NVIDIA GeForce RTX 5090 Mobile vs NVIDIA N1X 48SM Comparison

The Verdict

The recorded data presents a clear split between these two NVIDIA mobile parts. The GeForce RTX 5090 Mobile is a finished, shipping product with a full suite of benchmark scores, while the N1X 48SM is listed as Active but carries no benchmark entries, an average score of 0, and a 50th percentile placement versus all GPUs. The RTX 5090 Mobile sits at the 84th percentile with an average benchmark score of 45152. For anyone selecting a GPU strictly from the database, the RTX 5090 Mobile is the only one with verified performance data; the N1X 48SM cannot be evaluated on measured results, only on its architectural specifications.

The N1X 48SM, despite lacking benchmarks, offers unique attributes in the specification sheet: 128 GB of LPDDR5X memory versus 24 GB of GDDR7, a higher boost clock of 2346 MHz versus 1515 MHz, and more texture mapping units (384 versus 328). The RTX 5090 Mobile counters with a higher base clock (990 MHz versus 741 MHz), more shading units (10496 versus 6144), more RT cores (82 versus 48), more tensor cores (328 versus 192), and substantially higher memory bandwidth (896.0 GB/s versus 273.2 GB/s). The data implies the RTX 5090 Mobile is built for raw throughput and ray tracing workloads, while the N1X 48SM appears oriented toward capacity, likely for large datasets or memory-hungry compute tasks.

Architecture Differences

Both GPUs share the Blackwell 2.0 architecture and are fabricated on a 5 nm process at TSMC. The chips differ: the RTX 5090 Mobile uses GB203 with a die size of 378 mm² and 45,600 million transistors, achieving a transistor density of 120.6M per mm². The N1X 48SM uses GB20B with a slightly larger die at 382 mm², but its transistor count and density are listed as unknown. The RTX 5090 Mobile belongs to the GeForce 50 Mobile generation, while the N1X 48SM is part of the Blackwell IGP (N1x) family, and its API support is marked as N/A for DirectX, OpenGL, and Vulkan, suggesting a non-standard software stack or a different intended usage environment.

The memory subsystems diverge sharply. The RTX 5090 Mobile uses 24 GB of GDDR7 on a 256-bit bus, delivering 896.0 GB/s of bandwidth with a memory clock of 1750 MHz (28 Gbps effective). The N1X 48SM uses 128 GB of LPDDR5X on the same 256-bit bus width, but its bandwidth is only 273.2 GB/s, with a memory clock of 1067 MHz (8.5 Gbps effective). The RTX 5090 Mobile has 10496 shading units, 328 TMUs, 112 ROPs, 82 RT cores, and 328 tensor cores. The N1X 48SM has 6144 shading units, 384 TMUs, 48 ROPs, 48 RT cores, and 192 tensor cores. The RTX 5090 Mobile leads in pixel rate (169.7 GPixel/s versus 112.6 GPixel/s) but trails in texture rate (496.9 GTexel/s versus 900.9 GTexel/s), a direct result of the N1X 48SM having more TMUs.

The RTX 5090 Mobile lists a TDP of 95 W and uses no power connectors, matching the N1X 48SM, whose TDP is unknown but also uses no power connectors. Both are IGP slot width and PCIe 5.0 x16. The RTX 5090 Mobile has display outputs marked as Portable Device Dependent, while the N1X 48SM lists 1x HDMI. The RTX 5090 Mobile released on 2025-03-26, while the N1X 48SM has a release date of 2026-05-31, indicating a later arrival for the latter.

Head-to-Head Benchmarks

The head-to-head benchmark table is empty, so no direct comparative scores exist between these two GPUs. The RTX 5090 Mobile has its own benchmark set: 3DMark Steel Nomad DX12 scores 5871, Geekbench OpenCL scores 201834, Geekbench Vulkan scores 198405, Passmark DirectX 10 scores 183, Passmark DirectX 11 scores 269, Passmark DirectX 12 scores 138, Passmark DirectX 9 scores 324, Passmark G2D scores 1057, Passmark G3D scores 30034, and Passmark GPU Compute scores 13401. The N1X 48SM has zero recorded benchmarks, so no wins can be assigned; the wins counter for both is 0.

The RTX 5090 Mobile’s nearest rivals in the database provide context: the AMD Radeon Pro 5500 XT averages 45384 (a delta of -0.5% versus the RTX 5090 Mobile), the NVIDIA GeForce RTX 4070 Ti averages 44795 (delta +0.8%), the Intel Arc A730M averages 45592 (delta -1%), and the NVIDIA RTX 5880 Ada Generation averages 45972 (delta -1.8%). These deltas show the RTX 5090 Mobile sits within roughly 2% of its closest competitors, with the RTX 4070 Ti being the only one ahead (by 0.8%). The N1X 48SM has no nearest rivals listed, reinforcing its unmeasured status.

