Intel Arc B390 vs NVIDIA GeForce RTX 5090 SE Comparison

FAQ

Q: What is the performance difference between the Intel Arc B390 and the NVIDIA GeForce RTX 5090 SE?

A: The recorded data shows the Intel Arc B390 has an average benchmark score of 1482 in 3DMark Steel Nomad (DX12), placing it in the 9th percentile of all GPUs. The RTX 5090 SE has no recorded benchmark scores in the database, so a direct head-to-head comparison is not possible.

Q: How does the Intel Arc B390 compare to its closest rivals?

A: The Arc B390 is 1.3% ahead of the NVIDIA GeForce GT 520MX (score 1463), 1.5% ahead of the GeForce 800M (score 1460), 2.5% ahead of the GT 625 OEM (score 1446), and 2.7% ahead of the GT 710 (score 1443) in average benchmark score.

Q: What are the core specifications of each GPU?

A: The Arc B390 uses 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores on a 3 nm Intel process with a 300 MHz base and 2500 MHz boost clock. The RTX 5090 SE uses 14080 shading units, 440 TMUs, 160 ROPs, 110 RT cores, and 440 tensor cores on a 5 nm TSMC process with a 1740 MHz base and 2377 MHz boost clock.

Q: What memory configurations do the two GPUs use?

A: The Arc B390 uses system shared memory with system-dependent bandwidth. The RTX 5090 SE has 24 GB of GDDR7 memory on a 384-bit bus with 1.34 TB/s bandwidth.

Q: What is the power draw difference?

A: The Arc B390 has a TDP of 80 W and uses no power connectors, while the RTX 5090 SE has a TDP of 500 W and uses a single 16-pin connector with a suggested PSU of 900 W.

Q: What API support does each GPU offer?

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

Architecture Differences

The Intel Arc B390 and NVIDIA GeForce RTX 5090 SE represent fundamentally different design approaches. The Arc B390 is built on Intel's Xe3-LPG architecture using the Panther Lake chip, fabricated on a 3 nm process at Intel's own foundry. The RTX 5090 SE uses NVIDIA's Blackwell 2.0 architecture with the GB202 chip, fabricated on a 5 nm process at TSMC.

The die size difference is substantial. The RTX 5090 SE carries 92,200 million transistors on a 750 mm² die, yielding a transistor density of 122.9M per mm². The Arc B390 has an unknown transistor count and die size in the database, making direct density comparisons impossible.

The shading engine configurations diverge sharply. The Arc B390 packs 1536 shading units, 48 texture mapping units, and 24 raster output units. The RTX 5090 SE scales up to 14080 shading units, 440 TMUs, and 160 ROPs. This represents a 9.2x difference in shading units, a 9.2x difference in TMUs, and a 6.7x difference in ROPs.

Ray tracing hardware also differs. The Arc B390 includes 12 RT cores, while the RTX 5090 SE has 110 RT cores. Additionally, the RTX 5090 SE includes 440 tensor cores, a feature that the Arc B390 lacks entirely in the recorded specifications.

Clock behavior shows an interesting inversion. The Arc B390 has a lower base clock of 300 MHz but a higher boost clock of 2500 MHz. The RTX 5090 SE starts at 1740 MHz base and boosts to 2377 MHz. The Arc B390's boost clock is 123 MHz higher than the RTX 5090 SE's boost, but its base clock is 1440 MHz lower.

Memory architecture is another major divergence. The Arc B390 relies on system shared memory with system-dependent bandwidth, reflecting its integrated graphics positioning. The RTX 5090 SE uses dedicated 24 GB GDDR7 memory with a 384-bit bus and 1.34 TB/s bandwidth.

The power envelope separates the two dramatically. The Arc B390 operates at an 80 W TDP with no power connectors, classified as an IGP with a portable-device-dependent display output. The RTX 5090 SE draws 500 W, requires a 16-pin connector, a 900 W suggested PSU, and uses a dual-slot form factor with 1x HDMI 2.1b and 3x DisplayPort 2.1b outputs.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark results between the Intel Arc B390 and the NVIDIA GeForce RTX 5090 SE. The head-to-head benchmark array is empty, and the wins counter shows 0 for both GPUs. The RTX 5090 SE has no recorded benchmark scores at all, with an average benchmark score of 0 and no nearest rivals listed.

