Intel Xe DG1

The Intel Xe DG1 is a discrete graphics card built on the Generation 12.1 architecture, manufactured on Intel's 10 nm process. It packs 640 shading units, 40 texture mapping units, and 20 raster output units, with a base clock of 900 MHz and a boost clock of 1550 MHz. The card is equipped with 4 GB of LPDDR4X memory on a 128-bit bus, yielding a bandwidth of 68.26 GB/s. Its thermal design power is a modest 30 W, and it draws all power from the PCIe slot—no auxiliary power connectors are present. The DG1 is listed as end-of-life, with the "Graphics" line as its predecessor and "Alchemist" as its successor. In the benchmark database, it holds the 50th percentile among all GPUs, though its average benchmark score is recorded as 0.

Power and Cooling

The Xe DG1's power profile is exceptionally low for a discrete card. The 30 W TDP places it in the same thermal envelope as many integrated graphics solutions, yet it is delivered as a dual-slot card. This dual-slot design, with a length of 164 mm (6.5 inches), a height of 121 mm (4.8 inches), and a width of 22 mm (0.9 inches), suggests a substantial heatsink or shroud despite the minimal heat output. The absence of power connectors means that installation is straightforward: the card draws all required power from the PCIe 4.0 x8 slot, which can supply up to 75 W per specification. The suggested power supply is only 200 W, indicating that even entry-level PSUs can handle this card without concern. The low TDP also implies that cooling requirements are minimal, but the dual-slot form factor may be overkill for the 30 W dissipation; it could be a design choice to ensure compatibility with larger heatsinks or to maintain a uniform product line. The memory clock runs at 2133 MHz with an effective data rate of 4.3 Gbps, and the memory itself is LPDDR4X—a low-power variant—further contributing to the overall efficiency. In practice, this card should run quietly and coolly, making it suitable for small-form-factor or silent builds where power and thermal headroom are at a premium. The lack of external power connectors also simplifies cable management, as no 6-pin or 8-pin PCIe cables are required.

How It Compares

In the absence of direct rival benchmark data, the Xe DG1's position must be inferred from its database percentile and theoretical specifications. The 50th percentile among all GPUs indicates a median standing, but this is not corroborated by any actual benchmark score—the average benchmark score is 0, which suggests that no reliable performance results have been recorded for this part. This is a significant caveat: the percentile may be a placeholder or derived from limited data. The card's predecessor is the "Graphics" line, which is a broad category rather than a specific product, and its successor is "Alchemist," Intel's next discrete GPU generation. This places the DG1 as an early, exploratory discrete effort from Intel, likely aimed at validating the Xe architecture in a standalone form factor. Compared to the integrated graphics of the same era, the DG1 offers dedicated memory and higher clock speeds, but without benchmark results, a quantitative comparison is impossible. The card's 10 nm process and Generation 12.1 architecture are modern, yet the absence of ray tracing and tensor cores sets it apart from higher-tier offerings. Its 4 GB memory capacity and 128-bit bus are typical of entry-level discrete GPUs, but the 68.26 GB/s bandwidth is relatively narrow. The FP32 throughput of 1.984 TFLOPS and texture rate of 62.00 GTexel/s place it in a low-power segment, but again, these are theoretical peaks that do not translate directly into game performance. Given the end-of-life status, the DG1 is unlikely to receive major driver optimizations, which could further limit its real-world competitiveness. Overall, the data suggests a product that was positioned as a low-cost, low-power solution, but its actual standing relative to rivals cannot be established from the available facts.

