Following the launch of the Nvidia GeForce RTX 4090, a graphics card largely aimed at professional use, last month, today the new generation of GPUs has finally reached less affluent PC gamers in the form of the Nvidia GeForce RTX 4080 16GB Founders. Edition, which launches tomorrow for about 1,500 euros. This means that the entry point for the lofty performance claims of the RTX 40 series and DLSS 3 frame generation has dropped at least a bit. We are still at the beginning of this new generation of GPUs, but for now the RTX 4080 is strong.

What’s in a name

Before we get to the spec sheet and performance numbers, we need to talk about the RTX 4080 naming controversy and how Nvidia GPUs are often positioned, and why this generation is a little different from previous ones. For over a decade now, Nvidia GPUs bearing the moniker “-80” have been considered the flagship cards of the mainstream and are priced from $500 to $700.

You may wonder, then, why the RTX 4080 starts at almost double that range. The answer lies in the aforementioned naming fiasco. Originally, Nvidia planned two variants of the RTX 4080: one of 16 GB (the version that we analyze here) and another of 12 GB, with a price of closer to a thousand. It wasn’t the first time Nvidia had released cards with different variants of VRAM, but usually the amount of VRAM was the only difference, whereas in this case the two cards also had different core counts and clock speeds, differences that would have been justified before. a jump to another level (in this case, the RTX 4070).

People rightly complained about the confusion this was starting to cause, and to credit Nvidia they responded, opting not to release the 12GB RTX 4080; it is now rumored that those cards will be re-announced under the name “RTX 4070 Ti”, although nothing is official at the moment.

This is all well and good, but it leaves us with a “conventional” card carrying the price that would previously have been reserved for a card labeled “Ti”, which is a half-generation step. In other words, the typical generational comparisons are a bit skewed this time, so we will compare the RTX 4080 with the RTX 3080 Tiwhich also launched at a similar price in June 2021, instead of the RTX 3080.

Nvidia GeForce RTX 4080 – Design and features

The RTX 4090 is an absolutely huge card, both in size and performance. The RTX 4080, for its part, is not smaller. It carries the same triple-slot designation, and measures 304mm long, 137mm wide, and 61mm thick, meaning the exact same dimensions as the RTX 4090. It’s a big card. For comparison, the RTX 3080 measured 285mm long, 112mm wide, and 40mm thick, while the RTX 2080 and GTX 1080 were even smaller.

Most of that weight is due to the dual axis cooling solution needed to keep temperatures under control. The cooler design is very similar to the RTX 3090, but with larger fans and taller fins to achieve what Nvidia says is 15% more airflow at the same noise level. In practice, the RTX 4080 remained very quiet while keeping temperatures around 53-55ºC, with a peak of 57ºC, during a long stretch of tests.

Compared to the RTX 3080 Ti, the RTX 4080 has 9,728 CUDA cores (vs 10,240), 304 4th-gen Tensor cores (vs 320 3rd-gen) and 76 3rd-gen RT cores (vs 80 2nd-gen). In other words: it has more cores, but slightly fewer of them overall. The drop in count shouldn’t be alarming though, as the 4080 packs a clock speed of 2,505MHz vs. the RTX 3080 Ti’s 1,665MHz, not to mention 16GB of GDDR6X VRAM, vs. 12GB. of its “predecessor” of the 30 series.

Like the RTX 4090, the 4080 uses the controversial 16-pin 12VHPWR power connector, which has recently been in the news for reports of it overheating and melting. We haven’t had any issues with it in any of our tests, but we’ll certainly keep an eye on the situation as this generation of graphics cards matures.

Speaking of power, the RTX 4080 has a TDP of 320W, down from 350W for the RTX 3080 Ti. nvidia recommends use a power supply of at least 750 W. There’s also a 3x 8-pin adapter in the box for those whose power supplies don’t have the new connector.

As for the ports, the RTX 4080 has 3x DisplayPort 1.4a and 1x HDMI 2.1a. This is the typical layout for current generation graphics cards, although AMD’s recently announced RX 7900 XT and XTX use the new DisplayPort 2.1, which has more than triple the bandwidth and enables 4K resolution up to 480Hz. or 8K up to 165Hz, compared to 240Hz for 4K and 60Hz for 8K on DisplayPort 1.4. Most games and monitors won’t be able to take advantage of that bandwidth, so it’s a moot point, but AMD technically has the upper hand.

