Intel Arc Graphics 1 Xe Mobile vs Intel Arc Pro B70 Comparison

The Intel Arc Graphics 1 Xe Mobile and the Intel Arc Pro B70 occupy opposite ends of Intel's graphics spectrum, yet both share the Arc branding. The mobile part is a low-power integrated solution for laptops, while the Pro B70 is a high-end discrete workstation card. The database shows no direct head-to-head benchmark matches between these two, meaning comparisons must rely entirely on their architectural and specification sheets. The former uses Intel's own 3 nm process and Xe3-LPG architecture, while the latter uses a TSMC 5 nm node and Xe2-HPG architecture. Their performance profiles are dictated by massive disparities in compute resources, memory subsystems, and power envelopes.

Where Each One Wins

The Intel Arc Graphics 1 Xe Mobile wins in scenarios defined by extreme power efficiency and physical integration. Its 25 W TDP is the only power figure recorded, and it requires no external power connectors, operating purely as an IGP on the bus interface. This makes it the only choice for thin-and-light portable devices where a discrete card cannot fit. Its 128 shading units, 8 texture mapping units, and 4 raster output pipelines are sufficient for basic display output and light 2D workloads. The 300 MHz base clock and 2300 MHz boost clock indicate a design tuned for bursty, low-intensity tasks rather than sustained heavy loads. The system-shared memory model means bandwidth is entirely dependent on the host platform, with no dedicated VRAM allocation.

The Intel Arc Pro B70 wins in every performance-intensive category recorded. It delivers 22.94 TFLOPS of FP32 compute, which is roughly 39 times higher than the mobile part's 588.8 GFLOPS. Its 32 GB of GDDR6 memory on a 256-bit bus provides 608.0 GB/s of bandwidth, a figure that dwarfs any system-shared solution. The Pro B70 also carries 4096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores, versus 128, 8, 4, and 1 respectively on the mobile chip. Its 2800 MHz boost clock is 500 MHz higher than the mobile part's boost, and its base clock of 2280 MHz exceeds the mobile chip's maximum boost. The 230 W TDP and 1x 8-pin power connector confirm a design meant for desktop workstations with dedicated power delivery.

The Verdict

The data indicates that the Intel Arc Pro B70 is the superior choice for any workload requiring substantial graphics throughput, large memory capacity, or high-bandwidth data movement. Its 32 GB frame buffer and 608.0 GB/s bandwidth make it suitable for professional visualization, AI inference, or rendering tasks that exceed the mobile part's system-shared memory limits. The Pro B70 also supports PCIe 5.0 x16, providing a wide host interface for data transfer. Its dual-slot form factor and 267 mm length require a desktop chassis with adequate space, but the performance payoff is clear from the recorded specifications.

The Intel Arc Graphics 1 Xe Mobile is the correct selection for ultraportable systems where power draw and physical footprint are the primary constraints. Its 25 W TDP and IGP slot width mean it can be integrated directly into a processor package without additional cooling or power circuitry. It carries the same DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 API support as the Pro B70, so software compatibility is not a differentiator. Users who need basic graphics acceleration, video decode, or multi-display output on a battery-powered device will find the mobile part adequate. The Pro B70, with its 230 W TDP and 550 W suggested PSU, is not viable for such environments.

The performance gap is so wide that no use case bridges the two products. The mobile part's pixel rate of 9.200 GPixel/s and texture rate of 18.40 GTexel/s are functionally negligible compared to the Pro B70's 358.4 GPixel/s and 716.8 GTexel/s. The Pro B70 also offers 32 ray tracing cores versus a single RT core on the mobile chip, making ray-traced workloads impractical on the latter. The verdict is straightforward: choose the Pro B70 for compute, memory, and rendering; choose the mobile Xe part for portability and power efficiency.

Head-to-Head Benchmarks

The database contains no direct benchmark comparisons between these two GPUs, so the analysis relies on the raw specification sheet. The most significant disparity is in FP32 compute. The Pro B70 delivers 22.94 TFLOPS, while the mobile part provides 588.8 GFLOPS. This places the Pro B70 at approximately 39 times the raw floating-point throughput. In FP16, the Pro B70 reaches 45.88 TFLOPS (2:1), versus 1,177.6 GFLOPS (2:1) for the mobile chip, a similar 39-fold gap.

Memory bandwidth is another decisive factor. The Pro B70's 608.0 GB/s from 32 GB GDDR6 on a 256-bit bus is not just a larger pool, it is a dedicated high-speed resource. The mobile part's system-shared memory carries no fixed bandwidth figure, only "System Dependent," meaning its effective throughput is limited by the host laptop's memory configuration. For workloads that stream large datasets, such as texture-heavy rendering or large model inference, the Pro B70's dedicated bandwidth avoids contention with CPU memory traffic.

Pixel and texture throughput follow the same pattern. The Pro B70's 358.4 GPixel/s fill rate is 39 times the mobile part's 9.200 GPixel/s. Its 716.8 GTexel/s texture rate is 39 times the mobile part's 18.40 GTexel/s. These ratios align with the shading unit counts, as expected for architectures with similar per-unit efficiency. The mobile part's 4 ROPs limit final output resolution and framebuffer operations, while the Pro B70's 128 ROPs handle high-resolution multi-display outputs without bottlenecking.

