AMD Radeon RX 7900M vs NVIDIA GeForce RTX 5090 SE Comparison

Head-to-Head Benchmarks

The database contains recorded benchmark results for the AMD Radeon RX 7900M, while the NVIDIA GeForce RTX 5090 SE has no benchmark scores listed in the current dataset. This makes a direct head-to-head comparison of measured performance impossible. The RX 7900M shows an average benchmark score of 97,487 across its recorded tests, placing it in the 94th percentile of all GPUs in the database. Its nearest rivals include the AMD Radeon Pro VII at 97,131, which trails by 0.4%, and the NVIDIA Quadro RTX 6000 at 101,872, which leads by 4.3%. The AMD Radeon Instinct MI60 scores 92,466, a 5.4% deficit, and the NVIDIA RTX A4500 scores 91,671, a 6.3% deficit.

Looking at individual tests, the RX 7900M delivers 4,201 points in 3DMark Steel Nomad DX12, 129,499 points in Geekbench OpenCL, and 158,760 points in Geekbench Vulkan. The RTX 5090 SE, by contrast, has no recorded entries in the benchmark suite, so no comparison of these specific scores can be made. The absence of measured data for the RTX 5090 SE means any claims about its relative performance against the RX 7900M would lack support from the database. The RX 7900M, with a 94th percentile ranking, sits among the higher-performing GPUs in the database, but the RTX 5090 SE's 50th percentile ranking reflects its zero recorded scores rather than any measured capability.

The RTX 5090 SE does show a substantial theoretical advantage on paper. Its FP32 throughput of 66.94 TFLOPS is roughly 74% higher than the RX 7900M's 38.52 TFLOPS. The texture rate of 1,045.9 GTexel/s for the RTX 5090 SE compares with 601.9 GTexel/s for the RX 7900M, a 73.8% advantage. The pixel rate, however, tells a different story: the RX 7900M delivers 401.3 GPixel/s versus 380.3 GPixel/s for the RTX 5090 SE, an advantage of 5.5% for the AMD part. Memory bandwidth heavily favors the RTX 5090 SE at 1.34 TB/s compared with 576.0 GB/s for the RX 7900M, more than double the AMD figure.

Where Each One Wins

The RX 7900M wins in the only measured benchmark category where both GPUs have any data, but that is because the RTX 5090 SE has none. The RX 7900M's pixel rate of 401.3 GPixel/s exceeds the RTX 5090 SE's 380.3 GPixel/s, which suggests the AMD card can fill more pixels per second, a factor that can benefit certain rasterization workloads. The RX 7900M also has a higher base clock at 1825 MHz versus 1740 MHz for the RTX 5090 SE, though the NVIDIA card boosts higher at 2377 MHz versus 2090 MHz.

The RTX 5090 SE wins on raw compute throughput. Its FP32 figure of 66.94 TFLOPS is nearly double the RX 7900M's 38.52 TFLOPS, and its FP16 performance matches its FP32 at 66.94 TFLOPS with a 1:1 ratio, whereas the RX 7900M reaches 77.05 TFLOPS FP16 only through a 2:1 rate. The RTX 5090 SE has 440 tensor cores, a feature the RX 7900M lacks entirely. It also has 110 RT cores versus 72 for the AMD part. Memory capacity and bandwidth go to NVIDIA as well: 24 GB of GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth, compared with 16 GB of GDDR6 on a 256-bit bus with 576.0 GB/s.

The RTX 5090 SE also wins on shading units, TMUs, and ROPs count in most categories: 14,080 shading units versus 4,608, 440 TMUs versus 288, though the RX 7900M has 192 ROPs versus 160. The RX 7900M's higher ROP count contributes to its pixel rate advantage. For a workload that stresses pixel fill, the AMD card has the edge on paper. For compute-heavy tasks, the NVIDIA card has a decisive theoretical lead.

Architecture Differences

The two GPUs come from different architectural generations. The RX 7900M uses the RDNA 3.0 architecture on the Navi 31 chip, codenamed Plum Bonito, and belongs to the Navi Mobile generation within the RX 7000 series. The RTX 5090 SE uses the Blackwell 2.0 architecture on the GB202 chip and belongs to the GeForce 50 generation. Both are built on a 5 nm process at TSMC, but the die sizes differ: the RX 7900M measures 529 mm² with 57,700 million transistors, while the RTX 5090 SE measures 750 mm² with 92,200 million transistors. Transistor density is higher on the NVIDIA part at 122.9M per mm² versus 109.1M per mm².

The memory subsystems reflect different design priorities. The RX 7900M uses 16 GB of GDDR6 with a 256-bit bus and 576.0 GB/s bandwidth, while the RTX 5090 SE uses 24 GB of GDDR7 with a 384-bit bus and 1.34 TB/s bandwidth. The RTX 5090 SE has a 1x 16-pin power connector and a suggested PSU of 900 W, while the RX 7900M has no power connectors listed and uses an integrated form factor (IGP). The RX 7900M draws 180 W TDP, while the RTX 5090 SE draws 500 W TDP.

The bus interfaces differ: the RX 7900M uses PCIe 4.0 x16, and the RTX 5090 SE uses PCIe 5.0 x16. Display outputs also differ, with the RX 7900M listed as portable device dependent and the RTX 5090 SE offering 1x HDMI 2.1b and 3x DisplayPort 2.1b. The RTX 5090 SE is a dual-slot card measuring 267 mm in length, 111 mm in height, and 40 mm in width, while the RX 7900M has no listed dimensions and is designed for mobile integration. The RX 7900M supports 12 Ultimate (12_2) DirectX, OpenGL 4.6, and Vulkan 1.4, as does the RTX 5090 SE.

