NVIDIA GeForce RTX 2080 SUPER vs NVIDIA P106-100 Comparison

The NVIDIA GeForce RTX 2080 SUPER and the NVIDIA P106-100 occupy vastly different segments of the GPU landscape, despite sharing a manufacturer. The RTX 2080 SUPER is a fully-featured Turing architecture consumer card, while the P106-100 is a Pascal-based mining-oriented product with no display outputs. Benchmark data shows a decisive performance gap, with the RTX 2080 SUPER winning all three head-to-head tests by significant margins, yet the P106-100 holds a surprisingly close aggregate score due to its own niche positioning.

Where Each One Wins

The RTX 2080 SUPER wins every single benchmark in the head-to-head comparison. In 3DMark Steel Nomad (DX12), it scores 1882 against the P106-100’s 899, a delta of 109.3%. This indicates that for modern, DirectX 12 workloads typical of gaming and 3D rendering, the RTX 2080 SUPER is more than twice as fast. The compute-oriented tests show an even larger gap: in Geekbench OpenCL, the RTX 2080 SUPER scores 99226 versus 35951, a 176% lead, and in Geekbench Vulkan, it scores 111284 versus 32897, a 238.3% lead. This suggests the Turing card’s architectural advantages scale better with parallel compute and API-specific optimizations.

The P106-100, however, does not win any of the three compared benchmarks. Its only competitive advantage lies in its aggregate benchmark percentile: it sits at the 68th percentile of all GPUs, compared to the RTX 2080 SUPER’s 69th percentile. This near-identical percentile ranking, despite a 3.9% difference in average benchmark scores (23249 vs 24170), implies that the P106-100 is positioned similarly in the overall performance distribution. For a mining card with no display outputs, its strength is not in any specific workload but in its raw compute efficiency relative to its thermal and power constraints.

Architecture Differences

The architectural gap between these two GPUs is generational and fundamental. The RTX 2080 SUPER uses the TU104 chip on TSMC’s 12 nm process, packing 13,600 million transistors into a 545 mm² die with a transistor density of 25.0M per mm². The P106-100 uses the GP106 chip on TSMC’s 16 nm process, with 4,400 million transistors on a 200 mm² die, yielding a density of 22.0M per mm². This means the RTX 2080 SUPER has roughly three times the transistor count and a 345 mm² larger die, directly enabling its higher compute throughput.

The RTX 2080 SUPER features 3072 shading units, 192 TMUs, and 64 ROPs, along with 48 RT cores and 384 tensor cores—hardware that the P106-100 lacks entirely. The Pascal chip has 1280 shading units, 80 TMUs, and 48 ROPs. Clock speeds favor the newer card: the RTX 2080 SUPER boosts to 1815 MHz versus 1709 MHz for the P106-100, with base clocks of 1650 MHz and 1506 MHz respectively. Memory configurations are also starkly different: the RTX 2080 SUPER uses 8 GB of GDDR6 on a 256-bit bus, delivering 495.9 GB/s of bandwidth, while the P106-100 uses 6 GB of GDDR5 on a 192-bit bus, providing only 192.2 GB/s.

Compute output reflects these differences. The RTX 2080 SUPER delivers 11.15 TFLOPS of FP32 performance and 22.30 TFLOPS of FP16 (2:1 ratio), whereas the P106-100 manages 4.375 TFLOPS FP32 and a negligible 68.36 GFLOPS FP16 (1:64 ratio). Pixel and texture rates follow suit: 116.2 GPixel/s and 348.5 GTexel/s for the RTX 2080 SUPER versus 82.03 GPixel/s and 136.7 GTexel/s for the P106-100. The API support also differs, with the RTX 2080 SUPER supporting DirectX 12 Ultimate (12_2) and the P106-100 limited to DirectX 12 (12_1).

The Verdict

Based strictly on benchmark data, the RTX 2080 SUPER is the superior GPU for any workload that requires graphics output or modern rendering features. Its 109.3% lead in 3DMark Steel Nomad and 238.3% lead in Geekbench Vulkan are decisive for gaming, content creation, and general compute tasks. The presence of RT and tensor cores, combined with DirectX 12 Ultimate support, makes it future-proof for ray-traced applications, while the P106-100’s lack of display outputs means it cannot be used as a primary graphics card at all.

