NVIDIA GeForce RTX 3080 Mobile vs NVIDIA GeForce RTX 3090 Comparison

Head-to-Head Benchmarks

The head-to-head record is lopsided: the RTX 3090 wins 9 of 10 benchmarks, with the single exception being a striking reversal in Vulkan. The largest margin comes in compute workloads, where the RTX 3090 records 15,356 in Passmark GPU Compute against 7,276 for the RTX 3080 Mobile, a 111.1% advantage. That more than doubles the mobile part's compute throughput, which aligns with the desktop card's larger silicon and higher power budget.

In 3DMark Steel Nomad DX12, the RTX 3090 scores 5,118 versus 2,644, a 93.6% lead. This is the closest to a pure gaming workload in the data, and the delta suggests the desktop card sustains far higher frame rates under modern DirectX 12 rendering. The Geekbench OpenCL result shows a 64.8% gap (172,758 vs 104,831), indicating the RTX 3090 also dominates general-purpose compute tasks that rely on OpenCL.

Passmark's legacy DirectX tests tell a consistent story. In DirectX 9, the RTX 3090 leads 268 to 170 (57.6% ahead). DirectX 10 shows 182 vs 120 (51.7% ahead), DirectX 11 shows 220 vs 146 (50.7% ahead), and DirectX 12 shows 110 vs 72 (52.8% ahead). These are smaller relative gaps than the compute or Steel Nomad results, but still comfortable wins across every API generation. The Passmark G3D score (26,645 vs 16,321) is 63.3% higher for the RTX 3090, and the G2D score (1,063 vs 637) is 66.9% higher, suggesting even 2D composition tasks benefit from the desktop card's faster memory and pixel throughput.

The single reverse result is Geekbench Vulkan: the RTX 3080 Mobile scores 104,066 versus 53,927 for the RTX 3090, a 48.2% deficit for the desktop card. This is a curious outlier. Vulkan can be sensitive to driver scheduling and memory placement, and the mobile part's lower power draw may allow it to sustain boost clocks differently under this specific workload. The data offers no further explanation, but it does show that the mobile GPU is not universally slower; in one API, it holds a decisive edge.

Architecture Differences

Both cards share the Ampere architecture and are built on Samsung's 8 nm process, but they use different chips. The RTX 3090 is built around GA102, a 628 mm² die with 28,300 million transistors, yielding a density of 45.1M per mm². The RTX 3080 Mobile uses GA104, a 392 mm² die with 17,400 million transistors, at 44.4M per mm². The desktop chip is physically much larger, which explains its higher raw throughput: 10,496 shading units, 328 TMUs, and 112 ROPs versus 6,144 shading units, 192 TMUs, and 96 ROPs on the mobile part. It also carries 82 RT cores and 328 tensor cores, while the mobile GPU has 48 RT cores and 192 tensor cores.

Memory is another major divergence. The RTX 3090 ships with 24 GB of GDDR6X on a 384-bit bus, delivering 936.2 GB/s of bandwidth. The RTX 3080 Mobile has 8 GB of GDDR6 on a 256-bit bus, capping out at 448.0 GB/s. That is less than half the bandwidth, which matters for high-resolution textures and compute kernels that stream large datasets. The RTX 3090's memory clock is listed at 1,219 MHz with 19.5 Gbps effective, while the mobile part runs at 1,750 MHz with 14 Gbps effective. The effective rate is higher on the desktop card, but the bus width difference is the dominant factor.

Clock speeds also favor the desktop card. The RTX 3090 has a base clock of 1,395 MHz and a boost of 1,695 MHz. The RTX 3080 Mobile sits at 1,110 MHz base and 1,545 MHz boost. Those lower clocks, combined with fewer cores and less memory bandwidth, produce a peak FP32 throughput of 18.98 TFLOPS for the mobile chip versus 35.58 TFLOPS for the desktop card. FP16 is identical to FP32 for both, at a 1:1 ratio. The RTX 3090 also delivers a texture rate of 556.0 GTexel/s and a pixel rate of 189.8 GPixel/s, while the mobile part manages 296.6 GTexel/s and 148.3 GPixel/s.

Power consumption is a stark differentiator. The RTX 3090 is rated at 350 W TDP, requires a 12-pin power connector, and is a triple-slot card measuring 336 mm long, 140 mm tall, and 61 mm wide. The RTX 3080 Mobile has a 115 W TDP, uses no power connectors (it draws from the laptop's bus), and has no listed dimensions because it is portable-device dependent. The desktop card also lists a suggested PSU of 750 W, while the mobile part has none.

Where Each One Wins

The RTX 3090 wins everywhere except Vulkan. Its 111.1% compute lead makes it the obvious choice for GPU-accelerated rendering, scientific simulation, or any workload that leans on raw FP32 or tensor throughput. The 93.6% Steel Nomad lead shows it is also stronger for modern DirectX 12 gaming, and the consistent 50-57% wins across legacy DirectX APIs mean it handles older titles with ease.