Clock speeds also favor the Pro B70. Its base clock of 2280 MHz exceeds the mobile part's boost clock of 2300 MHz, and its boost reaches 2800 MHz. The mobile chip's 300 MHz base clock is exceptionally low, indicating a power-saving idle state that ramps up under load. In sustained workloads, the Pro B70's higher sustained clocks, combined with 32 times more shading units, produce a compute advantage that no clock scaling on the mobile part can overcome.

FAQ

Q: Which GPU has more shading units?

A: The Intel Arc Pro B70 has 4096 shading units, while the Intel Arc Graphics 1 Xe Mobile has 128 shading units.

Q: Can the Intel Arc Graphics 1 Xe Mobile handle ray tracing?

A: It has a single ray tracing core, but with only 128 shading units and a 25 W TDP, ray-traced workloads are severely limited. The Pro B70 has 32 ray tracing cores for comparison.

Q: What is the memory capacity difference?

A: The Pro B70 features 32 GB of GDDR6 memory with a 256-bit bus and 608.0 GB/s bandwidth. The mobile part uses system-shared memory with no fixed capacity or bandwidth, labeled as "System Dependent."

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.

Q: What power connector does the Pro B70 require?

A: The Pro B70 requires a single 8-pin power connector and has a 230 W TDP. The mobile part uses no power connectors and has a 25 W TDP.

Q: Which GPU has a higher boost clock?

A: The Pro B70 boosts to 2800 MHz, while the mobile part boosts to 2300 MHz. The Pro B70's base clock of 2280 MHz is also higher than the mobile part's boost clock.

Architecture Differences

The architectural split is fundamental. The Intel Arc Graphics 1 Xe Mobile uses the Xe3-LPG architecture, built on Intel's 3 nm process. It belongs to the Arc Graphics-M (Wildcat Lake) generation and uses the Wildcat Lake chip. The Xe3-LPG designation indicates a low-power graphics variant optimized for integrated use. The architecture is designed for minimal energy consumption, with a 25 W TDP and no external power delivery. Its predecessor is listed as HD Graphics-M, showing a lineage from Intel's earlier integrated graphics.

The Intel Arc Pro B70 uses the Xe2-HPG architecture, built on TSMC's 5 nm process. It belongs to the Battlemage (Pro Series) generation and uses the BMG-G31 chip. The Xe2-HPG designation indicates a high-performance graphics architecture for professional discrete cards. The die size is 368 mm², a substantial physical footprint for a desktop GPU. This architecture is designed for maximum throughput, with 4096 shading units and 32 ray tracing cores in a dual-slot form factor.

The process node difference is notable: 3 nm for the mobile part versus 5 nm for the Pro B70. Despite the smaller node, the mobile part's transistor density is not recorded, and its die size is unknown. The foundry also differs, with Intel fabricating the mobile chip and TSMC fabricating the Pro B70. This reflects a strategic split, with Intel's own fabs handling the low-power integrated part while outsourcing the high-performance discrete chip.

The API support is identical across both architectures: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. This means software written for one will run on the other, though performance will vary drastically. The ray tracing core counts differ by a factor of 32, indicating that the Xe3-LPG architecture includes RT capability but at a token level, while Xe2-HPG implements it as a primary feature.

Specification Differences

The specification sheets show divergence in nearly every measurable category. The mobile part has a base clock of 300 MHz and a boost clock of 2300 MHz. The Pro B70 has a base clock of 2280 MHz and a boost clock of 2800 MHz. The Pro B70's memory runs at 2375 MHz (19 Gbps effective), while the mobile part's memory is system-shared with no fixed clock.

Memory configuration differs completely. The Pro B70 uses 32 GB of GDDR6 on a 256-bit bus, yielding 608.0 GB/s. The mobile part uses system-shared memory with a "System Dependent" bandwidth figure and no dedicated bus width. This means the mobile part's memory performance is tied to the laptop's RAM configuration, which is typically dual-channel DDR5 at lower bandwidth than dedicated GDDR6.

Compute resources show a 32-fold difference in shading units, TMUs, and ROPs. The Pro B70 has 4096 shading units, 256 TMUs, and 128 ROPs. The mobile part has 128 shading units, 8 TMUs, and 4 ROPs. Ray tracing cores are 32 versus 1. Pixel rate is 358.4 GPixel/s versus 9.200 GPixel/s. Texture rate is 716.8 GTexel/s versus 18.40 GTexel/s. FP32 compute is 22.94 TFLOPS versus 588.8 GFLOPS. FP16 compute is 45.88 TFLOPS versus 1,177.6 GFLOPS.

Power and physical specifications also differ. The Pro B70 has a 230 W TDP, a dual-slot width, a 1x 8-pin power connector, and a suggested PSU of 550 W. It measures 267 mm in length, 110 mm in height, and 39 mm in width. The mobile part has a 25 W TDP, an IGP slot width, no power connectors, and no recorded dimensions. The Pro B70 uses a PCIe 5.0 x16 bus interface, while the mobile part uses an IGP interface. Display outputs differ as well: the Pro B70 provides 1x HDMI 2.1 and 3x DisplayPort 2.1, while the mobile part's outputs are "Portable Device Dependent."

The release dates are close, with the Pro B70 launching on 2026-03-25 and the mobile part on 2026-04-15. The Pro B70 has a launch MSRP of 949 USD, while the mobile part has no recorded launch MSRP. The production status for the mobile part is listed as "Active," while the Pro B70's status is not recorded. The Pro B70 has no predecessor or successor listed, while the mobile part's predecessor is HD Graphics-M.