Arc A750

Arc A750 Specifications

The Arc A750 is based on the Xe HPG architecture and manufactured using the 6 nm process node. It was launched approximately 2 years, 7 months ago, originally priced at $289. It features 3584 cores, including 40 compute units (SMs), 224 texture mapping units (TMUs), and 112 render output units (ROPs). For ray tracing and AI-accelerated tasks, it also includes 28 dedicated RT cores and 448 Tensor cores, enabling advanced rendering and upscaling features where supported. It has a base clock of 2050 MHz and a boost clock reaching up to 2400 MHz. The Arc A750 comes equipped with 8 GB of GDDR6 memory across a 256-bit bus. Memory runs at 2000 MHz and delivers a total memory bandwidth of 512 GB/s.

This GPU has a moderate thermal design power (TDP) of 225 W. A PC case with good airflow and a quality power supply are required for stable operation. Due to the moderate TDP, this GPU is suitable for small form factor builds with somewhat limited airflow.

Arc A750 Performance Review

Arc A750 is an entry-level graphics card with a relative gaming performance score of 25.85%. It performs well at 1080p and can handle 1440p with optimized settings or features like DLSS and FSR, though it's not suitable for 4K. Intended for for casual gamers or users who play occasionally on general-purpose PCs.

The Arc A750 has a moderate VRAM size of 8 GB, which aligns well with its performance class. It is well-suited for 1080p gaming at high settings and 1440p gaming at medium to high settings, depending on the game’s optimization and graphics demands. While capable in rasterized workloads, enabling ray tracing at higher resolutions can quickly consume available VRAM and reduce performance. Upscaling technologies like DLSS or FSR can help improve frame rates, but may come with a noticeable reduction in visual quality. The VRAM capacity is adequate for light to moderate content creation tasks, such as editing Full HD or QHD video, working with smaller assets in 3D environments, and using tools like Photoshop or Premiere Pro for general workloads. In the context of AI and machine learning, this GPU can handle lightweight inference tasks and basic model training. However, users working with large datasets or complex architectures may find the memory insufficient, leading to slowdowns or compatibility issues. This GPU remains a good option for gamers, casual creators, and students exploring AI.

Arc A750 provides excellent value at 99.2% efficiency. While its gaming performance is modest, it can still be a cost-effective solution for users with limited demands.