NVIDIA GeForce RTX 5090 vs NVIDIA N1X 48SM Comparison

The Verdict

The data clearly separates these two NVIDIA parts into entirely different roles. The GeForce RTX 5090 is a high-end discrete graphics card with a recorded average benchmark score of 79,842 and a 92nd percentile ranking among all GPUs. The N1X 48SM is an integrated graphics processor (IGP) with no recorded benchmark scores, a 50th percentile ranking, and an average score of zero in the database. The RTX 5090 is the only choice for any measurable graphics workload. The N1X 48SM, with its 128 GB memory allocation and IGP form factor, appears designed for a system-level role where raw rendering throughput is not the primary metric. The RTX 5090 targets gamers and workstation users who need maximum frame rates and compute throughput. The N1X 48SM targets a platform that needs unified memory capacity far beyond what the RTX 5090 offers, though its compute capabilities are drastically lower.

Architecture Differences

Both parts share the Blackwell 2.0 architecture and are fabricated on the same 5 nm process at TSMC. The similarity ends there. The RTX 5090 uses the GB202 chip with 92,200 million transistors on a 750 mm² die, giving a transistor density of 122.9M per mm². The N1X 48SM uses the GB20B chip on a 382 mm² die, with transistor count listed as unknown. The die size difference is substantial, with the RTX 5090's die nearly twice the area of the N1X 48SM's.

The shading unit count tells the performance story. The RTX 5090 carries 21,760 shading units, 680 texture mapping units, and 176 raster operation units. The N1X 48SM has 6,144 shading units, 384 TMUs, and 48 ROPs. The RTX 5090 has 170 ray tracing cores versus 48 on the N1X 48SM, and 680 tensor cores versus 192. These are not incremental differences; they represent a completely different class of silicon.

Clock behavior also diverges significantly. The RTX 5090 has a base clock of 2017 MHz and a boost clock of 2407 MHz. The N1X 48SM has a much lower base clock of 741 MHz but a boost clock of 2346 MHz, indicating a power-constrained design that relies on boosting to reach competitive frequencies. The memory subsystems are entirely different: the RTX 5090 uses 32 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth, while the N1X 48SM uses 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth. The N1X 48SM's bandwidth is less than one-sixth of the RTX 5090's, a direct consequence of the narrower bus and slower memory type.

API support also differs. The RTX 5090 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The N1X 48SM lists DirectX, OpenGL, and Vulkan as N/A, confirming it is not intended for standard graphics API workloads in the same way.

Where Each One Wins

The RTX 5090 wins in every measured graphics and compute benchmark category. Its recorded scores include 18,355 in 3DMark Steel Nomad DX12, 334,370 in Geekbench OpenCL, and 376,728 in Geekbench Vulkan. The PassMark suite shows 39,650 in G3D, 26,756 in GPU Compute, and 1,413 in G2D. The N1X 48SM has no recorded benchmark entries, so there is no data-driven basis for claiming any performance win.

The N1X 48SM wins in memory capacity. It offers 128 GB of LPDDR5X, which is four times the 32 GB on the RTX 5090. This capacity advantage points to use cases involving large datasets that must reside in unified memory, such as certain AI inference or data processing workloads. The N1X 48SM also wins on physical profile: it is an IGP with no power connectors and no slot width, while the RTX 5090 is a dual-slot card requiring a 16-pin connector and a 950 W suggested power supply.

The RTX 5090 wins on pixel rate at 423.6 GPixel/s versus 112.6 GPixel/s for the N1X 48SM, and on texture rate at 1,636.8 GTexel/s versus 900.9 GTexel/s. The FP32 compute figures are 104.8 TFLOPS versus 28.83 TFLOPS, and FP16 matches at a 1:1 ratio for both parts. The RTX 5090's FP32 output is roughly 3.6 times higher than the N1X 48SM's.

FAQ

Q: Which GPU has higher raw compute performance?

A: The RTX 5090 delivers 104.8 TFLOPS in both FP32 and FP16, while the N1X 48SM delivers 28.83 TFLOPS in both. The RTX 5090 is approximately 3.6 times faster in these metrics.

Q: Does the N1X 48SM have more memory than the RTX 5090?

A: Yes. The N1X 48SM has 128 GB of LPDDR5X on a 256-bit bus, while the RTX 5090 has 32 GB of GDDR7 on a 512-bit bus. The RTX 5090's bandwidth is 1.79 TB/s versus 273.2 GB/s for the N1X 48SM.

Q: What is the form factor difference?

A: The RTX 5090 is a dual-slot discrete card measuring 304 mm in length, 137 mm in height, and 40 mm in width, with a 16-pin power connector. The N1X 48SM is an IGP with no slot width, no dimensions listed, and no power connectors.

Q: Which GPU has better API support?

A: The RTX 5090 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The N1X 48SM lists all three APIs as N/A, indicating it is not designed for standard graphics API workloads.

Q: How do the clock speeds compare?

