After $1,599 was released last month Nvidia GeForce RTX4090A graphics card largely designed for professional use, today the next generation of GPUs has finally arrived in Nvidia form for slightly less wealthy PC gamers. GeForce RTX 4080 The 16GB Founders Edition will be available for $1,199 tomorrow. This means the high performance claims of the RTX 40 series and the entry point for frame-generating DLSS 3 have dropped at least a little. We’re just at the beginning of this new GPU generation, but so far the RTX 4080 is a strong show.
what’s in your name
Before we jump to the specs page and performance numbers, we should address the RTX 4080 naming confusion and talk a bit about how Nvidia’s GPUs are typically positioned and why this generation is a little different from previous years. With more than a decade of history, Nvidia GPUs are considered motherboards with the nickname “-80” and their prices range from $500 to $700. Looking more specifically at recent trends, the GTX 1080 launched at $599 and both the RTX 2080 and 3080 launched at $699.
Then you might be wondering why the RTX 4080 starts out with almost double that range. The answer lies in the naming fiasco mentioned above. Initially, Nvidia planned two variants of the RTX 4080 – 16GB for $1,199 (the version we’re reviewing here) and 12GB, priced at $899. This wouldn’t be the first time Nvidia has released VRAM-variant cards, but typically the only difference was the amount of VRAM, whereas in this case the two cards both carried different core counts and clock speeds – differences that would warrant earlier. switch to another layer (RTX 4070 in this case).
People have rightly complained about the confusion this is starting to cause, and to Nvidia’s credit, they responded by responding. “Start” RTX 4080 12GB; For now, these cards are said to be re-announced as the “RTX 4070 Ti”, but nothing official yet.
That’s all well and good, but still leaves us with a “mainstream” card carrying fancier tier pricing, representing a mid-generation step that was previously reserved for a card bearing the “Ti” label. In other words, typical generation comparisons are a bit skewed this time around, so we’ll mostly be comparing the RTX 4080 to the RTX 3080 Ti, which launched in June 2021 at $1,199 as opposed to the RTX 3080.
Nvidia GeForce RTX 4080 – Design and Features
if you read mine review You’ll remember that the RTX 4090 is an absolutely massive card, both in size and performance. By the way, the RTX 4080… no smaller. At 11.9 inches (304 mm) long, 5.4 inches (137 mm) wide and 2.4 inches (61 mm) thick, it carries the same triple slot designation with exactly the same dimensions as the RTX 4090. large card. For comparison, the RTX 3080 was 11.2 inches (285 mm) long, 4.4 inches (112 mm) wide and 1.5 inches (40 mm) thick, while the RTX 2080 and GTX 1080 were even smaller.
Much of that weight comes from the large, biaxial flow cooling solution needed to keep temperatures in check. The cooler design is mostly similar to that of the RTX 3090, but bigger fans and longer fins to achieve what Nvidia says is 15% more airflow at the same acoustic level. In practice, the RTX 4080 remained whisper quiet, keeping temperatures around 53-55C with a peak of 57C over a long benchmark period.
Compared to the RTX 3080 Ti, the RTX 4080 has 9,728 CUDA cores (down from 10,240), 304 fourth-generation Tensor cores (compared to the third-gen 320), and 76 third-gen RT cores (80 compared to the second-gen). In other words: it has newer cores, but slightly less overall. The drop in numbers shouldn’t be a concern, as the 4080 includes an increased clock speed of 2,505 MHz compared to the 1,665 MHz clock speed of the RTX 3080 Ti, not to mention 16GB of GDDR6X VRAM compared to the 12GB on the 30 series “predecessor.”
Like the RTX 4090, the 4080 uses the somewhat controversial 16-pin 12VHPWR power connector that has been in the news lately due to reports of overheating and melting. We didn’t have any issues with this in any of our tests, but we’ll definitely be monitoring the situation as this generation of graphics cards mature.
Speaking of power, the RTX 4080 has a TDP of 320W, up from the 350W on the RTX 3080 Ti. Nvidia recommends using a minimum 750W power supply. There is also a 3x 8-pin adapter in the box for people whose power supplies do not have the new connector.
