NVIDIA GeForce RTX 2080 vs NVIDIA GeForce RTX 5060 Comparison

NVIDIA’s GeForce RTX 5060 and RTX 2080 occupy very different moments in the company’s history, yet they are separated by more than just time. The recorded data shows a clear generational shift, with the RTX 5060 winning 9 of the 10 head-to-head benchmark comparisons. The RTX 2080’s sole victory comes in a legacy DirectX 10 test, a narrow 6.6% advantage that does little to offset the newer card’s dominance elsewhere. This is a straightforward case of modern architecture and efficiency overcoming an older, larger design.

Where Each One Wins

The RTX 5060 is the obvious choice for any modern workload that stresses current graphics APIs. Its win in 3DMark Steel Nomad, a DirectX 12 test, is the single largest margin in the comparison at 107.1% over the RTX 2080. The data also shows substantial leads in compute-oriented tasks: the RTX 5060 scores 38.5% higher in PassMark GPU Compute and 23.5% higher in Geekbench OpenCL. For gaming, the PassMark DirectX 11 result shows a 26.6% advantage, and even DirectX 12 shows a 6.9% lead. The RTX 5060 also excels in general 2D performance, with a 27.2% higher PassMark G2D score, which suggests better desktop compositing and window management.

The RTX 2080’s single win is in PassMark DirectX 10, where it scores 136 versus 127, a 6.6% margin. This is a legacy API that sees little use in current titles. The RTX 2080 also manages a nearly identical result in DirectX 9, losing by only 0.9%, and its PassMark G3D score is only 11.6% behind, indicating that in some older or lighter 3D loads the gap narrows. However, these are isolated bright spots. The RTX 2080’s average benchmark score of 22895 places it at the 68th percentile of all GPUs, while the RTX 5060’s average of 26331 puts it at the 72nd percentile. The newer card is not just faster in headline tests; it is consistently better across the board.

Architecture Differences

The two cards are built on fundamentally different foundations. The RTX 5060 uses the GB206 chip on the Blackwell 2.0 architecture, manufactured on a 5 nm process at TSMC. In contrast, the RTX 2080 uses the TU104 chip on the older Turing architecture, built on a 12 nm process. This node shrink is the primary driver of the performance and efficiency gap. The RTX 5060 packs 21,900 million transistors onto a die of 181 mm², yielding a transistor density of 121.0 million per mm². The RTX 2080 has fewer transistors overall, 13,600 million, but spreads them over a much larger 545 mm² die, resulting in only 25.0 million transistors per mm². The RTX 5060 is a smaller, denser chip that does more with less silicon.

Memory configurations are similar in capacity but different in technology. Both cards have 8 GB of VRAM, but the RTX 5060 uses GDDR7 with a 128-bit bus, while the RTX 2080 uses GDDR6 with a 256-bit bus. Interestingly, both achieve the exact same bandwidth of 448.0 GB/s. This means the RTX 5060 achieves the same memory throughput with half the bus width, evidence of the higher effective memory speed of GDDR7. The RTX 5060’s memory runs at 1750 MHz with 28 Gbps effective, while the RTX 2080’s runs at 1750 MHz with 14 Gbps effective.

The core configurations also tell a story. The RTX 5060 has 3840 shading units, 120 TMUs, and 48 ROPs. The RTX 2080 has 2944 shading units, 184 TMUs, and 64 ROPs. The RTX 2080 has more texture units and ROPs, but the RTX 5060 compensates with more shading units and higher clocks. The RTX 5060 boosts to 2497 MHz versus the RTX 2080’s 1710 MHz. This clock advantage is why the RTX 5060 achieves 19.18 TFLOPS of FP32 performance, nearly double the RTX 2080’s 10.07 TFLOPS. In FP16, the RTX 2080 reaches 20.14 TFLOPS with a 2:1 ratio, but the RTX 5060 matches its FP32 rate at 19.18 TFLOPS with a 1:1 ratio.

Head-to-Head Benchmarks

The most decisive result is in 3DMark Steel Nomad. The RTX 5060 scores 3628, which is 107.1% higher than the RTX 2080’s 1752. This is a massive lead that indicates the newer architecture handles modern DirectX 12 workloads with far greater efficiency. In compute, the RTX 5060’s PassMark GPU Compute score of 10899 is 38.5% ahead of the RTX 2080’s 7872. The Geekbench OpenCL result shows a similar trend, with the RTX 5060 at 112787 versus 91313, a 23.5% advantage. These are not marginal improvements; they represent a different performance class.

