Intel Arc B390 vs Intel Arc Pro A60 Comparison

The Intel Arc B390 and Intel Arc Pro A60 occupy very different positions in the recorded benchmark data. The B390, a Panther Lake integrated graphics solution on a 3 nm process, posts a single Steel Nomad DX12 score of 1482, placing it in the 9th percentile of all GPUs. The Pro A60, an Alchemist discrete workstation card on TSMC's 6 nm node, averages 60,326 across OpenCL and Vulkan tests, landing in the 88th percentile. These are not competing products in any conventional sense, but the database allows a direct comparison of their architectural and specification traits.

The Verdict

The data indicates that the Intel Arc Pro A60 is the overwhelmingly stronger performer. Its average benchmark score of 60,326 dwarfs the Arc B390's 1,482, a gap of roughly 40 times. The Pro A60's percentile ranking of 88 places it near the top of the database, while the B390's 9th percentile puts it near the bottom. The nearest rivals for the Pro A60 include the NVIDIA GeForce RTX 4090 (average score 60,347, a 0% delta), the AMD Radeon Pro W6600M (61,896, delta of -2.5% meaning the Pro A60 trails that card by 2.5%), and the AMD Radeon PRO V710 (58,657, where the Pro A60 leads by 2.8%). The Arc B390, by contrast, sits closest to the NVIDIA GeForce GT 520MX (1,463, delta of 1.3%), the GeForce 800M (1,460, delta 1.5%), the GT 625 OEM (1,446, delta 2.5%), and the GT 710 (1,443, delta 2.7%). The B390 edges out those legacy cards by small margins, but the performance class is fundamentally different.

For users selecting based on the recorded data, the Arc Pro A60 is the only choice for any compute-heavy workload. It delivers 8.397 TFLOPS FP32 and 16.79 TFLOPS FP16, while the B390 manages 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16. The Pro A60 also offers dedicated memory, a 192-bit bus, and 384.0 GB/s bandwidth, whereas the B390 relies on system shared memory with system-dependent bandwidth. The B390's only advantages are its lower 80 W TDP versus 130 W, its integrated form factor (IGP slot width, no power connectors), and a 2500 MHz boost clock versus 2050 MHz. However, raw clock speed does not compensate for the massive difference in shading units, texture units, and memory subsystem. The verdict is clear: the Arc Pro A60 dominates in every benchmark category, and the B390 should only be considered for ultra-low-power integrated scenarios where discrete graphics are not possible.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Intel Arc Pro A60 has an average benchmark score of 60,326, based on Geekbench OpenCL (63,485) and Vulkan (57,166) tests. The Intel Arc B390 has a single 3DMark Steel Nomad DX12 score of 1,482, making its average 1,482. The Pro A60's score is approximately 40 times higher.

Q: How do the two compare in terms of memory architecture?

A: The Arc Pro A60 uses 12 GB of GDDR6 memory on a 192-bit bus with 384.0 GB/s bandwidth. The Arc B390 uses system shared memory, with a bus width marked as "System Shared" and bandwidth described as "System Dependent." The Pro A60's dedicated memory provides deterministic bandwidth, while the B390's performance varies with the host system.

Q: What are the process node differences?

A: The Arc B390 is built on Intel's 3 nm process, while the Arc Pro A60 uses TSMC's 6 nm process. The Pro A60's die is 269 mm² with 11,500 million transistors and a transistor density of 42.8M per mm². The B390's transistor count and die size are unknown in the database.

Q: Which card has more shading units and texture units?

A: The Arc Pro A60 has 2,048 shading units, 128 texture mapping units (TMUs), and 64 raster output units (ROPs). The Arc B390 has 1,536 shading units, 48 TMUs, and 24 ROPs. The Pro A60 also has 16 ray tracing cores versus 12 on the B390.

Q: What is the power consumption difference?

