Intel Arc B390 vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded data contains a single shared benchmark between the Intel Arc B390 and the NVIDIA GeForce RTX 5090 D: the 3DMark Steel Nomad DX12 test. The results are decisive. The Intel Arc B390 scores 1482 points, while the NVIDIA GeForce RTX 5090 D scores 14326 points. This represents a delta of -89.7% for the Intel part, meaning the RTX 5090 D delivers roughly nine and a half times the raw performance in this specific workload. The win count in the head-to-head section is 1 for the RTX 5090 D and 0 for the Arc B390.

Context from the database places the Arc B390 in the 9th percentile of all GPUs. Its nearest rivals are all legacy NVIDIA parts: the GeForce GT 520MX with an average score of 1463 (a 1.3% delta), the GeForce 800M at 1460 (1.5%), the GeForce GT 625 OEM at 1446 (2.5%), and the GeForce GT 710 at 1443 (2.7%). The Arc B390 therefore sits at the very low end of the performance spectrum, within a few percentage points of GPUs that are more than a decade old in design intent.

The RTX 5090 D, by contrast, sits in the 92nd percentile of all GPUs. Its average benchmark score across all recorded tests is 77712, and its nearest rivals are all high-end parts: the AMD Radeon RX 6650M XT at 76904 (a 1.1% delta in favor of the RTX 5090 D), the RX 6850M XT at 78940 (-1.6%), the NVIDIA Tesla P100 PCIe 12 GB at 79396 (-2.1%), and the Tesla P100 PCIe 16 GB at 79605 (-2.4%). The RTX 5090 D is bracketed by some of the fastest mobile and datacenter GPUs in the database, and it is competitive with them, trading places within roughly two and a half percent.

The single shared benchmark tells a consistent story. The RTX 5090 D's 14326 points in 3DMark Steel Nomad DX12 is not merely a win; it is a class-defining result that places it among the top tier of the database. The Arc B390's 1482 points is a score that places it in the company of entry-level parts from over a decade ago. There is no overlap in their performance envelopes, and no benchmark in the shared set shows the Arc B390 with an advantage.

FAQ

Q: Which GPU wins the only shared benchmark?

A: The NVIDIA GeForce RTX 5090 D wins the 3DMark Steel Nomad DX12 test with a score of 14326 against the Intel Arc B390's 1482, a delta of -89.7%.

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

A: The Arc B390 has an average benchmark score of 1482. Its nearest rival is the NVIDIA GeForce GT 520MX at 1463 (1.3% slower), followed by the GeForce 800M at 1460 (1.5% slower), the GeForce GT 625 OEM at 1446 (2.5% slower), and the GeForce GT 710 at 1443 (2.7% slower).

Q: How does the RTX 5090 D compare to its nearest rivals?

A: The RTX 5090 D has an average benchmark score of 77712. It leads the AMD Radeon RX 6650M XT by 1.1%, but trails the RX 6850M XT by 1.6%, the Tesla P100 PCIe 12 GB by 2.1%, and the Tesla P100 PCIe 16 GB by 2.4%.

Q: What percentile does each GPU occupy?

A: The Intel Arc B390 is in the 9th percentile of all GPUs, while the NVIDIA GeForce RTX 5090 D is in the 92nd percentile.

Q: What is the RTX 5090 D's average benchmark score across all recorded tests?

A: The RTX 5090 D has an average benchmark score of 77712, which includes results from 3DMark Steel Nomad, Geekbench OpenCL, Geekbench Vulkan, and multiple Passmark tests.

Q: Does the Arc B390 have any additional benchmark data beyond the shared test?

A: No. The database records only one benchmark for the Arc B390, which is the 3DMark Steel Nomad DX12 test at 1482 points.

Architecture Differences

The two GPUs come from entirely different architectural lineages. The Intel Arc B390 uses the Xe3-LPG architecture built on a 3 nm process at Intel's own foundry, with the chip codenamed Panther Lake. It belongs to the Arc Graphics-M (Panther Lake) generation. The NVIDIA GeForce RTX 5090 D uses the Blackwell 2.0 architecture on a 5 nm process fabricated by TSMC, with the chip codenamed GB202. It belongs to the GeForce 50 generation.

The RTX 5090 D's chip is a large discrete design with 92,200 million transistors on a die size of 750 mm², giving a transistor density of 122.9 million per mm². The Arc B390's transistor count and die size are listed as unknown, which reflects its nature as an integrated graphics processor rather than a standalone chip. The Arc B390's slot width is listed as IGP, and its bus interface is also IGP, meaning it is designed to be part of a processor package rather than a separate card. The RTX 5090 D is a dual-slot expansion card using a PCIe 5.0 x16 interface.

The compute resources differ by an order of magnitude. The Arc B390 has 1536 shading units, 48 texture mapping units, 24 raster output units, and 12 ray tracing cores. The RTX 5090 D has 21760 shading units, 680 TMUs, 176 ROPs, 170 ray tracing cores, and 680 tensor cores. The Arc B390 has no tensor core field listed, while the RTX 5090 D's tensor cores are a major part of its design for AI-oriented workloads.

