Intel HD Graphics 620 Mobile

Benchmark Performance

The Intel HD Graphics 620 Mobile is an integrated graphics processor built on the Generation 9.5 architecture, manufactured on Intel's 14 nm++ process node. It is based on the Kaby Lake GT2 chip and sits in the HD Graphics-M (Kaby Lake) generation. With a 50th percentile ranking against all GPUs, this part occupies the exact middle of the performance distribution, indicating it is neither a standout performer nor a laggard in the broader GPU landscape.

The benchmark data shows no synthetic scores or average benchmark results for this product, meaning its performance must be inferred from its architectural characteristics rather than direct measurements. The GPU operates with a base clock of 300 MHz and a boost clock of 1000 MHz, a substantial 3.33x uplift between idle and peak states. This wide clock range suggests the processor dedicates significant thermal and power headroom to burst workloads while conserving energy during lighter tasks. In practice, the 1000 MHz boost ceiling limits raw throughput, as even modern entry-level discrete GPUs typically reach far higher clock speeds.

The shading unit configuration consists of 192 shading units, 24 texture mapping units, and only 3 raster output pipelines. This imbalance is telling: the texture rate of 24.00 GTexel/s is proportionally much higher than the pixel rate of 3.000 GPixel/s. The 8:1 ratio of TMUs to ROPs indicates geometry and texture-heavy workloads will fare better than fill-rate-bound scenarios. The FP32 compute throughput of 384.0 GFLOPS, with FP16 reaching 768.0 GFLOPS at a 2:1 ratio, places this GPU firmly in the entry-level integrated segment. For context, this means the GPU can handle basic 3D rendering and light productivity tasks, but complex shader workloads will quickly saturate its execution resources.

Since the nearestRivals array is empty, direct percentage comparisons against specific competing products cannot be derived from the data. However, the 50th percentile standing implies that roughly half of all GPUs in the database outperform this part, while the other half trail behind it. This mid-pack positioning is unusual for an integrated GPU, which typically ranks much lower; it suggests the database's comparison pool includes many older or lower-end parts that this Intel solution can match or exceed. The absence of an average benchmark score reinforces that this GPU's real-world performance is highly dependent on the host system's memory configuration, as the memory bandwidth is listed as "System Dependent."

Power and Cooling

The Intel HD Graphics 620 Mobile carries a thermal design power of 15 W, a figure that encompasses the entire graphics subsystem's power envelope. This low TDP classifies it as an integrated graphics processor (IGP) with a slot width designation of "IGP," meaning it occupies no expansion slot and draws power directly from the motherboard's power delivery system. The power connectors field is null, which is consistent with an IGP that requires no auxiliary power cabling.

The suggested PSU field is also null, which is logical for an integrated solution that shares the system's existing power supply without needing additional capacity. A 15 W TDP is negligible in the context of a full desktop or laptop power budget; the CPU and other components will consume far more power. Consequently, no dedicated power supply recommendation is warranted, and the cooling solution can be minimal — a passive heatsink or a low-profile fan is sufficient to manage the thermal output. The 14 nm++ manufacturing process contributes to this efficiency, allowing the GPU to operate within its modest power envelope while delivering the clock speeds specified above.

The display outputs are "Portable Device Dependent," meaning the physical connectors vary by laptop or mobile device implementation. This is typical for mobile IGPs, where the display interface is routed through the motherboard rather than directly from the GPU. For system builders, this means no additional power planning is needed beyond ensuring the motherboard supports the integrated graphics output.

How It Compares

With an empty nearestRivals array, the data provides no direct competitor comparisons for the Intel HD Graphics 620 Mobile. This absence of rival data means specific percentage deltas cannot be calculated or cited. The benchmark results section also contains no entries, leaving the percentile rank as the sole comparative metric.

The 50th percentile ranking against all GPUs is the primary positioning signal. This indicates that, in a database spanning integrated, entry-level, mid-range, and high-end discrete GPUs, the HD Graphics 620 sits at the median. It outperforms the lower half of the GPU population, which likely includes older integrated graphics, low-end mobile GPUs, and legacy discrete cards. It trails the upper half, which encompasses modern discrete graphics solutions from both major vendors.

