | 6,600 million |

| Die Size | 123 mm² | 284 mm² |

| Transistor Density | 24.4M / mm² | 23.2M / mm² |

| Base Clock | Not listed | 1530 MHz |

| Boost Clock | Not listed | 1725 MHz |

| Memory Clock | 1470 MHz, 5.9 Gbps effective | 1500 MHz, 12 Gbps effective |

| Memory Type | GDDR5 | GDDR6 |

| Memory Size | 4 GB | 4 GB |

| Bus Width | 128 bit | 128 bit |

| Bandwidth | 94.08 GB/s | 192.0 GB/s |

| Shading Units | 768 | 1,280 |

| TMUs | 48 | 80 |

| ROPs | 16 | 80 |

| ROPs | 16 | 32 |

| Pixel Rate | 14.51 GPixel/s | 55.20 GPixel/s |

| Texture Rate | 43.54 GTexel/s | 138.0 GTexel/s |

| FP32 | 1,393.2 GFLOPS | 4.416 TFLOPS |

| FP16 | 1,393.2 GFLOPS (1:1) | 8.832 TFLOPS (1:1) |

| TDP | 75 W | 100 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 6-pin |

| Suggested PSU | N/A | 300 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 3.0 x16 |

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

| DirectX Support | 12 (12_0) | 12 (12_1) |

| Vulkan Support | 1.3 | 1.4 |

| Release Date | 2018-07-15 | 2018-07-15 |

| Predecessor | None | 2019-11-21 |

| Successor | None | GeForce 10 |

| Launch MSRP | None | 20, 159 USD |