NVIDIA GeForce RTX 5090 SE vs NVIDIA N1X 48SM Comparison

Where Each One Wins

The recorded data paints a starkly different picture for these two NVIDIA parts, despite both being built on the Blackwell 2.0 architecture. The GeForce RTX 5090 SE is a dedicated graphics card built for maximum throughput, while the N1X 48SM is an integrated graphics processor (IGP) designed for a different class of system. There are no head-to-head benchmark wins recorded for either part in the database, as the benchmarks array is empty for both. The wins count is zero for each, meaning the comparison rests entirely on architectural and specification analysis rather than measured performance deltas.

The RTX 5090 SE is positioned for high-end desktop workloads. Its 24 GB of GDDR7 memory on a 384-bit bus delivers 1.34 TB/s of bandwidth, which is the kind of figure that supports heavy rasterization, ray tracing, and large texture sets. The N1X 48SM, by contrast, uses 128 GB of LPDDR5X on a 256-bit bus for 273.2 GB/s. That memory configuration suggests a very different role: one where capacity matters more than raw speed, likely for holding large datasets or many concurrent contexts in an integrated setting. The IGP has no power connector and is listed as "IGP" slot width, indicating it is soldered to a motherboard or platform rather than installed as a discrete card.

The practical split is clear. The RTX 5090 SE wins in any scenario that demands extreme fill rates, compute throughput, or memory bandwidth. The N1X 48SM wins in scenarios where a large memory pool and low physical footprint are the priorities, and where the absence of a discrete power connector is an advantage. The data shows a 2.32x advantage in FP32 throughput for the RTX 5090 SE (66.94 TFLOPS vs 28.83 TFLOPS), and a 4.9x advantage in memory bandwidth. The N1X 48SM counters with a 5.33x larger memory capacity (128 GB vs 24 GB). These are not competing products in the conventional sense; they serve different system architectures, but the specification comparison highlights where each part logically wins.

Architecture Differences

Both parts share the Blackwell 2.0 architecture and the same 5 nm process node from TSMC, but the implementations diverge sharply. The RTX 5090 SE uses the GB202 chip with a massive die size of 750 mm² and 92,200 million transistors, yielding a transistor density of 122.9 million per mm². The N1X 48SM uses the GB20B chip, which is considerably smaller at 382 mm². The transistor count for the GB20B is listed as "unknown" in the database, and the density is not recorded, so no direct transistor comparison is possible. The die size difference, however, is instructive: the GB202 is nearly twice the area of the GB20B, which aligns with the RTX 5090 SE having more than double the shading units (14080 vs 6144).

The memory subsystems are fundamentally different in type and purpose. The RTX 5090 SE pairs GDDR7 with a 384-bit interface, while the N1X 48SM uses LPDDR5X with a 256-bit interface. The clock speeds also differ: the RTX 5090 SE has a base clock of 1740 MHz and a boost of 2377 MHz, while the N1X 48SM runs at a much lower base of 741 MHz but boosts to a similar 2346 MHz. This suggests the IGP is designed to idle efficiently but can ramp up to near-discrete-card boost levels when needed. The memory clocks differ as well: 1750 MHz (28 Gbps effective) for the RTX 5090 SE versus 1067 MHz (8.5 Gbps effective) for the N1X 48SM.

API support is another major divergence. The RTX 5090 SE supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The N1X 48SM lists "N/A" for DirectX, OpenGL, and Vulkan support. This is a critical architectural difference: the IGP does not expose the same graphics API surface as the discrete card, which means it is not intended for the same class of gaming or graphics applications. The display outputs also differ, with the RTX 5090 SE offering 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the N1X 48SM has only a single HDMI output.

Head-to-Head Benchmarks

The database records no head-to-head benchmark results for these two parts. The headToHeadBenchmarks array is empty, and the winsA and winsB fields are both zero. The avgBenchmarkScore for each is also zero, and the percentileVsAllGpus is 50 for both, which is a neutral placeholder rather than a measured result. This absence of benchmark data means the comparison must be derived from the specification-level measurements recorded in the database.

The most significant measured differences come from the compute and memory figures. The RTX 5090 SE delivers 66.94 TFLOPS of FP32 performance, while the N1X 48SM delivers 28.83 TFLOPS. That is a 2.32x advantage for the discrete card. The texture rate tells a similar story: 1,045.9 GTexel/s for the RTX 5090 SE versus 900.9 GTexel/s for the N1X 48SM, a 1.16x difference that is much narrower than the FP32 gap. The pixel rate, however, shows a larger divergence: 380.3 GPixel/s versus 112.6 GPixel/s, a 3.38x advantage for the RTX 5090 SE.

The RT core and tensor core counts also differ substantially. The RTX 5090 SE has 110 RT cores and 440 tensor cores, while the N1X 48SM has 48 RT cores and 192 tensor cores. These are 2.29x and 2.29x differences, respectively, which align closely with the shading unit ratio. The memory bandwidth gap is the largest of all: 1.34 TB/s versus 273.2 GB/s, a 4.9x difference. This bandwidth differential is likely to be the most impactful in real-world workloads that are memory-bound. The N1X 48SM does not match the RTX 5090 SE on any recorded performance metric; its only numerical advantage is the memory capacity of 128 GB versus 24 GB.