The RTX 5090 SE has 440 tensor cores, which the RX 7900M does not have. It also has 110 RT cores versus 72. The RX 7900M has 4,608 shading units, 288 TMUs, and 192 ROPs, while the RTX 5090 SE has 14,080 shading units, 440 TMUs, and 160 ROPs. The RX 7900M's FP16 throughput of 77.05 TFLOPS exceeds its FP32 throughput due to the 2:1 ratio, while the RTX 5090 SE's FP16 matches its FP32 at 66.94 TFLOPS.

FAQ

Q: Which GPU has a higher average benchmark score in the database?

A: The AMD Radeon RX 7900M has an average benchmark score of 97,487, while the NVIDIA GeForce RTX 5090 SE has no recorded benchmark scores, giving it an average of 0.

Q: How does the RX 7900M compare with its nearest rivals?

A: The RX 7900M scores 0.4% higher than the AMD Radeon Pro VII, 5.4% higher than the AMD Radeon Instinct MI60, and 6.3% higher than the NVIDIA RTX A4500. The NVIDIA Quadro RTX 6000 leads it by 4.3%.

Q: What are the memory specifications of each GPU?

A: The RX 7900M has 16 GB of GDDR6 on a 256-bit bus with 576.0 GB/s bandwidth. The RTX 5090 SE has 24 GB of GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth.

Q: Which GPU has more RT cores?

A: The RTX 5090 SE has 110 RT cores, while the RX 7900M has 72 RT cores.

Q: What is the TDP of each GPU?

A: The RX 7900M has a TDP of 180 W. The RTX 5090 SE has a TDP of 500 W.

Q: Does the RTX 5090 SE have tensor cores?

A: Yes, the RTX 5090 SE has 440 tensor cores. The RX 7900M has no tensor cores listed.

The Verdict

The data in the database supports only one measured comparison, and it is incomplete. The RX 7900M has a 94th percentile ranking with an average benchmark score of 97,487, while the RTX 5090 SE has a 50th percentile ranking with no recorded scores. Anyone selecting between these two cards on the basis of recorded performance must choose the RX 7900M, because it is the only one with any benchmark data in the database.

But the paper specifications of the RTX 5090 SE indicate a much larger compute and memory footprint. Its FP32 throughput is 66.94 TFLOPS versus 38.52 TFLOPS, its memory bandwidth is 1.34 TB/s versus 576.0 GB/s, and it has 24 GB of memory versus 16 GB. The RTX 5090 SE also has 440 tensor cores, a feature absent from the RX 7900M, and 110 RT cores versus 72. Its power draw of 500 W and dual-slot, 267 mm physical format make it a desktop-oriented card, while the RX 7900M's 180 W TDP and IGP form factor target mobile systems.

The RX 7900M does hold a pixel rate advantage at 401.3 GPixel/s versus 380.3 GPixel/s, and it has more ROPs at 192 versus 160. Its base clock is higher at 1825 MHz versus 1740 MHz. For workloads that depend heavily on pixel fill, the AMD card has a theoretical edge. For compute, AI, and memory-bound tasks, the RTX 5090 SE has a clear theoretical lead. The RTX 5090 SE has a launch MSRP of 1,499 USD. The RX 7900M has no launch MSRP listed.

The recorded data favors the RX 7900M because it has actual scores. The theoretical specifications favor the RTX 5090 SE in most categories. The choice depends on whether the analysis prioritizes measured results or paper specifications. The RX 7900M is the only card with verified performance numbers in the database, and it sits in the 94th percentile of all GPUs. The RTX 5090 SE remains unmeasured, so its true standing cannot be confirmed. The RX 7900M is the safer pick for anyone who wants data-backed performance. The RTX 5090 SE is the pick for anyone who trusts its architectural specifications.

Specification Differences

| Specification | AMD Radeon RX 7900M | NVIDIA GeForce RTX 5090 SE |

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

| Architecture | RDNA 3.0 | Blackwell 2.0 |

| Chip | Navi 31 | GB202 |

| Process Node | 5 nm | 5 nm |

| Transistors | 57,700 million | 92,200 million |

| Die Size | 529 mm² | 750 mm² |

| Transistor Density | 109.1M / mm² | 122.9M / mm² |

| Base Clock | 1825 MHz | 1740 MHz |

| Boost Clock | 2090 MHz | 2377 MHz |

| Memory Size | 16 GB | 24 GB |

| Memory Type | GDDR6 | GDDR7 |

| Memory Bus Width | 256 bit | 384 bit |

| Memory Bandwidth | 576.0 GB/s | 1.34 TB/s |

| Shading Units | 4608 | 14080 |

| TMUs | 288 | 440 |

| ROPs | 192 | 160 |

| RT Cores | 72 | 110 |

| Tensor Cores | None | 440 |

| Pixel Rate | 401.3 GPixel/s | 380.3 GPixel/s |

| Texture Rate | 601.9 GTexel/s | 1,045.9 GTexel/s |

| FP32 | 38.52 TFLOPS | 66.94 TFLOPS |

| FP16 | 77.05 TFLOPS (2:1) | 66.94 TFLOPS (1:1) |

| TDP | 180 W | 500 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 16-pin |

| Suggested PSU | None | 900 W |

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

| Display Outputs | Portable Device Dependent | 1x HDMI 2.1b, 3x DisplayPort 2.1b |

| Dimensions | Not listed | 267 mm length, 111 mm height, 40 mm width |

| Release Date | 2023-10-18 | 2025-12-31 |

| Predecessor | Polaris Mobile | GeForce 40 |

| Successor | None | GeForce 60 |

| Production Status | Active | Active |