NVIDIA GeForce RTX 2070 Max-Q Refresh

The NVIDIA GeForce RTX 2070 Max-Q Refresh is a mobile Turing architecture GPU built on TSMC’s 12 nm process, with a die size of 445 mm² containing 10,800 million transistors. It is positioned as an end-of-life product in the GeForce 20 Mobile generation, succeeding the GeForce 10 Mobile line and preceding the GeForce 30 Mobile series. The chip, designated TU106B, operates with a base clock of 900 MHz and a boost clock of 1125 MHz, delivering a peak FP32 throughput of 5.184 TFLOPS. This data sheet shows no benchmark scores or nearest rivals are listed, meaning the analysis must rely on the core architectural specifications and feature set rather than comparative performance metrics. The GPU holds a 50th percentile ranking among all GPUs, indicating a mid-pack status in the broader hardware landscape, though this figure is not tied to any specific score.

Benchmark Performance

The absence of benchmark scores and nearest rival data in the fact pack means that performance interpretation must be derived from raw computational specifications. The FP32 throughput of 5.184 TFLOPS, combined with a texture rate of 162.0 GTexel/s and a pixel rate of 72.00 GPixel/s, suggests a GPU designed for balanced 1080p and 1440p gaming workloads. The shading units number 2304, with 144 texture mapping units and 64 render output units, a configuration that historically supports high-detail rasterization without extreme resolution demands. The boost clock of 1125 MHz is modest by desktop standards, but this is a Max-Q variant prioritizing thermal efficiency within a 115 W TDP envelope.

Given the percentile rank of 50, the data implies this GPU sits exactly at the median of all GPUs tracked, meaning half of the hardware population is faster and half is slower. For a mobile part, this position suggests it can handle esports titles and older AAA games at high settings, but modern demanding releases may require medium presets to maintain smooth frame rates. The FP16 performance of 10.37 TFLOPS (2:1 ratio) indicates a doubling of throughput for half-precision workloads, which can benefit certain compute tasks but has limited direct impact on gaming frame rates. The lack of rival comparisons prevents delta percentage calculations, so the analysis must emphasize the absolute numbers and their implications for typical usage scenarios.

Ray Tracing and Feature Set

The RTX 2070 Max-Q Refresh includes 36 ray tracing cores and 288 tensor cores, marking it as part of NVIDIA’s first generation of real-time ray tracing hardware. The presence of these dedicated cores enables hardware-accelerated ray tracing effects in supported games, though the performance headroom on a Max-Q part with a 1125 MHz boost clock is limited. The tensor cores provide AI-accelerated features such as DLSS, which can offset the performance cost of ray tracing by rendering at lower resolutions and upscaling with neural networks. The API support list includes DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, indicating full compliance with modern graphics standards and the ability to run current-generation titles that leverage these interfaces.

The DirectX 12 Ultimate designation confirms support for features like variable rate shading and mesh shaders, which are becoming standard in recent game engines. Vulkan 1.4 support offers cross-platform compatibility and potential performance improvements in Vulkan-based titles. The ray tracing cores, while present, are first-generation technology; the data suggests that enabling ray tracing at high resolutions will likely require significant compromises in other settings. The tensor cores, however, provide a counterbalancing capability through DLSS, which can help maintain playable frame rates while keeping visual quality acceptable. For users interested in ray tracing, the benchmark data indicates this GPU can run such effects, but the experience will be better suited to 1080p with DLSS enabled rather than native 4K.

Memory Subsystem

The memory subsystem consists of 8 GB of GDDR6 memory on a 256-bit bus, operating at 1375 MHz with 11 Gbps effective speed. This configuration yields a bandwidth of 352.0 GB/s, which is substantial for a mobile GPU of this era. The 256-bit bus width is a key advantage, allowing higher bandwidth per clock cycle compared to narrower interfaces. For 1080p gaming, 8 GB of VRAM is generally sufficient for current titles at high textures, though some demanding games with ultra texture packs may approach this limit. At 1440p, the capacity becomes more critical, and the 352.0 GB/s bandwidth helps maintain consistent frame pacing when streaming large textures.

The memory clock of 1375 MHz, while not extreme, is paired with the 256-bit interface to produce a bandwidth figure that supports high-resolution rendering without bottlenecking the GPU cores. The data suggests this memory configuration is well-matched to the 5.184 TFLOPS compute capability, avoiding a situation where the GPU starves for data. For future titles, the 8 GB capacity may become a limiting factor as texture sizes grow, but for the GPU’s intended lifespan, it provides adequate headroom. The effective bandwidth of 352.0 GB/s enables smooth performance in 1440p gaming, though the compute throughput may cap frame rates before memory bandwidth becomes a constraint in most scenarios.

