NVIDIA GeForce RTX 4090 Mobile vs NVIDIA P102-100 Comparison

Where Each One Wins

The recorded data splits cleanly along generational and workload lines. The NVIDIA P102-100, built on the Pascal architecture, holds its ground as a mining-era card with no display outputs, while the NVIDIA GeForce RTX 4090 Mobile is an active Ada Lovelace part designed for portable devices. Across the two directly comparable benchmark tests, the GeForce RTX 4090 Mobile wins both. That is a 2-0 sweep in the head-to-head comparison. The P102-100 does not secure a single win in any recorded test.

The P102-100 shows its strongest relative performance in the Geekbench Vulkan test, where it scores 67,454. The RTX 4090 Mobile scores 170,774 in the same test, a 60.5% advantage. In OpenCL, the gap widens further: the P102-100 posts 49,602, while the RTX 4090 Mobile reaches 180,831, a 72.6% lead. The database also records a broader benchmark suite for the RTX 4090 Mobile, including PassMark tests across DirectX 9, 10, 11, 12, G2D, G3D, and GPU compute. The P102-100 has no entries for those tests, so any use-case comparison must rely on the two Geekbench results and the architectural data.

For compute-heavy OpenCL workloads, the RTX 4090 Mobile is decisively ahead. For Vulkan-based rendering or compute, it still leads by a large margin, but the P102-100 is comparatively less far behind. The data suggests the P102-100 is not a general-purpose card: it has no display outputs and a PCIe 1.0 x4 bus interface, which limits its role to dedicated compute or mining tasks. The RTX 4090 Mobile, by contrast, is an integrated graphics processor (IGP) with portable-device-dependent outputs, making it suitable for laptop gaming and mobile workstations.

Architecture Differences

The two GPUs come from different process nodes, chip designs, and feature sets. The P102-100 uses the GP102 chip on a 16 nm TSMC process, with 11,800 million transistors on a 471 mm² die. That works out to a transistor density of 25.1 million per square millimeter. The RTX 4090 Mobile uses the AD103 chip on a 5 nm TSMC process, packing 45,900 million transistors into a 379 mm² die, for a density of 121.1 million per square millimeter. The density difference is stark: the Ada Lovelace part crams roughly 4.8 times more transistors per area.

Core counts also diverge heavily. The P102-100 has 3,200 shading units, 200 texture mapping units, and 80 raster output pipelines. The RTX 4090 Mobile has 9,728 shading units, 304 TMUs, and 112 ROPs. That is more than three times the shader count, 52% more TMUs, and 40% more ROPs. The RTX 4090 Mobile also adds dedicated ray tracing cores (76) and tensor cores (304), neither of which the P102-100 has. Those features enable hardware-accelerated ray tracing and AI workloads, which the Pascal card cannot handle natively.

Memory configurations differ as well. The P102-100 ships with 5 GB of GDDR5X on a 320-bit bus, delivering 440.3 GB/s of bandwidth. The RTX 4090 Mobile has 16 GB of GDDR6 on a 256-bit bus, but achieves higher bandwidth at 576.0 GB/s due to faster memory clocks (18 Gbps effective versus 11 Gbps effective). The memory clock figures are 1,376 MHz for the P102-100 and 2,250 MHz for the RTX 4090 Mobile, though the effective rates matter more for bandwidth. The RTX 4090 Mobile also doubles the bus interface to PCIe 4.0 x16, while the P102-100 is limited to PCIe 1.0 x4, a severe bottleneck for data transfer.

Clock speeds are closer than the rest of the spec sheet. The P102-100 has a base clock of 1,582 MHz and a boost of 1,683 MHz. The RTX 4090 Mobile has a lower base of 1,335 MHz but a slightly higher boost of 1,695 MHz. The Pascal card pushes higher raw clocks, but the Ada card compensates with far more cores and newer architecture. The result is a massive throughput difference: the P102-100 delivers 10.77 TFLOPS of FP32 and 168.3 GFLOPS of FP16 (at a 1:64 ratio). The RTX 4090 Mobile delivers 32.98 TFLOPS of FP32 and the same 32.98 TFLOPS of FP16 (at a 1:1 ratio). That is 3.06 times the FP32 throughput and roughly 196 times the FP16 throughput.

Power and physical design also separate them. The P102-100 draws 250 W, requires dual-slot cooling and two 8-pin power connectors, and has a suggested 600 W power supply. It measures 267 mm (10.5 inches) in length. The RTX 4090 Mobile is an IGP with a 120 W TDP, no power connectors, and no standalone dimensions, since it is integrated into a laptop motherboard. The P102-100 is end-of-life, released in February 2018, while the RTX 4090 Mobile is active and was released in January 2023. The RTX 4090 Mobile supports DirectX 12 Ultimate (12_2), while the P102-100 only reaches DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4.

The Verdict

The database clearly separates these two GPUs by role and era. The P102-100 is a mining-focused Pascal card with no display outputs, a 250 W power draw, and a PCIe 1.0 x4 interface. Its 88th percentile ranking among all GPUs reflects that it was a strong compute part in its time, but its average benchmark score of 58,528 places it near rivals like the AMD Radeon PRO V710 (58,657, 0.2% faster), AMD Radeon RX 6950 XT (58,392, 0.2% slower), Intel Arc A570M (58,239, 0.5% slower), and AMD Radeon RX 5600 OEM (58,085, 0.8% slower). Those margins are tiny, meaning the P102-100 sits in a tightly packed performance band.

