Intel Arc B390 vs Intel Arc Pro B60 Comparison

Where Each One Wins

The benchmark data splits sharply between these two Intel graphics parts. The Intel Arc B390 records a single 3DMark Steel Nomad DX12 score of 1482, while the Intel Arc Pro B60 reaches 2646 in the same test. That result gives the Pro B60 a 44% advantage in the only directly comparable workload, and it accounts for the 1-0 win tally in the head-to-head record.

Looking at the broader benchmark suite, the Arc Pro B60 accumulates scores across multiple test categories. Its PassMark G3D score of 14580 indicates strong overall 3D rendering capability. The G2D score of 763 shows 2D graphics performance, while the GPU compute score of 7029 reflects parallel processing throughput. API-specific PassMark results show 179 for DirectX 9, 122 for DirectX 11, 76 for DirectX 12, and 61 for DirectX 10. These figures demonstrate that the Pro B60 handles legacy DirectX workloads with varying efficiency, with the DirectX 9 score being the highest among the four API tests.

The Arc B390, by contrast, has no recorded PassMark results in the database. Its only benchmark entry is the 3DMark Steel Nomad DX12 score of 1482. This makes direct comparisons across multiple workloads impossible. The data indicates that the B390 is designed for a different form factor and power envelope, being an integrated graphics processor (IGP) with system-shared memory. Its performance profile in the single recorded test places it in the 9th percentile of all GPUs, while the Pro B60 sits in the 20th percentile.

The nearest rival comparisons reinforce the separation. The Arc B390's closest competitors are low-end NVIDIA parts: the GeForce GT 520MX (1463 average score, 1.3% delta), GeForce 800M (1460, 1.5%), GeForce GT 625 OEM (1446, 2.5%), and GeForce GT 710 (1443, 2.7%). The B390 leads all four by narrow margins. The Arc Pro B60's rivals include the Quadro P1000 (3163, 0.6% delta favoring the Pro B60), GeForce GT 640 (3210, 0.9% favoring the GT 640), GeForce 920M (3287, 3.2% favoring the 920M), and GeForce RTX 5080 SUPER (3075, 3.5% favoring the Pro B60). These deltas show the Pro B60 competing in a higher performance tier.

Architecture Differences

The two GPUs come from different architectural families and manufacturing processes. The Arc B390 uses the Xe3-LPG architecture on a 3 nm process, fabricated by Intel. Its chip is codenamed Panther Lake, and it belongs to the Arc Graphics-M (Panther Lake) generation. The Arc Pro B60 uses the Xe2-HPG architecture on a 5 nm process from TSMC, with the BMG-G21 chip, part of the Battlemage (Pro Series) generation.

Transistor and die details exist only for the Pro B60. It carries 19,600 million transistors on a 272 mm² die, yielding a transistor density of 72.1 million per mm². The B390's transistor count and die size are listed as unknown in the database.

Compute resources differ substantially. The B390 has 1536 shading units, 48 texture mapping units, 24 raster operation units, and 12 ray tracing cores. The Pro B60 scales up to 2560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores. This 67% increase in shading units and substantially higher TMU and ROP counts translate into the Pro B60's higher pixel rate of 192.0 GPixel/s versus 60.00 GPixel/s for the B390, and a texture rate of 384.0 GTexel/s versus 120.0 GTexel/s.

Clock behavior shows an interesting inversion. The B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The Pro B60 starts at 2000 MHz base and boosts to 2400 MHz. While the B390 has a higher peak boost, the Pro B60 operates at a much higher sustained base frequency. Floating-point throughput follows the shading unit counts: the B390 delivers 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 (2:1), while the Pro B60 delivers 12.29 TFLOPS FP32 and 24.58 TFLOPS FP16 (2:1).

Memory architecture separates the two fundamentally. The B390 uses system-shared memory with a system-dependent bus width and bandwidth. The Pro B60 has 24 GB of GDDR6 memory on a 192-bit bus, with a memory clock of 2375 MHz (19 Gbps effective) and bandwidth of 456.0 GB/s. This dedicated memory configuration gives the Pro B60 predictable performance independent of host system RAM.

Power and physical design also diverge. The B390 draws 80 W and is an integrated graphics processor with no power connectors and an IGP bus interface. The Pro B60 consumes 200 W, requires a dual-slot cooler, uses a single 8-pin power connector, and recommends a 550 W power supply. The Pro B60 connects via PCIe 5.0 x8 and measures 167 mm in length, 69 mm in height, and 40 mm in width. Display outputs for the Pro B60 are 4x mini-DisplayPort 2.1, while the B390's outputs depend on the portable device it is built into.

Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The B390 was released on 2026-01-26, while the Pro B60 launched earlier on 2025-09-04. Both are listed as Active in production status. The Pro B60 has a launch MSRP of 499 USD.

