NVIDIA GeForce RTX 2080 vs NVIDIA GeForce RTX 3070 Mobile Comparison

Where Each One Wins

The benchmark split between these two NVIDIA GPUs is unusually clear-cut. The desktop RTX 2080 takes the majority of the recorded tests, winning 8 out of 10 head-to-head comparisons, while the RTX 3070 Mobile wins only 2. However, the nature of those wins matters more than the raw count. The RTX 3070 Mobile's victories are concentrated in modern, compute-heavy workloads, while the RTX 2080 dominates across legacy DirectX and 2D scenarios.

The RTX 3070 Mobile's strongest showing comes in 3DMark Steel Nomad DX12, where it scores 2380 versus the RTX 2080's 1752, a 26.4% advantage. This is the single largest margin in either direction and points to the mobile chip's architectural efficiency in current-generation rendering. The Geekbench OpenCL result also favors the mobile part, 92939 versus 91313, a modest 1.7% edge that confirms its compute lead in a general-purpose API.

Conversely, the RTX 2080 is the clear pick for anyone dealing with older DirectX titles or 2D-heavy workloads. Its Passmark DirectX 9 score of 223 beats the mobile chip's 160, a 39.4% gap. The 2D test is even more lopsided: 907 versus 641, a 41.5% difference. These are not marginal wins; they represent a generational shift in optimization priorities. The RTX 2080 also wins Passmark G3D (18720 versus 15309, 22.3% ahead), GPU Compute (7872 versus 6827, 15.3% ahead), and every other DirectX variant tested: DX10 (136 vs 113, 20.4% ahead), DX11 (158 vs 138, 14.5% ahead), and DX12 (72 vs 64, 12.5% ahead).

The data suggests the RTX 2080 is better suited to legacy software compatibility and rasterization-heavy tasks, while the RTX 3070 Mobile pulls ahead specifically when the workload taps into modern DX12 features or OpenCL compute. For users mixing old and new games, the desktop card's broader consistency may be more valuable, but for pure latest-generation performance, the mobile chip wins where it counts.

Architecture Differences

The architectural gap between these two GPUs is substantial, stemming from a node and design philosophy shift. The RTX 2080 uses the TU104 chip on TSMC's 12 nm process, packing 13,600 million transistors into a 545 mm² die. The RTX 3070 Mobile uses the GA104 chip on Samsung's 8 nm node, fitting 17,400 million transistors into a smaller 392 mm² die. This yields a transistor density of 25.0M per mm² for the desktop part versus 44.4M per mm² for the mobile chip, a clear sign of the newer process's efficiency.

The compute resources tell a different story. The RTX 3070 Mobile has 5120 shading units, far more than the RTX 2080's 2944, and its FP32 throughput of 15.97 TFLOPS exceeds the desktop card's 10.07 TFLOPS. Yet the RTX 2080 has more TMUs (184 versus 160) and its texture rate of 314.6 GTexel/s outpaces the mobile chip's 249.6 GTexel/s. The RTX 2080 also has more RT cores (46 versus 40) and more tensor cores (368 versus 160), though the mobile part's tensor cores are of a newer generation. The RTX 3070 Mobile compensates with a higher pixel rate (124.8 GPixel/s versus 109.4 GPixel/s) and more ROPs (80 versus 64).

Memory is identical in configuration: 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth and 14 Gbps effective speed. The power envelopes diverge sharply, with the RTX 2080 rated at 215 W TDP and the mobile chip at 115 W TDP, which explains the latter's lower boost clock of 1560 MHz versus 1710 MHz despite a higher theoretical compute ceiling. The RTX 2080 also uses a dual-slot design with 1x 6-pin and 1x 8-pin power connectors and a 550 W suggested PSU, while the RTX 3070 Mobile has no external power connectors and is portable-device dependent.

Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, but the bus interface differs: PCIe 4.0 x16 on the mobile chip versus PCIe 3.0 x16 on the desktop part. The RTX 2080 offers a full array of display outputs (1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C), while the mobile GPU's outputs are entirely device-dependent. The FP16 capability also diverges: the RTX 2080 runs at 20.14 TFLOPS with a 2:1 ratio, while the RTX 3070 Mobile runs at 15.97 TFLOPS with a 1:1 ratio, indicating different compute prioritization.

The Verdict

The recorded data points to a straightforward choice based on workload. The NVIDIA GeForce RTX 2080 is the better option for users who prioritize legacy DirectX performance, 2D acceleration, and consistent wins across a broad benchmark suite. Its 68th percentile ranking among all GPUs, combined with an average benchmark score of 22895, places it in a competitive tier alongside the Intel Arc B580 (23021, -0.5% delta) and just ahead of the NVIDIA GeForce RTX 4060 Mobile (22729, 0.7% delta). The RTX 2080's 24.2% Vulkan lead over the RTX 3070 Mobile and its 41.5% 2D advantage are decisive for desktop productivity and older game libraries.

