Intel Arc G3 vs Intel Arc Pro B390 Comparison

FAQ

Q: What is the difference between the Intel Arc G3 and the Intel Arc Pro B390 in terms of architecture?

A: Both GPUs use the same Xe3-LPG architecture and both are built on Intel's 3 nm process node. They also share the same Panther Lake chip. The key architectural difference is in the number of execution units: the Arc G3 has 1280 shading units, while the Arc Pro B390 has 1536.

Q: How do the clock speeds compare between the two GPUs?

A: Both have the same base clock of 300 MHz. The boost clock differs: the Intel Arc G3 boosts up to 2400 MHz, while the Intel Arc Pro B390 boosts up to 2500 MHz.

Q: Does one GPU have more ray tracing capability than the other?

A: Yes. The Intel Arc G3 includes 10 ray tracing cores, while the Intel Arc Pro B390 includes 12 ray tracing cores. The Pro B390 has a 20% higher count in this area.

Q: What is the thermal design power (TDP) for each GPU?

A: The Intel Arc G3 has a TDP of 25 W. The Intel Arc Pro B390 has a TDP of 80 W, which is significantly higher.

Q: What is the release date for each product?

A: The Intel Arc Pro B390 was released earlier, on 2026-01-26. The Intel Arc G3 followed later, with a release date of 2026-05-31.

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. There is no difference in the API feature set between the two.

Architecture Differences

The Intel Arc G3 and the Intel Arc Pro B390 are built on the same fundamental architecture: Xe3-LPG. Both use the Panther Lake chip and are fabricated by Intel on a 3 nm process node. This means the underlying instruction set, memory model, and general design philosophy are identical. The differences are in scale and configuration.

The most significant architectural difference is the execution resource count. The Intel Arc Pro B390 has 1536 shading units, 48 texture mapping units (TMUs), and 24 render output units (ROPs). The Intel Arc G3 has 1280 shading units, 40 TMUs, and 20 ROPs. In each of these categories, the Pro B390 has exactly 20% more units than the G3. This consistent scaling suggests a larger slice of the Panther Lake die is enabled in the Pro B390.

The ray tracing hardware follows the same pattern. The Pro B390 has 12 ray tracing cores, while the G3 has 10. Again, this is a 20% increase.

Both GPUs share the same memory architecture: system shared memory, with system dependent bandwidth. Neither has dedicated VRAM, and both rely on the host system's memory. The memory type, bus width, and size are all listed as system shared, meaning the system configuration determines actual performance.

Clock behavior also differs. While the base clock is identical at 300 MHz, the boost clocks diverge. The Arc G3 boosts to 2400 MHz, while the Arc Pro B390 boosts to 2500 MHz. This 100 MHz difference, about 4.2%, adds to the Pro B390's already larger execution resource pool.

Both cards are integrated graphics products (IGP), use no power connectors, and have the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The display outputs are portable device dependent for both, which is typical for integrated graphics.

The production status for both is listed as active. The Intel Arc G3 does not list a predecessor, while the Intel Arc Pro B390 lists HD Graphics-WM as its predecessor.

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark results between the Intel Arc G3 and the Intel Arc Pro B390. The wins counter shows 0 for both sides, and there are no entries in the head-to-head benchmark list. As a result, the comparison must be derived from the specification data recorded in the database.

The theoretical peak performance figures provide the clearest comparison. The Intel Arc Pro B390 delivers 7.680 TFLOPS of FP32 compute, while the Intel Arc G3 delivers 6.144 TFLOPS. The Pro B390 is exactly 25% ahead of the G3 in FP32 throughput. This is the largest measurable performance gap between the two.

In FP16 compute, the Pro B390 reaches 15.36 TFLOPS (2:1), while the G3 reaches 12.29 TFLOPS (2:1). The same 25% advantage applies to the Pro B390. Both GPUs halve their FP32 throughput when performing FP16, as indicated by the 2:1 ratio.

The pixel fill rate follows the same proportional pattern. The Pro B390 achieves 60.00 GPixel/s, while the G3 achieves 48.00 GPixel/s. This is a 25% difference, consistent with the 24 ROPs versus 20 ROPs.

The texture fill rate also shows a 25% gap. The Pro B390 reaches 120.0 GTexel/s, while the G3 reaches 96.00 GTexel/s. This aligns with the 48 TMUs versus 40 TMUs.

The boost clock difference of 100 MHz (2500 MHz versus 2400 MHz) contributes a small additional advantage to the Pro B390, but the bulk of the performance gap comes from the larger execution resource pool. The Pro B390's shading unit, TMU, ROP, and ray tracing core counts are all 20% higher, and the higher boost clock compounds that advantage.

The power consumption difference is substantial. The Pro B390 has a TDP of 80 W, while the G3 is rated at 25 W. The Pro B390 uses 3.2 times the power budget of the G3. This is a much larger ratio than the 25% compute advantage, which has implications for thermal management and system design.

Both GPUs share the same percentile ranking of 50 against all GPUs in the database, and both have an average benchmark score of 0, reflecting the lack of recorded benchmark data.