The RTX 3080 Mobile's single win in Geekbench Vulkan (104,066 vs 53,927) is not enough to recommend it for general Vulkan gaming, but it does indicate that the mobile part's driver stack or memory subsystem behaves differently under that API. If a user's workload is exclusively Vulkan-based, the data suggests the mobile GPU may offer better performance per watt, but the desktop card still has far more total throughput available.

For 2D desktop work, the RTX 3090's G2D score of 1,063 versus 637 is a 66.9% lead, which is surprising for a gaming card but reflects the faster memory and higher pixel rate. For compute tasks like OpenCL, the 64.8% lead reinforces that the RTX 3090 is a workstation-class part in gaming clothing.

Specification Differences

| Specification | NVIDIA GeForce RTX 3090 | NVIDIA GeForce RTX 3080 Mobile |

|---|---|---|

| Chip | GA102 | GA104 |

| Process Node | 8 nm | 8 nm |

| Transistors | 28,300 million | 17,400 million |

| Die Size | 628 mm² | 392 mm² |

| Base Clock | 1,395 MHz | 1,110 MHz |

| Boost Clock | 1,695 MHz | 1,545 MHz |

| Memory Size | 24 GB | 8 GB |

| Memory Type | GDDR6X | GDDR6 |

| Memory Bus | 384 bit | 256 bit |

| Memory Bandwidth | 936.2 GB/s | 448.0 GB/s |

| Shading Units | 10,496 | 6,144 |

| TMUs | 328 | 192 |

| ROPs | 112 | 96 |

| RT Cores | 82 | 48 |

| Tensor Cores | 328 | 192 |

| Pixel Rate | 189.8 GPixel/s | 148.3 GPixel/s |

| Texture Rate | 556.0 GTexel/s | 296.6 GTexel/s |

| FP32 | 35.58 TFLOPS | 18.98 TFLOPS |

| TDP | 350 W | 115 W |

| Power Connectors | 1x 12-pin | None |

| Suggested PSU | 750 W | None |

| Dimensions | 336 x 140 x 61 mm | Portable Device Dependent |

| Display Outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a | Portable Device Dependent |

| Release Date | 2020-08-31 | 2021-01-11 |

| Launch MSRP | 1,499 USD | None |

FAQ

Q: Which GPU is faster in 3DMark Steel Nomad DX12?

A: The RTX 3090 scores 5,118 versus 2,644 for the RTX 3080 Mobile, a 93.6% lead.

Q: Does the RTX 3080 Mobile win any benchmark?

A: Yes, it wins Geekbench Vulkan with 104,066 points, beating the RTX 3090's 53,927 points by 48.2%.

Q: How much compute performance does the RTX 3090 have over the mobile part?

A: In Passmark GPU Compute, the RTX 3090 scores 15,356 versus 7,276, a 111.1% advantage.

Q: What is the memory bandwidth difference?

A: The RTX 3090 has 936.2 GB/s over a 384-bit bus with GDDR6X, while the RTX 3080 Mobile has 448.0 GB/s over a 256-bit bus with GDDR6.

Q: What are the TDP ratings?

A: The RTX 3090 is rated at 350 W, while the RTX 3080 Mobile is rated at 115 W.

Q: Which card has more shading units?

A: The RTX 3090 has 10,496 shading units, compared to 6,144 on the RTX 3080 Mobile.

The Verdict

The data is unambiguous for raw performance: the RTX 3090 dominates the RTX 3080 Mobile in 9 of 10 benchmarks, with margins ranging from 50.7% to 111.1%. It is the only choice for users who need maximum compute throughput, modern DX12 gaming performance, or large memory capacity. Its 24 GB of GDDR6X and 936.2 GB/s bandwidth are in a different class from the mobile part's 8 GB and 448.0 GB/s. The RTX 3090 also holds a 73rd percentile ranking among all GPUs in the database, versus 69th for the RTX 3080 Mobile.

The RTX 3080 Mobile is not without merit. Its 115 W TDP makes it feasible in a laptop, where the RTX 3090's 350 W TDP and triple-slot design are impossible. It also wins the Vulkan benchmark outright, which may matter for users locked into that API. But the average benchmark score of 23,628 for the mobile part versus 27,565 for the desktop card, and the latter's nearest rival being the RTX 4070 Mobile at only 0.5% away, shows the mobile GPU is competitive within its power class, not against the desktop flagship.

Pick the RTX 3090 for desktop workstations or high-end gaming rigs where power and size are not constraints. Pick the RTX 3080 Mobile for portable systems that need Ampere features in a laptop chassis, accepting the large performance gap in every other workload.