A: The Arc B390 has a TDP of 80 W and requires no power connectors, fitting an IGP slot. The Arc Pro A60 has a TDP of 130 W, uses a single-slot design, and requires a 300 W suggested power supply. The B390 is designed for portable devices, while the Pro A60 is a discrete PCIe 4.0 x16 card.

Q: How do the display outputs differ?

A: The Arc Pro A60 provides 4x DisplayPort 2.0 outputs. The Arc B390's display outputs are listed as "Portable Device Dependent," meaning they vary by the host device and are not standardized in the database.

Architecture Differences

The two Intel GPUs come from separate architectural lineages. The Arc B390 uses the Xe3-LPG architecture, a low-power graphics variant designed for integration into Panther Lake processors. The process node is 3 nm, fabricated by Intel. The architecture targets mobile and portable devices, as evidenced by its IGP slot width, lack of power connectors, and portable-device-dependent display outputs. The B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, identical API support to the Pro A60.

The Arc Pro A60 uses the Xe-HPG architecture, specifically the DG2-256 chip, which belongs to the Alchemist generation. This is a discrete workstation GPU fabricated by TSMC on a 6 nm process. The die measures 269 mm² and contains 11,500 million transistors, resulting in a density of 42.8M per mm². The larger die and dedicated memory interface (PCIe 4.0 x16) are hallmarks of a standalone card. The Pro A60's 12 GB GDDR6 memory with a 192-bit bus and 384.0 GB/s bandwidth reflects a professional-grade memory subsystem, whereas the B390's memory is system shared, meaning it borrows from the host's RAM without dedicated VRAM.

The B390's base clock is 300 MHz with a boost of 2500 MHz, while the Pro A60 runs at 900 MHz base and 2050 MHz boost. The B390's higher boost clock does not translate into higher throughput, as the Pro A60's larger count of execution units (2,048 shading units versus 1,536) and double the TMUs (128 versus 48) provide more parallel processing capacity. The B390 has 24 ROPs versus 64, and 12 ray tracing cores versus 16. The pixel rate on the Pro A60 is 131.2 GPixel/s and texture rate 262.4 GTexel/s, compared to 60.00 GPixel/s and 120.0 GTexel/s on the B390. In terms of compute, the Pro A60 delivers 8.397 TFLOPS FP32 and 16.79 TFLOPS FP16 (2:1), while the B390 achieves 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 (2:1). The Pro A60 leads in every throughput metric.

The B390's memory clock is listed as "System Shared," meaning there is no dedicated memory clock. The Pro A60 has a memory clock of 2000 MHz with 16 Gbps effective. The B390's bandwidth is "System Dependent," which means it relies on the host's memory bandwidth, a variable that can bottleneck performance. The Pro A60's fixed 384.0 GB/s bandwidth is a critical advantage for data-intensive workloads.

Specification Differences

The two cards differ across nearly every specification field. The Arc B390 uses a Panther Lake chip and Xe3-LPG architecture, while the Arc Pro A60 uses a DG2-256 chip and Xe-HPG architecture. The process nodes diverge: 3 nm for the B390, 6 nm for the Pro A60. The foundry also differs, with Intel fabricating the B390 and TSMC the Pro A60. The Pro A60 has documented transistors (11,500 million), die size (269 mm²), and density (42.8M per mm²), while the B390 lists these as unknown.

Clock speeds: the B390 runs at 300 MHz base and 2500 MHz boost, versus 900 MHz base and 2050 MHz boost for the Pro A60. The B390's memory clock is "System Shared," while the Pro A60 uses 2000 MHz with 16 Gbps effective. Memory capacity: the B390 has "System Shared" size, the Pro A60 has 12 GB. Memory type: system shared versus GDDR6. Bus width: system shared versus 192 bit. Bandwidth: system dependent versus 384.0 GB/s.

