NVIDIA GeForce GTX 965M vs NVIDIA P106-090 Comparison

# Head-to-Head Benchmarks

The benchmark data reveals a clear and consistent performance advantage for the NVIDIA P106-090 across every shared test. In the two head-to-head comparisons available, the P106-090 wins both, leaving the GeForce GTX 965M without a single victory. The most decisive margin appears in the Geekbench OpenCL test, where the P106-090 scores 21,304 against the GTX 965M’s 14,509. That represents a 31.9% lead, a substantial gap that places the two GPUs in different performance tiers despite their similar overall database rankings.

The Geekbench Vulkan test tells a similar story, though with a slightly narrower margin. The P106-090 posts 18,596 points, while the GTX 965M manages 14,299. The 23.1% delta in favor of the P106-090 confirms that the Pascal-based card’s advantage is not confined to a single API or workload type. Both OpenCL and Vulkan results point in the same direction: the P106-090 is the faster GPU in raw compute tasks.

Looking at the broader benchmark landscape, the average benchmark scores introduce a nuance worth examining. The GTX 965M holds a higher average benchmark score of 14,404, compared to the P106-090’s 13,470. This seems paradoxical given the head-to-head results, but it reflects the different benchmark suites each card has been tested with. The GTX 965M’s average is derived from its two Geekbench results (14,509 and 14,299), while the P106-090’s average includes a third test — 3DMark Steel Nomad DX12 — where it scores only 509. That low DX12 result drags the P106-090’s average down, even though it wins decisively in both Geekbench tests.

The percentile rankings align with this mixed picture. The GTX 965M sits at the 56th percentile of all GPUs, while the P106-090 ranks at the 54th percentile. The two-point difference is modest, suggesting that in the broader database context these GPUs occupy similar standing. Yet in direct competition, the P106-090’s higher clock speeds and newer architecture give it a consistent edge in compute workloads.

# FAQ

Q: Which GPU wins in Geekbench OpenCL performance?

A: The NVIDIA P106-090 wins decisively, scoring 21,304 compared to the GeForce GTX 965M’s 14,509. This represents a 31.9% advantage for the P106-090.

Q: Is the GeForce GTX 965M faster in any benchmark?

A: No. In the two shared head-to-head tests (Geekbench OpenCL and Geekbench Vulkan), the P106-090 wins both. The GTX 965M has zero wins in the head-to-head comparison.

Q: How do the average benchmark scores compare?

A: The GTX 965M has a higher average benchmark score of 14,404, versus 13,470 for the P106-090. However, this is influenced by the P106-090’s inclusion of a 3DMark Steel Nomad DX12 result of 509, which is not present in the GTX 965M’s benchmark set.

Q: What is the GTX 965M’s closest rival according to the database?

A: The AMD Radeon RX Vega 11 is the nearest rival, with an average score of 14,385 and a delta of just 0.1% compared to the GTX 965M’s 14,404.

Q: Which GPU has a higher percentile ranking?

A: The GeForce GTX 965M ranks at the 56th percentile of all GPUs, while the P106-090 sits at the 54th percentile. Despite losing both head-to-head tests, the GTX 965M holds a slightly higher percentile position.

Q: Does the P106-090 have any benchmark results outside of Geekbench?

A: Yes. The P106-090 has a 3DMark Steel Nomad DX12 score of 509, a test that the GTX 965M has no result for in the provided data.

# Where Each One Wins

The performance split between these two GPUs is unusually one-sided in raw compute benchmarks, yet the use-case analysis reveals more complexity than the win tally suggests.

The NVIDIA P106-090 is the clear winner for OpenCL and Vulkan compute workloads. Its 31.9% lead in OpenCL and 23.1% lead in Vulkan make it the stronger choice for GPU-accelerated applications that leverage these APIs — think compute-heavy tasks like rendering, data processing, or physics simulations. The P106-090’s higher boost clock of 1,531 MHz (versus 950 MHz on the GTX 965M) and its 192-bit memory bus with 192.2 GB/s bandwidth provide a substantial throughput advantage for memory-intensive compute operations. Its 2.352 TFLOPS of FP32 performance also exceeds the GTX 965M’s 1.946 TFLOPS, giving it more raw floating-point horsepower.