The memory subsystems are fundamentally different in kind. The Arc B390 uses system shared memory, with the size, type, bus width, and bandwidth all listed as system dependent. The RTX 5090 D has 32 GB of dedicated GDDR7 memory on a 512-bit bus, delivering 1.79 TB/s of bandwidth. The Arc B390's memory clock is also system shared, while the RTX 5090 D's memory runs at 1750 MHz with 28 Gbps effective data rate.

Clock behavior differs as well. The Arc B390 has a base clock of 300 MHz and a boost of 2500 MHz. The RTX 5090 D has a base of 2017 MHz and a boost of 2407 MHz. The Arc B390's boost clock is higher than the RTX 5090 D's, but with far fewer execution units, the higher clock does not translate into competitive throughput.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs differ: the Arc B390 is portable device dependent, while the RTX 5090 D provides 1x HDMI 2.1b and 3x DisplayPort 2.1b.

Specification Differences

The power envelope separates these two parts completely. The Arc B390 has a TDP of 80 W with no power connectors, since it is an integrated part. The RTX 5090 D has a TDP of 575 W, requires a single 16-pin power connector, and carries a suggested power supply rating of 950 W.

The compute throughput figures reflect the scale difference. The Arc B390 delivers 7.680 TFLOPS of FP32 performance and 15.36 TFLOPS of FP16 (at a 2:1 ratio). The RTX 5090 D delivers 104.8 TFLOPS of FP32 and 104.8 TFLOPS of FP16 at a 1:1 ratio. Pixel rates are 60.00 GPixel/s for the Arc B390 versus 423.6 GPixel/s for the RTX 5090 D. Texture rates are 120.0 GTexel/s versus 1,636.8 GTexel/s.

Physical dimensions are only listed for the RTX 5090 D: 304 mm in length, 137 mm in height, and 48 mm in width. The Arc B390 has no listed dimensions, consistent with its integrated status.

The RTX 5090 D has a launch MSRP of 2,299 USD. The Arc B390 has no launch MSRP listed.

Release timing differs by roughly a year. The RTX 5090 D was released on 2025-01-29, while the Arc B390 has a release date of 2026-01-26. The RTX 5090 D's predecessor is listed as GeForce 40 and its successor as GeForce 60. The Arc B390 has no predecessor or successor listed.

The Verdict

The database shows two GPUs that are not competitors in any practical sense. The Intel Arc B390 is an integrated graphics solution in the 9th percentile of all GPUs, with performance within 2.7% of the GeForce GT 710 and the GeForce GT 520MX. The NVIDIA GeForce RTX 5090 D is a discrete flagship in the 92nd percentile, trading benchmarks with the AMD Radeon RX 6850M XT and the NVIDIA Tesla P100 variants. The shared 3DMark Steel Nomad result shows the RTX 5090 D at 14326 points versus 1482 for the Arc B390, a gap of 89.7%. The RTX 5090 D's average benchmark score of 77712 across ten recorded tests is over 52 times the Arc B390's single recorded score of 1482.

The Arc B390's position as an IGP with system shared memory and an 80 W TDP places it in the segment of the database occupied by low-end legacy parts. The RTX 5090 D's 575 W TDP, 32 GB of GDDR7 memory, and 104.8 TFLOPS of FP32 performance place it among the fastest GPUs recorded. The data does not support any scenario where the Arc B390 outperforms the RTX 5090 D. The RTX 5090 D wins the only shared benchmark, holds a 92nd percentile ranking, and maintains a large margin over every recorded comparison.

Where Each One Wins

The RTX 5090 D wins in every measurable category in the database. It wins the shared 3DMark Steel Nomad DX12 test by 89.7%. It wins on raw compute, with 104.8 TFLOPS of FP32 versus 7.680 TFLOPS. It wins on memory, with 32 GB of dedicated GDDR7 at 1.79 TB/s versus system shared memory. It wins on pixel throughput at 423.6 GPixel/s versus 60.00 GPixel/s, and on texture throughput at 1,636.8 GTexel/s versus 120.0 GTexel/s. It has more shading units, more ROPs, more ray tracing cores, and the only tensor cores in the comparison. It also has a broader benchmark record, with ten recorded tests spanning DirectX 9 through 12, OpenCL, Vulkan, and compute workloads.

The Arc B390 has no benchmark wins in the database. Its advantages are limited to attributes that do not affect performance scores. It has a higher boost clock at 2500 MHz versus 2407 MHz. It uses a smaller 3 nm process node versus 5 nm. It draws 80 W versus 575 W. It requires no power connectors and occupies no expansion slot. It is portable device dependent for display output, and its release date is later. These are characteristics of an integrated part designed for low power and system integration, not for competitive frame rates.

For users selecting based on the recorded data, the RTX 5090 D is the only choice for any performance-oriented workload. The Arc B390's role is limited to systems where an integrated GPU is the only option, and even then, the data shows it performs at the level of legacy entry-level discrete GPUs from NVIDIA's catalog. The database records no scenario where the Arc B390 delivers a victory.