NVIDIA GeForce RTX 2070 SUPER vs NVIDIA GeForce RTX 2080 SUPER Comparison

FAQ

Q: Which GPU is faster in the recorded benchmark suite, the RTX 2080 SUPER or the RTX 2070 SUPER?

A: The RTX 2080 SUPER wins all 10 head-to-head benchmarks recorded in the database, with a perfect 10-0 record over the RTX 2070 SUPER.

Q: How much performance gap separates the two cards in the most demanding DirectX 12 test?

A: In the 3DMark Steel Nomad DX12 benchmark, the RTX 2080 SUPER scores 1882 versus 1651 for the RTX 2070 SUPER, a 14% advantage.

Q: What is the largest single-benchmark margin between the two cards?

A: The largest margin appears in Geekbench Vulkan, where the RTX 2080 SUPER scores 111284 against 90637, a 22.8% lead.

Q: Are the two cards built on the same architecture and chip?

A: Yes, both use the Turing architecture on the TU104 chip, fabricated at TSMC on a 12 nm process with 13,600 million transistors and a 545 mm² die.

Q: Do the cards share the same memory configuration?

A: They both have 8 GB of GDDR6 on a 256-bit bus, but the RTX 2080 SUPER runs at 15.5 Gbps effective for 495.9 GB/s bandwidth, while the RTX 2070 SUPER runs at 14 Gbps for 448.0 GB/s.

Q: How do their overall average benchmark scores compare?

A: The RTX 2080 SUPER averages 24170 across all recorded benchmarks, while the RTX 2070 SUPER averages 20282, a gap that places the former in the 69th percentile versus the latter's 65th percentile of all GPUs.

Architecture Differences

Both cards belong to the GeForce 20-series and share the same fundamental Turing architecture, TU104 chip, 12 nm TSMC process, 13,600 million transistors, and 545 mm² die size. The transistor density is identical at 25.0M per mm². The core configuration, however, diverges meaningfully. The RTX 2080 SUPER fields 3072 shading units, 192 texture mapping units, and 64 ROPs, while the RTX 2070 SUPER scales down to 2560 shading units, 160 TMUs, and the same 64 ROPs. Ray tracing and tensor core counts follow the same pattern: 48 RT cores and 384 tensor cores on the 2080 SUPER versus 40 RT cores and 320 tensor cores on the 2070 SUPER.

Clock speeds also differ. The 2080 SUPER has a base clock of 1650 MHz and a boost of 1815 MHz, while the 2070 SUPER operates at 1605 MHz base and 1770 MHz boost. These higher clocks, combined with the larger shader array, drive a substantial compute gap: FP32 throughput reaches 11.15 TFLOPS on the 2080 SUPER versus 9.062 TFLOPS on the 2070 SUPER. FP16 performance scales proportionally at 22.30 TFLOPS (2:1) versus 18.12 TFLOPS (2:1). Pixel fill rates are nearly identical at 116.2 GPixel/s versus 113.3 GPixel/s, but texture rate diverges more sharply at 348.5 GTexel/s versus 283.2 GTexel/s.

Memory architecture is another differentiating factor. Both use 8 GB of GDDR6 on a 256-bit bus, but the 2080 SUPER's memory clock of 1937 MHz (15.5 Gbps effective) yields 495.9 GB/s of bandwidth, a clear step above the 2070 SUPER's 1750 MHz (14 Gbps effective) and 448.0 GB/s. The TDP reflects the extra hardware: 250 W for the 2080 SUPER versus 215 W for the 2070 SUPER. The suggested PSU also scales accordingly, 600 W versus 550 W. Both cards are dual-slot, use the same 1x 6-pin plus 1x 8-pin power connectors, and share identical display outputs (1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C) and PCIe 3.0 x16 interface.

Where Each One Wins

The RTX 2080 SUPER wins in every recorded benchmark, so the use-case split is less about choosing a winner and more about understanding where the margin is largest or smallest. For gamers targeting high-refresh-rate 1440p or entry-level 4K, the 2080 SUPER's advantages in shading units, texture rate, and memory bandwidth make it the stronger choice, particularly in DirectX 12 workloads. The 3DMark Steel Nomad DX12 test shows a 14% lead, and DirectX 12 in Passmark shows 11.9%. Those gains translate directly to modern titles that leverage DX12's lower-level API access.

