AMD Radeon 840M vs NVIDIA GeForce RTX 5090 Mobile Comparison

The AMD Radeon 840M and the NVIDIA GeForce RTX 5090 Mobile represent two entirely different segments of the mobile graphics market. The database records no direct head-to-head benchmark matches between them, but the broader measurements for the NVIDIA part are extensive, while the AMD IGP has no recorded benchmark entries. The data indicates a massive performance gulf, with the RTX 5090 Mobile scoring in the 84th percentile of all GPUs, while the Radeon 840M sits at the 50th percentile with an average benchmark score of zero. This analysis walks through the recorded specifications and benchmark data to clarify where each part fits.

Where Each One Wins

The AMD Radeon 840M wins in efficiency and integration for ultra-portable systems. Its recorded power draw is 15 W, a figure that allows it to operate without any power connectors, functioning purely as an integrated graphics processor (IGP). It uses system-shared memory, meaning there is no dedicated VRAM allocation, and its memory bandwidth is listed as system dependent. This part is designed for devices where space and thermal budgets are minimal, and it delivers a base clock of 400 MHz with a boost clock of 2900 MHz. The Radeon 840M also holds a process node advantage at 4 nm, compared to the 5 nm node used by the NVIDIA part.

The NVIDIA GeForce RTX 5090 Mobile wins in raw performance and features. Its recorded benchmark scores are substantial: a 3DMark Steel Nomad DX12 score of 5871, a Geekbench OpenCL score of 201834, and a Geekbench Vulkan score of 198405. The PassMark G3D score reaches 30034, and the GPU compute score is 13401. The average benchmark score across all recorded tests is 45152. The RTX 5090 Mobile also carries the higher percentile ranking at 84 versus 50. In terms of hardware, it ships with 24 GB of GDDR7 memory on a 256-bit bus, delivering 896.0 GB/s of bandwidth, a figure that dwarfs the system-dependent memory of the AMD part. The NVIDIA part also includes 82 ray tracing cores and 328 tensor cores, features that the AMD IGP does not list.

Architecture Differences

The two GPUs come from different architectural generations. The AMD Radeon 840M uses RDNA 3.5 architecture, built on a 4 nm process at TSMC, and belongs to the Navi III IGP generation for Strix Point Mobile. Its chip is designated Krackan Point. The NVIDIA GeForce RTX 5090 Mobile uses Blackwell 2.0 architecture, built on a 5 nm process at TSMC, with the GB203 chip. The transistor counts diverge sharply: the RTX 5090 Mobile has 45,600 million transistors on a 378 mm² die, with a transistor density of 120.6M per mm². The Radeon 840M has an unknown transistor count and die size.

The compute resources differ by an order of magnitude. The Radeon 840M has 256 shading units, 16 texture mapping units (TMUs), and 8 raster operation pipelines (ROPs). The RTX 5090 Mobile has 10496 shading units, 328 TMUs, and 112 ROPs. Ray tracing hardware also differs: the AMD part has 4 ray tracing cores, while the NVIDIA part features 82. Tensor cores are present only on the NVIDIA side, with 328 units. The Radeon 840M lists no tensor cores at all. The memory architecture is fundamentally different, with the AMD part using system-shared memory and the NVIDIA part using dedicated GDDR7.

The bus interface also separates the two. The Radeon 840M connects via PCIe 4.0 x8, while the RTX 5090 Mobile uses PCIe 5.0 x16. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is not a differentiator. The NVIDIA part has a higher power draw at 95 W, but it still uses no power connectors, indicating it is designed for a soldered mobile implementation rather than a modular card.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark results between the Radeon 840M and the RTX 5090 Mobile. The Radeon 840M has no recorded benchmark scores, while the RTX 5090 Mobile has ten separate entries. This absence of direct comparison data means the analysis must rely on the recorded scores for the NVIDIA part and its nearest rivals to contextualize performance.

The RTX 5090 Mobile delivers a 3DMark Steel Nomad DX12 score of 5871, which reflects its ability to handle modern DirectX 12 workloads. The Geekbench OpenCL score of 201834 and Vulkan score of 198405 are nearly identical, indicating balanced compute performance across those two APIs. The PassMark suite shows a G3D score of 30034, which is the highest among the PassMark tests, while the GPU compute score is 13401. Lower scores appear in the legacy DirectX tests: DirectX 10 at 183, DirectX 11 at 269, DirectX 12 at 138, and DirectX 9 at 324. The G2D score of 1057 is notably lower than the G3D, reflecting that the part is heavily optimized for 3D workloads rather than 2D rendering.