A: The RTX 5090 has a base clock of 2017 MHz and a boost clock of 2407 MHz. The N1X 48SM has a base clock of 741 MHz and a boost clock of 2346 MHz. The N1X 48SM's lower base clock suggests a power-limited design.

Q: What is the release timeline difference?

A: The RTX 5090 was released on 2025-01-29, while the N1X 48SM has a release date of 2026-05-31. The RTX 5090 has a predecessor in the GeForce 40 series and a successor in the GeForce 60 series, while the N1X 48SM lists neither.

Head-to-Head Benchmarks

The head-to-head benchmark table is empty in the database, and the N1X 48SM has zero recorded benchmark scores. Therefore, direct comparisons must rely on the RTX 5090's absolute scores and the architectural differences between the two parts.

The RTX 5090's strongest recorded result is in Geekbench Vulkan with a score of 376,728, followed by Geekbench OpenCL at 334,370. These compute-oriented tests highlight the card's tensor and shading capabilities. The 3DMark Steel Nomad DX12 score of 18,355 reflects real-time rendering performance. PassMark results show 39,650 in G3D, which aggregates multiple DirectX tests, and 26,756 in GPU Compute. The individual PassMark DirectX scores are 395 in DX9, 341 in DX11, 226 in DX10, and 185 in DX12. The G2D score of 1,413 indicates strong 2D acceleration as well.

The N1X 48SM's only recorded statistic is an average benchmark score of 0 with a 50th percentile ranking. Its nearest rivals list is empty, which means the database has no comparable parts for it. The RTX 5090's nearest rivals show it sitting slightly above the NVIDIA Tesla P100 PCIe 16 GB (0.3% higher average score) and the Tesla P100 PCIe 12 GB (0.6% higher), while being 1.1% above the AMD Radeon RX 6850M XT and 1.4% below the AMD Radeon Pro Vega 64X. These deltas are small, meaning the RTX 5090's average score of 79,842 is closely clustered with those parts in the database's aggregate ranking.

The FP32 throughput gap between the two parts is the clearest head-to-head measurement available. At 104.8 TFLOPS versus 28.83 TFLOPS, the RTX 5090 holds a 3.6x advantage. Pixel throughput shows a 3.8x gap (423.6 GPixel/s versus 112.6 GPixel/s), and texture throughput shows a 1.8x gap (1,636.8 GTexel/s versus 900.9 GTexel/s). Memory bandwidth shows the largest relative difference, with the RTX 5090's 1.79 TB/s being 6.5 times the N1X 48SM's 273.2 GB/s.

These figures confirm that the RTX 5090 is designed for maximum throughput across every graphics pipeline stage, while the N1X 48SM makes a different tradeoff: it prioritizes memory capacity and integration over raw speed.

Specification Differences

The two parts differ across nearly every specification field. The chip identifiers are GB202 for the RTX 5090 and GB20B for the N1X 48SM. The generation field lists GeForce 50 for the RTX 5090 and Blackwell IGP (N1x) for the N1X 48SM. Die size is 750 mm² versus 382 mm², and transistor density is 122.9M per mm² for the RTX 5090 while the N1X 48SM has no listed density.

Clock speeds differ in base frequency: 2017 MHz versus 741 MHz. Boost clocks are closer at 2407 MHz versus 2346 MHz. Memory clock is 1750 MHz (28 Gbps effective) for the RTX 5090 and 1067 MHz (8.5 Gbps effective) for the N1X 48SM. Memory size is 32 GB GDDR7 versus 128 GB LPDDR5X, with bus widths of 512-bit versus 256-bit and bandwidth of 1.79 TB/s versus 273.2 GB/s.

Compute unit counts show the RTX 5090 with 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The N1X 48SM has 6,144 shading units, 384 TMUs, 48 ROPs, 48 RT cores, and 192 tensor cores. Pixel rate is 423.6 GPixel/s versus 112.6 GPixel/s, and texture rate is 1,636.8 GTexel/s versus 900.9 GTexel/s. FP32 and FP16 are 104.8 TFLOPS versus 28.83 TFLOPS, both at a 1:1 ratio.

Power specifications differ sharply. The RTX 5090 has a TDP of 575 W, a dual-slot form factor, one 16-pin connector, and a suggested PSU of 950 W. The N1X 48SM has unknown TDP, an IGP form factor, no power connectors, and no suggested PSU. The RTX 5090 measures 304 mm by 137 mm by 40 mm, while the N1X 48SM has no listed dimensions.

Display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b for the RTX 5090, versus a single HDMI output for the N1X 48SM. Both use PCIe 5.0 x16 interfaces. The RTX 5090 has a launch MSRP of 1,999 USD, while the N1X 48SM has no launch MSRP recorded. Release dates are 2025-01-29 and 2026-05-31 respectively. The RTX 5090 lists a predecessor (GeForce 40) and successor (GeForce 60), while the N1X 48SM lists neither. Production status is Active for both.