Nvidia Geforce RTX 4080 – Performance

Starting with our synthetic benchmarks, the RTX 4080 comes out on top in 3D Mark Fire Strike Ultra with a 17% improvement over the RTX 3090 Ti and a 28% improvement over AMD’s RX 3950 XT – the two best GPUs of the previous generation. – and 35% compared to its generational predecessor of equivalent price, the RTX 3080 Ti. Unsurprisingly, though, it falls well short of the RTX 4090, scoring 16,255 to the RTX 4090’s 21,872, which makes perfect sense when you consider that card costs significantly more. .

Moving on to Unigine Heaven, the RTX 4080 outperforms the RTX 3090 Ti and RX 6950 XT at 1080p and 1440p, but falls short of both cards at 4K. Against the RTX 3080 Ti, however, it wins consistently with a 13% advantage at 1080p, 14% at 1440p, and a slight 4% at 4K.

The ray tracing results are more spectacular. The RTX 4080 offers an average improvement of 28% compared to the RTX 3090 Ti in our three tests, and of course completely shreds the RX 6950 XT, which lacks the ray tracing capabilities of Nvidia’s hardware. Comparison with the 3080 Ti offers even more impressive results, with an average improvement of 45% over that card.

Moving on to our gaming benchmarks, the RTX 4080 again performs strongly at all three resolutions tested. At this point, our benchmark tests are basically CPU-bound at 1080p, with the RTX 4080 moving on the meter alongside the more powerful RTX 4090. At 1440p the result is relatively similar, with the card showing big improvements over the latest generation in non-CPU-bound tests, and of course matching the best in CPU-bound tests.

However, given the high-end nature of this hardware, the real story is in 4K. (If you’re not going to be gaming at 4K or higher resolutions, you shouldn’t spend as much money on a GPU.) Expanding our test suite a bit, you can see that the RTX 4080 offers considerable gains over the previous generation, with an average improvement of 27% over the RTX 3090 Ti and 45% over the RTX 3080 Ti. Keep in mind that these latter cards launched at the same price, while the former had an RRP of over €2,000 when it launched earlier this year (although prices have now dropped to what is should pay for an RTX 4080 straight off the shelf).

Those are impressive gains, but they don’t go out of the ordinary if we take into account that it is a new generation of graphics. If we go back to the RTX 3080, that card offered a 50-70% improvement over its generational predecessor, the RTX 2080 Super. Not to write off the RTX 4080 (4K frame rates above 60fps in the most demanding games will be buzzwords for years to come), but I think it’s important to remember that this is a high-end price point we’re talking about. Not to mention enthusiast-priced, so my expectations are through the roof.

Finally, I want to talk about Nvidia’s new DLSS 3 frame generation technology. The GPU looks at two sequential frames, calculates the difference between them, and then uses the AI ​​to generate a frame between them. As with the RTX 4090, I tested DLSS 3 and frame generation in Cyberpunk 2077.

DLSS once again offered an impressive improvement, bringing the framerate of the RTX 4080 up to 73 without framerate, and 108 with it. Those are impressive numbers for one of the most technically demanding games available on PC today, and keep in mind that this benchmark runs at 4K with maximum settings and ray tracing on. RTX 30-series cards, meanwhile, receive fewer DLSS enhancements and do not have access to frame generation.

Of course, DLSS 3 is still a new technology, and game support is limited at the moment. However, it is being extended to more games, such as Microsoft Flight Simulator, A Plague Tale: Requiem, and Marvel’s Spider-Man Remastered. If support continues to grow as expected and performance gains continue to be as tremendous, DLSS 3 will be the decisive feature that makes it worthwhile to upgrade to a 40-series card for high-res, high-speed gaming.

The GeForce RTX 4080 shows itself to be admirable and offers performance levels worthy of its high price. It outperforms even the best of the previous generation, offering nearly a 50% average improvement over its comparably priced predecessor, the RTX 3080 Ti. That said, we’re still early on in this new generation, and we won’t have a good idea of ​​the relative value of the RTX 4080 until the low-end and mid-range 40-series cards (and AMD’s new RDNA 3 cards) arrive in the next few months. But until then, aside from its big brother, the RTX 4080 is the best graphics card you can buy, and it’s worth the outlay if you’re looking to build a high-end gaming PC that can handle demanding games at 4K.