The Arc B390's only recorded benchmark is the 3DMark Steel Nomad DX12 test, where it scores 1482. This places it in the 9th percentile of all GPUs. Its nearest rivals in the database are all older NVIDIA integrated or entry-level parts: the GT 520MX (1463, 1.3% behind), the GeForce 800M (1460, 1.5% behind), the GT 625 OEM (1446, 2.5% behind), and the GT 710 (1443, 2.7% behind).

The RTX 5090 SE sits in the 50th percentile of all GPUs, but this percentile is based on no actual benchmark data. The percentile value of 50 with zero recorded scores indicates a placeholder position rather than a measured performance level.

Given the absence of overlapping benchmark data, no quantitative comparison can be established between the two GPUs. The performance gap suggested by the specification differences, such as the 8.7x difference in FP32 throughput (7.680 TFLOPS for the Arc B390 versus 66.94 TFLOPS for the RTX 5090 SE), cannot be confirmed by benchmark results in the database.

Texture and pixel rates show similarly large gaps. The Arc B390 delivers 120.0 GTexel/s and 60.00 GPixel/s. The RTX 5090 SE delivers 1,045.9 GTexel/s and 380.3 GPixel/s. These are 8.7x and 6.3x differences, respectively, but without benchmark validation, these remain theoretical specifications.

Specification Differences

The two GPUs differ across nearly every recorded specification field. The process node differs: 3 nm for the Arc B390 versus 5 nm for the RTX 5090 SE. The foundry also differs: Intel for the Arc B390 versus TSMC for the RTX 5090 SE.

Transistor counts and die sizes are recorded only for the RTX 5090 SE. It has 92,200 million transistors on a 750 mm² die with a density of 122.9M per mm². The Arc B390 lists these fields as unknown.

Clock speeds differ in both directions. The Arc B390 has a 300 MHz base clock and a 2500 MHz boost clock. The RTX 5090 SE has a 1740 MHz base clock and a 2377 MHz boost clock. Memory clocks also differ: the Arc B390 uses system shared memory, while the RTX 5090 SE uses 1750 MHz with 28 Gbps effective speed.

Memory capacity, type, bus width, and bandwidth all differ. The Arc B390 uses system shared memory, while the RTX 5090 SE has 24 GB GDDR7 on a 384-bit bus with 1.34 TB/s bandwidth.

Core counts differ as follows: 1536 versus 14080 shading units, 48 versus 440 TMUs, 24 versus 160 ROPs, and 12 versus 110 RT cores. The RTX 5090 SE has 440 tensor cores; the Arc B390 has none listed.

Pixel rate is 60.00 GPixel/s for the Arc B390 versus 380.3 GPixel/s for the RTX 5090 SE. Texture rate is 120.0 GTexel/s versus 1,045.9 GTexel/s. FP32 compute is 7.680 TFLOPS versus 66.94 TFLOPS. FP16 compute is 15.36 TFLOPS (2:1 ratio) for the Arc B390 versus 66.94 TFLOPS (1:1 ratio) for the RTX 5090 SE.

TDP differs: 80 W versus 500 W. Slot width differs: IGP versus dual-slot. Power connectors differ: none versus 1x 16-pin. The suggested PSU is listed only for the RTX 5090 SE at 900 W. The bus interface differs: IGP versus PCIe 5.0 x16.

Display outputs differ: portable device dependent for the Arc B390 versus 1x HDMI 2.1b and 3x DisplayPort 2.1b for the RTX 5090 SE. Dimensions are recorded only for the RTX 5090 SE: 267 mm length, 111 mm height, and 40 mm width.

Release dates differ: the Arc B390 launched on 2026-01-26, while the RTX 5090 SE launched on 2025-12-31. The RTX 5090 SE has a predecessor (GeForce 40) and successor (GeForce 60); the Arc B390 has neither listed. The RTX 5090 SE has a launch MSRP of 1,499 USD; the Arc B390 has no launch MSRP recorded.

