Intel Arc B390 vs Intel Arc Graphics 64EU Comparison

Head-to-Head Benchmarks

The recorded data includes a single direct comparison: the 3DMark Steel Nomad DX12 test. The Intel Arc B390 scores 1482 points, while the Intel Arc Graphics 64EU scores 733 points. This represents a 102.2% advantage for the B390, meaning it more than doubles the performance of the 64EU in this workload. The B390 takes the only win in this head-to-head matchup, giving it a 1 to 0 record.

To contextualize the B390 score, the database shows it sits at the 9th percentile among all GPUs. Its nearest rivals are older discrete mobile parts: the NVIDIA GeForce GT 520MX (1463 points, 1.3% slower), the NVIDIA GeForce 800M (1460 points, 1.5% slower), the NVIDIA GeForce GT 625 OEM (1446 points, 2.5% slower), and the NVIDIA GeForce GT 710 (1443 points, 2.7% slower). The B390 leads each of these by a small margin, indicating that while it outpaces the 64EU decisively, it operates in a similar performance tier to those legacy discrete GPUs.

The Intel Arc Graphics 64EU, in contrast, holds only the 3rd percentile among all GPUs. Its closest rivals include the Intel Arc Graphics 32EU (733 points, 0% difference) and the Intel Arc Graphics 24EU (733 points, 0% difference), which shows that the 64EU delivers essentially identical performance to its lower-EU siblings in this test. The AMD Radeon HD 6470M trails slightly at 723 points (1.4% behind), while the NVIDIA GeForce GT 415M leads at 751 points (2.4% ahead). The 64EU's position among these older low-end parts confirms that its performance ceiling is modest, and the gap to the B390 is substantial.

Benchmark results indicate that the B390's advantage stems from a combination of a higher boost clock and a far larger execution resource pool. The B390 reaches 2500 MHz boost versus 1900 MHz for the 64EU, a 31.6% clock advantage. More importantly, the B390 carries 1536 shading units against 512 for the 64EU, a threefold difference. These resources translate directly into the 102.2% score delta, showing that the B390's architectural scaling delivers near-linear gains in this DX12 workload.

FAQ

Q: How large is the performance gap between the Intel Arc B390 and the Intel Arc Graphics 64EU?

A: In the 3DMark Steel Nomad DX12 test, the B390 scores 1482 versus 733 for the 64EU, a 102.2% difference. The B390 produces more than double the benchmark score.

Q: Which GPU has the higher boost clock?

A: The Intel Arc B390 boosts to 2500 MHz, while the Intel Arc Graphics 64EU boosts to 1900 MHz. Both have a 300 MHz base clock.

Q: How do the shading unit counts compare?

A: The Intel Arc B390 contains 1536 shading units, while the Intel Arc Graphics 64EU contains 512. The B390 also has 48 texture mapping units versus 32, and 24 render output units versus 16.

Q: What is the performance percentile ranking for each GPU?

A: The Intel Arc B390 sits at the 9th percentile among all GPUs, while the Intel Arc Graphics 64EU sits at the 3rd percentile. This places the B390 in a higher overall performance tier.

Q: Do both GPUs support the same graphics APIs?

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

Q: Which GPU has a higher pixel rate?

A: The Intel Arc B390 delivers 60.00 GPixel/s, while the Intel Arc Graphics 64EU delivers 30.40 GPixel/s. The B390's rate is roughly double.

Architecture Differences

The Intel Arc B390 uses the Xe3-LPG architecture built on the Panther Lake chip, manufactured by Intel on a 3 nm process. It belongs to the Arc Graphics-M (Panther Lake) generation. The Intel Arc Graphics 64EU uses the Xe-LPG architecture on the Arrow Lake-S chip, manufactured by TSMC on a 3 nm process, and belongs to the Arc Graphics (Arrow Lake) generation. Both share the 3 nm node, but the foundries differ: Intel for the B390, TSMC for the 64EU.

The B390 integrates 12 ray tracing cores, while the 64EU has no listed ray tracing cores. This gives the B390 a dedicated hardware path for ray-traced workloads that the 64EU lacks. Neither part lists tensor cores in the database.

The B390's bus interface is listed as IGP, and its power connector requirement is none. The 64EU uses a Ring Bus interface, and its power connector field is not specified. Display outputs for the B390 are portable device dependent, while the 64EU's are motherboard dependent, reflecting their different integration contexts.