The GeForce GTX 965M, however, has its own strengths that the benchmark scores do not fully capture. It carries a higher average benchmark score (14,404 vs. 13,470) and a better percentile ranking (56th vs. 54th). More practically, the GTX 965M is a mobile GPU with display outputs — listed as "Portable Device Dependent" — making it suitable for laptop use where the P106-090, with no display outputs at all, cannot function as a primary graphics solution. The GTX 965M also belongs to the GeForce 900M generation, a line designed for gaming laptops, whereas the P106-090 is explicitly categorized under "Mining GPUs."

For gaming in laptops, the GTX 965M is the only viable option of the two, as the P106-090 lacks any video output capability. For compute workloads where display output is irrelevant, the P106-090 is the stronger performer by a significant margin. The 3DMark Steel Nomad DX12 result of 509 for the P106-090, while low in absolute terms, indicates some DirectX 12 capability, but without a comparable GTX 965M result, direct gaming comparisons are not possible from this data.

# Specification Differences

The two GPUs differ substantially across nearly every specification category, reflecting their different design goals and architectures.

The memory configurations are notably distinct. The GTX 965M offers 2 GB of GDDR5 memory on a 128-bit bus, delivering 80.19 GB/s of bandwidth. The P106-090 steps up to 3 GB of GDDR5 on a 192-bit bus, more than doubling bandwidth to 192.2 GB/s. Memory clocks also differ: the GTX 965M runs at 1,253 MHz (5 Gbps effective), while the P106-090 operates at 2,002 MHz (8 Gbps effective).

Clock speeds show a significant divergence. The GTX 965M has a base clock of 924 MHz and a boost clock of 950 MHz. The P106-090 runs considerably faster, with a 1,354 MHz base and 1,531 MHz boost. This 581 MHz boost-clock advantage is a major factor in the P106-090’s performance lead.

The compute unit configurations also differ in interesting ways. The GTX 965M has more shading units (1,024 vs. 768) and more texture mapping units (64 vs. 48), but the P106-090 counters with more ROPs (48 vs. 32) and higher fill rates. Pixel rate for the P106-090 is 73.49 GPixel/s versus 30.40 GPixel/s for the GTX 965M. Texture rate similarly favors the P106-090 at 73.49 GTexel/s, compared to 60.80 GTexel/s for the older card.

Power and physical specifications reveal their different intended environments. The GTX 965M is an MXM Module with no power connectors and a slot width listed as "MXM Module" — a mobile form factor. The P106-090 is a dual-slot desktop card measuring 250 mm (9.8 inches) in length, with a 75 W TDP, a single 6-pin power connector, and a suggested PSU of 250 W. The bus interfaces differ as well: the GTX 965M uses MXM-B (3.0), while the P106-090 uses PCIe 1.0 x1 — a notably limited interface for a desktop card.

# Architecture Differences

The architectural divide between these two GPUs spans two generations of NVIDIA design. The GeForce GTX 965M is built on the GM204 chip using the Maxwell 2.0 architecture, fabricated on a 28 nm process at TSMC. The P106-090 uses the GP106 chip with the newer Pascal architecture, also from TSMC but on a more advanced 16 nm node.

The process shrink from 28 nm to 16 nm brings significant transistor-density improvements. The GTX 965M’s GM204 die measures 398 mm² and contains 5,200 million transistors, yielding a density of 13.1 million transistors per square millimeter. The P106-090’s GP106 die is much smaller at 200 mm² but holds 4,400 million transistors, achieving a density of 22.0 million per square millimeter — a 68% improvement in density that reflects the more modern fabrication process.

Despite having fewer transistors overall, the P106-090 achieves higher performance through its architecture and clock speeds. The Pascal architecture also introduces FP16 support at a ratio of 1:64 (36.74 GFLOPS), a feature absent from the Maxwell 2.0-based GTX 965M, which has no listed FP16 capability.

Both GPUs share the same API support for DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, and neither has dedicated ray tracing or tensor cores. The production status for both is end-of-life, though their release dates differ by about two and a half years — the GTX 965M launched in January 2015 as part of the GeForce 900M generation, while the P106-090 arrived in July 2017 under the "Mining GPUs" generation.

The GTX 965M has a clear lineage with a predecessor in the GeForce 800M series and a successor in the GeForce 10 Mobile line. The P106-090, as a mining-focused product, has neither a listed predecessor nor successor. The P106-090’s display output is listed as "No outputs," confirming its purpose as a compute-only device, whereas the GTX 965M’s outputs are "Portable Device Dependent," indicating its role as a mobile graphics solution.