Intel Arc A380M

The Intel Arc A380M is a mobile graphics module built on the Xe-HPG architecture, specifically the DG2-128 chip, fabricated on a 6 nm process at TSMC. It occupies the 50th percentile in the global GPU performance distribution, indicating it sits at the exact midpoint of all benchmarked graphics processors, with an average benchmark score of zero in the current database. This places it in a peculiar position: it is not a low-end part, but its performance ceiling is strictly defined by its modest hardware configuration, which includes 1024 shading units and a 96-bit memory interface.

Benchmark Performance

The Arc A380M’s raw computational output is anchored by a base clock of 1550 MHz and a boost clock of 2000 MHz. This yields a peak FP32 throughput of 4.096 TFLOPS, which is the primary metric for traditional rasterization workloads. In terms of texture processing, the card delivers 128.0 GTexel/s, while pixel fill rate reaches 64.00 GPixel/s. These figures suggest that the card is engineered for 1080p gaming with medium to high settings, though the data does not support claims beyond that threshold.

The absence of a nearestRivals array in the fact pack means there are no direct percentage deltas to cite against specific competitors. However, the 50th percentile ranking is informative: it indicates that the A380M outperforms half of all GPUs in the database, yet lags behind the other half. This is a typical profile for a mobile chip designed for thin-and-light laptops, where thermal and power constraints limit absolute performance. The FP16 throughput of 8.192 TFLOPS (2:1 ratio) suggests that the card can double its rate in half-precision workloads, which may benefit certain compute tasks but does not translate directly into gaming performance.

The boost clock of 2000 MHz is notably high for a 35 W part, indicating Intel’s engineers prioritized frequency scaling within a tight power envelope. This is a double-edged sword: it helps maintain responsiveness in bursty workloads, but sustained loads may cause the chip to throttle if the cooling solution is inadequate. The texture rate of 128.0 GTexel/s, when divided by the 64 TMUs, confirms the theoretical per-TMU throughput aligns with the boost clock, so the architecture is balanced internally. Pixel rate of 64.00 GPixel/s, however, is constrained by the 32 ROPs, which could become a bottleneck at higher resolutions where fill-rate demands increase.

Ray Tracing and Feature Set

The Arc A380M includes 8 dedicated ray tracing cores, a feature that was once reserved for high-end desktop parts. This hardware support means the card can accelerate ray-traced effects in games that leverage DirectX 12 Ultimate, specifically the 12_2 feature level. The presence of these cores does not guarantee playable frame rates in heavy RT titles, but it does mean the card is not strictly limited to rasterization. The architecture supports Vulkan 1.4 and OpenGL 4.6, which covers the vast majority of modern PC games and emulation workloads.

The ray tracing cores are paired with Xe-HPG’s scalable design, which allows for variable RT workloads. In practice, the 8 cores on a 35 W chip will handle simple reflections and shadows, but complex global illumination scenes will likely strain the hardware. The card’s FP32 throughput of 4.096 TFLOPS is the baseline for any RT calculations, since the RT cores offload bounding volume hierarchy traversal but not the shading that follows a ray hit. This means that enabling ray tracing will reduce effective rasterization performance, as the shading units are shared between traditional and RT effects.

The API support is comprehensive for a mobile part: DirectX 12 Ultimate ensures compatibility with the latest titles, while Vulkan 1.4 and OpenGL 4.6 provide broad legacy and cross-platform coverage. The display outputs are listed as "Portable Device Dependent," which means the actual connectors vary by laptop model, so users cannot assume a standard set of ports. This is a notable caveat for external display connectivity, as some implementations may limit output options.

Who Should Consider It

Benchmark results position the Arc A380M as a 1080p-oriented GPU. The 64.00 GPixel/s pixel rate and 128.0 GTexel/s texture rate are sufficient for 1920×1080 resolution at medium settings in most modern titles. For esports games like Valorant or Counter-Strike 2, the 4.096 TFLOPS FP32 throughput will likely exceed the requirements for high frame rates, since these titles are CPU-bound at lower resolutions. Conversely, demanding AAA titles at 1440p will expose the 186.0 GB/s memory bandwidth limitation, potentially causing texture streaming issues or frame pacing drops.

The 50th percentile ranking suggests this is a mainstream part, not a budget or enthusiast piece. Users who prioritize portability and efficiency over absolute performance will find it suitable for casual gaming and content creation on the go. The card’s 6 GB VRAM is adequate for 1080p with high-resolution texture packs, but it may fall short for 4K texture loads or heavy modding scenarios. The lack of a launch MSRP field means no pricing guidance is available, so the decision should be based solely on performance needs.

For users who demand ray tracing, the 8 RT cores provide a baseline capability, but the expectation should be modest: enabling RT will likely require lowering resolution or settings to maintain playable frame rates. The card is not designed for high-refresh-rate 1440p gaming, nor is it a professional visualization tool. It is a balanced mobile solution for gamers who want a step above integrated graphics without sacrificing battery life or chassis size.