The NVIDIA GeForce RTX 3070 Mobile, despite its lower 65th percentile and average score of 20534, is the more future-proof choice for modern DX12 titles. Its 26.4% lead in 3DMark Steel Nomad and 1.7% edge in OpenCL show that when software targets current APIs, the Ampere architecture's higher shading unit count and FP32 throughput win out. The mobile chip's nearest rivals include the Intel Arc B570 (20556, -0.1% delta) and Intel Arc A750 (20582, -0.2% delta), placing it in a tight mid-range pack. Its 115 W TDP also makes it the only one of these two viable for thin-and-light machines, though the database does not list dimensions for the mobile part.

For a desktop user with an existing PCIe 3.0 platform and a mix of old and new games, the RTX 2080's broader win profile is safer. For someone buying a laptop or building a system around PCIe 4.0 with a focus on the latest DX12 releases, the RTX 3070 Mobile's concentrated modern performance is the better bet. Neither card is universally superior; the data simply says the RTX 2080 is the more versatile legacy performer, while the RTX 3070 Mobile is the specialized modern one.

FAQ

Q: Which GPU wins in Vulkan performance?

A: The NVIDIA GeForce RTX 2080 scores 107797 in Geekbench Vulkan versus 86768 for the RTX 3070 Mobile, a 24.2% advantage.

Q: Is there any test where the RTX 3070 Mobile beats the RTX 2080 significantly?

A: Yes, in 3DMark Steel Nomad DX12, the RTX 3070 Mobile scores 2380 against the RTX 2080's 1752, a 26.4% lead. It also edges ahead in Geekbench OpenCL by 1.7%.

Q: How do these GPUs compare in raw compute benchmarks?

A: The RTX 2080 wins Passmark GPU Compute with 7872 points versus 6827, a 15.3% margin. The RTX 3070 Mobile's FP32 throughput is higher at 15.97 TFLOPS versus 10.07 TFLOPS, but this does not translate to a compute benchmark win.

Q: What are the memory specifications for both cards?

A: Both use 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth and 14 Gbps effective memory speed. There is no difference in memory configuration.

Q: Which GPU has a higher boost clock?

A: The RTX 2080 boosts to 1710 MHz, while the RTX 3070 Mobile boosts to 1560 MHz. The desktop part runs 150 MHz higher despite the mobile chip's newer architecture.

Q: How do their average benchmark scores compare to close rivals?

A: The RTX 2080 averages 22895, sitting 0.5% below the Intel Arc B580 and 0.7% above the RTX 4060 Mobile. The RTX 3070 Mobile averages 20534, nearly matching the Intel Arc B570 at -0.1% delta.

Head-to-Head Benchmarks

The largest win for the RTX 3070 Mobile is in 3DMark Steel Nomad DX12, where it posts 2380 points against the RTX 2080's 1752. That 26.4% delta is the single biggest swing in the entire comparison and highlights the Ampere architecture's strength in modern, geometry-heavy DX12 workloads. The mobile chip's second win, Geekbench OpenCL, is far narrower: 92939 versus 91313, only 1.7% ahead, suggesting a marginal compute advantage that does not scale to all APIs.

The RTX 2080 answers with its own dominant victories. The biggest is Passmark G2D, where 907 points dwarfs the mobile chip's 641, a 41.5% gap. This is followed closely by Passmark DirectX 9, with 223 versus 160, a 39.4% lead. These two tests alone show the desktop card's strength in legacy and 2D workloads, areas where the mobile chip's newer design offers no benefit. The RTX 2080 also takes Passmark G3D by a substantial 22.3% (18720 versus 15309), which is the more meaningful overall gaming metric.

In the middle range, the RTX 2080 wins Passmark DirectX 10 by 20.4% (136 versus 113) and DirectX 11 by 14.5% (158 versus 138). The DirectX 12 Passmark result is closer at 12.5% (72 versus 64), but still favors the desktop part. Geekbench Vulkan shows a 24.2% advantage for the RTX 2080 (107797 versus 86768), and Passmark GPU Compute reveals a 15.3% edge (7872 versus 6827). Across all ten tests, the RTX 2080's wins are consistently larger in percentage terms than the RTX 3070 Mobile's two victories, except for the Steel Nomad outlier. The data paints a clear picture: the RTX 2080 is the superior all-rounder in this matchup, while the RTX 3070 Mobile is a specialist that excels only in the newest DX12 compute scenarios.