Intel Arc Pro B390 vs NVIDIA GeForce RTX 4060 Max-Q Comparison

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark entries for the Intel Arc Pro B390 against the NVIDIA GeForce RTX 4060 Max-Q. Both products have an empty benchmark array, and the wins counter shows zero for each side. This absence of measured performance data means the comparison must rely entirely on the architectural specifications recorded in the database.

The raw compute figures show a measurable gap. The Intel Arc Pro B390 delivers 7.680 TFLOPS of FP32 throughput, while the NVIDIA GeForce RTX 4060 Max-Q reaches 9.032 TFLOPS. That places the NVIDIA part approximately 17.6% ahead in single-precision floating-point work. The pixel throughput tells a similar story: the RTX 4060 Max-Q sustains 70.56 GPixel/s against 60.00 GPixel/s for the Arc Pro B390, a 17.6% advantage. Texture rate follows the same pattern, with the NVIDIA part at 141.1 GTexel/s versus 120.0 GTexel/s for the Intel part, again a 17.6% lead.

The FP16 comparison is more complex. The Intel Arc Pro B390 lists 15.36 TFLOPS with a 2:1 ratio, meaning it processes half-precision at twice the FP32 rate. The NVIDIA GeForce RTX 4060 Max-Q lists 9.032 TFLOPS with a 1:1 ratio, meaning its FP16 throughput equals its FP32 throughput. In raw FP16 numbers, the Intel part is ahead by 70.1%. However, the NVIDIA architecture uses its tensor cores for mixed-precision workloads, which changes how these figures apply in real applications.

Clock behavior differs substantially. The Intel Arc Pro B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The NVIDIA GeForce RTX 4060 Max-Q starts at 1140 MHz base and boosts to 1470 MHz. The Intel part's boost clock is 70.1% higher than its base, while the NVIDIA part's boost is only 28.9% above its base. This suggests the Intel design relies on aggressive power scaling to reach its rated performance.

The memory subsystem separates the two decisively. The Intel Arc Pro B390 uses system shared memory with system-dependent bandwidth and no dedicated VRAM allocation. The NVIDIA GeForce RTX 4060 Max-Q carries 8 GB of GDDR6 on a 128-bit bus, delivering 256.0 GB/s of dedicated bandwidth. The memory clock for the NVIDIA part is 2000 MHz with 16 Gbps effective transfer. For workloads that depend on repeated memory access, the NVIDIA part has a structural advantage that cannot be mitigated by the Intel part's higher boost clock.

Where Each One Wins

The Intel Arc Pro B390 wins in raw FP16 throughput. Its 15.36 TFLOPS half-precision figure exceeds the NVIDIA part's 9.032 TFLOPS by 70.1%. This matters for workloads that can use the 2:1 FP16 path, such as certain AI inference tasks and media processing that operate natively in half precision.

The Intel part also wins on power envelope flexibility in one narrow sense. Its TDP is 80 W, while the NVIDIA part is rated at 35 W. The higher TDP means the Intel design can sustain higher absolute power draw, which supports its 2500 MHz boost clock. However, this is not a performance win in isolation; it is a thermal budget allocation.

The NVIDIA GeForce RTX 4060 Max-Q wins in FP32 compute, delivering 9.032 TFLOPS versus 7.680 TFLOPS, a 17.6% advantage. It also wins in pixel rate at 70.56 GPixel/s versus 60.00 GPixel/s, and in texture rate at 141.1 GTexel/s versus 120.0 GTexel/s. These are the core rasterization throughput metrics.

The NVIDIA part wins decisively on memory architecture. It has 8 GB of dedicated GDDR6 with 256.0 GB/s of bandwidth. The Intel part has no dedicated memory at all, relying on system shared memory with bandwidth that is system dependent. For any benchmark that stresses memory bandwidth, the NVIDIA part has a structural lead that no shared-memory configuration can match.

The NVIDIA part also wins on resource counts. It has 3072 shading units against 1536 for the Intel part, exactly double. It has 96 texture mapping units versus 48, also double. It has 48 ROPs versus 24, again double. It has 24 ray tracing cores versus 12, double. It has 96 tensor cores, which the Intel part does not list at all. These doubled resource counts align with the NVIDIA part's higher FP32 throughput.

The process node favors the Intel part. The Arc Pro B390 uses a 3 nm process from Intel, while the RTX 4060 Max-Q uses a 5 nm process from TSMC. The Intel part also lists the foundry as Intel, while the NVIDIA part lists TSMC as its foundry.

Architecture Differences

The Intel Arc Pro B390 uses the Panther Lake chip with the Xe3-LPG architecture, part of the Arc Graphics-WM generation. The NVIDIA GeForce RTX 4060 Max-Q uses the AD107 chip with the Ada Lovelace architecture, part of the GeForce 40 Mobile generation. These are different design philosophies entirely.