For compute-heavy tasks like OpenCL workloads, the 2080 SUPER's 11.15 TFLOPS FP32 throughput and higher tensor core count give it a 19% advantage in Geekbench OpenCL. Vulkan workloads show the largest gap at 22.8%, indicating that the 2080 SUPER's extra execution resources scale well with Vulkan's explicit multi-threading model. The RTX 2070 SUPER, meanwhile, remains a capable option for those prioritizing lower power draw (215 W versus 250 W) and a lower suggested PSU rating (550 W versus 600 W), while still delivering solid DirectX 9 and 2D performance where the margins are narrowest (1.8% and 4.8% respectively).

The 2070 SUPER also holds value in scenarios where the extra 35 W of power headroom matters, such as compact builds with limited thermal capacity, though it sacrifices 7.3% in Passmark G3D and 9.7% in GPU compute to achieve that efficiency. For users who prioritize raw rasterization and compute throughput above all else, the 2080 SUPER is the definitive pick; for those who want a slightly more power-frugal Turing card with the same feature set, the 2070 SUPER covers the essentials at a lower performance ceiling.

Specification Differences

| Specification | RTX 2080 SUPER | RTX 2070 SUPER |

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

| Base Clock | 1650 MHz | 1605 MHz |

| Boost Clock | 1815 MHz | 1770 MHz |

| Memory Clock | 1937 MHz (15.5 Gbps effective) | 1750 MHz (14 Gbps effective) |

| Memory Bandwidth | 495.9 GB/s | 448.0 GB/s |

| Shading Units | 3072 | 2560 |

| TMUs | 192 | 160 |

| RT Cores | 48 | 40 |

| Tensor Cores | 384 | 320 |

| Pixel Rate | 116.2 GPixel/s | 113.3 GPixel/s |

| Texture Rate | 348.5 GTexel/s | 283.2 GTexel/s |

| FP32 | 11.15 TFLOPS | 9.062 TFLOPS |

| FP16 | 22.30 TFLOPS (2:1) | 18.12 TFLOPS (2:1) |

| TDP | 250 W | 215 W |

| Suggested PSU | 600 W | 550 W |

| Launch MSRP | 699 USD | 499 USD |

The two cards share identical ROP counts (64), die size, transistor count, process node, memory size (8 GB GDDR6), bus width (256 bit), slot width, power connectors, display outputs, and API support (DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4). Both are end-of-life products with the same dimensions (267 mm x 116 mm x 35 mm) and were released within two weeks of each other in July 2019.

Head-to-Head Benchmarks

The database records 10 benchmark comparisons between these two GPUs, and the RTX 2080 SUPER wins every single one. The largest margin comes from Geekbench Vulkan, where the 2080 SUPER scores 111284 against 90637, a 22.8% advantage. This result highlights the 2080 SUPER's superior shader throughput and memory bandwidth in an API that scales well with raw compute resources. Geekbench OpenCL follows closely at 19% (99226 versus 83358), confirming that compute-heavy workloads favor the larger chip.

The 3DMark Steel Nomad DX12 benchmark, a demanding modern rasterization test, shows the 2080 SUPER leading 1882 to 1651, a 14% gap. This is a meaningful margin for a card that sits one tier higher in NVIDIA's lineup. Passmark's DirectX 12 test shows a similar pattern at 11.9% (75 versus 67), while DirectX 11 comes in at 9.3% (165 versus 151). These results suggest the 2080 SUPER's extra 512 shading units and 32 TMUs pay dividends across API generations, with the gap widening as APIs become more compute-intensive.

The smallest margins appear in legacy and 2D workloads. Passmark DirectX 9 shows only a 1.8% difference (227 versus 223), indicating that older APIs are less sensitive to the 2080 SUPER's additional resources. Passmark G2D shows 4.8% (920 versus 878), a modest gain for a card with higher clocks and more memory bandwidth. Passmark DirectX 10 sits at 6.1% (140 versus 132), and Passmark G3D at 7.3% (19490 versus 18169), the latter being a strong overall rasterization indicator. Passmark GPU Compute shows 9.7% (8290 versus 7557), reinforcing the 2080 SUPER's compute advantage.

The average benchmark score in the database tells the same story: the RTX 2080 SUPER averages 24170 across all tests, while the RTX 2070 SUPER averages 20282. The 2080 SUPER's nearest rivals in the database include the GTX 780 Ti (delta -0.3%), GTX 1630 (delta -0.4%), and RX 6800S (delta +0.4%), placing it in a tight performance cluster. The 2070 SUPER's nearest rivals include the Quadro M4000M (delta -1%), RTX 3070 Mobile (delta -1.2%), and Arc B570 (delta -1.3%), showing it competes in a lower bracket. The 2080 SUPER's 69th percentile versus the 2070 SUPER's 65th percentile confirms that the performance gap is consistent across the entire recorded benchmark suite, with no workload where the smaller card manages to close the deficit.