Intel Arc B390 vs NVIDIA N1X 48SM Comparison

Head-to-Head Benchmarks

The recorded data presents an unusual comparison. The Intel Arc B390 has a single benchmark entry, the 3DMark Steel Nomad DX12 test, where it scores 1482. The NVIDIA N1X 48SM has no benchmark entries in the database, which means a direct head-to-head comparison using identical workloads is impossible. The Intel part sits at the 9th percentile among all GPUs, while the NVIDIA part is placed at the 50th percentile, a positioning that comes from factors other than measured performance.

The Intel Arc B390's nearest rivals in the database are all entry-level NVIDIA parts. It sits 1.3% above the GeForce GT 520MX, which averages 1463. It is 1.5% ahead of the GeForce 800M at 1460. The gap widens slightly against the GT 625 OEM, which averages 1446, a 2.5% delta. The GT 710 averages 1443, putting it 2.7% behind the Arc B390. These are narrow margins, indicating that the Arc B390 lands at the very bottom of the performance spectrum according to the database's scoring system.

The NVIDIA N1X 48SM presents a different situation. With no benchmark scores recorded, its 50th percentile placement suggests mid-pack positioning, but the data does not reveal its scoring methodology. The Intel part's 9th percentile placement, by contrast, is anchored to a concrete score. The absence of a score for the NVIDIA part means any direct numerical comparison between the two would be speculative, and the data does not support such a claim.

The Intel part's 1482 score against rivals within 1.3% to 2.7% shows clustering at the low end. The performance deltas are small, and the recorded data indicates that the Arc B390 is not far from these older, low-end NVIDIA parts. The N1X 48SM, without a score, cannot be placed on the same scale. The percentile gap, from 9 to 50, hints at a substantial difference, but the database's own numbers do not provide the intermediary scores needed to quantify it.

The Verdict

The data supports only one clear conclusion: the Intel Arc B390 is a low-end part, scoring 1482 in 3DMark Steel Nomad DX12 and placing at the 9th percentile. Its nearest competitors, the GT 520MX, GeForce 800M, GT 625 OEM, and GT 710, all sit within 2.7% of its score. This is a tightly grouped cluster of low-performance hardware.

The NVIDIA N1X 48SM cannot be given a performance verdict from benchmark data because no scores exist. Its 50th percentile placement is the only quantitative signal, and that places it far above the Arc B390 in the database's ranking. However, the percentile field is not a direct benchmark score, so the magnitude of the difference remains unknown.

The Intel part's 7.680 TFLOPS FP32 throughput and 120.0 GTexel/s texture rate are the concrete performance ceilings recorded. The NVIDIA part's 28.83 TFLOPS FP32 and 900.9 GTexel/s are substantially higher on paper. These are specification-derived figures, not benchmark results, but they align with the percentile gap. The Arc B390 delivers roughly a quarter of the FP32 throughput of the N1X 48SM based on the recorded numbers.

For users looking at the database's measured data alone, the Arc B390 is a known quantity with a clear low-end position. The N1X 48SM is an unknown quantity in terms of measured performance, but its specifications and percentile placement indicate it operates in a different class. The data does not justify choosing the Intel part for any performance-driven scenario, as every indicator points to the NVIDIA part being stronger.

Where Each One Wins

The Intel Arc B390 wins in the category of having recorded benchmark data. Its 3DMark Steel Nomad DX12 score of 1482 is the only direct performance measurement available in this comparison. It also supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, whereas the NVIDIA N1X 48SM lists N/A for DirectX, OpenGL, and Vulkan. For software compatibility in those specific API areas, the Intel part has documented support.

The NVIDIA N1X 48SM wins on raw specification counts. It has 6144 shading units against 1536, 384 TMUs against 48, 48 ROPs against 24, 48 RT cores against 12, and 192 tensor cores where the Intel part has none recorded. Its FP32 throughput of 28.83 TFLOPS is 3.75 times the Arc B390's 7.680 TFLOPS. Its FP16 throughput is 28.83 TFLOPS at a 1:1 ratio, while the Intel part reaches 15.36 TFLOPS at a 2:1 ratio, meaning the NVIDIA part does not rely on packed math to reach its number.

Memory is another clear win for the NVIDIA part. It has 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s of bandwidth. The Intel part uses system-shared memory with bandwidth described as system dependent. The NVIDIA part's memory clock is recorded as 1067 MHz with 8.5 Gbps effective, while the Intel part's memory clock is listed as system shared.

The pixel rate favors the NVIDIA part at 112.6 GPixel/s versus 60.00 GPixel/s for Intel. The texture rate favors NVIDIA at 900.9 GTexel/s versus 120.0 GTexel/s. These are fill-rate metrics derived from clock speeds and unit counts, and they all point in the same direction.

The NVIDIA part also uses PCIe 5.0 x16 as its bus interface, while the Intel part is listed as IGP. Both are integrated-style parts with no power connectors, but the NVIDIA part has a dedicated 382 mm² die from TSMC on a 5 nm process. The Intel part uses a 3 nm process from Intel's own foundry.

FAQ

Q: Does the NVIDIA N1X 48SM outperform the Intel Arc B390 in benchmarks?

A: The database has no benchmark scores for the NVIDIA N1X 48SM. Its average benchmark score is recorded as 0, and it has no nearest rivals listed. The Intel Arc B390 has a 3DMark Steel Nomad DX12 score of 1482. Direct benchmark comparison is not possible from the recorded data.

Q: What is the Intel Arc B390's position relative to other GPUs?

A: The Arc B390 sits at the 9th percentile among all GPUs. Its nearest rivals are the GeForce GT 520MX at 1463 (1.3% behind), the GeForce 800M at 1460 (1.5% behind), the GT 625 OEM at 1446 (2.5% behind), and the GT 710 at 1443 (2.7% behind).

