NVIDIA Quadro FX 380M

The NVIDIA Quadro FX 380M is a legacy mobile workstation graphics solution built on the Tesla 2.0 architecture, utilizing the GT218S chip manufactured on a 40 nm process at TSMC. With a transistor count of 260 million on a 57 mm² die, this part occupies a distinct, low-power niche in the mobile professional GPU landscape. Its benchmark percentile ranking places it at the 50th percentile of all GPUs, indicating a baseline level of performance that is neither entry-level nor competitive with modern parts. The data shows a product designed for a specific era of mobile computing, where power efficiency and basic professional features took precedence over raw computational throughput.

Who Should Consider It

Given its performance profile, the Quadro FX 380M is strictly for legacy system repairs or for running period-specific software that requires a certified mobile workstation GPU. The FP32 compute throughput is 46.98 GFLOPS, which is a figure that modern integrated graphics far surpass, indicating that this card is not suitable for any contemporary 3D rendering, CAD, or video editing workload. The pixel rate of 2.424 GPixel/s and texture rate of 4.848 GTexel/s further cement its status as a part that will struggle with any resolution above the most basic 1024x768 or 1366x768 panels of its time. At 720p with low settings, it might handle very old games (pre-2008) or 2D applications, but benchmark results indicate that any attempt at 1080p gaming or modern GPU-accelerated tasks will result in unplayable frame rates.

The 512 MB VRAM capacity is a hard limitation for texture-heavy applications; any modern game or professional application requiring more than this will fail to load or run with severe stuttering. The target user is someone who needs a drop-in replacement for a failing GPU in an older Dell Precision or Lenovo ThinkPad workstation, not someone building a new system. For professional tasks, the 50th percentile ranking against all GPUs means it sits in the absolute middle of the historical performance curve, which in practical terms translates to being effectively unusable for any current software that relies on GPU acceleration. The data suggests this is a part for preservation or basic 2D productivity, not for any task that leverages the GPU's parallel processing capabilities.

Power and Cooling

The power profile of the Quadro FX 380M is exceptionally modest, with a TDP of 25 W. This low figure means that a dedicated cooling solution is not required; a simple heat pipe and small fan, or even a passive heatsink in a well-ventilated chassis, will suffice. The power connectors are listed as "None," which means the card draws all its power directly from the PCIe 2.0 x16 slot, eliminating the need for any auxiliary power cables from the system power supply. Consequently, there is no suggested PSU wattage provided in the data, as the card's draw is negligible within any system that has a functional PCIe slot.

This low power draw is a direct consequence of the 40 nm process node and the limited 16 shading units operating at the given clocks. For system integrators or technicians, this simplifies installation considerably: no PSU upgrade is required, and thermal management is a non-issue even in thin and light mobile chassis. The data shows that the card's thermal design is so conservative that it can be considered a near-drop-in replacement for any GPU with a similar or higher TDP, provided the cooling solution is physically compatible. The absence of a slot width specification and the "Portable Device Dependent" display output field indicate that this is a mobile-only part, soldered or mounted via a proprietary MXM-style interface, so end-user cooling modifications are not applicable.

Memory Subsystem

The memory subsystem is a critical bottleneck for this GPU. It features 512 MB of GDDR3 memory on a 64-bit bus, yielding a total memory bandwidth of 12.64 GB/s. This combination of small capacity and narrow bus width is the primary reason the card cannot handle high resolutions. The memory clock runs at 790 MHz, translating to 1580 Mbps effective, which was standard for the era but is now several orders of magnitude slower than modern GDDR6 or HBM solutions. For high-resolution textures, the 512 MB capacity is instantly exhausted, forcing the system to swap data to system RAM over the PCIe bus, which causes severe frame pacing issues and stutters.

The 64-bit bus width is particularly limiting; it means that even if the memory clock were higher, the data throughput would still be capped by the narrow interface. In practical terms, this bandwidth is insufficient for 1080p gaming, where even a 2010-era game would require textures to be streamed from disk constantly. The 12.64 GB/s figure is roughly an order of magnitude lower than what is needed for smooth 1080p operation in modern titles. For professional applications like 3D modeling, the limited VRAM prevents loading complex scenes, and the low bandwidth slows down any GPU-accelerated computation that requires frequent data transfers. Benchmark results indicate that this memory configuration is the single most limiting factor for any task beyond basic 2D desktop rendering.

How It Compares

The FACT PACK provides no nearestRivals data for this GPU, meaning there are no comparative scores or deltaPct values available in the database. As such, a quantitative comparison against specific rival products is not possible from the given facts. The only positional data point is the 50th percentile ranking vs all GPUs, which places it at the exact median of the historical performance distribution.

Without rival data, the analysis must rely on the absolute performance figures. The 46.98 GFLOPS FP32 throughput is a baseline that indicates the card is over 20 years behind modern entry-level discrete GPUs, which typically offer several teraflops. The architecture, Tesla 2.0, is a predecessor to Fermi, and the chip is a derivative of the consumer GeForce 210 series, which was the lowest tier of its generation. The 16 shading units and 4 ROPs are minimal counts that confirm its position at the bottom of the professional stack. The data shows that any modern iGPU with a similar TDP will outperform this card by a significant margin, but those figures are not in the FACT PACK and cannot be cited. The conclusion is that this GPU has no competitive standing in the current market, and its only comparison is against its own historical baseline.

Ray Tracing and Feature Set

The Quadro FX 380M has no dedicated RT cores or Tensor cores, as these were introduced in later architectures. The data lists these fields as null, confirming that hardware-accelerated ray tracing and AI-based tensor operations are entirely absent. The feature set is defined by its API support: DirectX 11.1 (10_1) and OpenGL 3.3. Vulkan support is null, meaning the card cannot run any Vulkan-based titles or applications. The DirectX 11.1 (10_1) designation is notable; it indicates that while the card supports the DirectX 11.1 API, it is limited to feature level 10_1, which means it cannot execute DirectX 11 shader models and is effectively a DirectX 10.1 class part.

This API limitation is critical for software compatibility. Any game or application that requires DirectX 11 feature level 11_0 will not run. The OpenGL 3.3 support is similarly dated, restricting compatibility with modern CAD and content creation tools that now require OpenGL 4.6 or higher. The display outputs are "Portable Device Dependent," meaning the card relies on the laptop's internal LVDS or eDP connection and does not offer external display outputs directly from the GPU. The bus interface is PCIe 2.0 x16, which is electrically backward and forward compatible, but the card's low bandwidth requirements mean it will not saturate even a PCIe 2.0 x1 link. The data shows a feature set that is wholly obsolete for modern workloads, with no acceleration for ray tracing, no tensor operations, and API support that caps out at a decade-old standard.