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

FAQ

Q: How do the two GPUs compare in overall compute performance?

A: The NVIDIA GeForce RTX 5090 D delivers substantially higher compute output. Its average benchmark score is 77,112, while the RTX 5070 Mobile averages 29,928. The 5090 D wins all nine recorded head-to-head benchmark tests.

Q: Which GPU has higher memory bandwidth and capacity?

A: The RTX 5090 D offers 32 GB of GDDR7 memory on a 512-bit bus, providing 1.79 TB/s bandwidth. The RTX 5070 Mobile has 8 GB of GDDR7 on a 128-bit bus, reaching 384.0 GB/s. The 5090 D also runs faster effective memory at 28 Gbps versus 24 Gbps.

Q: What are the power and physical requirements for each?

A: The RTX 5070 Mobile is rated at 50 W TDP and uses an integrated form factor (IGP) with no power connectors. The RTX 5090 D consumes 575 W, uses a dual-slot design with a 1x 16-pin connector, and requires a 950 W suggested PSU. Its dimensions are 304 mm in length, 137 mm in height, and 48 mm in width.

Q: How do they rank among all GPUs in the database?

A: The RTX 5090 D sits in the 92nd percentile of all GPUs, while the RTX 5070 Mobile ranks in the 75th percentile. The 5090 D's nearest rivals are the AMD Radeon RX 6650M XT (1.1% slower) and RX 6850M XT (1.6% faster), whereas the 5070 Mobile's closest competitor is the RTX 3070 Ti at 0.1% faster.

Q: What API support do they share?

A: Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. They are built on the same Blackwell 2.0 architecture and use TSMC's 5 nm process node.

Q: When were they released?

A: The RTX 5090 D was released on January 29, 2025. The RTX 5070 Mobile followed on April 14, 2025. The 5090 D has a successor, the GeForce 60 series, while the 5070 Mobile does not.

Architecture Differences

Both GPUs share the Blackwell 2.0 architecture and TSMC's 5 nm process, but their chip designs diverge sharply. The RTX 5070 Mobile uses the GB206 chip, which contains 21,900 million transistors on a 181 mm² die, yielding a transistor density of 121.0M per mm². The RTX 5090 D employs the GB202 chip with 92,200 million transistors on a 750 mm² die, reaching 122.9M per mm². The 5090 D's die is more than four times larger in area and holds over four times the transistor count.

Compute resources scale accordingly. The RTX 5070 Mobile has 4,608 shading units, 144 texture mapping units, and 48 raster output pipelines. The RTX 5090 D increases those to 21,760 shading units, 680 TMUs, and 176 ROPs. Ray tracing cores jump from 36 to 170, and tensor cores rise from 144 to 680. These differences translate into raw throughput: the 5070 Mobile delivers 13.13 TFLOPS FP32 and FP16 (1:1), while the 5090 D reaches 104.8 TFLOPS in both precision formats.

Clock speeds also differ. The RTX 5070 Mobile has a base clock of 907 MHz and a boost of 1425 MHz. The RTX 5090 D runs at 2017 MHz base and 2407 MHz boost. Pixel fill rate for the 5070 Mobile is 68.40 GPixel/s, compared to 423.6 GPixel/s for the 5090 D. Texture rate measures 205.2 GTexel/s versus 1,636.8 GTexel/s.

Memory architecture is another major split. Both use GDDR7, but the 5070 Mobile has 8 GB on a 128-bit bus with 384.0 GB/s bandwidth. The 5090 D has 32 GB on a 512-bit bus with 1.79 TB/s bandwidth. The 5090 D also runs its memory at 28 Gbps effective, while the 5070 Mobile operates at 24 Gbps effective.

Form factor and power delivery reflect their different roles. The RTX 5070 Mobile is an integrated GPU (IGP) with no power connectors and a 50 W TDP, designed for portable devices with device-dependent display outputs. The RTX 5090 D is a dual-slot desktop card measuring 304 mm by 137 mm by 48 mm, using a single 16-pin connector and a 575 W TDP. It offers 1x HDMI 2.1b and 3x DisplayPort 2.1b outputs. Both use a PCIe 5.0 x16 bus interface.

Head-to-Head Benchmarks

The RTX 5090 D wins every recorded comparison, but the margins vary across workloads. The largest gap appears in compute-heavy tests. In passmark_gpu_compute, the 5090 D scores 28,396 versus 8,279 for the 5070 Mobile, a 70.8% advantage. Vulkan performance is nearly as lopsided: 376,915 against 116,960, or 69% higher. Geekbench OpenCL shows the 5090 D at 310,674 versus 122,238, a 60.7% lead.

Graphics API tests follow a similar pattern with slightly smaller deltas. In passmark_directx_12, the 5090 D scores 219 against 93, a 57.5% difference. DirectX 11 results show 371 versus 192, a 48.2% gap. DirectX 9 performance favors the 5090 D by 50.7%, with scores of 434 and 214. DirectX 10 yields 231 versus 129, a 44.2% margin.

