NVIDIA GeForce RTX 3070 Mobile vs NVIDIA GeForce RTX 3080 Comparison

Head-to-Head Benchmarks

The recorded data shows a decisive overall victory for the desktop RTX 3080, which wins 9 of the 10 head-to-head benchmarks. The largest margin appears in Passmark GPU Compute, where the RTX 3080 scores 14,397 against the RTX 3070 Mobile's 6,827, a delta of 110.9%. This near-doubling of compute throughput suggests the desktop part's raw execution resources translate directly into heavier GPGPU workloads.

In 3DMark Steel Nomad DX12, the RTX 3080 posts 4,407 versus 2,380 for the mobile chip, an 85.2% advantage. This is the closest proxy in the set for modern gaming load behavior, and the gap is substantial. The desktop card also leads in Passmark DX9 by 61.3% (258 versus 160), DX12 by 56.3% (100 versus 64), and DX11 by 50% (207 versus 138). These consistent wins across API generations indicate the RTX 3080's advantage is not tied to any single rendering path.

Geekbench OpenCL shows a 64% lead for the RTX 3080, with scores of 152,423 versus 92,939. Passmark G3D (25,086 versus 15,309) and Passmark G2D (1,054 versus 641) yield deltas of 63.9% and 64.4% respectively. Even the DX10 test, often considered legacy, favors the desktop part by 50.4% (170 versus 113).

The single exception is Geekbench Vulkan, where the RTX 3070 Mobile wins decisively: 86,768 versus 33,620, a delta of -61.3% from the desktop card's perspective. This is a striking outlier. The mobile part's Vulkan score is more than double the desktop's, which suggests driver-level optimization or measurement conditions that favor the mobile configuration in this specific API test. It is importantly this is the only benchmark where the mobile chip wins, yet the magnitude of the win is large enough to complicate any simple narrative.

Average benchmark scores reflect the overall trend: the RTX 3080 averages 23,172 across the suite, while the RTX 3070 Mobile averages 20,534. The desktop card ranks in the 68th percentile of all GPUs in the database, the mobile part in the 65th. The nearest rivals for the RTX 3080 include the NVIDIA P106-100 and AMD Radeon Pro Vega 16, both within 0.3% of its average score. For the RTX 3070 Mobile, the nearest rivals are the Intel Arc B570 (-0.1%), Intel Arc A750 (-0.2%), and NVIDIA Quadro M4000M (+0.3%). These proximity values show both cards sit in crowded performance bands relative to the rest of the database.

Architecture Differences

Both GPUs share the Ampere architecture and are fabricated on Samsung's 8 nm process, but they are fundamentally different chips. The RTX 3080 uses the GA102 die with 28,300 million transistors on a 628 mm² footprint, yielding a density of 45.1 million transistors per mm². The RTX 3070 Mobile uses the GA104 die with 17,400 million transistors on 392 mm², a density of 44.4 million per mm². The desktop chip is physically larger and packs roughly 63% more transistors, which explains much of its performance headroom.

The RTX 3080 deploys 8,704 shading units, 272 texture mapping units, and 96 raster operation units. It also carries 68 RT cores and 272 tensor cores. The mobile part is cut down to 5,120 shading units, 160 TMUs, 80 ROPs, 40 RT cores, and 160 tensor cores. This means the desktop GPU has 70% more shading units and 70% more tensor cores, which aligns with its compute benchmark dominance.

Memory subsystems diverge sharply. The RTX 3080 uses 10 GB of GDDR6X on a 320-bit bus, delivering 760.3 GB/s of bandwidth. The RTX 3070 Mobile uses 8 GB of GDDR6 on a 256-bit bus, with 448.0 GB/s. The memory clock difference is also notable: the desktop part runs at 1188 MHz (19 Gbps effective), while the mobile part runs at 1750 MHz (14 Gbps effective). The higher bus width and newer memory type give the desktop card a 69.7% bandwidth advantage, which is critical for high-resolution textures and compute workloads.

