NVIDIA GeForce RTX 5070 SUPER vs NVIDIA GeForce RTX 5090 Mobile Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct benchmark comparison between the NVIDIA GeForce RTX 5070 SUPER and the NVIDIA GeForce RTX 5090 Mobile. In the 3DMark Steel Nomad DX12 test, the RTX 5090 Mobile delivers a score of 5871, while the RTX 5070 SUPER scores 2690. This represents a 54.2% advantage for the mobile part, a substantial margin that places these two GPUs in entirely different performance classes.

The RTX 5090 Mobile's victory in this test is consistent with its overall standing in the database. Its average benchmark score of 45152 places it at the 84th percentile among all GPUs. The nearest rivals to the RTX 5090 Mobile include the AMD Radeon Pro 5500 XT with an average score of 45384 (0.5% higher), the NVIDIA GeForce RTX 4070 Ti with 44795 (0.8% lower), the Intel Arc A730M with 45592 (1% higher), and the NVIDIA RTX 5880 Ada Generation with 45972 (1.8% higher). These deltas are all within roughly two percent, indicating that the RTX 5090 Mobile sits in a tightly contested performance band.

The RTX 5070 SUPER, by contrast, holds an 18th percentile ranking among all GPUs. Its nearest rivals are the NVIDIA Quadro K1100M with an average score of 2664 (1% higher), the NVIDIA GeForce GT 1030 with 2662 (1.1% higher), the Intel Arc Pro B50 with 2660 (1.1% higher), and the NVIDIA GeForce GT 440 with 2645 (1.7% higher). The RTX 5070 SUPER's average benchmark score of 2690 places it just ahead of these older or lower-tier parts, but the margin over its closest competitors is narrow.

The single head-to-head result shows a decisive win for the RTX 5090 Mobile. The delta of 54.2% is not a marginal difference; it represents more than double the raw score of the RTX 5070 SUPER in the Steel Nomad workload. For a desktop-oriented GPU such as the RTX 5070 SUPER, this outcome is notable because the mobile part achieves it while operating within a much lower power envelope.

Architecture Differences

Both GPUs share the Blackwell 2.0 architecture and are fabricated on a 5 nm process at TSMC. The similarities end there, as the chips differ significantly in scale and configuration. The RTX 5070 SUPER uses the GB205 chip with 31,100 million transistors on a 263 mm² die, resulting in a transistor density of 118.3 million per square millimeter. The RTX 5090 Mobile uses the larger GB203 chip with 45,600 million transistors on a 378 mm² die, yielding a density of 120.6 million per square millimeter. The mobile chip packs roughly 46% more transistors into a die that is about 44% larger by area.

Clock speeds present a striking contrast. The RTX 5070 SUPER runs with a base clock of 2325 MHz and a boost clock of 2512 MHz. The RTX 5090 Mobile operates at a base clock of 990 MHz and a boost clock of 1515 MHz. Despite the mobile part's much lower clocks, it achieves higher raw performance in the benchmark, indicating that the sheer scale of execution resources compensates for the frequency deficit.

The memory subsystems differ in capacity and bandwidth. The RTX 5070 SUPER comes with 18 GB of GDDR7 memory on a 192-bit bus, delivering 672.0 GB/s of bandwidth. The RTX 5090 Mobile offers 24 GB of GDDR7 memory on a 256-bit bus, with bandwidth reaching 896.0 GB/s. Both parts use the same memory clock of 1750 MHz with 28 Gbps effective data rate, so the bandwidth advantage for the mobile GPU comes entirely from the wider bus.

Compute resources scale in favor of the RTX 5090 Mobile. The mobile part contains 10496 shading units, 328 texture mapping units, 112 render output units, 82 ray tracing cores, and 328 tensor cores. The RTX 5070 SUPER counters with 6400 shading units, 200 TMUs, 80 ROPs, 50 ray tracing cores, and 200 tensor cores. The mobile GPU holds a 64% advantage in shading units and a 64% advantage in tensor cores. Curiously, the FP32 throughput figures are nearly identical: 32.15 TFLOPS for the RTX 5070 SUPER versus 31.80 TFLOPS for the RTX 5090 Mobile. The desktop part achieves this parity through higher clocks, while the mobile part relies on its larger execution array.

Power and physical characteristics diverge sharply. The RTX 5070 SUPER carries a TDP of 275 W, requires a dual-slot cooler, and uses a single 16-pin power connector. The RTX 5090 Mobile has a TDP of 95 W, is an integrated graphics package (IGP) with no dedicated power connectors, and has no listed physical dimensions. The 180 W difference in TDP explains why the mobile part can operate within a laptop chassis while the desktop part needs a full expansion slot and external power delivery.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The display outputs differ: the RTX 5070 SUPER provides 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the RTX 5090 Mobile's outputs are described as portable device dependent. Both use a PCIe 5.0 x16 bus interface.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce RTX 5090 Mobile has an average benchmark score of 45152, while the NVIDIA GeForce RTX 5070 SUPER has an average benchmark score of 2690.

Q: How much faster is the RTX 5090 Mobile in the 3DMark Steel Nomad DX12 test?

A: The RTX 5090 Mobile scores 5871 compared to 2690 for the RTX 5070 SUPER, a difference of 54.2% in favor of the mobile part.

Q: What are the memory capacities of these two GPUs?

A: The RTX 5070 SUPER has 18 GB of GDDR7 memory, while the RTX 5090 Mobile has 24 GB of GDDR7 memory.