Where Each One Wins

The Intel Arc B390 wins in power efficiency. Its 80 W TDP with no power connectors makes it suitable for integrated graphics implementations in portable devices. The 300 MHz base clock rising to 2500 MHz boost shows a wide dynamic range for power management.

The Arc B390 also wins on process technology. The 3 nm node is more advanced than the 5 nm node used by the RTX 5090 SE. This process advantage partially explains the Arc B390's lower power draw despite a similar boost clock.

The Arc B390 wins in the only recorded benchmark comparison available. Its 1482 score in 3DMark Steel Nomad places it slightly ahead of four NVIDIA rivals in the database. However, this benchmark result says nothing about performance relative to the RTX 5090 SE.

The NVIDIA GeForce RTX 5090 SE wins on raw compute specifications. It delivers 66.94 TFLOPS FP32 performance, which is 8.7x the Arc B390's 7.680 TFLOPS. Its texture rate of 1,045.9 GTexel/s and pixel rate of 380.3 GPixel/s dwarf the Arc B390's corresponding 120.0 GTexel/s and 60.00 GPixel/s.

The RTX 5090 SE wins on memory capacity and bandwidth. Its 24 GB GDDR7 with 1.34 TB/s bandwidth provides 384-bit bus width, far exceeding the system shared memory approach of the Arc B390. The RTX 5090 SE also wins on feature set with 440 tensor cores for AI workloads and 110 RT cores for ray tracing.

The RTX 5090 SE wins on expansion capabilities. Its PCIe 5.0 x16 interface, dual-slot cooling, and dedicated display outputs (HDMI 2.1b and DisplayPort 2.1b) support discrete graphics configurations. The Arc B390's IGP bus interface and portable-device-dependent outputs limit it to integrated scenarios.

The RTX 5090 SE wins on physical size and cooling options. Its 267 mm length, 111 mm height, and 40 mm width with dual-slot design allow for substantial cooling hardware, consistent with its 500 W TDP.

The Verdict

The recorded data presents two GPUs with almost no overlapping measurement points. The Intel Arc B390 has exactly one benchmark result: 1482 in 3DMark Steel Nomad, placing it in the 9th percentile of all GPUs. The NVIDIA GeForce RTX 5090 SE has zero benchmark results, an average score of 0, and no nearest rivals listed.

For users selecting between these two, the specification sheet favors the RTX 5090 SE in nearly every compute category. The 14080 shading units, 66.94 TFLOPS FP32 throughput, 24 GB GDDR7 memory, and 1.34 TB/s bandwidth represent a substantially larger hardware configuration. The 110 RT cores and 440 tensor cores add capabilities that the Arc B390 does not list.

The Arc B390's advantages are power consumption and integration. An 80 W TDP with no power connectors and an IGP form factor suits portable devices where the RTX 5090 SE's 500 W TDP, 16-pin connector, and 900 W suggested PSU would be impractical.

The process node comparison favors the Arc B390 at 3 nm versus 5 nm, but the RTX 5090 SE compensates with a much larger die: 750 mm² with 92,200 million transistors.

The release timing shows the RTX 5090 SE launched on 2025-12-31, about four weeks before the Arc B390 on 2026-01-26. The RTX 5090 SE has a launch MSRP of 1,499 USD; the Arc B390 has no MSRP recorded.

The database cannot confirm a performance relationship between these two GPUs because no shared benchmark exists. The Arc B390's nearest rivals are low-end NVIDIA parts from previous generations, all within 2.7% of its score. The RTX 5090 SE's 50th percentile placement is unverified by any benchmark score.

In practical terms, the RTX 5090 SE targets discrete, high-power desktop graphics with dedicated memory and extensive compute resources. The Arc B390 targets integrated graphics in portable devices, relying on system memory and drawing a fraction of the power. The data supports choosing the RTX 5090 SE for applications demanding maximum compute throughput and memory bandwidth, and the Arc B390 for power-constrained, space-constrained integrated implementations.