Q: How much faster is the NVIDIA N1X 48SM in raw compute?

A: The NVIDIA part records 28.83 TFLOPS FP32, which is 3.75 times the Intel part's 7.680 TFLOPS. In FP16, the NVIDIA part also records 28.83 TFLOPS at a 1:1 ratio, while the Intel part reaches 15.36 TFLOPS at a 2:1 ratio.

Q: What memory configurations do the two GPUs use?

A: The NVIDIA N1X 48SM uses 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth. The Intel Arc B390 uses system-shared memory with a system-shared bus width and system-dependent bandwidth.

Q: Which GPU has more rendering units?

A: The NVIDIA N1X 48SM has 6144 shading units, 384 TMUs, 48 ROPs, 48 RT cores, and 192 tensor cores. The Intel Arc B390 has 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores, with no tensor cores recorded.

Q: Do both GPUs support the same graphics APIs?

A: No. The Intel Arc B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA N1X 48SM lists N/A for DirectX, OpenGL, and Vulkan in the database.

Architecture Differences

The two GPUs come from different architectural lineages. The Intel Arc B390 uses the Xe3-LPG architecture on the Panther Lake chip, part of the Arc Graphics-M generation. The NVIDIA N1X 48SM uses the Blackwell 2.0 architecture on the GB20B chip, part of the Blackwell IGP generation.

Process nodes differ. Intel's part is built on a 3 nm process at Intel's own foundry. NVIDIA's part uses TSMC's 5 nm process. The NVIDIA die is recorded at 382 mm², while the Intel die size is unknown. Transistor counts are unknown for both.

The NVIDIA part has a significant unit advantage across the board. Its 6144 shading units are four times the Intel part's 1536. Its 384 TMUs are eight times Intel's 48. Its 48 ROPs double Intel's 24. Its 48 RT cores are four times Intel's 12. The NVIDIA part has 192 tensor cores; the Intel part has none recorded.

The NVIDIA part's FP16 throughput matches its FP32 at 28.83 TFLOPS, indicating a 1:1 ratio. The Intel part reaches 15.36 TFLOPS FP16 at a 2:1 ratio, meaning it uses packed execution to double its FP32 rate. This is a fundamental architectural difference in how each part handles half-precision math.

Clock behavior differs as well. The Intel part has a 300 MHz base clock and a 2500 MHz boost. The NVIDIA part has a 741 MHz base and a 2346 MHz boost. Despite the lower boost clock, the NVIDIA part achieves higher fill rates, which indicates the unit count, not the clock, drives its throughput. The NVIDIA pixel rate of 112.6 GPixel/s versus 60.00 GPixel/s for Intel, and texture rate of 900.9 GTexel/s versus 120.0 GTexel/s, both confirm this.

The NVIDIA part's memory architecture is discrete in nature: 128 GB of LPDDR5X on a 256-bit bus. The Intel part relies on system-shared memory. The NVIDIA part connects via PCIe 5.0 x16, while the Intel part is an IGP. Both are listed as IGP slot width with no power connectors.

The Intel part has a recorded TDP of 80 W. The NVIDIA part's TDP is unknown. The Intel part's display outputs are described as portable device dependent, while the NVIDIA part has 1x HDMI.

Specification Differences

The recorded specifications show clear separation between the two parts, with the NVIDIA N1X 48SM holding advantages in nearly every measurable field.

The Intel Arc B390 uses a 3 nm process at Intel's foundry; the NVIDIA N1X 48SM uses a 5 nm process at TSMC. The NVIDIA die is 382 mm²; the Intel die size is unknown.

Base clocks: Intel at 300 MHz, NVIDIA at 741 MHz. Boost clocks: Intel at 2500 MHz, NVIDIA at 2346 MHz. The NVIDIA part has a recorded memory clock of 1067 MHz with 8.5 Gbps effective; the Intel part's memory clock is system shared.

Memory: NVIDIA has 128 GB of LPDDR5X on a 256-bit bus with 273.2 GB/s bandwidth. Intel has system-shared memory with system-dependent bandwidth.

Shading units: 1536 for Intel, 6144 for NVIDIA. TMUs: 48 versus 384. ROPs: 24 versus 48. RT cores: 12 versus 48. Tensor cores: none recorded for Intel, 192 for NVIDIA.

Pixel rate: 60.00 GPixel/s for Intel, 112.6 GPixel/s for NVIDIA. Texture rate: 120.0 GTexel/s for Intel, 900.9 GTexel/s for NVIDIA. FP32: 7.680 TFLOPS versus 28.83 TFLOPS. FP16: 15.36 TFLOPS at 2:1 for Intel, 28.83 TFLOPS at 1:1 for NVIDIA.

TDP: 80 W for Intel, unknown for NVIDIA. Bus interface: IGP for Intel, PCIe 5.0 x16 for NVIDIA. Display outputs: portable device dependent for Intel, 1x HDMI for NVIDIA.

APIs: Intel supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. NVIDIA lists N/A for all three.

Release dates differ: the Intel part is dated 2026-01-26, the NVIDIA part 2026-05-31. Both are marked as active production. Neither has a recorded launch MSRP.

The percentile placement, 9 for Intel and 50 for NVIDIA, summarizes the gap. The Intel part's benchmark score of 1482 in 3DMark Steel Nomad DX12 anchors its low percentile. The NVIDIA part has no score, but its 50th percentile placement and specification advantages indicate a fundamentally different performance class. The data consistently shows the NVIDIA N1X 48SM as the stronger part on paper, with the Intel Arc B390 occupying the entry-level segment.