Ray Tracing and Feature Set

The Xe DG1 does not include dedicated ray tracing cores or tensor cores. This is a critical omission for modern workloads that rely on hardware-accelerated ray tracing or AI-based features such as DLSS. The card's feature set is limited to traditional rasterization, with support for DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. These APIs provide a solid foundation for current games, but the lack of RT acceleration means that any ray-traced effects will be handled in software, which is generally too slow for real-time use. The absence of tensor cores also precludes any form of hardware-accelerated machine learning inference, which is increasingly used in upscaling and post-processing. The FP32 performance is rated at 1.984 TFLOPS, while FP16 reaches 3.968 TFLOPS with a 2:1 ratio—this indicates that the card can perform mixed-precision work, but without dedicated tensor hardware, the efficiency is not comparable to NVIDIA's Tensor Core offerings. The card does support a modern feature set in terms of API compatibility, including Vulkan 1.4, which is the latest version as of the data. However, the lack of RT and tensor cores means that the DG1 is best suited for conventional rendering pipelines. The pixel rate is 31.00 GPixel/s and the texture rate is 62.00 GTexel/s, which are modest figures that align with its low-power design. The display outputs include one HDMI port and three DisplayPort connections, allowing for multi-monitor setups. The bus interface is PCIe 4.0 x8, which provides adequate bandwidth for the 4 GB memory pool, though the x8 lane count is unusual for a discrete card—most use x16. This could be a cost-saving measure or a reflection of the card's limited bandwidth requirements.

Who Should Consider It

The Xe DG1 is aimed at users who need a discrete GPU for basic tasks without the power or space requirements of a full-size card. Its 30 W TDP and lack of power connectors make it an ideal drop-in upgrade for pre-built systems with minimal PSU headroom, such as office PCs or small-form-factor desktops. The 4 GB memory capacity is sufficient for light gaming at lower resolutions, but the 128-bit bus and 68.26 GB/s bandwidth will likely limit performance in modern titles. The card's 20 ROPs and 40 TMUs are entry-level figures, suggesting that it can handle 1080p gaming at reduced settings, but concrete resolution and quality recommendations cannot be made without benchmark data. The lack of ray tracing and tensor cores means that users who prioritize those features should look elsewhere. For productivity tasks such as video playback, office applications, or casual web browsing, the DG1 offers a significant upgrade over integrated graphics, providing smooth acceleration for multiple displays. The dual-slot form factor, while larger than necessary for the TDP, ensures compatibility with standard PCIe slots and may include a passive or low-noise cooler. The card's end-of-life status means that it is no longer in production, so potential buyers would need to source it from existing inventory or the second-hand market. Given its low power draw, it is also suitable for systems with limited airflow, as the thermal load is minimal. However, the absence of benchmark scores and the lack of driver support for future titles are significant drawbacks. The card's 50th percentile ranking, if taken at face value, suggests it performs around the median of all GPUs, but this is unverified. In summary, the Xe DG1 is a niche product for users who require a low-power, no-frills discrete GPU, but it is not suited for demanding gaming or professional workloads.

Benchmark Performance

The benchmark data for the Intel Xe DG1 is notably absent. The database lists an average benchmark score of 0, which is an anomaly—it likely indicates that no benchmark results have been submitted or that the card failed to complete any tests. The percentile vs all GPUs is 50, but this figure is not derived from actual performance data; it is a neutral placeholder. Consequently, any quantitative comparison against other GPUs is impossible from the available facts. The card's theoretical peak FP32 rate is 1.984 TFLOPS, which is a measure of raw compute capability. Its texture rate is 62.00 GTexel/s, and its pixel rate is 31.00 GPixel/s, both of which are derived from the clock speeds and unit counts. These numbers suggest that the DG1 is capable of basic 3D rendering, but they do not predict real-world frame rates. The memory bandwidth of 68.26 GB/s is a bottleneck for high-resolution textures, and the 4 GB capacity may cause stuttering in games that exceed this limit. The FP16 performance of 3.968 TFLOPS (2:1) indicates that the card can double its throughput in half-precision workloads, which could benefit certain compute applications but is not relevant for gaming. The lack of ray tracing and tensor cores further narrows its performance envelope. Given the end-of-life status, it is unlikely that future driver optimizations will improve performance. The card's 50th percentile standing, if it were based on real data, would place it in the middle of the pack, but the zero average score suggests that this ranking is unreliable. In the absence of benchmark results, the only objective performance indicators are the theoretical specifications, which point to a low-power entry-level GPU. The card's 10 nm process and Generation 12.1 architecture are modern, but the implementation is limited by its narrow memory bus and lack of specialized hardware. Overall, the benchmark performance of the Xe DG1 cannot be assessed with any confidence from the available data, and any claims about its gaming capabilities would be speculative.