Clock speeds tell a nuanced story. The RTX 3080 has a base clock of 1440 MHz and a boost of 1710 MHz. The RTX 3070 Mobile starts lower at 1110 MHz base and boosts to 1560 MHz. Despite the mobile part's lower clock speeds, its memory clock is higher in raw MHz (1750 versus 1188), though effective data rate is lower. The desktop card's pixel rate is 164.2 GPixel/s versus 124.8 GPixel/s for the mobile chip, and texture rate is 465.1 GTexel/s versus 249.6 GTexel/s. FP32 throughput is 29.77 TFLOPS for the RTX 3080 versus 15.97 TFLOPS for the RTX 3070 Mobile, a near 2:1 ratio.

Power and physical design differ as expected. The RTX 3080 carries a TDP of 320 W, uses a dual-slot cooler, requires a 1x 12-pin power connector, and recommends a 700 W PSU. Its dimensions are 285 mm long, 112 mm tall, and 40 mm wide. The RTX 3070 Mobile has a TDP of 115 W, no dedicated power connectors, and no slot width or dimensions listed, reflecting its integration into laptops. The mobile part's display outputs are listed as "Portable Device Dependent," while the desktop card offers 1x HDMI 2.1 and 3x DisplayPort 1.4a.

Both support PCIe 4.0 x16, DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The desktop card was released on 2020-08-31, the mobile part on 2021-01-11. Both are end-of-life in production status. The RTX 3080's predecessor is GeForce 20 and successor is GeForce 40; the RTX 3070 Mobile's predecessor is GeForce 20 Mobile with no successor listed.

Where Each One Wins

The RTX 3080 dominates across virtually every measured category. Its largest wins are in compute-heavy workloads, with the 110.9% lead in Passmark GPU Compute being the standout. This makes it the clear choice for tasks like rendering, scientific simulation, or any workload that exercises raw FP32 throughput. The 85.2% lead in 3DMark Steel Nomad DX12 indicates strong performance in modern game engines that use DirectX 12 with heavy geometry and effects. The consistent 50% to 64% margins across DX9, DX10, DX11, and DX12 suggest the desktop card will simply outperform in most gaming scenarios regardless of API.

The RTX 3070 Mobile's lone victory in Geekbench Vulkan is notable but narrow in scope. With a score of 86,768 versus 33,620, the mobile part is more than 2.5 times faster in this specific test. This could point to a scenario where the mobile chip's driver stack is better optimized for Vulkan, or where the test's workload characteristics favor the mobile memory layout. However, this single result does not offset the desktop card's wins in OpenCL and all Passmark tests. The mobile part's average score of 20,534 places it in the 65th percentile, just three points below the desktop's 68th percentile, indicating that in the broader database context, both are mid-to-high performers.

For gaming, the RTX 3080 is the safer bet given its 3DMark and Passmark DX12 results. For compute, the Passmark GPU Compute delta is decisive. The mobile part's Vulkan win is interesting for developers targeting that API specifically, but it is an isolated data point. The desktop card's 64 GB/s bandwidth advantage and higher shading unit count suggest it will handle texture-heavy scenes and large compute kernels more gracefully.

FAQ

Q: Which GPU wins the majority of head-to-head benchmarks?

A: The NVIDIA GeForce RTX 3080 wins 9 out of 10 head-to-head benchmarks against the RTX 3070 Mobile. The only exception is Geekbench Vulkan, where the mobile part wins by a delta of -61.3%.

Q: What is the largest performance gap between the two cards?

A: The largest gap is in Passmark GPU Compute, where the RTX 3080 scores 14,397 versus 6,827 for the RTX 3070 Mobile, a delta of 110.9% in favor of the desktop card.

Q: How do the memory types and bandwidth compare?

A: The RTX 3080 uses 10 GB of GDDR6X on a 320-bit bus with 760.3 GB/s bandwidth. The RTX 3070 Mobile uses 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s. The desktop card has a 69.7% bandwidth advantage.

