NVIDIA GeForce RTX 2070 SUPER vs NVIDIA GeForce RTX 2080 Comparison

Head-to-Head Benchmarks

The benchmark data is unambiguous: the RTX 2080 wins every single recorded head-to-head test against the RTX 2070 SUPER, with a clean 10-0 sweep. The largest margin comes in Geekbench Vulkan, where the RTX 2080 posts 107,797 points against the 2070 SUPER's 90,637, a substantial 18.9% advantage. This is the widest gap in the entire comparison and highlights a meaningful difference in compute-heavy API workloads.

The next biggest win is in Geekbench OpenCL, where the RTX 2080 scores 91,313 versus 83,358, a 9.5% lead. This confirms the pattern: the RTX 2080's higher raw compute throughput translates directly into better results in general-purpose GPU workloads. In the 3DMark Steel Nomad DX12 test, the RTX 2080 scores 1752 against 1651, a 6.1% edge, showing that the lead persists even in modern, API-heavy gaming workloads.

The remaining wins are smaller but consistent. In Passmark DirectX 12, the RTX 2080 leads 72 to 67, a 7.5% difference. DirectX 11 shows a 4.6% gap (158 vs 151), while DirectX 10 and DirectX 9 are closer: 136 vs 132 (3%) and 223 vs 223 (0%), respectively. The 3% lead in Passmark G3D (18,720 vs 18,169) and the 4.2% edge in GPU Compute (7,872 vs 7,557) round out the picture. The G2D test shows 907 vs 878, a 3.3% win.

Notably, the RTX 2080 also holds a higher overall average benchmark score of 22,895 compared to 20,282 for the 2070 SUPER. The percentile placement reflects this: the RTX 2080 sits at the 68th percentile across all GPUs, while the 2070 SUPER sits at the 65th. The delta between them is roughly 12.9% in average score, which is larger than any single head-to-head test except the Vulkan result. That suggests the RTX 2080's advantage is broad-based rather than concentrated in one specific workload.

The closest rival comparisons in the database put these cards into context. The RTX 2080's nearest rivals include the Intel Arc B580 (23021 avg, -0.5% delta), the AMD Radeon RX 580 2048SP (23061 avg, -0.7%), the RTX 4060 Mobile (22729 avg, +0.7%), and the RTX 3080 (23172 avg, -1.2%). The 2070 SUPER's nearest rivals include the Quadro M4000M (20480 avg, -1%), the RTX 3070 Mobile (20534 avg, -1.2%), the Intel Arc B570 (20556 avg, -1.3%), and the Intel Arc A750 (20582 avg, -1.5%). In other words, the 2070 SUPER sits in a lower performance tier, and the RTX 2080 is positioned closer to the RTX 3080 and Arc B580 class of cards.

FAQ

Q: Which card wins in Vulkan compute workloads?

A: The RTX 2080 wins decisively. In Geekbench Vulkan, it scores 107,797 against the 2070 SUPER's 90,637, a 18.9% advantage. This is the single largest margin in any head-to-head test between the two.

Q: Is the RTX 2070 SUPER competitive in any benchmark?

A: No. The recorded data shows the RTX 2080 winning all 10 head-to-head tests. The closest result is Passmark DirectX 9, where both cards tie at 223 points. The 2070 SUPER does not win any test outright.

Q: How do the average benchmark scores compare?

A: The RTX 2080 has an average benchmark score of 22,895, while the RTX 2070 SUPER averages 20,282. That is a gap of roughly 2,613 points, or about 12.9% in favor of the RTX 2080.

Q: What does the percentile ranking tell us?

A: The RTX 2080 ranks at the 68th percentile of all GPUs, while the 2070 SUPER ranks at the 65th. This means the RTX 2080 sits in a higher performance class relative to the entire GPU landscape.

Q: Are these cards similar in memory configuration?

A: Yes. Both have 8 GB of GDDR6 memory on a 256-bit bus, with 448.0 GB/s bandwidth and 1750 MHz memory clock (14 Gbps effective). Memory is not a differentiator between them.

Q: What about power and board design?

A: Both cards share the same TDP of 215 W, use a dual-slot design, require the same 1x 6-pin + 1x 8-pin power connectors, and list a suggested PSU of 550 W. The physical dimensions are identical: 267 mm length, 116 mm height, 35 mm width.

Architecture Differences

Both cards are built on the same Turing architecture, using the TU104 chip fabricated at TSMC on a 12 nm process. The die size is identical at 545 mm², and both pack 13,600 million transistors, yielding a transistor density of 25.0M / mm². The architectural family, GeForce 20-series, is the same, and both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither card has a codename listed in the database.

The key architectural difference lies in the execution resources. The RTX 2080 has 2,944 shading units, 184 texture mapping units (TMUs), and 64 raster output units (ROPs). The RTX 2070 SUPER has fewer shading units at 2,560, fewer TMUs at 160, but the same 64 ROPs. This means the RTX 2080 has roughly 15% more shading units and 15% more TMUs, which directly contributes to its higher fillrate and compute throughput.

Ray tracing and tensor core counts also differ. The RTX 2080 carries 46 RT cores and 368 tensor cores, while the RTX 2070 SUPER has 40 RT cores and 320 tensor cores. This is a 15% and 15% difference, respectively, consistent with the shading unit gap. These resources matter for ray-traced workloads and DLSS-style tensor operations, though the database does not include specific RT or tensor benchmarks for these cards.