Specification Differences

The following fields differ between the Intel Arc G3 and the Intel Arc Pro B390:

Boost Clock: The Arc G3 boosts to 2400 MHz, while the Arc Pro B390 boosts to 2500 MHz.

Shading Units: The Arc G3 has 1280, while the Arc Pro B390 has 1536.

Texture Mapping Units: The Arc G3 has 40, while the Arc Pro B390 has 48.

Render Output Units: The Arc G3 has 20, while the Arc Pro B390 has 24.

Ray Tracing Cores: The Arc G3 has 10, while the Arc Pro B390 has 12.

Pixel Rate: The Arc G3 delivers 48.00 GPixel/s, while the Arc Pro B390 delivers 60.00 GPixel/s.

Texture Rate: The Arc G3 delivers 96.00 GTexel/s, while the Arc Pro B390 delivers 120.0 GTexel/s.

FP32 Performance: The Arc G3 delivers 6.144 TFLOPS, while the Arc Pro B390 delivers 7.680 TFLOPS.

FP16 Performance: The Arc G3 delivers 12.29 TFLOPS (2:1), while the Arc Pro B390 delivers 15.36 TFLOPS (2:1).

TDP: The Arc G3 is rated at 25 W, while the Arc Pro B390 is rated at 80 W.

Generation: The Arc G3 is part of Arc Graphics-M (Panther Lake), while the Arc Pro B390 is part of Arc Graphics-WM (Panther Lake).

Release Date: The Arc Pro B390 released on 2026-01-26, while the Arc G3 released on 2026-05-31.

Predecessor: The Arc G3 has no listed predecessor, while the Arc Pro B390 lists HD Graphics-WM.

All other recorded fields are identical: process node (3 nm), foundry (Intel), base clock (300 MHz), memory configuration (system shared), API support (DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4), slot width (IGP), power connectors (none), bus interface (IGP), and display outputs (portable device dependent).

The Verdict

The recorded data shows a clear performance hierarchy. The Intel Arc Pro B390 is the more capable GPU in every measurable compute category. It delivers 25% higher FP32 throughput, 25% higher FP16 throughput, 25% higher pixel fill rate, and 25% higher texture fill rate than the Intel Arc G3. It also has 20% more shading units, 20% more TMUs, 20% more ROPs, and 20% more ray tracing cores.

The Intel Arc G3 is the more power efficient option. Its 25 W TDP is less than one third of the Pro B390's 80 W TDP. For systems where power delivery and thermal output are tightly constrained, the G3 is the appropriate choice. The compute per watt ratio favors the G3 substantially: it delivers 6.144 TFLOPS within a 25 W envelope, while the Pro B390 delivers 7.680 TFLOPS within an 80 W envelope.

The release dates indicate the Pro B390 came first, launching on 2026-01-26, while the G3 followed on 2026-05-31. Both are currently marked as active production products.

The database places both GPUs at the 50th percentile against all GPUs, which indicates they occupy a mid-range position in the overall GPU landscape. Neither is a top-tier performer, but both are above the midpoint of the recorded distribution.

There is no benchmark data to validate real-world performance, so the comparison rests entirely on the theoretical specifications. The consistency of the 25% gap across all compute metrics makes it likely that the Pro B390 will lead in most workloads, but the G3's power efficiency is a strong counterweight for power-sensitive applications.

Where Each One Wins

Integrated, low-power systems: The Intel Arc G3 wins here. Its 25 W TDP is dramatically lower than the Pro B390's 80 W TDP. For portable devices or compact systems with limited cooling, the G3 is the only viable option between the two. The 300 MHz base clock and 2400 MHz boost clock still provide respectable compute, and the system shared memory design means the G3 does not require dedicated memory components.

Compute-heavy workloads: The Intel Arc Pro B390 wins in raw throughput. Its 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 give it a 25% advantage over the G3 in every compute metric. Applications that are limited by shading units, texture throughput, or pixel output will favor the Pro B390.

Ray tracing workloads: The Pro B390 has 12 ray tracing cores versus 10 on the G3. Any workload that scales with ray tracing core count will see an advantage on the Pro B390. The 20% higher core count is a meaningful margin.

Power-constrained environments: The G3's 25 W TDP makes it suitable for systems where the Pro B390's 80 W TDP would exceed thermal or power budgets. The G3 delivers 6.144 TFLOPS of FP32 compute within that 25 W envelope, which is a strong efficiency ratio.

High-resolution or high-fill-rate scenarios: The Pro B390's 60.00 GPixel/s pixel rate versus 48.00 GPixel/s on the G3 gives it an edge in scenarios where rasterization output is the bottleneck. The texture rate of 120.0 GTexel/s versus 96.00 GTexel/s provides a similar advantage in texture-heavy scenes.

General purpose integrated graphics: The G3 is the safer default for most integrated systems due to its lower power draw. The Pro B390 is the choice when the system can accommodate the higher TDP and the application demands maximum compute from the integrated GPU.

The database records no direct benchmark wins for either GPU, so these allocations are based entirely on the specification deltas. The Pro B390 dominates on absolute performance, while the G3 dominates on efficiency. The choice between them depends on which constraint, performance or power, is more binding in the target system.