FAQ

Q: What is the performance percentile of the Intel Arc A380M?

A: It ranks in the 50th percentile of all GPUs in the database, meaning it outperforms exactly half of the benchmarked graphics cards.

Q: How much VRAM does the Arc A380M have and what type is it?

A: It has 6 GB of GDDR6 memory on a 96-bit bus, providing a total bandwidth of 186.0 GB/s.

Q: Does the Arc A380M support hardware ray tracing?

A: Yes, it includes 8 dedicated ray tracing cores, which are compatible with DirectX 12 Ultimate (12_2) workloads.

Q: What is the boost clock speed of this GPU?

A: The boost clock is 2000 MHz, while the base clock is 1550 MHz.

Q: What is the thermal design power (TDP) of the Arc A380M?

A: The TDP is rated at 35 W, which is a low-power specification for a mobile module.

Q: What is the FP32 performance in TFLOPS?

A: The peak FP32 throughput is 4.096 TFLOPS, with FP16 reaching 8.192 TFLOPS at a 2:1 ratio.

Power and Cooling

The Arc A380M is rated for a TDP of 35 W, which is exceptionally low for a discrete GPU. This power envelope allows for fanless or low-noise cooling solutions in thin laptops, but it also caps sustained performance. The card is delivered as an MXM Module with a slot width of "MXM Module" and a bus interface of MXM-A (3.1), meaning it is not a solder-down chip but a replaceable module. However, no power connector details are provided, and no suggested PSU rating is listed, so the power delivery is entirely dependent on the host laptop’s design.

The 35 W TDP is the total board power, which includes the memory and VRM losses. This low figure suggests that the card can operate in devices with modest cooling systems, such as 14-inch ultrabooks or compact gaming laptops. The absence of a suggested PSU is irrelevant for mobile users, but for any theoretical MXM upgrade scenario, the host system’s power delivery must be verified. The production status is "Active," meaning the module is still in production, though the release date of 2023-01-23 indicates it has been on the market for a while.

Under sustained load, the 2000 MHz boost clock may not be maintainable if the thermal solution cannot dissipate the 35 W efficiently. The 6 nm process node from TSMC helps with power efficiency, but the 7,200 million transistors on a 157 mm² die (transistor density of 45.9M/mm²) are packed densely, which can create localized hot spots. Users should expect the card to run warm but not hot, provided the laptop’s cooling is properly designed.

Memory Subsystem

The memory configuration consists of 6 GB of GDDR6 running at an effective speed of 15.5 Gbps, with a memory clock of 1937 MHz. The 96-bit bus width is the limiting factor, as it yields a total bandwidth of 186.0 GB/s. This is a moderate figure by modern standards, but it is sufficient for 1080p gaming where texture sizes are manageable. At 1440p, the bandwidth may become a bottleneck, particularly in scenes with large draw distances or high-resolution shadows.

The 6 GB capacity is adequate for current games at 1080p with high texture settings, but it is not future-proof. Titles that require more than 6 GB of VRAM at 1080p will force the card to fall back to system memory, which will cause stuttering. The 96-bit bus is narrower than the 128-bit interfaces found on many desktop parts, which means the card trades bandwidth for lower power consumption and smaller die area. The effective 15.5 Gbps memory speed is high, partially compensating for the narrow bus, but the overall bandwidth is still the card’s Achilles heel.

For compute workloads that rely on memory bandwidth, such as some AI inference tasks, the 186.0 GB/s figure will limit performance. The FP16 throughput of 8.192 TFLOPS suggests the card can handle some machine learning tasks, but the memory subsystem will likely be the constraint. In gaming, high-resolution texture packs for 4K are out of the question, but 1080p ultra settings should fit within the 6 GB frame buffer.

How It Compares

Since the nearestRivals array is empty, the Arc A380M cannot be directly compared to specific competitor models using numeric deltas. The 50th percentile ranking is the only positional reference available. This means the card sits in the middle of the performance distribution, which is a crowded segment occupied by various mobile and low-power desktop parts. Without rival names or scores, the analysis must rely on the internal metrics: the 4.096 TFLOPS FP32 throughput is the primary comparison point.

The card’s 35 W TDP is a defining characteristic, as it places the A380M in the efficiency tier rather than the performance tier. Its 6 GB VRAM and 186.0 GB/s bandwidth are typical for this class. The 8 RT cores are a differentiator, as many competing parts in this power class lack hardware ray tracing entirely. The DirectX 12 Ultimate support ensures feature parity with higher-end cards, but the execution will be slower due to fewer shading units and lower bandwidth.

The 2023 release date suggests the A380M is a current-generation part, but the "Active" production status indicates it is still being manufactured. The lack of a successor or predecessor in the data means it occupies a standalone position in Intel’s mobile lineup. For users, this means it is a mid-tier option that competes on efficiency and features rather than raw speed. The absence of a launch MSRP further complicates any value assessment, so the card must be judged purely on its benchmark-derived capabilities.