FAQ

Q: What is the transistor count and die size of the RTX 2070 Max-Q Refresh?

A: The GPU contains 10,800 million transistors on a 445 mm² die, fabricated on TSMC’s 12 nm process.

Q: How many ray tracing and tensor cores does this GPU have?

A: It includes 36 ray tracing cores and 288 tensor cores, enabling hardware-accelerated ray tracing and AI-based features like DLSS.

Q: What is the memory bandwidth and bus width?

A: The 8 GB GDDR6 memory is connected via a 256-bit bus, providing 352.0 GB/s of bandwidth at an effective speed of 11 Gbps.

Q: Which APIs are supported?

A: The GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, covering modern graphics standards.

Q: What is the TDP and power connector requirement?

A: The TDP is 115 W, and the power connector is listed as “None,” meaning it draws power through the MXM module interface.

Q: When was this product released and what is its production status?

A: It was released on 2020-03-03 and is now end-of-life, with the GeForce 30 Mobile series as its successor.

How It Compares

The fact pack lists no nearest rivals, so comparisons must be made against the broader GPU landscape using the percentile rank. At the 50th percentile, this GPU sits exactly at the median of all GPUs, suggesting it outperforms lower-tier integrated graphics and older discrete cards but lags behind high-end desktop and newer mobile parts. Within the GeForce 20 Mobile lineup, the Max-Q Refresh variant is positioned below the full-power RTX 2070, trading clock speed for lower power consumption. The 115 W TDP is a defining characteristic, differentiating it from non-Max-Q versions that typically consume more power and achieve higher boost clocks.

Against the predecessor GeForce 10 Mobile series, the architecture shift to Turing brings ray tracing and tensor cores, which were absent in the older generation. The successor GeForce 30 Mobile series likely offers improved performance and efficiency, but without specific data, the analysis can only note the generational progression. The 50th percentile ranking implies that this GPU is not a high-end performer by contemporary standards, but it remains capable for its intended mobile form factor. Users upgrading from a GTX 10-series laptop will gain access to ray tracing and DLSS, while those expecting desktop-level performance should look to higher-tier GPUs.

Who Should Consider It

Given the 5.184 TFLOPS FP32 performance and 8 GB VRAM, this GPU is best suited for gamers targeting 1080p resolution with high settings in most titles. The data shows it can handle 1440p gaming, but users should expect to adjust settings to medium or high rather than ultra in demanding AAA games. The presence of 36 ray tracing cores suggests that ray tracing is possible, but at 1080p with DLSS enabled to maintain playable frame rates. Users who prioritize battery life and thermals in a slim laptop chassis will appreciate the 115 W TDP, which is lower than full-power variants.

For content creators, the 10.37 TFLOPS FP16 performance and 288 tensor cores offer acceleration for AI-based workflows and certain compute tasks. The 352.0 GB/s bandwidth supports video editing and 3D rendering workloads that benefit from high memory throughput. However, the end-of-life status means software optimizations for newer titles may be limited over time. The GPU is not recommended for 4K gaming, as the 8 GB VRAM and compute throughput will likely struggle to maintain high frame rates at that resolution. It fits a specific niche: a portable laptop GPU capable of solid 1080p gaming and entry-level ray tracing, with enough memory and bandwidth for moderate creative work.

Power and Cooling

The TDP is rated at 115 W, which is a moderate power draw for a mobile GPU, allowing for thinner laptop designs with adequate cooling solutions. The power connector is listed as “None,” indicating that the GPU draws power directly through the MXM module slot rather than requiring external PCIe power cables. This design simplifies integration into laptops but also means cooling and power delivery are entirely dependent on the laptop’s internal design. The slot width is specified as MXM Module, confirming the form factor for replaceable mobile graphics modules.

With a 115 W TDP, the thermal solution must dissipate this heat effectively to maintain the boost clock of 1125 MHz. The base clock of 900 MHz provides a lower performance floor when thermals are constrained. The data suggests that sustained gaming sessions will cause the GPU to operate near its TDP limit, and laptops with adequate cooling will sustain boost clocks better than those with weaker thermal designs. The lack of a suggested PSU rating is irrelevant for a mobile part, as power is drawn from the laptop’s battery and adapter. Users should expect that heavy workloads will drain battery life quickly, and the GPU’s performance is contingent on the laptop’s ability to manage heat within the chassis.