The Intel part is an integrated graphics processor, listed as IGP in slot width and bus interface. It has no power connectors and no standalone memory. Its display outputs are portable device dependent. The NVIDIA part is also listed as IGP in slot width, but it uses a PCIe 4.0 x8 bus interface. Both are mobile-class parts intended for portable devices.

The transistor counts differ dramatically. The NVIDIA part has 18,900 million transistors on a 159 mm² die, giving a transistor density of 118.9 million per square millimeter. The Intel part has unknown transistor count and die size in the database. The NVIDIA part's dedicated silicon footprint is substantially larger than what an integrated GPU would typically allocate.

Memory architecture is the largest divergence. The Intel Arc Pro B390 uses system shared memory for both size and type, with a system shared bus width and system dependent bandwidth. The NVIDIA GeForce RTX 4060 Max-Q has 8 GB of GDDR6 on a 128-bit bus with 256.0 GB/s of bandwidth. The NVIDIA memory clock is 2000 MHz with 16 Gbps effective transfer. This is a fundamental difference: one part depends on the host system's memory subsystem, the other has its own dedicated memory pool.

The compute feature sets differ in FP16 handling. The Intel part lists 15.36 TFLOPS FP16 with a 2:1 ratio, meaning it uses the same hardware units for both FP32 and FP16, with FP16 running at twice the rate. The NVIDIA part lists 9.032 TFLOPS FP16 with a 1:1 ratio, meaning its FP16 rate matches its FP32 rate. The NVIDIA part compensates with 96 tensor cores, which the Intel part does not list. These tensor cores handle matrix operations that the Intel part would process through its regular FP32 or FP16 paths.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The API support is identical across both products.

The release timeline shows the NVIDIA part launched on 2023-01-02, while the Intel part launched on 2026-01-26. The NVIDIA part has a predecessor in GeForce 30 Mobile and a successor in GeForce 50 Mobile. The Intel part has a predecessor in HD Graphics-WM and no successor listed.

Power characteristics differ by a factor of 2.3. The Intel Arc Pro B390 is rated at 80 W TDP, while the NVIDIA GeForce RTX 4060 Max-Q is rated at 35 W TDP. The NVIDIA part achieves higher FP32 throughput while drawing less than half the power budget. The Intel part uses its higher power allocation to reach a 2500 MHz boost clock, which is 70.1% higher than its 300 MHz base clock.

The Verdict

The recorded data points to the NVIDIA GeForce RTX 4060 Max-Q as the stronger performer for most workloads. Its FP32 throughput exceeds the Intel part by 17.6%, its pixel rate and texture rate each exceed the Intel part by 17.6%, and it carries 8 GB of dedicated GDDR6 memory with 256.0 GB/s of bandwidth. The doubled resource counts in shading units, TMUs, ROPs, and ray tracing cores reinforce this position. The 35 W TDP achieves these results at less than half the power draw of the Intel part's 80 W TDP.

The Intel Arc Pro B390 has one clear advantage in the recorded data: FP16 throughput. Its 15.36 TFLOPS half-precision rate is 70.1% higher than the NVIDIA part's 9.032 TFLOPS. The 3 nm process node and the 2500 MHz boost clock are also notable, though the boost clock advantage does not translate into higher FP32 throughput in the recorded figures.

For buyers who prioritize rasterization performance, dedicated memory bandwidth, and power efficiency, the NVIDIA GeForce RTX 4060 Max-Q is the better choice based on the database records. For workloads that can exploit the 2:1 FP16 path, the Intel Arc Pro B390 holds a mathematical advantage, provided the system memory subsystem can feed it adequately, since its bandwidth is system dependent.

FAQ

Q: Which GPU has higher FP32 performance?

A: The NVIDIA GeForce RTX 4060 Max-Q delivers 9.032 TFLOPS FP32, which is 17.6% higher than the Intel Arc Pro B390's 7.680 TFLOPS.

Q: How much dedicated memory does each GPU have?

A: The NVIDIA GeForce RTX 4060 Max-Q has 8 GB of GDDR6 on a 128-bit bus with 256.0 GB/s bandwidth. The Intel Arc Pro B390 uses system shared memory with system dependent bandwidth.

Q: What is the power consumption difference?

A: The Intel Arc Pro B390 is rated at 80 W TDP, while the NVIDIA GeForce RTX 4060 Max-Q is rated at 35 W TDP.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce RTX 4060 Max-Q has 3072 shading units, exactly double the Intel Arc Pro B390's 1536 shading units.

Q: How do the FP16 rates compare?

A: The Intel Arc Pro B390 has 15.36 TFLOPS FP16 with a 2:1 ratio, which is 70.1% higher than the NVIDIA GeForce RTX 4060 Max-Q's 9.032 TFLOPS with a 1:1 ratio.

Q: What process nodes do the two GPUs use?

A: The Intel Arc Pro B390 uses a 3 nm process from Intel, while the NVIDIA GeForce RTX 4060 Max-Q uses a 5 nm process from TSMC.