Is the RTX 4080 better than the 3090?

Will it be worth upgrading when the new card comes out?

Updated: Oct 25, 2022 2:13 pm

Table of Contents Table of Contents

The launch of the Nvidia RTX 4000 series has officially replaced the 30-series. The RTX 4000 series graphics cards have introduced the new Lovelace architecture, third-generation ray tracing, and DLSS 3 features.

The RTX 3090 was launched in September 2020 with a launch price of $1499, but now after two years of launch, many third-party websites are selling it for around $1000. The RTX 4080 will be available for sale in November 2022, but no dates have been specified.

The Lovelace architecture in the 4090 offers 76 billion transistors, while the Ampere architecture present in the RTX 3090 offers 28 billion transistors. The more transistors correspond to the higher speed of the card. When it comes to memory, the RTX 3090 has 24GB of memory which is a lot more than the 16GB of the 4080. Potentially giving it a good lead in the higher resolutions.

The RTX 4080 has less number of CUDA cores as compared to the RTX 3090, so we can assume that the 4080 won’t be as good at computation as the RTX 3090. Cuda cores assist in playing demanding games by doing more calculations at once. But, the Lovelace architecture offers better overall performance than the Ampere architecture.

The 3rd generation ray tracing and DLSS 3 features in 4080 provide high-quality visual graphics for gamers looking for visually intensive games. Since all the cards have not been appropriately reviewed yet, we cannot claim which one will offer the best performance.

Until and unless the chips don’t arrive for testing, we can tell in detail which card is the better one. But after looking at the architecture and specifications, we can expect that the RTX 4080 is the upcoming tough competition for the RTX 3090.

Is 4080 faster than 3090?

The Nvidia RTX 4080 and RTX 4090 were officially introduced at the Nvidia GTC Keynote 2022, taking the place of the RTX 3000 Series. Along with additional features unique to the RTX 4000 Series, it also adds the most recent Lovelace architecture, third-generation ray tracing, and DLSS 3.

Is the RTX 3090 still worthwhile, considering all these additional features and the new architecture? To learn how the RTX 4080 and RTX 3090 compare, continue reading. 

The Ampere architecture used in the Nvidia RTX 3090 makes use of Samsung’s 8nm manufacturing process. It has 10,496 CUDA Cores and Micron’s 24GB GDDR6X video memory in a single configuration. The RTX 4080 model uses the 4nm process node from TSMC as part of its innovative Lovelace architecture.

In comparison to the 28 billion transistors on the Ampere cards, Nvidia has managed to fit 76 billion transistors on this node. Since data can be transported and processed more quickly with more transistors, performance is typically faster in the Lovelace architecture.

The RTX 4080 16GB contains 9,728 CUDA Cores, compared to 7,680 in the 12GB version. This demonstrates that the RTX 3090 outperforms both RTX 4080 configurations in terms of CUDA Cores while being in the next-generation range. But in this case, we can’t say for sure which graphics card performs best because we haven’t had a chance to review either of them yet. 

But we know that the RTX 4080 will have a boost clock of up to 2.51 GHz to Nvidia’s spec. Whilst the RTX 3090 only could reach up to 1.695 GHz. So it is much faster but until benchmarks, we won’t know if the performance is up to scratch as well.