Intel Arc A310E vs Intel Arc Pro B390 Comparison

FAQ

Q: What are the core architectural identities of the Intel Arc A310E and the Intel Arc Pro B390?

A: The Arc A310E is built on the Xe-HPG architecture using the DG2-128 chip, part of the Alchemist (Arc 3) generation, fabricated on a 6 nm process by TSMC. The Arc Pro B390 uses the Xe3-LPG architecture with the Panther Lake chip, part of the Arc Graphics-WM (Panther Lake) generation, fabricated on a 3 nm process by Intel.

Q: How do the two GPUs differ in memory configuration?

A: The Arc A310E has 4 GB of dedicated GDDR6 memory on a 64-bit bus, delivering 124.0 GB/s of bandwidth. The Arc Pro B390 uses system shared memory, with bus width and bandwidth described as system dependent, meaning it relies on the host platform's memory subsystem.

Q: What is the difference in shader and ray tracing resources?

A: The Arc A310E has 768 shading units, 32 texture mapping units, 16 raster operation units, and 6 ray tracing cores. The Arc Pro B390 doubles these counts to 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores.

Q: What are the clock speed profiles of the two cards?

A: The Arc A310E has a base clock of 2000 MHz and a boost clock of 2000 MHz, with memory clocked at 1937 MHz (15.5 Gbps effective). The Arc Pro B390 has a much lower base clock of 300 MHz but a higher boost clock of 2500 MHz.

Q: What are the power and physical form factor differences?

A: The Arc A310E has a 75 W TDP, is single-slot, requires no power connectors, and has a suggested PSU of 250 W. The Arc Pro B390 has an 80 W TDP, is an integrated graphics processor (IGP), uses no power connectors, and has no suggested PSU listed.

Q: What is the production status and release timing for each?

A: The Arc A310E is end-of-life, released on 2024-03-31, with a predecessor of Xe Graphics and a successor of Battlemage. The Arc Pro B390 is active, released on 2026-01-26, with a predecessor of HD Graphics-WM and no listed successor.

Architecture Differences

The Intel Arc A310E and Intel Arc Pro B390 represent two distinct architectural generations from the same vendor. The A310E is built on the Xe-HPG architecture, specifically the DG2-128 chip, which places it in the Alchemist (Arc 3) generation. This architecture is manufactured on a 6 nm process at TSMC, with a transistor count of 7,200 million across a die size of 157 mm², yielding a transistor density of 45.9M per mm².

In contrast, the Arc Pro B390 uses the Xe3-LPG architecture with the Panther Lake chip, belonging to the Arc Graphics-WM (Panther Lake) generation. This part is fabricated on a 3 nm process at Intel, though the transistor count and die size are listed as unknown in the database. The process node shift from 6 nm to 3 nm represents a significant manufacturing advancement, potentially enabling higher density and efficiency, though the data does not quantify this directly.

The compute resources differ substantially. The A310E provides 768 shading units, 32 TMUs, 16 ROPs, and 6 RT cores. The B390 doubles the shading units to 1536, increases TMUs to 48, ROPs to 24, and RT cores to 12. This doubling of core resources indicates a fundamentally larger execution engine in the B390, which directly impacts throughput capabilities.

Clock behavior is notably different between the two. The A310E runs at a flat 2000 MHz for both base and boost clocks, suggesting a fixed operating point. The B390 has a 300 MHz base clock and a 2500 MHz boost clock, a much wider dynamic range that allows for aggressive power management and burst performance when thermals permit.

Memory architecture is a critical differentiator. The A310E uses 4 GB of dedicated GDDR6 memory on a 64-bit bus with 124.0 GB/s bandwidth. The B390 relies entirely on system shared memory, making its bandwidth and capacity dependent on the host platform. This architectural choice fundamentally changes the performance envelope, particularly for memory-intensive workloads.