The RTX 4090 Mobile, despite winning every head-to-head test, has a lower percentile rank at 84. Its average benchmark score of 43,667 is dragged down by the PassMark DirectX tests, which score very low (173 for DirectX 10, 262 for DirectX 11, 107 for DirectX 12, 310 for DirectX 9). Those numbers are likely from a specific driver or mobile implementation and do not reflect gaming performance. Its nearest rivals include the NVIDIA Quadro M6000 (43,301, 0.8% slower), GeForce RTX 5050 Mobile (43,268, 0.9% slower), Quadro M6000 24 GB (43,262, 0.9% slower), and RTX A6000 (44,075, 0.9% faster). The RTX 4090 Mobile sits between those cards, slightly behind the RTX A6000 and slightly ahead of the Quadro parts.

Who should pick which? The data says the RTX 4090 Mobile is the only choice for any workload involving ray tracing, tensor operations, or high-throughput FP16 compute, given its 76 RT cores, 304 tensor cores, and 1:1 FP16 ratio. It also wins decisively in both recorded Geekbench tests, making it the better all-around compute GPU. The P102-100, however, is a niche product: it has no display outputs, so it cannot drive a monitor. Its only practical use is as a secondary compute or mining card, and even then, its PCIe 1.0 x4 interface and 250 W power draw make it inefficient by modern standards. The RTX 4090 Mobile, at 120 W, delivers more than three times the FP32 throughput at half the power.

FAQ

Q: Which GPU is faster in OpenCL?

A: The NVIDIA GeForce RTX 4090 Mobile scores 180,831 in Geekbench OpenCL, while the NVIDIA P102-100 scores 49,602. That is a 72.6% lead for the RTX 4090 Mobile.

Q: Does the P102-100 support ray tracing?

A: No. The P102-100 has no RT cores and no tensor cores. The RTX 4090 Mobile has 76 RT cores and 304 tensor cores.

Q: Which GPU has higher memory bandwidth?

A: The RTX 4090 Mobile has 576.0 GB/s, compared to 440.3 GB/s for the P102-100. The RTX 4090 Mobile achieves this with a narrower 256-bit bus but faster GDDR6 memory at 18 Gbps effective.

Q: Can the P102-100 be used for display output?

A: No. The P102-100 has no display outputs. The RTX 4090 Mobile has outputs that are portable-device dependent.

Q: What is the transistor density difference?

A: The P102-100 has 25.1 million transistors per square millimeter, while the RTX 4090 Mobile has 121.1 million. The RTX 4090 Mobile is built on a 5 nm process versus 16 nm for the P102-100.

Q: Which GPU has a higher percentile ranking?

A: The P102-100 ranks in the 88th percentile among all GPUs, while the RTX 4090 Mobile ranks in the 84th percentile. Despite that, the RTX 4090 Mobile wins every head-to-head benchmark.

Head-to-Head Benchmarks

The database includes two directly comparable tests: Geekbench OpenCL and Geekbench Vulkan. The RTX 4090 Mobile wins both, but the margin differs significantly.

In Geekbench OpenCL, the RTX 4090 Mobile scores 180,831 against the P102-100's 49,602. The delta is 72.6% in favor of the RTX 4090 Mobile. This is the larger of the two gaps. OpenCL is a general-purpose compute API, and the RTX 4090 Mobile's 9,728 shading units and 32.98 TFLOPS FP32 throughput dominate the P102-100's 3,200 shaders and 10.77 TFLOPS. The RTX 4090 Mobile also has 16 GB of VRAM with 576.0 GB/s bandwidth, which likely helps in memory-intensive OpenCL workloads.

In Geekbench Vulkan, the RTX 4090 Mobile scores 170,774, while the P102-100 scores 67,454. The delta is 60.5% in favor of the RTX 4090 Mobile. This is a smaller lead than OpenCL, but still decisive. Vulkan is closer to a gaming or rendering API, and the P102-100's higher base clock (1,582 MHz versus 1,335 MHz) may help it narrow the gap slightly. Still, the RTX 4090 Mobile's architectural advantages, including dedicated RT cores and a much larger shader array, keep it well ahead.

The RTX 4090 Mobile also has a suite of PassMark results that the P102-100 lacks entirely. Those scores are 173 for DirectX 10, 262 for DirectX 11, 107 for DirectX 12, 310 for DirectX 9, 984 for G2D, 27,212 for G3D, and 12,347 for GPU compute. The G3D score of 27,212 is notably high and suggests strong rasterization performance, while the low DirectX scores likely reflect specific driver or workload conditions. The P102-100 has no comparable entries, so no direct comparison is possible there.

Looking at the overall average benchmark scores, the P102-100 actually has a higher average (58,528) than the RTX 4090 Mobile (43,667). This is because the RTX 4090 Mobile's PassMark DirectX scores are extremely low, pulling its average down. But the head-to-head data is unambiguous: in the two tests where both GPUs have results, the RTX 4090 Mobile wins by 60.5% and 72.6%. The P102-100's nearest rivals in the database are all within 0.8% of its score, showing it is a mid-pack performer among its peers. The RTX 4090 Mobile's nearest rivals are similarly tight, with the RTX A6000 being 0.9% faster and the Quadro M6000 being 0.8% slower.

The verdict from the numbers is clear. The RTX 4090 Mobile is the superior compute and graphics part, with more cores, more memory, higher bandwidth, and newer architecture. The P102-100 is an older mining card with no display output, a narrow PCIe interface, and a higher power draw. It is not competitive in any recorded metric against the RTX 4090 Mobile.