Q: What is the FP32 compute difference?

A: The RTX 3080 delivers 29.77 TFLOPS of FP32 performance, while the RTX 3070 Mobile delivers 15.97 TFLOPS. This is a ratio of approximately 1.86:1 in favor of the desktop card.

Q: Are there any benchmarks where the RTX 3070 Mobile is faster?

A: Yes, in Geekbench Vulkan the RTX 3070 Mobile scores 86,768 versus 33,620 for the RTX 3080, a delta of -61.3%. This is the only head-to-head test the mobile part wins.

Q: What are the nearest rivals for each card based on average score?

A: For the RTX 3080, the nearest rivals are the NVIDIA P106-100 (-0.3%), AMD Radeon Pro Vega 16 (-0.3%), AMD Radeon RX 6600M (-0.4%), and AMD Radeon R9 M290X (-0.4%). For the RTX 3070 Mobile, the nearest rivals are the Intel Arc B570 (-0.1%), Intel Arc A750 (-0.2%), NVIDIA Quadro M4000M (+0.3%), and AMD Radeon R9 M390X (-0.6%).

Specification Differences

| Field | RTX 3080 | RTX 3070 Mobile |

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

| Chip | GA102 | GA104 |

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

| Die Size | 628 mm² | 392 mm² |

| Transistor Density | 45.1M / mm² | 44.4M / mm² |

| Base Clock | 1440 MHz | 1110 MHz |

| Boost Clock | 1710 MHz | 1560 MHz |

| Memory Clock | 1188 MHz (19 Gbps effective) | 1750 MHz (14 Gbps effective) |

| Memory Size | 10 GB | 8 GB |

| Memory Type | GDDR6X | GDDR6 |

| Memory Bus | 320 bit | 256 bit |

| Bandwidth | 760.3 GB/s | 448.0 GB/s |

| Shading Units | 8704 | 5120 |

| TMUs | 272 | 160 |

| ROPs | 96 | 80 |

| RT Cores | 68 | 40 |

| Tensor Cores | 272 | 160 |

| Pixel Rate | 164.2 GPixel/s | 124.8 GPixel/s |

| Texture Rate | 465.1 GTexel/s | 249.6 GTexel/s |

| FP32 | 29.77 TFLOPS | 15.97 TFLOPS |

| TDP | 320 W | 115 W |

| Slot Width | Dual-slot | None |

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

| Suggested PSU | 700 W | None |

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

| Dimensions | 285 mm x 112 mm x 40 mm | None |

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

| Predecessor | GeForce 20 | GeForce 20 Mobile |

| Successor | GeForce 40 | None |

| Launch MSRP | 699 USD | None |

The Verdict

The data presents a clear picture: the RTX 3080 is the stronger GPU in nearly every measurable way. Its wins in 9 of 10 benchmarks, including an 85.2% margin in 3DMark Steel Nomad and a 110.9% margin in compute, establish it as the superior choice for gaming and compute alike. The desktop card's higher transistor count, larger die, wider memory bus, and 29.77 TFLOPS FP32 throughput all contribute to this dominance. The RTX 3070 Mobile's 15.97 TFLOPS and 448 GB/s bandwidth are respectable for a laptop part, but they fall short of the desktop card's capabilities.

The single Vulkan win for the mobile part is intriguing but should not be overweighed. It may indicate an optimization quirk or driver advantage in that specific API, but the OpenCL and Passmark results suggest the desktop card is otherwise faster across the board. For users who prioritize raw performance, the RTX 3080 is the clear pick. For those who need a mobile solution, the RTX 3070 Mobile still offers solid 65th percentile performance, but the benchmark data shows it cannot match the desktop card's output. The RTX 3080's launch MSRP of 699 USD reflects its desktop positioning, while the mobile part has no listed MSRP. Ultimately, the choice depends on the use case: desktop builders should choose the RTX 3080, while laptop buyers should accept the RTX 3070 Mobile's lower scores as the tradeoff for portability.