The DirectX 11 test shows the RTX 5060 scoring 200 versus 158, a 26.6% lead. This is the most relevant result for a large library of current games. The PassMark G3D score, a broad measure of 3D capability, has the RTX 5060 at 20891 and the RTX 2080 at 18720, an 11.6% difference. Even in Vulkan, where the RTX 2080 performs relatively well, the RTX 5060 still leads by 5.1%, scoring 113321 to 107797. The only reverse result is in DirectX 10, where the RTX 2080 scores 136 and the RTX 5060 scores 127, a 6.6% gap in favor of the older card. The DirectX 9 test is nearly a tie, with the RTX 5060 winning by just 0.9% (225 to 223). DirectX 12 also favors the RTX 5060 by 6.9% (77 to 72).

FAQ

Q: Is the RTX 5060 faster than the RTX 2080 in modern games?

A: Yes. The RTX 5060 wins the DirectX 12 Steel Nomad test by 107.1% and the DirectX 11 test by 26.6%. These are the most relevant APIs for current titles.

Q: What is the main architectural difference between the two?

A: The RTX 5060 uses the Blackwell 2.0 architecture on a 5 nm process, while the RTX 2080 uses Turing on a 12 nm process. The RTX 5060 has a much higher transistor density at 121.0M per mm² versus 25.0M per mm².

Q: Do both cards have the same memory bandwidth?

A: Yes, both achieve 448.0 GB/s. The RTX 5060 does this with GDDR7 on a 128-bit bus, while the RTX 2080 uses GDDR6 on a 256-bit bus.

Q: In which benchmark does the RTX 2080 win?

A: The RTX 2080 wins only in PassMark DirectX 10, scoring 136 versus 127, a 6.6% advantage. It nearly ties in DirectX 9, losing by just 0.9%.

Q: How do their compute performances compare?

A: The RTX 5060 leads by 38.5% in PassMark GPU Compute and by 23.5% in Geekbench OpenCL. Its FP32 throughput is 19.18 TFLOPS versus 10.07 TFLOPS for the RTX 2080.

Q: Which card has a higher average benchmark score?

A: The RTX 5060 has an average score of 26331, placing it at the 72nd percentile. The RTX 2080 averages 22895, at the 68th percentile.

The Verdict

The data is unambiguous. The RTX 5060 is the superior card for anyone building a system today. It wins 9 of 10 benchmarks, with the only loss in a legacy API that is irrelevant for modern software. Its leads are largest in the most demanding tests: 107.1% in 3DMark Steel Nomad and 38.5% in compute. The RTX 2080’s average score is 13% lower, and its percentile ranking is 4 points behind. If you are choosing between these two, the RTX 5060 offers a decisive performance advantage across all meaningful workloads.

The RTX 2080 should only be considered if you have a very specific need for its legacy DirectX 10 performance, which is unlikely. Its higher power draw of 215 W versus 145 W, and its suggested 550 W PSU versus 300 W, further reinforce that it is an older, less efficient design. The RTX 5060 achieves its results with a smaller die, fewer transistors, and lower power consumption. For any practical purpose, the RTX 5060 is the better investment in performance and efficiency.

Specification Differences

The two cards differ significantly in their core specifications. The RTX 5060 has a base clock of 2280 MHz and a boost clock of 2497 MHz, while the RTX 2080 has a base clock of 1515 MHz and a boost of 1710 MHz. The RTX 5060 features 3840 shading units, 120 TMUs, and 48 ROPs, compared to the RTX 2080’s 2944 shading units, 184 TMUs, and 64 ROPs. The RTX 5060 has 30 RT cores and 120 tensor cores, while the RTX 2080 has 46 RT cores and 368 tensor cores. The RTX 5060’s FP32 performance is 19.18 TFLOPS, with FP16 at the same rate, while the RTX 2080 achieves 10.07 TFLOPS FP32 and 20.14 TFLOPS FP16. The RTX 5060 uses GDDR7 memory on a 128-bit bus, while the RTX 2080 uses GDDR6 on a 256-bit bus, both with 448.0 GB/s bandwidth. The RTX 5060 has a TDP of 145 W and requires a 300 W PSU, while the RTX 2080 has a TDP of 215 W and needs a 550 W PSU. The RTX 5060 uses a PCIe 5.0 x8 interface, while the RTX 2080 uses PCIe 3.0 x16. The RTX 5060 has a die size of 181 mm² and 21,900 million transistors, while the RTX 2080 has a die size of 545 mm² and 13,600 million transistors. The RTX 5060’s launch MSRP is 299 USD.