Intel Arc B390 vs Intel Arc Graphics 24EU Comparison

Head-to-Head Benchmarks

The recorded database contains a single direct comparison between the Intel Arc B390 and the Intel Arc Graphics 24EU: the 3DMark Steel Nomad DX12 test. In this benchmark, the Intel Arc B390 scores 1482 points, while the Intel Arc Graphics 24EU scores 733 points. This produces a decisive victory for the Intel Arc B390, with a delta of 102.2% higher performance. The margin is substantial; the B390 more than doubles the output of the 24EU part in this workload.

Looking at the broader database percentile rankings, the Intel Arc B390 sits at the 9th percentile of all GPUs, while the Intel Arc Graphics 24EU sits at the 3rd percentile. This indicates that the B390 is positioned significantly higher in the overall performance distribution, even though both are integrated graphics solutions. The B390’s nearest rivals in the database include the NVIDIA GeForce GT 520MX (average score 1463, delta of 1.3% against the B390), the NVIDIA GeForce 800M (1460, delta 1.5%), the NVIDIA GeForce GT 625 OEM (1446, delta 2.5%), and the NVIDIA GeForce GT 710 (1443, delta 2.7%). The B390 leads each of these by a small margin, confirming that its 1482 score places it just above a cluster of older discrete mobile and entry-level desktop GPUs.

The Intel Arc Graphics 24EU’s nearest rivals paint a different picture. Its score of 733 ties exactly with the Intel Arc Graphics 32EU and the Intel Arc Graphics 64EU, both showing a delta of 0%. This suggests that in this particular test, the 24EU execution unit configuration does not separate itself from those higher-EU variants, at least within the recorded data. The AMD Radeon HD 6470M trails slightly with an average score of 723 (delta 1.4% in favor of the 24EU), while the NVIDIA GeForce GT 415M leads the 24EU with a score of 751 (delta -2.4%, meaning the 24EU is 2.4% slower). The head-to-head comparison shows the B390 with exactly 1 win and the 24EU with 0 wins.

Architecture Differences

The two processors come from different architectural generations and chip designs. The Intel Arc B390 uses the Xe3-LPG architecture built on a Panther Lake chip, fabricated on a 3 nm process at Intel’s foundry. Its generation is listed as Arc Graphics-M (Panther Lake). The Intel Arc Graphics 24EU uses the Xe-LPG architecture on an Arrow Lake-S chip, also on a 3 nm process, but fabricated by TSMC. Its generation is Arc Graphics (Arrow Lake). The process node matches at 3 nm, but the foundry differs, with Intel handling the B390 and TSMC handling the 24EU.

The B390 carries significantly more execution resources. It has 1536 shading units, 48 texture mapping units, and 24 ROPs. The 24EU has only 192 shading units, 12 TMUs, and 6 ROPs. This represents an 8x difference in shading units and a 4x difference in both TMUs and ROPs. The B390 also includes 12 ray tracing cores, while the 24EU has no ray tracing cores listed in the database. Neither part lists tensor cores.

Clock speeds differ as well. The B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The 24EU has the same 300 MHz base clock but a lower boost of 2000 MHz. The higher boost clock on the B390, combined with the much larger execution resource pool, produces the large performance gap observed in the benchmark.

The B390’s recorded pixel rate is 60.00 GPixel/s, and its texture rate is 120.0 GTexel/s. The 24EU records 12.00 GPixel/s and 24.00 GTexel/s. Floating point throughput shows a similar ratio: the B390 delivers 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 (2:1), while the 24EU delivers 768.0 GFLOPS FP32 and 1.536 TFLOPS FP16 (2:1). The B390 is exactly 10x higher in FP32 and FP16 throughput.

Power draw also differs. The B390 has a TDP of 80 W, while the 24EU has a TDP of 65 W. Both are integrated graphics parts with no dedicated slot width beyond IGP designation, and both use system shared memory with system dependent bandwidth. The B390 lists no power connectors, while the 24EU does not list a power connector field at all. The B390’s bus interface is IGP, while the 24EU’s is Ring Bus.

The 24EU provides transistor and die information: 17,800 million transistors, a 243 mm² die size, and a transistor density of 73.3M per mm². The B390 does not list transistor count or die size in the database. The 24EU’s predecessor is listed as HD Graphics, while the B390 has no predecessor listed. The 24EU was released on 2024-10-23, and the B390 on 2026-01-26.

The Verdict

The benchmark data clearly favors the Intel Arc B390 in raw performance. Its 1482 score against 733 represents a 102.2% advantage in the 3DMark Steel Nomad DX12 test. The B390 also occupies a higher percentile rank (9th versus 3rd), meaning it outperforms a larger fraction of the database’s GPU population. For tasks that rely on shader throughput, texture processing, or ray tracing, the B390’s architecture provides 8x the shading units, 4x the TMUs, 4x the ROPs, and 12 dedicated ray tracing cores where the 24EU has none.

