Skip to main content

Nvidia RTX 2070 vs. GTX 1080

Image used with permission by copyright holder

There’s no denying that the top Nvidia RTX graphics cards are the most powerful consumer cards in the world right now (even if AMD is giving them a lot of competition), but with availability what it is, sometimes buying an older card secondhand is a more affordable option. Two of the best cards from their respective generations were the RTX 2070 and the GTX 1080. But how do these cards compare, and if given the chance, which should you buy?

Availability and pricing

The GTX 1080 was originally released in May 2016 with a suggested price of $599. The RTX 2070 debuted in October 2018, priced at $500. In 2021, both of these cards haven’t been on sale for a long time, with the GTX 1080 replaced by the RTX 2080, and the RTX 2070 replaced by the RTX 2070 Super. Those again have been replaced by cards in the RTX 3000 series.

However, due to a global silicon shortage, import tariffs, and cryptocurrency mining demand in 2021, all GPUs are almost impossible to buy at fair retail prices. The GTX 1080 is going for over $400 on secondhand markets, which, for a nearly five-year-old card, is completely off the charts. The RTX 2070 is even worse, often retailing for $600 secondhand — more than it sold for when new over two years ago.

Performance

At launch, the RTX 2070 performed roughly comparably to the GTX 1080, but with newer games, and as support for ray tracing and deep learning super sampling have become more common, not only is it the more powerful card, but it can produce better-looking games at higher frame rates too.

RTX 2070 GTX 1080
Architecture Turing Pascal
CUDA cores 2,304 2,560
Texture units 144 160
Render output units 64 64
RT cores 36 0
Tensor cores 288 0
Base speed 1,410MHz 1,607MHz
Boost speed 1,620MHz 1,733MHx
Boost speed (FE model) 1,710MHz 1,800MHz
Memory 8GB GDDR6 8GB GDDR5X
Memory speed 14Gbps 10Gbps
Bandwidth 448GBps 320GBps
Transistors 10.8 billion 7.2 billion

By the numbers, both cards have a slight leg up over the other in certain specifications, and the Founders Editions don’t do anything to close those gaps. The 1080 has 2,560 CUDA cores, while the 2070 has 2,304 — although they’re based on the younger Turing architecture. The 1080 also has slightly higher clock speeds and a few more texture units. However, the 2070 sports GDDR6 memory, giving it a near 50% greater memory bandwidth. It also has a larger die size, with 3.6 billion extra transistors. That has its own efficiency and performance improvements.

This equates to a sub-10% frame rate advantage to the RTX 2070, making it the more capable card. In 2021, these cards are mostly suited to AAA games at 1440p or high frame rates at 1080p, depending on your preference and the types of games you’re playing. If you’re only playing games that demand traditional rasterization performance, either card will do a good job. But if it’s a newer game, and particularly if it has ray tracing or supports DLSS, the RTX 2070 has a huge advantage thanks to its dedicated hardware accelerators.

Ray tracing and DLSS

Nvidia

Alongside its newer architecture, the RTX 2070 also helps introduce RT and Tensor cores that enable ray tracing and DLSS. Of these two cards, those features are only really available on the 2070. While the 1080 can technically perform ray tracing calculations without the hardware acceleration of RT cores, it’s only going to be able to offer that at 1080p at 30 fps or less.

The list of supporting games for either or both of these technologies has grown a lot since the launch of the RTX 2070, making it a must-have feature for new GPUs. Ray tracing has full support with all of Nvidia’s latest-generation Ampere GPUs as well as AMD’s RX 6000 series. DLSS isn’t as broadly supported, and AMD doesn’t have a competing solution just yet, but its effect can’t be denied. When enabled, DLSS 2.0 makes it possible to run a game at 1440p but look very close to, or sometimes even better than, 4K. That leads to a great performance leap over native 4K rendering and makes cards that support it, like the RTX 2070, a much more attractive buy if you play or plan to play games that can enable it.

The 2070 is the new 1080 … if you can find it

The biggest difference between the 2070 and 1080 when the former first launched was price. With global GPU shortages the way they are, that’s still the case today, but in the secondhand market. You won’t be able to buy these cards new, and even if you could, you shouldn’t be paying hundreds of dollars for them. With far faster newer-generation cards like the RTX 3060 Ti, and 6700 XT sporadically available, grab one of those instead if you can.

If you are looking at either of these cards secondhand and the price isn’t too ridiculous, the RTX 2070 is by far the better buy. It’s more powerful, and its features can far better keep up with modern games and their developments than the aged 1080.

Jon Martindale
Jon Martindale is the Evergreen Coordinator for Computing, overseeing a team of writers addressing all the latest how to…
Wait, what? Nvidia’s RTX 50-series might be ready to go this month
RTX 4090.

Update: According to MEGAsizeGPU on X (formerly Twitter), the following might be a mistranslation. The source reportedly talks about Nvidia finalizing the design for Blackwell in September instead of launching the card. No details about the release date have been confirmed. The original article follows below.

Nvidia's RTX 50-series graphics cards are coming -- we know that for a fact. But when? The release dates of these graphics cards have been the topic of much speculation. Early leakers predicted that they'd launch in late 2024, but the general consensus slowly shifted toward an early 2025 release instead.

Read more
The RTX 5090 might decimate your power supply
Fans on the Nvidia RTX 3080.

If you thought the best graphics cards already drew a ton of power, you're in for a rude awakening. A series of claims surrounding Nvidia's upcoming RTX 50-series GPUs say that the next-gen cards will push power limits even further, with a flagship card like the RTX 5090 drawing as much as 600 watts.

Nvidia has yet to even announce RTX 50-series GPUs, but we've already seen some troubles with the Blackwell architecture the cards will use in the data center. Official details on the cards are few and far between, but a handful of sources now claim the RTX 5090 will push power limits beyond the 450W we saw with the RTX 4090 in the previous generation. The most recent speculation comes with well-known leaker kopite7kimi, who claimed on X (formerly Twitter) that the RTX 5090 will go up to 600W, while the RTX 5080 will require 400W.

Read more
Nvidia is launching a new GPU, but read this before buying
Two RTX 4070 graphics cards sitting side by side.

Nvidia is launching a new version of one of its best GPUs -- the RTX 4070 (non-Super). Normally, this would be great news. However, there's one downside: Nvidia is equipping the new RTX 4070 with GDDR6 memory, which is a step down from the GDDR6X that the GPU usually comes with. This wouldn't be the end of the world, but there are a couple of red flags.

For starters, Nvidia itself lists the GPU alongside its previous version. It seems like the two RTX 4070s will only differ by memory type, all the while retaining the same clock speeds, bus width, and CUDA core counts. This is good news, but switching from GDDR6X to GDDR6 means a drop in bandwidth.

Read more