The physical implementation also diverges sharply. The A310E is a discrete, single-slot card with dimensions of 168 mm length, 69 mm height, and 20 mm width, using a PCIe 4.0 x8 interface. The B390 is an integrated graphics processor with no discrete dimensions, no bus interface beyond IGP, and display outputs dependent on the portable device. The A310E provides 4x mini-DisplayPort 2.0 outputs, while the B390's display connectivity is unspecified.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, indicating parity in API feature levels despite the architectural differences.

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark results between the Intel Arc A310E and Intel Arc Pro B390. With zero benchmark entries for either GPU and no nearest rivals listed, the comparison relies entirely on the recorded specification data and calculated throughput metrics.

The peak pixel rate differs significantly. The A310E delivers 32.00 GPixel/s, while the B390 reaches 60.00 GPixel/s. This represents an 87.5% advantage for the B390 in pixel throughput, directly reflecting the combination of higher clock speed and doubled ROP count. Texture rate shows a similar pattern: the A310E provides 64.00 GTexel/s, while the B390 achieves 120.0 GTexel/s, an 87.5% improvement.

Floating-point performance shows the largest gap. The A310E delivers 3.072 TFLOPS FP32 and 6.144 TFLOPS FP16 (2:1). The B390 delivers 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 (2:1). This represents a 150% advantage in both FP32 and FP16 compute for the B390, driven by the doubled shading units and the higher boost clock of 2500 MHz versus 2000 MHz.

The ray tracing resources also scale with the core counts. The B390's 12 RT cores versus the A310E's 6 RT cores suggests a proportional doubling in ray tracing workload capacity, though no benchmark data quantifies the real-world impact.

Memory bandwidth presents a different story. The A310E's dedicated GDDR6 provides a fixed 124.0 GB/s. The B390's system shared memory bandwidth is listed as system dependent, meaning it cannot be directly compared without knowing the host platform configuration. In scenarios where the host memory bandwidth exceeds 124.0 GB/s, the B390 could potentially offer higher bandwidth, but this is not guaranteed and depends entirely on the system design.

The A310E's memory clock of 1937 MHz (15.5 Gbps effective) is a fixed specification, while the B390's memory performance is variable. This makes the A310E more predictable in memory-bound workloads, while the B390's performance will scale with the platform.

Specification Differences

| Specification | Intel Arc A310E | Intel Arc Pro B390 |

|---|---|---|

| Chip | DG2-128 | Panther Lake |

| Architecture | Xe-HPG | Xe3-LPG |

| Generation | Alchemist (Arc 3) | Arc Graphics-WM (Panther Lake) |

| Process Node | 6 nm | 3 nm |

| Foundry | TSMC | Intel |

| Transistors | 7,200 million | unknown |

| Die Size | 157 mm² | unknown |

| Transistor Density | 45.9M / mm² | null |

| Base Clock | 2000 MHz | 300 MHz |

| Boost Clock | 2000 MHz | 2500 MHz |

| Memory Clock | 1937 MHz 15.5 Gbps effective | System Shared |

| Memory Size | 4 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Bus Width | 64 bit | System Shared |