The B390's transistor count and die size are not recorded, whereas the 64EU has a documented transistor count of 17,800 million on a 243 mm² die, giving a transistor density of 73.3M per mm². The B390's FP32 compute is 7.680 TFLOPS, compared to 1.946 TFLOPS for the 64EU. FP16 rates follow the same 2:1 ratio pattern: 15.36 TFLOPS for the B390 and 3.891 TFLOPS for the 64EU.

The B390's texture rate is 120.0 GTexel/s against 60.80 GTexel/s for the 64EU. Its pixel rate is 60.00 GPixel/s versus 30.40 GPixel/s. These figures confirm that the B390 doubles the 64EU's fillrate capabilities, matching the benchmark delta closely.

Specification Differences

The two GPUs differ across several recorded specification fields. The shading unit count is 1536 for the B390 versus 512 for the 64EU. Texture mapping units are 48 against 32. Render output units are 24 against 16. Ray tracing cores are present on the B390 (12) but absent from the 64EU.

Boost clocks differ: 2500 MHz for the B390, 1900 MHz for the 64EU. Base clocks are identical at 300 MHz. The B390's TDP is 80 W, while the 64EU's is 65 W. The B390's FP32 throughput is 7.680 TFLOPS versus 1.946 TFLOPS, and FP16 is 15.36 TFLOPS versus 3.891 TFLOPS.

Pixel rate is 60.00 GPixel/s for the B390 and 30.40 GPixel/s for the 64EU. Texture rate is 120.0 GTexel/s versus 60.80 GTexel/s. The bus interface differs: IGP for the B390, Ring Bus for the 64EU. Display outputs are portable device dependent for the B390 and motherboard dependent for the 64EU.

Memory parameters are identical in type: both use system shared memory with system dependent bandwidth and a system shared bus width. The power connector field is "None" for the B390 and not specified for the 64EU. Release dates differ: the B390 launched on 2026-01-26, while the 64EU launched on 2024-10-23. The 64EU has a listed predecessor (HD Graphics), while the B390 has none recorded.

Where Each One Wins

The Intel Arc B390 wins in every measured category. It dominates the 3DMark Steel Nomad DX12 test with a 102.2% higher score. Its compute throughput is nearly four times higher in FP32 (7.680 TFLOPS versus 1.946 TFLOPS), and its fillrate is exactly double in both pixel rate and texture rate. The B390's ray tracing cores provide hardware acceleration that the 64EU cannot match. The B390 also carries triple the shading units and double the texture mapping and render output units.

The B390's higher boost clock of 2500 MHz versus 1900 MHz gives it a per-clock advantage on top of its larger execution resources. Its 80 W TDP allows for this additional performance within an integrated form factor, though it draws 15 W more than the 64EU's 65 W.

The Intel Arc Graphics 64EU does not win any benchmark in the recorded data. Its only advantages are a lower TDP, an earlier release date, and a documented die size and transistor count. For workloads that rely solely on the CPU's integrated graphics with minimal power overhead, the 64EU's lower power draw may be preferable, but in pure graphics performance, the database shows no category where it leads.

The 64EU's nearest rivals include other Intel Arc low-EU variants (32EU and 24EU) with identical scores, plus older AMD and NVIDIA mobile parts. This places it in a legacy performance bracket. The B390, by contrast, sits alongside slightly older NVIDIA discrete mobile GPUs, all within a narrow 1.3% to 2.7% range.

The Verdict

The recorded data indicates a clear performance hierarchy. The Intel Arc B390 delivers 102.2% more performance than the Intel Arc Graphics 64EU in the sole benchmark available. The B390's 9th percentile ranking versus the 64EU's 3rd percentile confirms that the B390 belongs to a higher performance class overall.

For users prioritizing graphics throughput in a compact integrated package, the B390 is the stronger choice. Its triple shading units, dedicated ray tracing cores, higher clocks, and doubled fillrate all contribute to its decisive lead. The B390's 80 W TDP is modest enough for an integrated part, and its 2026 release date makes it the newer design.

The Intel Arc Graphics 64EU serves a different role. It consumes less power at 65 W, integrates into a Ring Bus architecture, and uses a TSMC-manufactured die with known transistor density. Its performance matches other low-EU Intel Arc parts exactly, indicating that the 64EU configuration does not translate into a measurable advantage over its smaller siblings in the tested workload. For applications where the absolute lowest power draw matters more than graphics capability, the 64EU remains a viable option, but the data shows no performance scenario where it overtakes the B390.

The B390's nearest rivals are all NVIDIA discrete mobile GPUs from older generations, and it leads each by a modest margin. The 64EU's nearest rivals include its own lower-EU variants, with identical scores, plus legacy discrete parts. This contrast reinforces the B390's position as the more capable integrated graphics solution in this comparison.