The P106-100, however, has a rationale for existence. Its 68th percentile ranking versus the RTX 2080 SUPER’s 69th shows that, in aggregate, it performs nearly as well relative to the entire GPU market. With a 120 W TDP and a single 6-pin power connector, it is far more power-efficient in terms of performance per watt—though the data does not directly measure this ratio. For a mining or compute-only scenario where display output is irrelevant, the P106-100 could be a lower-cost alternative, but its 4.375 TFLOPS FP32 and 192.2 GB/s bandwidth limit it to less demanding workloads. The RTX 2080 SUPER is the clear choice for anyone needing a complete, high-performance solution.

FAQ

Q: Which GPU has higher raw compute performance?

A: The RTX 2080 SUPER achieves 11.15 TFLOPS FP32 and 22.30 TFLOPS FP16, compared to the P106-100’s 4.375 TFLOPS FP32 and 68.36 GFLOPS FP16, making the RTX 2080 SUPER roughly 2.5 times faster in FP32.

Q: Can the P106-100 be used for gaming?

A: No, because it has no display outputs, making it impossible to connect a monitor. Its DirectX 12 (12_1) support is also a generation behind the RTX 2080 SUPER’s 12 Ultimate (12_2).

Q: How do their memory subsystems compare?

A: The RTX 2080 SUPER has 8 GB of GDDR6 on a 256-bit bus with 495.9 GB/s bandwidth, while the P106-100 has 6 GB of GDDR5 on a 192-bit bus with 192.2 GB/s, a 2.6x bandwidth advantage for the RTX card.

Q: What is the performance gap in Vulkan?

A: In Geekbench Vulkan, the RTX 2080 SUPER scores 111284 versus 32897 for the P106-100, a 238.3% lead, indicating the Turing architecture handles Vulkan far more efficiently.

Q: Are their aggregate scores close?

A: Yes, the RTX 2080 SUPER has an average benchmark score of 24170 (69th percentile), while the P106-100 scores 23249 (68th percentile), a 3.9% difference despite the individual test gaps.

Q: Which card has better DirectX 12 support?

A: The RTX 2080 SUPER supports DirectX 12 Ultimate (12_2), while the P106-100 only reaches DirectX 12 (12_1), meaning the RTX card can handle newer DX12 features.

Head-to-Head Benchmarks

The most striking result is in 3DMark Steel Nomad (DX12), where the RTX 2080 SUPER scores 1882 against the P106-100’s 899. This 109.3% delta means the RTX card more than doubles the P106-100’s performance in a modern DX12 workload. The gap suggests that the Turing architecture’s dedicated RT and tensor cores, combined with its higher memory bandwidth, provide a massive advantage in geometry-heavy and compute-intensive scenes.

Geekbench OpenCL shows an even more pronounced split: 99226 for the RTX 2080 SUPER versus 35951 for the P106-100, a 176% difference. This test measures general-purpose compute, and the RTX 2080 SUPER’s 3072 shading units and 384 tensor cores clearly outperform the P106-100’s 1280 shading units. The result is that the RTX card is nearly three times faster in OpenCL, making it far more suitable for scientific computing, machine learning inference, and video encoding workloads.

The largest gap emerges in Geekbench Vulkan, where the RTX 2080 SUPER scores 111284 and the P106-100 scores 32897, a 238.3% delta. Vulkan is a low-overhead API that benefits from modern hardware features; the RTX 2080 SUPER’s newer architecture and higher clock speeds (1815 MHz boost vs 1709 MHz) allow it to execute more draw calls and shader operations per frame. This result confirms that for any Vulkan-based game or application, the RTX 2080 SUPER is over three times as capable.

Across all three benchmarks, the RTX 2080 SUPER’s smallest win is 109.3% in DX12, and its largest is 238.3% in Vulkan. The P106-100’s 48 ROPs and 136.7 GTexel/s texture rate cannot compete with the RTX card’s 64 ROPs and 348.5 GTexel/s, explaining the consistent defeats. While the P106-100’s 68th percentile aggregate ranking is respectable, it is clear that the RTX 2080 SUPER is not just faster—it is in a different performance class entirely.