Clock speeds tell a more nuanced story. The RTX 2070 SUPER actually runs higher clocks: base 1605 MHz and boost 1770 MHz, versus the RTX 2080's 1515 MHz base and 1710 MHz boost. Despite this clock advantage, the RTX 2080 still wins every benchmark, which indicates that its larger execution resource pool (more cores and TMUs) overcomes the lower clock speeds. The 2070 SUPER's higher clocks are not enough to close the gap.

Pixel and texture rates reflect this. The RTX 2080 achieves 109.4 GPixel/s and 314.6 GTexel/s, while the 2070 SUPER posts 113.3 GPixel/s and 283.2 GTexel/s. The 2070 SUPER actually has a slight pixel rate advantage (113.3 vs 109.4), but the RTX 2080 dominates in texture rate (314.6 vs 283.2) and in FP32 throughput: 10.07 TFLOPS vs 9.062 TFLOPS. FP16 performance scales similarly: 20.14 TFLOPS (2:1) for the RTX 2080 versus 18.12 TFLOPS (2:1) for the 2070 SUPER.

Both cards are end-of-life products, with the RTX 2080 released on 2018-09-19 and the RTX 2070 SUPER released on 2019-07-08. Both share the same predecessor (GeForce 10) and successor (GeForce 30) in the database. The launch MSRP for the RTX 2080 was 699 USD, and for the RTX 2070 SUPER it was 499 USD.

Specification Differences

The following fields differ between the two cards:

• Base clock: RTX 2080 at 1515 MHz, RTX 2070 SUPER at 1605 MHz
• Boost clock: RTX 2080 at 1710 MHz, RTX 2070 SUPER at 1770 MHz
• Shading units: 2944 vs 2560
• TMUs: 184 vs 160
• RT cores: 46 vs 40
• Tensor cores: 368 vs 320
• Pixel rate: 109.4 GPixel/s vs 113.3 GPixel/s
• Texture rate: 314.6 GTexel/s vs 283.2 GTexel/s
• FP32: 10.07 TFLOPS vs 9.062 TFLOPS
• FP16: 20.14 TFLOPS (2:1) vs 18.12 TFLOPS (2:1)
• Release date: 2018-09-19 vs 2019-07-08
• Launch MSRP: 699 USD vs 499 USD

All other hardware specifications are identical: 8 GB GDDR6, 256-bit bus, 448.0 GB/s bandwidth, 64 ROPs, 215 W TDP, dual-slot width, same power connectors, same suggested PSU, same bus interface (PCIe 3.0 x16), same display outputs (1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C), and same dimensions (267 mm x 116 mm x 35 mm). Both are listed as end-of-life.

The Verdict

The data is clear: the RTX 2080 is the faster card in every measured category. It wins all 10 head-to-head benchmarks, has a higher average score (22,895 vs 20,282), and sits at a higher percentile (68th vs 65th). The RTX 2070 SUPER's higher clock speeds do not compensate for its reduced core count, TMUs, RT cores, and tensor cores. The RTX 2080's FP32 throughput advantage of 10.07 TFLOPS versus 9.062 TFLOPS is the underlying reason it pulls ahead in compute-heavy tests like Vulkan and OpenCL.

However, the RTX 2070 SUPER is not a weak card. It shares the same memory subsystem (8 GB GDDR6, 448.0 GB/s) and the same power envelope (215 W). It also has a slightly higher pixel rate (113.3 GPixel/s) and boost clock (1770 MHz), which could matter in specific pixel-bound scenarios. But the benchmark record shows no test where that translates into a win.

For a buyer choosing between these two, the RTX 2080 is the straightforward pick if raw performance is the only criterion. The 2070 SUPER's lower launch MSRP (499 USD vs 699 USD) indicates a different market position, but the performance gap is consistent across all workloads. The RTX 2080 is also positioned closer to higher-tier rivals like the RTX 3080 in the database, while the 2070 SUPER sits near mid-range cards like the Intel Arc A750.

Where Each One Wins

RTX 2080 wins in:

• Vulkan compute (18.9% lead)
• OpenCL compute (9.5% lead)
• DirectX 12 workloads (7.5% lead)
• DirectX 11 workloads (4.6% lead)
• GPU compute (4.2% lead)
• 2D graphics (3.3% lead)
• DirectX 10 (3% lead)
• DirectX 9 (tie, 0%)
• 3D graphics (3% lead)
• Overall average benchmark score (22,895 vs 20,282)

RTX 2070 SUPER wins in:

• No recorded benchmark wins. The card does, however, have a higher base clock, boost clock, and pixel rate, which are specification advantages but do not translate into test victories in the database.

For use cases, the RTX 2080 is the better choice for any workload that leverages compute parallelism, such as OpenCL-based rendering, Vulkan gaming, or DirectX 12 titles. The 2070 SUPER's higher pixel rate suggests it could be competitive in fill-rate-limited scenarios, but no benchmark in the record confirms this. The RTX 2080's larger RT core and tensor core counts also position it better for ray tracing and DLSS-style features, though no specific RT benchmark scores are recorded. In practical terms, the RTX 2080 is the card to pick for maximum performance across the board, while the RTX 2070 SUPER should only be considered if the lower launch MSRP matters more than the measured performance gap.