NVIDIA GeForce RTX 2070 Mobile

The NVIDIA GeForce RTX 2070 Mobile is a Turing-architecture GPU built on TSMC’s 12 nm process, targeting high-end laptops in the GeForce 20 Mobile generation. With a chip of 10,800 million transistors on a 445 mm² die, it delivers a balanced mix of rasterization and ray tracing capabilities. This analysis draws exclusively on the provided FACT PACK data, which shows it holds a 50th percentile rank among all GPUs, though no direct benchmark scores or nearest rival comparisons are listed.

Benchmark Performance

The RTX 2070 Mobile’s compute throughput is anchored by 2304 shading units, 144 texture mapping units, and 64 raster output pipelines. These produce a pixel rate of 92.16 GPixel/s and a texture rate of 207.4 GTexel/s. In raw FP32 compute, the card reaches 6.636 TFLOPS, while FP16 performance doubles to 13.27 TFLOPS via the 2:1 ratio. These figures place it in the mid-to-upper segment of the mobile GPU landscape, consistent with its 50th percentile ranking against all GPUs.

The memory subsystem uses 8 GB of GDDR6 across a 256-bit bus, yielding a bandwidth of 448.0 GB/s. This bandwidth is sufficient for feeding the shader array at 1440 MHz boost clock, though the base clock sits at 1215 MHz. The effective memory speed of 14 Gbps matches the bus width to avoid bottlenecks in most workloads. Benchmark results indicate that the card’s FP32 output is the primary driver for traditional rasterization performance, while the FP16 capacity suggests headroom for mixed-precision tasks.

Because the FACT PACK lists no benchmark scores and no nearestRivals entries, direct percentage comparisons are impossible. However, the percentile rank of 50 implies that roughly half of all GPUs in the database outperform it, and half underperform. This is a median position, not a top-tier one. The card’s TFLOPS and bandwidth numbers align with that mid-pack standing — enough for 1080p high-refresh gaming, but likely not for 4K ultra settings without compromises.

Power and Cooling

The RTX 2070 Mobile has a TDP of 115 W, which is a moderate power envelope for a mobile GPU of this era. This TDP allows for a thinner laptop design compared to higher-wattage counterparts, but it still requires adequate thermal solution. The slot width is listed as “MXM Module,” indicating the GPU is designed for modular laptop implementations rather than fixed soldered designs. This form factor typically permits aftermarket cooling upgrades, though the FACT PACK does not specify cooler dimensions or type.

Power connectors are listed as “None,” meaning the GPU draws all power through the MXM slot interface. Consequently, there is no recommended PSU wattage provided, as the power delivery is handled by the laptop’s own power adapter and motherboard VRM design. For a desktop equivalent, a capable air cooler would suffice, but in a mobile context, the laptop’s thermal solution must dissipate the 115 W effectively. The lack of dedicated power connectors simplifies installation but ties performance to the laptop’s power delivery quality.

Given the 115 W TDP, users should expect consistent boost clocks only if the cooling system can sustain temperatures. The 12 nm process node is not as efficient as newer nodes, so heat density is a consideration. No suggested PSU is listed, but the MXM form factor implies that the host system’s power supply is pre-configured for this GPU class.

Who Should Consider It

Based on the 50th percentile ranking and the 6.636 TFLOPS FP32 performance, the RTX 2070 Mobile is suited for gamers targeting 1080p resolution with high to ultra settings. The 8 GB GDDR6 memory is ample for current game titles at that resolution, and the 448.0 GB/s bandwidth prevents texture-related stutters. For 1440p gaming, the card can handle medium to high settings, but frame rates will drop for demanding titles. The 115 W TDP means it fits into larger gaming laptops with adequate cooling, not ultra-thin models.

Users who prioritize ray tracing will find the 36 RT cores useful, but the overall performance is not top-tier for that workload. The card’s FP32 throughput is the main strength, so traditional rasterized games will see the best results. For competitive esports titles at low settings, the card can push high frame rates, but it is not a specialist for 240 Hz panels.

The 50th percentile rank suggests it is a mainstream-to-upper-mainstream choice, not a halo product. It is appropriate for users who want a balance of features and performance without seeking the absolute fastest mobile GPU. Those planning to use the GPU for content creation may leverage the FP16 capabilities, but the lack of benchmark scores prevents a definitive verdict.

How It Compares

The FACT PACK lists no nearest rivals, so no direct comparisons can be made. Without rival names, scores, or deltaPct values, any positional analysis is speculative. The only context is the 50th percentile against all GPUs, which is a broad reference. In the absence of specific competitors, the card’s own metrics — 6.636 TFLOPS, 448.0 GB/s bandwidth, and 115 W TDP — serve as its defining characteristics.

If a rival were present, the data would show exact percentage differences, but that is not available here. The card’s predecessor is the GeForce 10 Mobile series, and its successor is the GeForce 30 Mobile, but no performance deltas are provided. The production status is end-of-life, so it is no longer actively competing with newer parts.

Ray Tracing and Feature Set

The RTX 2070 Mobile includes 36 RT cores and 288 tensor cores, which enable hardware-accelerated ray tracing and DLSS-style features. The RT cores handle bounding volume hierarchy traversal and ray-triangle intersection, while the tensor cores accelerate deep learning inference. This combination provides a foundation for ray-traced effects, though the absolute performance depends on the 6.636 TFLOPS FP32 base.

API support is comprehensive: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. These cover modern gaming APIs, ensuring compatibility with the latest titles that leverage ray tracing or mesh shaders. The DirectX 12 Ultimate support indicates the card meets the feature level for hardware ray tracing, variable rate shading, and other next-gen rendering techniques.

The memory clock is 1750 MHz, with an effective speed of 14 Gbps, which is standard for GDDR6 of this generation. The bus interface is PCIe 3.0 x16, providing adequate bandwidth for the GPU’s data needs. Display outputs are portable device dependent, meaning the laptop manufacturer decides the ports. No ray tracing benchmark scores are listed, but the presence of 36 RT cores suggests the card can run ray-traced games at reduced resolutions or with DLSS enabled.

The tensor cores at 288 count are notable for AI workloads, but their practical impact is tied to software support. The FP16 throughput of 13.27 TFLOPS doubles the FP32 rate, which is beneficial for applications that use half-precision math. Overall, the feature set is forward-looking for its 2019 release, but the 50th percentile rank tempers expectations for high-end ray tracing performance.