Q: How do the thermal design powers compare?

A: The RTX 5070 SUPER has a TDP of 275 W, whereas the RTX 5090 Mobile has a TDP of 95 W.

Q: Which GPU has more shading units?

A: The RTX 5090 Mobile has 10496 shading units, significantly more than the 6400 shading units found on the RTX 5070 SUPER.

Q: What is the transistor count difference between the two chips?

A: The RTX 5090 Mobile's GB203 chip contains 45,600 million transistors, compared to 31,100 million on the RTX 5070 SUPER's GB205 chip.

Specification Differences

The two GPUs differ across nearly every technical specification in the database. The process node is identical at 5 nm from TSMC, and both use the Blackwell 2.0 architecture. The chips themselves are distinct: GB205 for the RTX 5070 SUPER versus GB203 for the RTX 5090 Mobile.

The RTX 5090 Mobile holds the advantage in transistor count (45,600 million versus 31,100 million), die size (378 mm² versus 263 mm²), and transistor density (120.6M per mm² versus 118.3M per mm²). Clock speeds favor the RTX 5070 SUPER, with a base clock of 2325 MHz versus 990 MHz and a boost clock of 2512 MHz versus 1515 MHz.

Memory configuration favors the mobile part in capacity and bandwidth. The RTX 5090 Mobile provides 24 GB across a 256-bit bus for 896.0 GB/s, while the RTX 5070 SUPER offers 18 GB on a 192-bit bus for 672.0 GB/s. Both use GDDR7 memory at 1750 MHz with 28 Gbps effective speed.

Compute unit counts all favor the RTX 5090 Mobile: shading units (10496 versus 6400), TMUs (328 versus 200), ROPs (112 versus 80), ray tracing cores (82 versus 50), and tensor cores (328 versus 200). The FP32 and FP16 throughput figures are nearly equal, with the RTX 5070 SUPER posting 32.15 TFLOPS and the RTX 5090 Mobile posting 31.80 TFLOPS in both precision modes.

Pixel and texture rates differ. The RTX 5070 SUPER achieves 201.0 GPixel/s and 502.4 GTexel/s, while the RTX 5090 Mobile achieves 169.7 GPixel/s and 496.9 GTexel/s. The desktop part wins on these fill-rate metrics despite having fewer ROPs and TMUs, again due to its higher clock speeds.

Power and physical specifications are fundamentally different. The RTX 5070 SUPER has a TDP of 275 W, a dual-slot form factor, a 16-pin power connector, and dimensions of 245 mm by 115 mm by 40 mm. The RTX 5090 Mobile has a TDP of 95 W, is classified as an IGP, requires no power connectors, and has no listed dimensions. Release dates also differ, with the RTX 5090 Mobile appearing on 2025-03-26 and the RTX 5070 SUPER appearing on 2025-12-31.

Where Each One Wins

The benchmark data gives the RTX 5090 Mobile a clear victory in the only direct comparison available. The 3DMark Steel Nomad DX12 score of 5871 versus 2690 represents a 54.2% lead, and the mobile part's 84th percentile ranking versus the 18th percentile ranking for the RTX 5070 SUPER reinforces this separation. The RTX 5090 Mobile also benefits from more memory (24 GB versus 18 GB), higher bandwidth (896.0 GB/s versus 672.0 GB/s), and a wider memory bus (256-bit versus 192-bit). These factors make it the stronger choice for memory-intensive workloads and modern game titles that demand large frame buffers.

The RTX 5070 SUPER wins on clock speed and fill rate. Its boost clock of 2512 MHz dwarfs the 1515 MHz of the RTX 5090 Mobile, and its pixel rate of 201.0 GPixel/s exceeds the 169.7 GPixel/s of the mobile part. The desktop GPU also posts higher texture rate, 502.4 GTexel/s versus 496.9 GTexel/s, though the margin is slim. Its FP32 output of 32.15 TFLOPS edges out the 31.80 TFLOPS of the RTX 5090 Mobile, meaning the desktop part delivers comparable compute throughput with fewer execution units.

Power efficiency favors the RTX 5090 Mobile. With a TDP of 95 W versus 275 W, the mobile part achieves a higher benchmark score while consuming less than half the power budget. The RTX 5090 Mobile's 5871 Steel Nomad score divided by its 95 W TDP yields roughly 61.8 points per watt, while the RTX 5070 SUPER manages only 9.8 points per watt. This efficiency gap is dramatic and positions the mobile GPU as the better option for constrained thermal environments.

The RTX 5070 SUPER offers a conventional desktop form factor with dual-slot cooling, a 16-pin power connector, and explicit display outputs including 1x HDMI 2.1b and 3x DisplayPort 2.1b. The RTX 5090 Mobile is an IGP with no dedicated power connectors and output configuration that depends on the host device. For systems requiring standardized display connectivity and independent GPU installation, the desktop part fits the bill.

The RTX 5090 Mobile's predecessor designation of GeForce 40 Mobile indicates an established product line, while the RTX 5070 SUPER has no predecessor listed. The mobile GPU also has a broader benchmark profile in the database, with results spanning OpenCL, Vulkan, DirectX 9 through 12, and compute workloads, whereas the RTX 5070 SUPER only has the single Steel Nomad result. The mobile part's PassMark G3D score of 30034 and compute score of 13401 suggest strong general-purpose and compute performance, though the RTX 5070 SUPER lacks comparable data points for direct comparison.