NVIDIA Quadro RTX 3000 Mobile Refresh

The NVIDIA Quadro RTX 3000 Mobile Refresh is a professional mobile GPU built on the Turing architecture, fabricated on TSMC's 12 nm process and released in May 2019. Now marked as end-of-life, it sits at the 50th percentile of all GPUs in the benchmark database, with no benchmark scores recorded. This MXM module carries an 80 W TDP, 6 GB of GDDR6 memory on a 192-bit bus, and 1920 shading units, while supporting DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4.

Who Should Consider It

This card is designed for mobile workstations that require certified drivers and Turing-specific features without top-tier performance. Its 6 GB VRAM and 336.0 GB/s bandwidth are well suited for 1080p and 1440p professional workloads such as CAD, 3D modeling, and GPU-accelerated rendering. The 80 W TDP allows it to fit into laptops that balance performance with power efficiency, and the MXM form factor means it is a replaceable module in high-end portable systems. Because the card is end-of-life, prospective users will be looking at refurbished or existing machines, and its 50th percentile ranking indicates a mid-range position relative to all GPUs tracked. For tasks that demand more than 6 GB of memory, such as large simulation datasets or 4K texture work, this card may fall short, but for standard professional applications at 1080p and 1440p, it provides a capable foundation. The absence of a launch MSRP prevents any cost discussion, but the specification profile clearly targets the professional segment rather than consumer gaming.

Memory Subsystem

The Quadro RTX 3000 Mobile Refresh is equipped with 6 GB of GDDR6 memory on a 192-bit bus. The memory operates at 1750 MHz, yielding an effective data rate of 14 Gbps and a total bandwidth of 336.0 GB/s. This bandwidth is sufficient to feed the 1920 shading units at 1080p and 1440p resolutions, but it may become a bottleneck at 4K or when handling very large textures and datasets. The 6 GB capacity is modest by modern standards, and while it handles typical professional applications, users working with multi-GPU setups or extremely large scenes might find it limiting. The 192-bit bus width is a middle-ground configuration, more bandwidth than 128-bit parts, but less than 256-bit or 384-bit designs. In practice, the memory subsystem is balanced for the card's compute throughput: 5.299 TFLOPS FP32, 165.6 GTexel/s texture fill rate, and 88.32 GPixel/s pixel rate. The pixel rate indicates the card can drive high refresh rates at lower resolutions, but at 4K the pixel throughput will be a limiting factor, and the 336 GB/s bandwidth will be the first constraint when using high-resolution textures that exceed the 6 GB frame buffer.

Ray Tracing and Feature Set

This card includes 30 ray tracing cores and 240 tensor cores, which are part of the Turing architecture's hardware acceleration for ray-traced effects and AI-based features. The tensor cores enable AI-accelerated workflows such as DLSS, although specific DLSS support is not listed in the fact pack. The card supports DirectX 12 Ultimate (feature level 12_2), OpenGL 4.6, and Vulkan 1.4, ensuring compatibility with modern graphics APIs and features like variable rate shading, mesh shaders, and ray tracing when exposed by applications. The 30 RT cores provide hardware acceleration for ray-traced shadows, reflections, and global illumination, which is a key selling point for professional visualization and rendering tasks. The 240 tensor cores can accelerate machine learning inference and training, making the card suitable for AI-assisted content creation. The combination of RT and tensor cores distinguishes this card from older Pascal-generation Quadros, which lacked dedicated hardware for these features. The 12 nm process node and 10,800 million transistors on a 445 mm² die indicate a relatively large chip, typical for professional-grade parts. The power connector is listed as "None," suggesting it draws all power from the MXM slot, and its display outputs are portable-device dependent, meaning they vary by laptop model.

FAQ

Q: What is the memory capacity and type of the Quadro RTX 3000 Mobile Refresh?

A: It has 6 GB of GDDR6 memory on a 192-bit bus, with a bandwidth of 336.0 GB/s.

Q: What architecture is this GPU based on?

A: It is based on the Turing architecture, using the TU106 chip, and is fabricated on a 12 nm process at TSMC.

Q: Does it support ray tracing?

A: Yes, it includes 30 ray tracing cores, providing hardware acceleration for ray-traced effects.

Q: What is the TDP of this mobile GPU?

A: The TDP is 80 W, and it is designed as an MXM module.

Q: What is the production status?

A: It is marked as end-of-life, with a release date of May 26, 2019, and a successor in the Ampere-MW line.

Q: What is the card's performance percentile?

A: It sits at the 50th percentile of all GPUs in the benchmark database.

Benchmark Performance

The fact pack for this GPU lists no benchmark scores, and its average benchmark score is zero. This indicates that no standardized performance measurements are available in the database for this specific card. However, we can derive theoretical performance from its specifications. The card delivers 5.299 TFLOPS of FP32 compute, 10.60 TFLOPS of FP16 (with a 2:1 ratio), and texture and pixel rates of 165.6 GTexel/s and 88.32 GPixel/s respectively. These figures place it in the mid-range of professional mobile GPUs. Its 50th percentile ranking confirms that it sits exactly in the middle of the performance distribution of all GPUs tracked by the database. Without rival scores, we cannot compute exact percentage deltas, but the theoretical throughput suggests it is suitable for 1080p and 1440p gaming and professional workloads, though not for extreme 4K or high-refresh-rate scenarios. The 1920 shading units, 120 TMUs, and 64 ROPs are typical for a TU106-based part. The memory bandwidth of 336.0 GB/s is a crucial factor for performance at higher resolutions; at 4K, the pixel rate of 88.32 GPixel/s becomes a bottleneck, as the card must process a large number of pixels per second, and with heavy shading and memory traffic, the card may struggle. The FP32 throughput of 5.299 TFLOPS is modest compared to high-end desktop cards, but for a mobile 80 W part, it represents a reasonable balance. The card's performance is further influenced by its memory subsystem; the 192-bit bus and 336 GB/s bandwidth are sufficient for the compute capability, but not overprovisioned. In terms of feature set, the 30 RT cores and 240 tensor cores provide dedicated hardware for ray tracing and AI, which can offload work from the shader units. The card's support for DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 ensures compatibility with modern APIs. Because there are no benchmark entries, we cannot provide a comparative analysis against specific rivals. The 50th percentile ranking is the only quantitative performance indicator available. That ranking implies that the card outperforms half of all GPUs in the database and underperforms the other half, which is a neutral position. For users seeking a professional mobile GPU with Turing features, this card offers a balanced set of specifications, but its end-of-life status and lack of benchmark data make it a less attractive option compared to newer parts. The absence of a launch MSRP prevents any cost analysis, but the specifications alone indicate a mid-tier professional product.