Intel Arc Graphics 24EU vs Intel Arc Pro B390 Comparison

Head-to-Head Benchmarks

The recorded data contains a single benchmark result for the Intel Arc Graphics 24EU: a score of 733 in 3DMark Steel Nomad DX12. The Intel Arc Pro B390 has no recorded benchmark scores in the database at this time, so a direct head-to-head comparison is not possible from measured results. The Arc Graphics 24EU’s score places it in the 3rd percentile among all GPUs, meaning it outperforms only a small fraction of the database’s tracked devices.

For context, the Arc Graphics 24EU’s 733 score is exactly matched by two other Intel integrated parts: the Arc Graphics 32EU and the Arc Graphics 64EU, both also scoring 733, a delta of 0%. This indicates that, in this particular DX12 workload, the 24EU configuration performs identically to those with more execution resources. The nearest non-Intel rival is the AMD Radeon HD 6470M, which scores 723, placing the Arc Graphics 24EU 1.4% ahead. On the other side, the NVIDIA GeForce GT 415M scores 751, meaning the Arc Graphics 24EU trails by 2.4%.

The Arc Pro B390, with no benchmark entries, cannot be placed relative to these numbers. Its percentile ranking of 50 is derived from its specification profile, not from measured performance, and its average benchmark score is listed as 0 due to the absence of recorded tests. Therefore, any performance comparison between the two items must rely on architectural and specification differences rather than direct benchmark deltas.

Where Each One Wins

Based on the available data, the Intel Arc Graphics 24EU wins in the only measured benchmark category: 3DMark Steel Nomad DX12, where it posts a score of 733. This is a functional result, indicating the part can complete the workload. The Arc Pro B390 currently has no wins in any recorded test, as no benchmark results exist for it.

The Arc Pro B390’s advantage lies in its theoretical capability profile. Its specification sheet shows a substantially larger execution resource pool: 1536 shading units, 48 texture mapping units, and 24 raster output units, compared to the Arc Graphics 24EU’s 192 shading units, 12 TMUs, and 6 ROPs. The B390 also includes 12 ray tracing cores, which the 24EU lacks entirely. The B390’s peak pixel rate is 60.00 GPixel/s versus 12.00 GPixel/s for the 24EU, and its texture rate is 120.0 GTexel/s versus 24.00 GTexel/s. Floating-point throughput is 7.680 TFLOPS for the B390 against 768.0 GFLOPS for the 24EU, a tenfold difference.

In real-world terms, the 24EU’s measured score suggests it is suitable for lightweight or legacy 3D workloads, given its low percentile standing. The B390, by contrast, is positioned for more demanding tasks based on its resource counts, but without benchmark data, its actual performance cannot be confirmed. The database shows no wins for the B390 in any test, but its specification-driven percentile of 50 indicates a mid-pack theoretical standing, far above the 24EU’s 3rd percentile.

FAQ

Q: What is the only recorded benchmark score for the Intel Arc Graphics 24EU?

A: The Intel Arc Graphics 24EU scores 733 in 3DMark Steel Nomad DX12. This is its only listed benchmark result, and it places the GPU in the 3rd percentile among all tracked GPUs.

Q: Does the Intel Arc Pro B390 have any benchmark scores in the database?

A: No. The Arc Pro B390 has an empty benchmarks list, an average benchmark score of 0, and no nearest rivals. Its 50th percentile ranking is based on its specification profile, not on measured performance.

Q: How does the Arc Graphics 24EU compare to its nearest Intel rivals?

A: The Arc Graphics 24EU scores 733, identical to the Arc Graphics 32EU and Arc Graphics 64EU, both of which also score 733. The delta between the 24EU and each of these parts is 0%.

Q: What is the closest non-Intel competitor to the Arc Graphics 24EU?

A: The AMD Radeon HD 6470M scores 723, which is 1.4% lower than the Arc Graphics 24EU’s 733. The NVIDIA GeForce GT 415M scores 751, which is 2.4% higher.

Q: What are the key performance-oriented differences in shading and texture resources?

A: The Arc Pro B390 has 1536 shading units, 48 TMUs, and 24 ROPs. The Arc Graphics 24EU has 192 shading units, 12 TMUs, and 6 ROPs. The B390 also has 12 ray tracing cores, while the 24EU has none.

Q: What are the peak throughput numbers for each GPU?

A: The Arc Pro B390 delivers 60.00 GPixel/s pixel rate, 120.0 GTexel/s texture rate, and 7.680 TFLOPS FP32. The Arc Graphics 24EU delivers 12.00 GPixel/s, 24.00 GTexel/s, and 768.0 GFLOPS FP32.

Specification Differences

The two GPUs differ across nearly every specification field. The most significant differences are in execution resources and clock speeds.