The narrowest wins for the 5090 D appear in 2D and general graphics workloads. Passmark G2D scores 1,487 versus 896, a 39.7% advantage. Passmark G3D shows 44,065 against 20,355, a 53.8% lead. Even at its smallest margin, the 5090 D remains over a third faster than the 5070 Mobile.

The average benchmark scores reinforce the hierarchy. The 5090 D averages 77,112, placing it in the 92nd percentile. The 5070 Mobile averages 29,928, putting it in the 75th percentile. The 5090 D's closest competitor is the AMD Radeon RX 6650M XT at 76,904, just 1.1% behind. The 5070 Mobile's nearest rival is the RTX 3070 Ti at 29,945, only 0.1% ahead. The 5090 D also outperforms the RTX 2080 Ti (29,783) and AMD RX 6800 (30,095), which are among the 5070 Mobile's comparison set.

The Verdict

The benchmark data indicates that the RTX 5090 D is the dominant part in absolute performance, winning all nine head-to-head tests. Its average score of 77,112 is roughly 2.6 times that of the RTX 5070 Mobile's 29,928. The 5090 D also achieves a higher percentile rank (92nd versus 75th) and delivers a massive lead in compute, Vulkan, and DirectX 12 workloads, ranging from 57.5% to 70.8%.

The RTX 5070 Mobile occupies a different segment entirely. Its 50 W TDP, integrated form factor, and lack of power connectors indicate a mobile-focused design for thin, portable systems. Its 8 GB memory and 384.0 GB/s bandwidth suit lighter workloads, while the 5090 D's 32 GB and 1.79 TB/s target high-end desktop rendering and compute. The 5070 Mobile's nearest rivals are desktop GPUs like the RTX 3070 Ti and RTX 2080 Ti, suggesting it competes near the upper midrange despite its mobile positioning.

Selecting between them depends on the deployment context. The RTX 5090 D is the clear choice for workloads demanding maximum throughput, large memory capacity, and high-bandwidth access. The RTX 5070 Mobile is the appropriate option for systems constrained by power and physical space, where its 50 W envelope and IGP form factor fit portable chassis. The data does not support treating them as interchangeable; the 5090 D's 575 W TDP and dual-slot dimensions place it in an entirely different power class.

The 5090 D's release date of January 29, 2025, and its successor in the GeForce 60 series mark it as a current-generation flagship. The 5070 Mobile, released April 14, 2025, has no successor listed, indicating it remains a current product. Both share the same architecture generation, but the 5090 D's launch MSRP of 2,299 USD reflects its flagship positioning, a detail not available for the 5070 Mobile.

Specification Differences

The two GPUs differ across almost every measured specification. The RTX 5070 Mobile uses the GB206 chip, while the RTX 5090 D uses GB202. Transistor counts are 21,900 million versus 92,200 million, and die sizes are 181 mm² versus 750 mm². The 5090 D has a slightly higher transistor density at 122.9M per mm², compared to 121.0M per mm² for the 5070 Mobile.

Clock speeds diverge: the 5070 Mobile runs at 907 MHz base and 1425 MHz boost, while the 5090 D runs at 2017 MHz base and 2407 MHz boost. Memory clocks differ as well, with the 5070 Mobile at 1500 MHz (24 Gbps effective) and the 5090 D at 1750 MHz (28 Gbps effective). Memory capacity is 8 GB versus 32 GB, bus width is 128 bit versus 512 bit, and bandwidth is 384.0 GB/s versus 1.79 TB/s.

Compute unit counts are uniformly higher on the 5090 D: 4,608 shading units versus 21,760, 144 TMUs versus 680, 48 ROPs versus 176, 36 RT cores versus 170, and 144 tensor cores versus 680. Pixel rate is 68.40 GPixel/s versus 423.6 GPixel/s, and texture rate is 205.2 GTexel/s versus 1,636.8 GTexel/s. FP32 and FP16 both measure 13.13 TFLOPS on the 5070 Mobile and 104.8 TFLOPS on the 5090 D.

Power and physical specs are reversed in scale. The 5070 Mobile has a 50 W TDP, IGP slot width, and no power connectors, while the 5090 D has a 575 W TDP, dual-slot width, and a 1x 16-pin connector. The 5090 D specifies a 950 W suggested PSU; the 5070 Mobile lists none. Dimensions exist only for the 5090 D: 304 mm length, 137 mm height, and 48 mm width. Display outputs are portable-device dependent for the 5070 Mobile, while the 5090 D offers 1x HDMI 2.1b and 3x DisplayPort 2.1b. Release dates differ by about two and a half months, and the 5090 D lists a successor (GeForce 60) while the 5070 Mobile does not.