This mid-tier position is noteworthy because integrated graphics typically fall into the bottom quartile. The HD Graphics 620's ability to reach the median suggests that for basic computing tasks — office productivity, video playback, and lightweight gaming at low settings — it offers adequate performance. However, the lack of benchmark scores means this assessment is qualitative rather than quantitative, and actual performance will hinge on system memory speed and configuration. The "System Shared" memory type and "System Dependent" bandwidth underscore that the GPU's performance scales with the host's RAM capabilities, making it faster in systems with dual-channel high-speed memory and slower in single-channel configurations.

FAQ

Q: What is the manufacturing process for this GPU?

A: The Intel HD Graphics 620 Mobile is built on Intel's 14 nm++ process node, which is a refined version of the 14 nm manufacturing technology.

Q: What are the base and boost clock speeds?

A: The base clock is 300 MHz and the boost clock is 1000 MHz, providing a 3.33x range between idle and maximum performance states.

Q: How much VRAM does this GPU have?

A: The memory size is "System Shared," meaning the GPU uses a portion of the system's main RAM rather than dedicated video memory. The memory type and bus width are likewise "System Shared."

Q: What is the thermal design power of this processor?

A: The TDP is 15 W, which is typical for an integrated graphics solution and requires no dedicated power connectors or supplementary cooling.

Q: Which API versions does this GPU support?

A: It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3, covering modern graphics APIs for gaming and compute workloads.

Q: What is the production status of this product?

A: The production status is marked as "End-of-life," and the release date is 2016-08-29, indicating it has been discontinued from active manufacturing.

Ray Tracing and Feature Set

The Intel HD Graphics 620 Mobile does not include dedicated ray tracing cores, as the rtCores field is null. Similarly, it lacks tensor cores, with the tensorCores field also null. This places the GPU in the pre-DXR era of graphics hardware, where ray tracing was not yet a standard feature in consumer GPUs. The architecture, Generation 9.5, predates dedicated ray tracing acceleration, meaning any ray-traced workloads would fall back to compute shaders on the 192 shading units, which is inefficient for such algorithms.

The API support, however, is more forward-looking. DirectX 12 (12_1) provides feature level 12_1, which includes conservative rasterization, rasterizer-ordered views, and other advanced rendering techniques that do not require dedicated RT hardware. OpenGL 4.6 and Vulkan 1.3 support indicate compatibility with modern cross-platform graphics APIs, enabling the GPU to run titles that leverage these APIs. Vulkan 1.3 in particular offers improved multi-threading and lower driver overhead, which can help mitigate the GPU's limited compute resources.

The FP16 throughput of 768.0 GFLOPS, double the FP32 rate, suggests support for half-precision compute, which can accelerate certain workloads such as machine learning inference and image processing. However, without tensor cores, these workloads run on general-purpose shading units and will not achieve the performance of dedicated AI accelerators. The feature set is thus oriented toward compatibility and efficiency rather than high-end graphics effects, which is consistent with its integrated positioning and 15 W TDP.

Memory Subsystem

The memory subsystem of the Intel HD Graphics 620 Mobile is entirely system-dependent. The memory size is "System Shared," meaning the GPU dynamically allocates from the host's RAM, with no dedicated VRAM pool. The memory type is likewise "System Shared," which in practice means the GPU uses DDR3L or DDR4 memory from the system, depending on the platform. The bus width is also "System Shared," reflecting that the GPU accesses memory over the same bus as the CPU rather than a dedicated memory interface.

The bandwidth is listed as "System Dependent," which is the most critical factor for this GPU's performance. In a system with dual-channel high-speed memory, the available bandwidth could reach tens of gigabytes per second, but in a single-channel or low-speed configuration, bandwidth will be severely constrained. This dependency means that benchmark results can vary widely between systems, even with identical GPU clocks. For high-resolution workloads, this memory architecture is a significant bottleneck: at 1080p and above, the GPU must repeatedly fetch textures and geometry from system memory, and the shared bus competes with CPU traffic. The 3.000 GPixel/s pixel rate further limits high-resolution performance, as fill-rate requirements scale with resolution. For 4K gaming or rendering, the combination of shared memory bandwidth and limited ROPs makes playable frame rates unlikely. At lower resolutions like 720p, the memory subsystem's constraints are less pronounced, allowing the GPU to operate closer to its theoretical peak.

The 192 shading units and 24 TMUs can be kept reasonably busy at lower resolutions, but the memory bottleneck will dominate as resolution increases. The "System Dependent" bandwidth designation is a warning: this GPU's real-world performance is only as good as the memory it is paired with, making it a poor choice for users who cannot ensure fast dual-channel system memory.