For ports, the RTX 4080 has 3x DisplayPort 1.4a and 1x HDMI 2.1a. AMD’s recently announced RX 7900 XT and XTX are a typical layout for current generation graphics cards, although they use the new DisplayPort 2.1, which has more than three times the bandwidth and provides 4K resolutions up to 480Hz or 8K resolutions up to 165Hz. Against 240Hz at 4K and 60Hz at 8K for DisplayPort 1.4. Most games and monitors won’t be able to take advantage of that bandwidth so this is kind of a moot point, but AMD has the advantage technically.
Nvidia Geforce RTX 4080 – Performance
Starting off with our synthetic tests, the RTX 4080 wobbles 17% over the two best GPUs of the previous generation, the RTX 3090 Ti, and 28% over AMD’s RX 3950 XT and 3D Mark Fire Strike Ultra. % increase over the RTX 3080 Ti, its 3rd Gen price-equivalent predecessor. As you’d expect, it falls well behind the RTX 4090, but with a score of 16,255 compared to the RTX 4090’s 21,872, which makes complete sense considering the card is $400 more expensive.
Moving on to Unigine Heaven, the RTX 4080 outperforms the RTX 3090 Ti and RX 6950 XT at 1080p and 1440p, but actually lags behind both cards at 4K. However, it consistently wins against the RTX 3080 Ti, with 13% at 1080p, 14% at 1440p, and a slight 4% at 4K.
Ray tracing synthetics are more dramatic. The RTX 4080 offers an average of 28% increase over the RTX 3090 Ti in our three tests, and it certainly knocks out the RX 6950 XT, which of course lacks the ray tracing parts of Nvidia’s hardware. Comparing it to the 3080 Ti offers even more impressive results, with an average of 45% improvement over that card.
Moving on to our gaming tests, 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 pinging the meter along with the more powerful RTX 4090. The 1440p is relatively similar, with the card showing big gains over the last generation in non-CPU-bound tests. and of course it matches the best in tests.
But given the high-end nature of this hardware, the real story is in 4K. (If you don’t play at 4K or higher resolutions, you shouldn’t spend that much on a GPU.) Expanding our test suite a bit, you can see that the RTX 4080 offers significant gains over the previous generation at an average value. 27% improvement over RTX 3090 Ti and 45% improvement over RTX 3080 Ti. Note that the second of these cards launched for the same $1,199 price, while the first one carried an MSRP of $2,000 when it launched earlier this year (although prices have now dropped to what you should expect to pay for a new RTX 4080 on the shelves).
These are impressive gains, but not really unusual when you consider that this is a new generation of graphics. when we look back for us RTX 3080 review, this card provides a 50% to 70% improvement over the previous generation RTX 2080 Super. That’s not discounting the RTX 4080 – its 4K framerates well above 60fps in the most demanding games will raise eyebrows for the next few years – I just think it’s important to remember that we’re talking about high-end, if not fanciful. Pricing here so my expectations are sky high.
Finally, I would like to mention Nvidia’s new DLSS 3 frame rendering technology. See my RTX 4090 review for a more detailed explanation, but briefly the GPU looks at two consecutive frames, calculates the difference between them, and then uses AI to render a frame between them. I tested DLSS 3 and frame rendering on Cyberpunk 2077 as well as on the RTX 4090.
DLSS again offered a dramatic upgrade, increasing the RTX 4080’s frame rate to 73 without frame rendering and 108 with it. Those are great numbers for one of the most technically demanding games available on PC today, and remember, this benchmark runs at 4K with max settings and ray tracing enabled. The 30-series RTX cards, meanwhile, both receive less upgrades than DLSS and do not have access to co-framing.
Of course, DLSS 3 is still a new technology and game support is limited for now. That said, Microsoft Flight Simulator is continually expanding into more games, including A Plague Tale: Requiem and Marvel’s Spider-Man Remastered. If support continues to increase as expected, and the performance gains remain staggering, DLSS 3 will be the top feature for high-resolution, high-framerate gaming that really makes it worth an upgrade to a 40-series card.