| Bandwidth | 124.0 GB/s | System Dependent |

| Shading Units | 768 | 1536 |

| TMUs | 32 | 48 |

| ROPs | 16 | 24 |

| RT Cores | 6 | 12 |

| Pixel Rate | 32.00 GPixel/s | 60.00 GPixel/s |

| Texture Rate | 64.00 GTexel/s | 120.0 GTexel/s |

| FP32 | 3.072 TFLOPS | 7.680 TFLOPS |

| FP16 | 6.144 TFLOPS (2:1) | 15.36 TFLOPS (2:1) |

| TDP | 75 W | 80 W |

| Slot Width | Single-slot | IGP |

| Power Connectors | None | None |

| Suggested PSU | 250 W | null |

| Bus Interface | PCIe 4.0 x8 | IGP |

| Display Outputs | 4x mini-DisplayPort 2.0 | Portable Device Dependent |

| Dimensions | 168 mm 6.6 inches, 69 mm 2.7 inches, 20 mm 0.8 inches | null |

| Production Status | End-of-life | Active |

| Release Date | 2024-03-31 | 2026-01-26 |

| Predecessor | Xe Graphics | HD Graphics-WM |

| Successor | Battlemage | null |

The Verdict

The recorded data indicates a clear performance hierarchy between these two Intel GPUs. The Arc Pro B390 holds a substantial advantage in every compute throughput metric: 150% higher FP32, 150% higher FP16, 87.5% higher pixel rate, and 87.5% higher texture rate. Its shading unit count is double, and its ray tracing core count is double. The boost clock is 25% higher at 2500 MHz versus 2000 MHz.

The Arc A310E's advantages are concentrated in its memory architecture and physical form factor. It provides dedicated GDDR6 memory with a fixed 124.0 GB/s bandwidth, while the B390 depends on system shared memory with system dependent bandwidth. The A310E is a discrete single-slot card with defined dimensions and a PCIe 4.0 x8 interface, while the B390 is an integrated processor with no discrete dimensions or interface beyond IGP.

The release timeline places the A310E in the Alchemist generation with an end-of-life status, while the B390 is active and belongs to the newer Arc Graphics-WM generation. The B390's 3 nm Intel process versus the A310E's 6 nm TSMC process indicates a more advanced manufacturing node for the newer part. The B390's TDP of 80 W is only slightly higher than the A310E's 75 W, despite the significantly higher compute throughput.

For users requiring dedicated, predictable memory bandwidth and a discrete card form factor, the A310E offers those features but at a lower compute ceiling. For workloads that prioritize raw shader throughput, ray tracing resources, and the efficiency of a newer architecture, the B390 is the stronger candidate based on the specification-derived metrics.

Where Each One Wins

The Arc Pro B390 wins in compute-intensive workloads. Its FP32 throughput of 7.680 TFLOPS versus 3.072 TFLOPS gives it a 150% advantage in general-purpose shader workloads. The FP16 performance of 15.36 TFLOPS versus 6.144 TFLOPS provides the same margin for half-precision compute tasks. The doubled RT core count from 6 to 12 positions the B390 ahead for ray-traced rendering, assuming software uses these resources effectively.

The B390 also wins in fill-rate-bound scenarios. Its pixel rate of 60.00 GPixel/s versus 32.00 GPixel/s and texture rate of 120.0 GTexel/s versus 64.00 GTexel/s make it more capable for high-resolution rasterization and texture-heavy effects. The higher boost clock of 2500 MHz versus 2000 MHz contributes to these throughput advantages.

The Arc A310E wins in memory predictability. Its dedicated 4 GB GDDR6 on a 64-bit bus provides a fixed 124.0 GB/s bandwidth, whereas the B390's system shared memory has bandwidth that is system dependent. In platforms where the shared memory bandwidth is insufficient, the A310E could outperform the B390 in memory-bound tasks, though the database does not provide a direct comparison.

The A310E wins in discrete form factor utility. It offers 4x mini-DisplayPort 2.0 outputs, a PCIe 4.0 x8 interface, and dimensions of 168 mm by 69 mm by 20 mm, making it a self-contained add-in card. The B390 is an IGP with display outputs dependent on the portable device, meaning its connectivity is not directly specified.

The A310E also wins on power simplicity. Its 75 W TDP with no power connectors and a 250 W suggested PSU makes it straightforward for small builds. The B390's 80 W TDP is marginally higher, but as an IGP it draws from the host platform's power delivery, which is not directly comparable.

The B390 wins on production status and longevity. It is active, while the A310E is end-of-life. The B390's release on 2026-01-26 is nearly two years after the A310E's 2024-03-31 release, and the B390 has no successor listed, while the A310E's successor is Battlemage. For new designs, the active B390 presents less obsolescence risk.