The Intel Arc Graphics 24EU, by contrast, is positioned much lower in the performance spectrum. Its score ties with the 32EU and 64EU variants in the same benchmark, which suggests that in this specific workload, the execution unit count does not translate into a measurable advantage among those Intel integrated parts. The 24EU trails the NVIDIA GeForce GT 415M by 2.4% and only leads the AMD Radeon HD 6470M by 1.4%. This places it in a very modest performance tier.

The B390’s nearest rivals are all NVIDIA parts with scores between 1443 and 1463. The B390 leads them by 1.3% to 2.7%, which is a narrow margin. This indicates that while the B390 decisively beats the 24EU, it is only slightly ahead of those older discrete GPUs in the database. The choice between the two Intel parts depends entirely on the workload. The B390 delivers roughly double the performance and includes ray tracing support, but it also draws 15 W more power. The 24EU uses less power and is a much smaller implementation, but it lacks ray tracing cores and delivers a fraction of the throughput.

FAQ

Q: What is the performance difference between the Intel Arc B390 and the Intel Arc Graphics 24EU in the recorded benchmark?

A: In the 3DMark Steel Nomad DX12 test, the Intel Arc B390 scores 1482 points, while the Intel Arc Graphics 24EU scores 733 points. The B390 leads by 102.2%.

Q: How does each part rank against all GPUs in the database?

A: The Intel Arc B390 sits at the 9th percentile, while the Intel Arc Graphics 24EU sits at the 3rd percentile.

Q: Which part has ray tracing cores?

A: The Intel Arc B390 has 12 ray tracing cores. The Intel Arc Graphics 24EU has no ray tracing cores listed.

Q: What are the shading unit counts for each?

A: The Intel Arc B390 has 1536 shading units, while the Intel Arc Graphics 24EU has 192 shading units.

Q: What are the TDP values?

A: The Intel Arc B390 has a TDP of 80 W, and the Intel Arc Graphics 24EU has a TDP of 65 W.

Q: Which part has a higher boost clock?

A: The Intel Arc B390 has a boost clock of 2500 MHz, while the Intel Arc Graphics 24EU has a boost clock of 2000 MHz. Both have a base clock of 300 MHz.

Where Each One Wins

The Intel Arc B390 wins in every recorded benchmark category. Its FP32 throughput of 7.680 TFLOPS is exactly 10x that of the 24EU’s 768.0 GFLOPS. Its pixel rate of 60.00 GPixel/s is 5x the 24EU’s 12.00 GPixel/s, and its texture rate of 120.0 GTexel/s is 5x the 24EU’s 24.00 GTexel/s. The B390 also has 12 ray tracing cores, which the 24EU lacks entirely, making it the only option for ray-traced workloads in this comparison. In the 3DMark Steel Nomad DX12 benchmark, the B390’s 1482 score is more than double the 24EU’s 733.

The Intel Arc Graphics 24EU wins in power efficiency relative to its own performance envelope. Its TDP of 65 W is lower than the B390’s 80 W, and it still manages to match the B390’s base clock of 300 MHz. The 24EU also benefits from a known transistor count of 17,800 million and a die size of 243 mm², which gives it a defined physical footprint, while the B390’s transistor and die data are not recorded. In the database’s nearest rival comparison, the 24EU ties exactly with the Intel Arc Graphics 32EU and 64EU at a score of 733, which indicates that within this benchmark, the 24EU performs identically to those larger EU configurations. This is not a performance win, but it shows that the 24EU achieves parity with those variants in this specific test.

Specification Differences

The two parts differ in nearly every specification field except for base clock (300 MHz for both), memory configuration (both system shared with system dependent bandwidth), API support (both DirectX 12 Ultimate 12_2, OpenGL 4.6, Vulkan 1.4), slot width (both IGP), and production status (both Active). The process node is the same at 3 nm, but the foundry differs: Intel for the B390 and TSMC for the 24EU.

The Intel Arc B390 uses the Xe3-LPG architecture on a Panther Lake chip, with a generation of Arc Graphics-M (Panther Lake). The Intel Arc Graphics 24EU uses Xe-LPG on an Arrow Lake-S chip, with a generation of Arc Graphics (Arrow Lake). The B390’s boost clock is 2500 MHz versus 2000 MHz for the 24EU. Shading units are 1536 versus 192, TMUs are 48 versus 12, and ROPs are 24 versus 6. The B390 has 12 ray tracing cores; the 24EU has none. Pixel rate is 60.00 GPixel/s versus 12.00 GPixel/s, texture rate is 120.0 GTexel/s versus 24.00 GTexel/s, and FP32 is 7.680 TFLOPS versus 768.0 GFLOPS. FP16 is 15.36 TFLOPS versus 1.536 TFLOPS. TDP is 80 W versus 65 W. The B390’s bus interface is IGP, while the 24EU’s is Ring Bus. Display outputs are portable device dependent for the B390 and motherboard dependent for the 24EU. The B390 lists no power connectors, while the 24EU does not record a power connector field. The 24EU has a predecessor (HD Graphics) and a release date of 2024-10-23, while the B390 has no predecessor and a release date of 2026-01-26. The 24EU also records transistors (17,800 million), die size (243 mm²), and transistor density (73.3M per mm²); the B390 does not provide those figures.