Specification Differences

The two parts differ across nearly every recorded specification field. The process node is identical at 5 nm from TSMC, but the chips are different: GB202 for the RTX 5090 SE and GB20B for the N1X 48SM. The die sizes are 750 mm² versus 382 mm², and the transistor count is 92,200 million for the RTX 5090 SE while the N1X 48SM's transistor count is unknown. The base clocks are 1740 MHz versus 741 MHz, while the boost clocks are 2377 MHz versus 2346 MHz. The memory types are GDDR7 versus LPDDR5X, with capacities of 24 GB versus 128 GB, bus widths of 384 bit versus 256 bit, and bandwidths of 1.34 TB/s versus 273.2 GB/s.

The compute units show the RTX 5090 SE with 14080 shading units, 440 TMUs, 160 ROPs, 110 RT cores, and 440 tensor cores. The N1X 48SM has 6144 shading units, 384 TMUs, 48 ROPs, 48 RT cores, and 192 tensor cores. The pixel rates are 380.3 GPixel/s versus 112.6 GPixel/s, and the texture rates are 1,045.9 GTexel/s versus 900.9 GTexel/s. The FP32 and FP16 figures are both 66.94 TFLOPS for the RTX 5090 SE and 28.83 TFLOPS for the N1X 48SM, with both listed as 1:1 ratios.

The power and physical specifications differ as well. The RTX 5090 SE has a TDP of 500 W, a dual-slot form factor, a 1x 16-pin power connector, and a suggested PSU of 900 W. The N1X 48SM has an unknown TDP, an IGP form factor, no power connectors, and no suggested PSU. The RTX 5090 SE has dimensions of 267 mm length, 111 mm height, and 40 mm width, while the N1X 48SM has no recorded dimensions. The display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b for the RTX 5090 SE versus a single HDMI for the N1X 48SM. The API support also differs, with the RTX 5090 SE supporting DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while the N1X 48SM lists N/A for all three. The release dates are 2025-12-31 for the RTX 5090 SE and 2026-05-31 for the N1X 48SM, and the launch MSRP for the RTX 5090 SE is 1,499 USD while the N1X 48SM has no launch MSRP recorded.

FAQ

Q: Which part has higher FP32 compute performance?

A: The NVIDIA GeForce RTX 5090 SE delivers 66.94 TFLOPS of FP32 throughput, while the NVIDIA N1X 48SM delivers 28.83 TFLOPS. The RTX 5090 SE is 2.32x ahead on this metric.

Q: What is the memory capacity difference between the two?

A: The N1X 48SM has 128 GB of LPDDR5X memory, which is 5.33x larger than the RTX 5090 SE's 24 GB of GDDR7 memory. However, the RTX 5090 SE has 1.34 TB/s of bandwidth versus 273.2 GB/s for the N1X 48SM.

Q: Do both parts support the same graphics APIs?

A: No. The RTX 5090 SE supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The N1X 48SM lists N/A for DirectX, OpenGL, and Vulkan support in the database.

Q: What are the physical form factors of these parts?

A: The RTX 5090 SE is a dual-slot discrete card with a 1x 16-pin power connector and dimensions of 267 mm length, 111 mm height, and 40 mm width. The N1X 48SM is an IGP with no power connectors and no recorded dimensions.

Q: Which part has a higher boost clock?

A: The RTX 5090 SE has a boost clock of 2377 MHz, while the N1X 48SM has a boost clock of 2346 MHz. The difference is 31 MHz, less than 1.4%. The base clocks differ more significantly at 1740 MHz versus 741 MHz.

Q: What is the transistor count for each chip?

A: The RTX 5090 SE uses the GB202 chip with 92,200 million transistors. The N1X 48SM uses the GB20B chip, and its transistor count is recorded as unknown in the database.

The Verdict

The data indicates that the RTX 5090 SE is the superior choice for any workload that depends on raw graphics throughput, memory bandwidth, or compute density. It leads the N1X 48SM by 2.32x in FP32, 3.38x in pixel rate, and 4.9x in memory bandwidth. It also has more than double the shading units, RT cores, and tensor cores. The launch MSRP is 1,499 USD, which positions it as a high-end discrete solution. The N1X 48SM cannot match any of these performance figures.

The N1X 48SM, however, offers a memory capacity that is 5.33x larger at 128 GB, and it does so without any power connectors in an IGP form factor. It also has a slightly later release date of 2026-05-31 compared to the RTX 5090 SE's 2025-12-31. For systems where the priority is holding a very large memory footprint in an integrated package, the N1X 48SM has a clear specification advantage. The absence of API support for DirectX, OpenGL, and Vulkan, however, means it is not positioned for conventional gaming or graphics applications.

The recorded data does not include any benchmark results for either part, so the verdict rests on specification analysis. The RTX 5090 SE is the pick for discrete graphics performance, while the N1X 48SM is the pick for integrated memory capacity. The 500 W TDP and 900 W suggested PSU for the RTX 5090 SE contrast with the unknown TDP and no PSU requirement for the N1X 48SM, which further reinforces their different intended system roles.