NVIDIA GeForce RTX 2070 SUPER vs NVIDIA GeForce RTX 3070 Mobile Comparison

The NVIDIA GeForce RTX 3070 Mobile and NVIDIA GeForce RTX 2070 SUPER represent two distinct generations of NVIDIA graphics, and the benchmark data reveals a fascinating split. The RTX 3070 Mobile, built on the newer Ampere architecture, decisively wins the modern DirectX 12 3DMark test. Conversely, the older Turing-based RTX 2070 SUPER, a desktop part, dominates nearly every legacy PassMark test. The data shows a clear generational and market-segment divide, with the mobile part winning 2 of the 10 head-to-head benchmarks and the desktop part winning 8.

Head-to-Head Benchmarks

The most significant victory for the RTX 3070 Mobile comes in the 3DMark Steel Nomad DX12 test, a modern, API-heavy workload. Here, the mobile GPU scores 2380, a massive 44.2% ahead of the RTX 2070 SUPER’s 1651. This is the largest delta in the entire comparison and signals a major advantage in current-generation gaming engines that leverage DirectX 12 Ultimate features. The RTX 3070 Mobile also wins in Geekbench OpenCL, scoring 92939 compared to 83358, a more modest but still substantial 11.5% margin. This indicates a strong lead in general-purpose compute that is not tied to a specific graphics API.

The RTX 2070 SUPER, however, demonstrates its strengths in a variety of DirectX legacy tests. In PassMark DirectX 9, it scores 223, which is 28.3% higher than the RTX 3070 Mobile’s 160. This is the second-largest delta in the comparison. It also holds a significant edge in PassMark G3D, scoring 18169 versus 15309, a 15.7% difference. The desktop card continues its winning streak with an 8.6% lead in PassMark DirectX 11 (151 vs. 138) and a 9.7% lead in PassMark GPU Compute (7557 vs. 6827). In other tests, the margins are smaller but consistent: the RTX 2070 SUPER leads by 14.4% in PassMark DirectX 10, 4.5% in PassMark DirectX 12, and 27% in the PassMark G2D test, which measures 2D graphics performance.

The overall average benchmark scores are nearly identical, with the RTX 3070 Mobile averaging 20534 and the RTX 2070 SUPER averaging 20282. This places them in the same performance tier, with the RTX 3070 Mobile’s nearest rivals including the Intel Arc B570 (average 20556) and the RTX 2070 SUPER itself (average 20282). The RTX 2070 SUPER’s nearest rivals list confirms this near-equivalence, showing a -1.2% delta relative to the RTX 3070 Mobile. Despite the RTX 2070 SUPER winning more individual tests, the overall average is essentially a tie, underscoring how the results are heavily dependent on the specific workload. The RTX 3070 Mobile’s 44.2% win in one test is offset by the RTX 2070 SUPER’s consistent, if smaller, wins across many others.

Where Each One Wins

The RTX 3070 Mobile is the clear winner for modern, next-generation gaming and compute workloads. Its dominant 3DMark Steel Nomad DX12 score suggests it is better optimized for current-generation game engines and DirectX 12 Ultimate features. The 11.5% victory in Geekbench OpenCL further cements its position for applications that utilize general-purpose GPU computing, such as content creation or scientific simulations that rely on OpenCL. If the workload is modern and API-intensive, the data indicates the RTX 3070 Mobile is the superior choice.

The RTX 2070 SUPER, on the other hand, is the winner for a vast library of existing games and applications. Its 28.3% lead in PassMark DirectX 9 is particularly notable, as many older titles and e-sports games still rely on this API. The 15.7% win in PassMark G3D, a synthetic overall 3D graphics score, suggests a more well-rounded performance in a variety of 3D tasks. The RTX 2070 SUPER also wins in compute with a 9.7% lead in PassMark GPU Compute, which is a different compute benchmark than Geekbench OpenCL. For users with a library of older games or applications that use DirectX 10, 11, or 9, the RTX 2070 SUPER’s benchmark results show a consistent advantage. The 27% lead in PassMark G2D, while less relevant for gaming, indicates a significant edge in 2D desktop rendering tasks.

Architecture Differences

The fundamental difference between these two GPUs lies in their underlying architectures. The RTX 3070 Mobile is built on the Ampere architecture, using the GA104 chip fabricated on an 8 nm process by Samsung. In contrast, the RTX 2070 SUPER uses the Turing architecture, with the TU104 chip on a 12 nm process from TSMC. This process difference is significant, as the 8 nm node allows for a much higher transistor density of 44.4M / mm² on the RTX 3070 Mobile, compared to 25.0M / mm² on the RTX 2070 SUPER. The RTX 3070 Mobile packs 17,400 million transistors into a 392 mm² die, while the RTX 2070 SUPER has 13,600 million transistors on a larger 545 mm² die.