The nearest rivals in the database for the RTX 5090 Mobile provide a reference frame. The AMD Radeon Pro 5500 XT has an average score of 45384, which is 0.5% lower than the RTX 5090 Mobile. The NVIDIA GeForce RTX 4070 Ti scores 44795, which is 0.8% lower. The Intel Arc A730M scores 45592, which is 1% higher than the RTX 5090 Mobile. The NVIDIA RTX 5880 Ada Generation scores 45972, which is 1.8% higher. These deltas place the RTX 5090 Mobile in a tight performance band, slightly behind the Intel Arc A730M and the RTX 5880 Ada Generation, but ahead of the Radeon Pro 5500 XT and the RTX 4070 Ti. The average benchmark score of 45152 confirms this positioning.

The Verdict

The data suggests two distinct use cases. The AMD Radeon 840M is suited for devices where power consumption is the primary constraint. Its 15 W power draw, system-shared memory, and IGP slot width indicate it is meant for thin and light laptops that do not require dedicated graphics. The 4 nm process node and 400 MHz base clock confirm a focus on efficiency rather than peak throughput. The 256 shading units and 4 ray tracing cores are enough for basic rendering and light gaming, but the absence of recorded benchmark scores means no performance validation exists in the database.

The NVIDIA GeForce RTX 5090 Mobile is for systems that demand maximum graphics capability. Its 10496 shading units, 82 ray tracing cores, and 328 tensor cores provide substantial compute headroom. The 24 GB GDDR7 memory with 896.0 GB/s bandwidth supports large datasets and high-resolution textures. The 95 W power draw is higher than the AMD part, but still connector-free, indicating a mobile design that prioritizes performance within a laptop chassis. The benchmark results confirm its place among high-end desktop and mobile GPUs, with the average score of 45152 placing it just behind the Intel Arc A730M and the RTX 5880 Ada Generation.

Users who need ray tracing and tensor core acceleration should select the RTX 5090 Mobile, as the Radeon 840M lacks tensor cores entirely and has far fewer ray tracing cores. Users who prioritize battery life and low heat output from a processor with minimal power requirements should consider the Radeon 840M. The percentile ranking reinforces this split: 84th percentile for the NVIDIA part versus 50th for the AMD part.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 5090 Mobile has an average benchmark score of 45152, while the AMD Radeon 840M has an average benchmark score of 0.

Q: How does the RTX 5090 Mobile compare to its nearest rivals?

A: The RTX 5090 Mobile scores 0.5% higher than the AMD Radeon Pro 5500 XT, 0.8% higher than the NVIDIA GeForce RTX 4070 Ti, 1% lower than the Intel Arc A730M, and 1.8% lower than the NVIDIA RTX 5880 Ada Generation.

Q: What memory configurations do the two GPUs use?

A: The AMD Radeon 840M uses system-shared memory with system-dependent bandwidth. The NVIDIA GeForce RTX 5090 Mobile uses 24 GB of GDDR7 memory on a 256-bit bus with 896.0 GB/s bandwidth.

Q: Do both GPUs support the same APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce RTX 5090 Mobile has 10496 shading units, while the AMD Radeon 840M has 256 shading units.

Q: What is the power draw difference?

A: The AMD Radeon 840M draws 15 W, while the NVIDIA GeForce RTX 5090 Mobile draws 95 W. Neither uses power connectors.

Specification Differences

| Specification | AMD Radeon 840M | NVIDIA GeForce RTX 5090 Mobile |

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

| Chip | Krackan Point | GB203 |

| Architecture | RDNA 3.5 | Blackwell 2.0 |

| Generation | Navi III IGP (Strix Point Mobile) | GeForce 50 Mobile |

| Process Node | 4 nm | 5 nm |

| Transistors | unknown | 45,600 million |

| Die Size | unknown | 378 mm² |

| Transistor Density | null | 120.6M / mm² |

| Base Clock | 400 MHz | 990 MHz |

| Boost Clock | 2900 MHz | 1515 MHz |

| Memory Size | System Shared | 24 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 256 bit |

| Memory Bandwidth | System Dependent | 896.0 GB/s |

| Shading Units | 256 | 10496 |

| TMUs | 16 | 328 |

| ROPs | 8 | 112 |

| Ray Tracing Cores | 4 | 82 |

| Tensor Cores | null | 328 |

| Pixel Rate | 23.20 GPixel/s | 169.7 GPixel/s |

| Texture Rate | 46.40 GTexel/s | 496.9 GTexel/s |

| FP32 Performance | 1,484.8 GFLOPS | 31.80 TFLOPS |

| FP16 Performance | 1,484.8 GFLOPS (1:1) | 31.80 TFLOPS (1:1) |

| TDP | 15 W | 95 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 5.0 x16 |

| Release Date | 2025-02-28 | 2025-03-26 |

| Predecessor | Navi II IGP | GeForce 40 Mobile |

| Percentile vs All GPUs | 50 | 84 |

| Average Benchmark Score | 0 | 45152 |