Execution Resources: The Arc Pro B390 contains 1536 shading units, 48 texture mapping units, and 24 raster output units. The Arc Graphics 24EU contains 192 shading units, 12 texture mapping units, and 6 raster output units. The B390 also includes 12 ray tracing cores; the 24EU has none.

Clock Speeds: The Arc Graphics 24EU has a base clock of 300 MHz and a boost clock of 2000 MHz. The Arc Pro B390 also has a 300 MHz base clock but boosts to 2500 MHz, a 500 MHz higher peak.

Power and Thermal: The Arc Graphics 24EU has a TDP of 65 W. The Arc Pro B390 has a TDP of 80 W. The B390 lists no power connectors, while the 24EU lists none either, though the B390’s slot width is listed as IGP for both.

Process and Foundry: Both use a 3 nm process node, but the Arc Graphics 24EU is fabricated by TSMC, while the Arc Pro B390 is fabricated by Intel. The 24EU has a recorded transistor count of 17,800 million and a die size of 243 mm², with a transistor density of 73.3M / mm². The B390 lists transistors, die size, and density as unknown or null.

Memory and Bus: Both use System Shared memory, with System Shared type, bus width, and System Dependent bandwidth. The bus interface differs: the 24EU uses a Ring Bus, while the B390 uses IGP.

Display Outputs: The Arc Graphics 24EU’s display outputs are Motherboard Dependent. The Arc Pro B390’s display outputs are Portable Device Dependent.

Release Dates: The Arc Graphics 24EU was released on 2024-10-23. The Arc Pro B390 was released on 2026-01-26, over a year later.

Predecessors: The 24EU lists HD Graphics as its predecessor. The B390 lists HD Graphics-WM.

Production Status: Both are listed as Active.

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

Benchmark Data: The 24EU has one benchmark result (733 in 3DMark Steel Nomad DX12) and a 3rd percentile ranking. The B390 has no benchmark results and a 50th percentile ranking.

Architecture Differences

The architecture gap between the two parts is substantial, reflecting different generations and design goals.

Architecture Generation: The Arc Graphics 24EU is based on the Xe-LPG architecture, part of the Arc Graphics (Arrow Lake) generation. The Arc Pro B390 uses the Xe3-LPG architecture, part of the Arc Graphics-WM (Panther Lake) generation. This is a jump from the first Xe-LPG iteration to the third-generation Xe3-LPG design.

Chip and Die: The 24EU uses the Arrow Lake-S chip, fabricated by TSMC on a 3 nm process. The B390 uses the Panther Lake chip, fabricated by Intel on the same 3 nm node. The 24EU has a known transistor count of 17,800 million and a die size of 243 mm²; the B390’s transistor count and die size are unknown in the database.

Execution Pipeline: The B390’s architecture provides eight times the shading units (1536 vs 192), four times the TMUs (48 vs 12), and four times the ROPs (24 vs 6) compared to the 24EU. The B390 also adds 12 ray tracing cores, a feature entirely absent from the 24EU. This indicates the B390 is designed for hardware-accelerated ray tracing, while the 24EU lacks any dedicated RT hardware.

Throughput Implications: The B390’s pixel rate of 60.00 GPixel/s is five times the 24EU’s 12.00 GPixel/s. Its texture rate of 120.0 GTexel/s is five times the 24EU’s 24.00 GTexel/s. FP32 compute is 7.680 TFLOPS for the B390 versus 768.0 GFLOPS for the 24EU, a tenfold increase. FP16 performance follows the same ratio: 15.36 TFLOPS for the B390 versus 1.536 TFLOPS for the 24EU, both with a 2:1 ratio.

Clock Behavior: Both parts have a 300 MHz base clock, but the B390’s boost clock of 2500 MHz exceeds the 24EU’s 2000 MHz by 25%. This higher boost ceiling, combined with the larger execution resource pool, gives the B390 a significantly higher peak throughput ceiling.

Power and Packaging: The B390’s 80 W TDP is 15 W higher than the 24EU’s 65 W, consistent with its larger resource count. Both are integrated graphics processors (IGP) with no dedicated power connectors, and both rely on System Shared memory. The B390’s display outputs are tied to a portable device, while the 24EU’s are tied to the motherboard, reflecting different target platforms.

API Support: Both parts expose the same API set: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The architectural differences do not affect the software-facing feature level in the recorded data.

Release Timing: The B390 launched on 2026-01-26, more than a year after the 24EU’s 2024-10-23 release. This timeline aligns with the newer Xe3-LPG architecture. The 24EU’s predecessor is HD Graphics, while the B390’s predecessor is HD Graphics-WM, indicating a distinct product lineage.

The benchmark data shows the 24EU as a low-tier integrated part, sitting at the 3rd percentile. The B390’s 50th percentile, despite no measured scores, reflects its much larger specification footprint. Without benchmark results for the B390, the architectural advantages cannot be translated into measured performance deltas, but the resource counts and clock speeds provide a clear theoretical hierarchy.