NVIDIA GeForce RTX 2080 SUPER Max-Q

NVIDIA GeForce RTX 2080 SUPER Max-Q is a mobile Turing-architecture part built on a 12 nm process at TSMC, featuring 13,600 million transistors on a 545 mm² die. It belongs to the GeForce 20 Mobile generation, released in April 2020, and is now end-of-life. With a 50th percentile ranking among all GPUs, this chip sits squarely in the middle of the performance distribution, though its specific benchmark scores and nearest rival data are not available in the current record. The analysis below focuses on its architectural capabilities, memory subsystem, and power characteristics based solely on the provided facts.

Benchmark Performance

The FACT PACK lists no benchmark scores, no average benchmark score (0), and no nearest rivals with delta percentages. Consequently, there is no quantitative performance data to interpret against competing GPUs. The only anchor is the percentileVsAllGpus field, which places this GPU at the 50th percentile. This means that, in a hypothetical distribution of all GPUs, half would be slower and half would be faster, but without actual score values or rival comparisons, no specific percentage deltas can be stated.

What the data does indicate is raw compute throughput. The shading units (3072) combined with a boost clock of 975 MHz yield an FP32 performance of 5.990 TFLOPS. This is a moderate figure for a mobile part from the 20-series, reflecting the Max-Q design's emphasis on efficiency over peak clocks. The FP16 rate doubles to 11.98 TFLOPS (2:1 ratio), which is useful for workloads that leverage mixed-precision arithmetic, though the absence of benchmark scores prevents a direct statement on how this translates into real-world gaming or productivity results.

The pixel rate is 62.40 GPixel/s, derived from 64 ROPs at the boost clock. Texture rate stands at 187.2 GTexel/s, powered by 192 TMUs. These figures suggest a GPU that can handle 1440p gaming comfortably, but without scores or rival data, any claim about relative performance would be speculative. The benchmark section must remain descriptive rather than comparative, noting that the silicon's theoretical limits are documented but its measured performance is not recorded in the database.

Ray Tracing and Feature Set

This GPU includes dedicated ray tracing hardware. The FACT PACK specifies 48 RT cores, which are the Turing-generation units responsible for accelerating ray-traced lighting and shadow effects. Additionally, there are 384 tensor cores, which handle AI-based tasks such as DLSS (Deep Learning Super Sampling) and other neural network operations. The inclusion of both core types marks this as a feature-complete Turing product.

API support is broad and current. DirectX 12 Ultimate (12_2) is listed, which includes hardware support for features like variable rate shading and mesh shaders, though the exact feature level is not broken down further. OpenGL 4.6 and Vulkan 1.4 are both supported, giving compatibility with modern titles and engines across multiple graphics APIs. The data does not mention specific driver-level ray tracing performance, only the presence of the RT cores and API conformance.

The tensor cores enable AI-accelerated features, but the FACT PACK does not specify which applications or games leverage them beyond the implicit DLSS capability. The RT core count of 48 is notable for a mobile chip, but without benchmark data, it is impossible to state how it compares to desktop or rival mobile GPUs in ray tracing workloads. The feature set is technically complete, but measured outcomes are absent.

How It Comprises

The nearestRivals array is empty, meaning there are no direct competitor comparisons available in the FACT PACK. As such, no paragraphs can be written about specific rival GPUs, their scores, or delta percentages. The instruction to compare against rivals cannot be fulfilled because no rival names or data are present.

The only positional data is the 50th percentile ranking. This places the RTX 2080 SUPER Max-Q in the middle of the entire GPU landscape, which is a reasonable position for a premium mobile part from 2020, but it is not a substitute for direct rival analysis. Without rival names, scores, or deltas, this section must state the absence of data rather than fabricate comparisons. The GPU's position is thus only known in aggregate terms, not relative to specific products.

FAQ

Q: What is the FP32 compute performance of this GPU?

A: The FP32 performance is 5.990 TFLOPS, based on 3072 shading units at a boost clock of 975 MHz.

Q: Does this GPU support hardware ray tracing?

A: Yes, it includes 48 RT cores dedicated to ray tracing acceleration, along with 384 tensor cores for AI tasks.

Q: Which APIs are supported?

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

Q: How much VRAM does it have, and what type?

A: It has 8 GB of GDDR6 memory on a 256-bit bus, providing a bandwidth of 352.0 GB/s.

Q: What is the memory clock speed?

A: The memory clock is 1375 MHz, which translates to 11 Gbps effective data rate.

Q: What is the thermal design power (TDP)?

A: The TDP is 80 W, and the GPU uses an MXM module form factor with no power connectors listed.

Memory Subsystem

The memory configuration consists of 8 GB of GDDR6 VRAM, connected via a 256-bit bus. The memory clock is 1375 MHz, yielding an effective data rate of 11 Gbps and a total bandwidth of 352.0 GB/s. This bandwidth is a critical factor for high-resolution gaming, as modern textures and frame buffers at 4K resolutions demand substantial data throughput.

For a mobile GPU, 8 GB is a solid amount for 1080p and 1440p gaming, and it remains sufficient for most 4K titles at medium-to-high settings, though the FACT PACK does not provide specific resolution performance data. The 256-bit bus width is wider than many mobile parts, which typically use 128-bit or 192-bit interfaces, and this contributes directly to the 352.0 GB/s figure. The effective 11 Gbps memory speed is standard for GDDR6 of this era.

The pixel rate of 62.40 GPixel/s and texture rate of 187.2 GTexel/s are tied to the memory subsystem's ability to feed the ROPs and TMUs. At 4K, the pixel rate becomes a bottleneck only if the GPU's compute is underutilized, but the memory bandwidth is sufficient to avoid starvation in most scenarios. Without benchmark scores, it is not possible to state a specific performance ceiling, but the combination of 8 GB and 352.0 GB/s places this GPU in a capable position for high-resolution workloads.

Power and Cooling

The thermal design power (TDP) is 80 W, which is a modest figure for the compute capability on offer. This is characteristic of the Max-Q design philosophy, which prioritizes power efficiency and thermal management over raw clock speeds. The base clock is 735 MHz, and the boost clock is 975 MHz, both relatively low for a 20-series part, but this is necessary to keep power draw within the 80 W envelope.

The GPU uses an MXM Module slot width, which is a standardized form factor for laptop graphics. Notably, the power connectors field is listed as "None," indicating that this module draws power directly from the laptop's motherboard rather than requiring external PCIe power cables. There is no suggested PSU rating provided, which is typical for mobile components where the system's power delivery is pre-engineered.

Cooling requirements are not specified beyond the TDP, but an 80 W part is generally manageable with a capable air cooler in a laptop chassis. The absence of a suggested PSU is expected, as desktop power supply recommendations do not apply to mobile MXM modules. The low TDP also means that thinner and lighter laptops can accommodate this GPU, though the FACT PACK does not provide dimensional or weight data to confirm this.