Compute resources: the B390 has 1,536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. The Pro A60 has 2,048 shading units, 128 TMUs, 64 ROPs, and 16 ray tracing cores. Pixel rate: 60.00 GPixel/s versus 131.2 GPixel/s. Texture rate: 120.0 GTexel/s versus 262.4 GTexel/s. FP32: 7.680 TFLOPS versus 8.397 TFLOPS. FP16: 15.36 TFLOPS versus 16.79 TFLOPS.

Power and form factor: the B390 has an 80 W TDP, IGP slot width, no power connectors, and a bus interface of IGP. The Pro A60 has a 130 W TDP, single-slot width, a suggested PSU of 300 W, and a PCIe 4.0 x16 interface. Display outputs: the B390 is "Portable Device Dependent," while the Pro A60 has 4x DisplayPort 2.0. Release dates: the B390 was released on 2026-01-26, the Pro A60 on 2023-06-05. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither has a launch MSRP in the database.

Head-to-Head Benchmarks

The database includes no direct head-to-head benchmark results between the Arc B390 and Arc Pro A60. Instead, each card has its own distinct test set. The B390 was tested with 3DMark Steel Nomad DX12, scoring 1,482. The Pro A60 was tested with Geekbench OpenCL (63,485) and Geekbench Vulkan (57,166), averaging 60,326. Because the test types differ, a direct apples-to-apples comparison is not possible from the recorded data. However, the average scores and percentile rankings provide a clear hierarchy.

The Pro A60's 88th percentile versus the B390's 9th percentile is the strongest signal. The Pro A60's average score of 60,326 is 60,326 minus 1,482, a difference of 58,844, which is over 39 times higher. In the nearest rival comparisons, the B390 competes with GPUs from over a decade ago, such as the GT 520MX and GT 710, and only leads by 1.3% to 2.7%. The Pro A60 competes with modern high-end parts, trading blows with the RTX 4090 (0% delta) and Radeon Pro W6600M (trailing by 2.5%), while leading the Radeon PRO V710 by 2.8% and the Radeon Pro Vega 48 by 0.3%.

The B390's single benchmark result indicates it is a low-end integrated solution. Its 9th percentile means it outperforms only about 9% of all GPUs in the database. The Pro A60's 88th percentile means it outperforms 88% of all GPUs, placing it in the upper tier. The B390's wins are limited to efficiency metrics: lower TDP (80 W versus 130 W), higher boost clock (2500 MHz versus 2050 MHz), and integrated form factor. These are not performance wins but design choices.

In terms of compute throughput, the Pro A60 delivers 8.397 TFLOPS FP32, which is 8.397 minus 7.680, a 0.717 TFLOPS advantage over the B390's 7.680 TFLOPS. That is roughly a 9% lead in raw FP32. The FP16 gap is 16.79 versus 15.36, a 1.43 TFLOPS lead. The pixel rate difference is more pronounced: 131.2 versus 60.00 GPixel/s, a 2.19 times advantage. The texture rate difference is 262.4 versus 120.0 GTexel/s, a 2.19 times advantage. These ratios underscore the Pro A60's superior rasterization throughput.

Memory bandwidth is where the largest proportional gap exists. The Pro A60 has 384.0 GB/s dedicated bandwidth, while the B390 has system-dependent bandwidth with no fixed figure. The B390's reliance on shared memory means its effective bandwidth is whatever the host system provides, which is typically far lower than dedicated GDDR6. The Pro A60's 192-bit bus and 12 GB capacity provide a stable foundation for professional workloads.

The data shows no scenario where the B390 outperforms the Pro A60 in a benchmark. The B390's only advantages are its lower power draw and integrated nature, which are relevant for portable devices but not for raw performance. The Pro A60's nearest rival list includes the RTX 4090, a flagship consumer card, indicating it belongs to a completely different performance tier. The B390's nearest rivals are entry-level GPUs from prior generations. The verdict from the database is unambiguous: the Arc Pro A60 is the superior GPU by a wide margin, and the Arc B390 is a minimal-performance integrated solution suitable only for basic display tasks.