The data implies the RTX 5090 Mobile, despite its high average score of 45152, is not dominant over its immediate peers; it trails three of four rivals in average score, though by margins under 2%. The N1X 48SM, with no benchmarks, cannot be compared on any metric beyond specifications, making its actual performance unknown.

FAQ

Q: Which GPU has more memory bandwidth?

A: The RTX 5090 Mobile has 896.0 GB/s of bandwidth from its GDDR7 memory, while the N1X 48SM has 273.2 GB/s from LPDDR5X, a difference of roughly 3.3 times in favor of the RTX 5090 Mobile.

Q: Which GPU has more shading units?

A: The RTX 5090 Mobile has 10496 shading units, versus 6144 on the N1X 48SM, giving the former a 71% higher count.

Q: Does the N1X 48SM support DirectX or Vulkan?

A: The database lists DirectX, OpenGL, and Vulkan as N/A for the N1X 48SM, while the RTX 5090 Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the release date difference?

A: The RTX 5090 Mobile has a release date of 2025-03-26, while the N1X 48SM is dated 2026-05-31, over a year later.

Q: Which GPU has more RT cores?

A: The RTX 5090 Mobile has 82 RT cores, compared to 48 on the N1X 48SM, a 71% advantage.

Q: What is the average benchmark score for each?

A: The RTX 5090 Mobile has an average benchmark score of 45152 with an 84th percentile placement, while the N1X 48SM has an average score of 0 with a 50th percentile placement.

Where Each One Wins

The RTX 5090 Mobile wins on raw compute metrics: higher FP32 performance (31.80 TFLOPS versus 28.83 TFLOPS), higher FP16 performance (31.80 TFLOPS versus 28.83 TFLOPS, both at 1:1 ratio), more shading units, more RT cores, more tensor cores, more ROPs, and higher pixel rate (169.7 GPixel/s versus 112.6 GPixel/s). Its memory bandwidth advantage (896.0 GB/s versus 273.2 GB/s) makes it suited for high-resolution rendering, ray tracing, and compute workloads that saturate memory. The RTX 5090 Mobile also has a higher base clock (990 MHz versus 741 MHz), though the N1X 48SM has a higher boost clock (2346 MHz versus 1515 MHz), implying the latter can reach a higher peak clock under load.

The N1X 48SM wins on memory capacity and texture throughput. Its 128 GB of LPDDR5X dwarfs the 24 GB on the RTX 5090 Mobile, making it a candidate for workloads requiring massive in-memory datasets, such as large model inference or scientific computing. Its texture rate of 900.9 GTexel/s exceeds the RTX 5090 Mobile’s 496.9 GTexel/s by 81%, and its 384 TMUs outnumber the 328 on the RTX 5090 Mobile. The N1X 48SM also has a later release date, suggesting a newer product cycle, but its lack of API support for DirectX, OpenGL, and Vulkan limits its applicability in standard gaming or graphics environments.

The RTX 5090 Mobile is the only one with benchmark evidence, so it wins on any measured performance metric by default. The N1X 48SM, without scores, cannot claim a verified win in any test, but its specifications point to a specialized role: memory-bound compute with high texture throughput, rather than general-purpose graphics.

Specification Differences

The two GPUs differ in nearly every core specification category. The chips are distinct: GB203 for the RTX 5090 Mobile versus GB20B for the N1X 48SM. The RTX 5090 Mobile has 45,600 million transistors and a 378 mm² die; the N1X 48SM has unknown transistors and a 382 mm² die. Base clocks are 990 MHz versus 741 MHz, and boost clocks are 1515 MHz versus 2346 MHz. Memory differs in size (24 GB versus 128 GB), type (GDDR7 versus LPDDR5X), and bandwidth (896.0 GB/s versus 273.2 GB/s), though both use a 256-bit bus. Shading units are 10496 versus 6144, TMUs are 328 versus 384, ROPs are 112 versus 48, RT cores are 82 versus 48, and tensor cores are 328 versus 192. Pixel rates are 169.7 GPixel/s versus 112.6 GPixel/s, and texture rates are 496.9 GTexel/s versus 900.9 GTexel/s. FP32 and FP16 are 31.80 TFLOPS for both on the RTX 5090 Mobile, and 28.83 TFLOPS for both on the N1X 48SM. TDP is 95 W for the RTX 5090 Mobile, unknown for the N1X 48SM. Display outputs are Portable Device Dependent versus 1x HDMI. API support is full for the RTX 5090 Mobile, N/A for the N1X 48SM. Release dates are 2025-03-26 versus 2026-05-31. The RTX 5090 Mobile lists a predecessor (GeForce 40 Mobile), while the N1X 48SM has none. Both share the same process node, foundry, slot width, power connectors, and bus interface.