The Verdict

The recorded data points to distinct roles for each GPU. The Intel Arc Pro B60 delivers roughly 78% higher raw score in the shared 3DMark Steel Nomad DX12 benchmark (2646 versus 1482). It also provides dedicated 24 GB GDDR6 memory, a 192-bit bus, and 456.0 GB/s bandwidth, all of which enable sustained workloads without host memory contention. The Pro B60's pixel rate of 192.0 GPixel/s and texture rate of 384.0 GTexel/s place it in a different performance class entirely.

The Intel Arc B390, with its integrated design, 80 W power draw, and system-shared memory, occupies a mobile or compact-device segment. Its single benchmark score of 1482 places it just ahead of the GeForce GT 520MX and GeForce 800M, with margins of 1.3% and 1.5% respectively. The B390's boost clock of 2500 MHz is higher than the Pro B60's 2400 MHz, but the B390's 300 MHz base clock and lower resource counts limit sustained performance.

For workloads requiring dedicated graphics memory, high pixel throughput, and multiple display outputs, the Pro B60 is the clear choice based on the data. For integrated, low-power deployment in portable systems, the B390 fits that niche. The percentile rankings confirm the gap: the B390 sits in the 9th percentile of all GPUs, while the Pro B60 reaches the 20th percentile. The Pro B60's average benchmark score of 3182 compares to the B390's 1482, a factor of roughly 2.1x.

FAQ

Q: Which GPU has the higher 3DMark Steel Nomad DX12 score?

A: The Intel Arc Pro B60 scores 2646, while the Intel Arc B390 scores 1482. The Pro B60 leads by 44%.

Q: How much memory does each GPU have?

A: The Intel Arc Pro B60 has 24 GB of GDDR6 memory on a 192-bit bus with 456.0 GB/s bandwidth. The Intel Arc B390 uses system-shared memory with system-dependent bandwidth.

Q: What are the power requirements for each GPU?

A: The Intel Arc B390 draws 80 W and has no power connectors. The Intel Arc Pro B60 draws 200 W, requires a single 8-pin power connector, and recommends a 550 W power supply.

Q: Which GPU has more shading units?

A: The Intel Arc Pro B60 has 2560 shading units. The Intel Arc B390 has 1536 shading units.

Q: What is the process node for each chip?

A: The Intel Arc B390 uses a 3 nm process from Intel. The Intel Arc Pro B60 uses a 5 nm process from TSMC.

Q: How do the nearest rivals compare for each GPU?

A: The Arc B390 leads the GeForce GT 520MX by 1.3%, the GeForce 800M by 1.5%, the GeForce GT 625 OEM by 2.5%, and the GeForce GT 710 by 2.7%. The Arc Pro B60 leads the Quadro P1000 by 0.6% and the GeForce RTX 5080 SUPER by 3.5%, but trails the GeForce GT 640 by 0.9% and the GeForce 920M by 3.2%.

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is 3DMark Steel Nomad DX12. The Intel Arc Pro B60 scores 2646, and the Intel Arc B390 scores 1482. The delta is 44% in favor of the Pro B60. This single result constitutes the entire head-to-head record, with the Pro B60 taking the sole win.

Beyond the shared test, the Pro B60 has a broader benchmark footprint. Its PassMark results show a G3D score of 14580, which is the highest among its recorded metrics. The G2D score of 763 and GPU compute score of 7029 indicate balanced 2D and compute capabilities. Among the DirectX API tests, DirectX 9 leads at 179, followed by DirectX 11 at 122, DirectX 12 at 76, and DirectX 10 at 61. These numbers suggest the Pro B60's driver stack handles older APIs more efficiently than newer ones in PassMark's specific workloads, though the 3DMark Steel Nomad result (a DirectX 12 test) shows strong modern-API performance at 2646.

The B390 has no PassMark entries, so its performance in those categories is unrecorded. Its lone 3DMark result of 1482 places it near the bottom of the database's percentile ranking at 9%. The Pro B60's 20th percentile ranking, while still modest, represents a significant step up. The average benchmark scores reflect this: the Pro B60 averages 3182 across its eight recorded benchmarks, while the B390 averages 1482 across its single test.

The nearest rival data provides context for each GPU's standing. The B390's competitors are all low-end NVIDIA parts from previous generations, and the B390 beats each by a small margin. The Pro B60's field includes the Quadro P1000, a professional workstation card, and the GeForce RTX 5080 SUPER, a much newer discrete GPU. The Pro B60 edges out the Quadro P1000 by 0.6% and the RTX 5080 SUPER by 3.5%, while trailing the older GeForce GT 640 and GeForce 920M by less than 3.3% in both cases. These mixed results indicate the Pro B60 sits in a crowded mid-range performance band.

In summary, the database shows a clear performance hierarchy. The Pro B60 delivers more than 1.7 times the 3DMark score of the B390, offers dedicated memory, and supports professional display outputs. The B390 provides integrated graphics with a higher boost clock and lower power draw, suited to compact systems where the Pro B60's dual-slot, 200 W design would not fit.