These architectural changes also affect core composition. The RTX 3070 Mobile has 5120 shading units, double the 2560 found on the RTX 2070 SUPER. Both have 160 TMUs, but the RTX 3070 Mobile has 80 ROPs compared to 64 on the RTX 2070 SUPER. Ray tracing cores are identical in count at 40, but the Tensor core configuration differs: the RTX 3070 Mobile has 160 Tensor cores, while the RTX 2070 SUPER has 320. This is a notable inversion, as the newer card has fewer Tensor cores. The FP32 compute performance is 15.97 TFLOPS for the RTX 3070 Mobile, significantly higher than the 9.062 TFLOPS of the RTX 2070 SUPER. For FP16, the RTX 3070 Mobile operates at a 1:1 ratio (15.97 TFLOPS), while the RTX 2070 SUPER operates at a 2:1 ratio, providing 18.12 TFLOPS.

Specification Differences

Aside from the core architectural changes, several key specifications differ between the two cards. The most obvious is the process node, with the RTX 3070 Mobile using 8 nm and the RTX 2070 SUPER using 12 nm. The bus interface also differs, with the RTX 3070 Mobile supporting PCIe 4.0 x16, a newer standard than the PCIe 3.0 x16 on the RTX 2070 SUPER. The power profile is drastically different: the RTX 3070 Mobile has a TDP of 115 W and requires no power connectors, reflecting its mobile design. The RTX 2070 SUPER, a desktop card, has a TDP of 215 W and requires a 1x 6-pin + 1x 8-pin power connector setup. The RTX 2070 SUPER also has a suggested PSU of 550 W, while the RTX 3070 Mobile has none listed.

Clock speeds are another point of divergence. The RTX 3070 Mobile has a base clock of 1110 MHz and a boost clock of 1560 MHz. The RTX 2070 SUPER has a much higher base clock of 1605 MHz and a boost clock of 1770 MHz. This higher clock speed helps the Turing card in many benchmarks, despite its lower core count. The memory configuration is identical in size, type, bus width, and bandwidth: both use 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth and a memory clock of 1750 MHz (14 Gbps effective). The display outputs differ, with the RTX 3070 Mobile being listed as "Portable Device Dependent," while the RTX 2070 SUPER offers a fixed set of outputs including 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C. The RTX 2070 SUPER is a physical, dual-slot card with dimensions of 267 mm in length, 116 mm in height, and 35 mm in width. The RTX 3070 Mobile does not have listed dimensions, as it is designed to be integrated into laptops. The RTX 2070 SUPER launched on 2019-07-08 with a launch MSRP of 499 USD, while the RTX 3070 Mobile was released later on 2021-01-11.

FAQ

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

A: The NVIDIA GeForce RTX 3070 Mobile is significantly faster, scoring 2380 compared to the RTX 2070 SUPER’s 1651, a 44.2% advantage.

Q: How do the two GPUs compare in overall average benchmark score?

A: They are effectively tied. The RTX 3070 Mobile has an average benchmark score of 20534, while the RTX 2070 SUPER has an average of 20282. The RTX 2070 SUPER’s nearest rivals list shows a -1.2% delta against the RTX 3070 Mobile.

Q: Does the RTX 2070 SUPER win any benchmarks?

A: Yes, the RTX 2070 SUPER wins 8 of the 10 head-to-head benchmarks, including PassMark DirectX 9 (223 vs. 160), PassMark G3D (18169 vs. 15309), and PassMark GPU Compute (7557 vs. 6827).

Q: What are the key architectural differences between the two cards?

A: The RTX 3070 Mobile uses the Ampere architecture (GA104 chip) on an 8 nm Samsung process, while the RTX 2070 SUPER uses the Turing architecture (TU104 chip) on a 12 nm TSMC process. The RTX 3070 Mobile has 5120 shading units and 80 ROPs, compared to 2560 shading units and 64 ROPs on the RTX 2070 SUPER.

Q: Do the two GPUs have the same memory configuration?

A: Yes, the memory specifications are identical. Both have 8 GB of GDDR6 memory on a 256-bit bus, with a bandwidth of 448.0 GB/s and a memory clock of 1750 MHz (14 Gbps effective).

Q: Which GPU has a higher TDP?

A: The NVIDIA GeForce RTX 2070 SUPER has a much higher TDP of 215 W, while the NVIDIA GeForce RTX 3070 Mobile has a TDP of 115 W. The desktop card also requires a 1x 6-pin + 1x 8-pin power connector, whereas the mobile card requires none.