The RTX 4080 has lastly arrived and as anticipated, it’s a significantly potent GPU. It’s formally the second strongest GPU cash can purchase. That will usually be excellent news for players, however its $1,199 value places it in an odd house throughout the market. The RTX 4090 provides much more efficiency, and also you’re already within the ballpark so far as pricing goes with the 4080. This can go away lots of players in a conundrum over which GPU to purchase, assuming they’ll truly purchase both of them. Nonetheless, the RTX 4080 is undeniably highly effective, and an enormous step up from playing cards just like the RTX 3080. Nonetheless, Nvidia has tacked $500 onto the value tag in contrast with its x80 predecessor, which is an enormous leap. Additionally, AMD’s RX 7900 GPUs are ready within the wings at $999 and $899, which makes Nvidia’s pricing look much more questionable.
Though Nvidia endowed the RTX 3090 and 3080 with the identical “large” GA102 die for Ampere, it took a special strategy with Ada Lovelace. Solely the RTX 4090 will get the massive die, with the RTX 4080 sporting a a lot smaller AD103 chip. It’s roughly half the dimensions – 379mm² versus 609mm² – of that within the RTX 4090. As you possibly can think about this leaves fairly a efficiency hole between the 2 GPUs. Thus, it’s simple to see Nvidia finally introducing an RTX 4080 Ti with a neutered AD102 die. Nvidia additionally clipped the RTX 4080’s wings so far as its reminiscence bus goes. The RTX 3080 had a 320-bit bus, whereas the 4080 is restricted to 256-bit. It compensates a bit by providing extra L2 cache than the RTX 3080 although. It has 64MB in contrast with the 3080’s paltry 5MB.
The RTX 4080 is rocking 9,728 CUDA cores with 16GB of GDDR6X reminiscence. Its base clock is 2.21GHz with a lift clock of two.51GHz. The GPU is fabricated on TSMC’s 4nm course of and has a goal energy consumption of 320W. That’s a bit simpler to swallow than the RTX 4090’s 450W, and the identical TDP as the unique RTX 3080. It’s spectacular that Nvidia stored the facility goal the identical between generations, regardless of offering vital efficiency uplift. That’s anticipated with a node shrink, nonetheless. It’s the identical measurement bodily because the RTX 4090 Founders Version. One small change is the adapter cable is a three-into-one design as a substitute of a four-into-one for PCIe energy cables. It’ll be attention-grabbing to see if this nonetheless leads to cables melting.
For benchmarking, our colleagues at PCMag.com ran the cardboard by means of its paces. Its check bench consisted of an Asus ROG Maximus Z690 Hero motherboard with a inventory Core i9-12900K. It’s working 32GB of DDR5 reminiscence clocked at 5,600MHz. A 1TB PCIe 4.0 SSD and 1,500W PSU from Corsair spherical out the system. The CPU can be liquid-cooled through a Corsair cooler.
The Benchmarks: Artificial Assessments
It’s been mentioned there are often few surprises in GPU launches. You have a look at the spec sheet and that’s how the playing cards line up, barring issues like boosts from DLSS. In artificial exams, that’s precisely what we see. The RTX 4090 and 4080 tower over the competitors in each check. In Time Spy Excessive’s Graphic rating, the 4080 is 60 p.c quicker than the RTX 3080 it’s changing. The identical delta exists between the high-end AMD GPUs as nicely. Nonetheless, the RTX 4090 has a major benefit over its sibling, to the tune of 40 p.c extra efficiency. Regardless, the 4080 has a commanding lead over each different GPU examined. In these exams, the 4080 is round 40 p.c slower than a 4090 whereas being 40 to 50 p.c quicker than the RTX 3080.
AAA Gaming Benchmarks w/DLSS & FSR
As soon as once more, there aren’t any surprises right here: the 4080 is the second-fastest GPU on the block. The identical delta we noticed in artificial exams is current, with a 40 p.c delta to the 4090, and in reverse to the 4080. Nonetheless, Deep Studying Tremendous Sampling injects some attention-grabbing efficiency outcomes. Basically, the RTX 4080 with DLSS can match an RTX 4090, no less than in Guardians of the Galaxy. It additionally confirmed CPU bottlenecking at 1080p, which isn’t shocking.
As soon as once more, not many surprises right here in normal AAA gaming. At 1080p we see some CPU bottlenecking. Then at 1440p the RTX 4080 begins to hole the sector. Then at 4K it actually stretches its legs. Each the 4080 and 4090 are a transparent step above each different GPU within the discipline proper now. The 4080 is 50 p.c quicker than the RTX 3080 at greater resolutions. As at all times, there’s a 30 to 40 p.c hole to the RTX 4090 as nicely. Nonetheless, the RTX 4080 goes from round 60fps with high-end Ampere to 100 fps and better at 4K in each recreation besides Whole Battle: Three Kingdoms.
Energy Consumption and Thermals
Most of our fears about Ada Lovelace GPUs resembling nuclear reactors have been assuaged at this level. Sure, they’ve large coolers, however that retains them cool and quiet. Plus, the RTX 4080 has a 320W TDP, which is in keeping with earlier high-end GPUs. Total, it looks like Nvidia did some wonderful engineering on its cooler. Even within the FurMark torture check, it peaked at 62C. Within the precise recreation of Guardians of the Galaxy, it hit 63C. The full system energy draw was 447W, which appears affordable in contrast with the 4090’s 577W. It’s additionally quite a bit lower than the RTX 3080s 535W, displaying Ada Lovelace’s effectivity features over Ampere. Total, that is one class the place Nvidia’s GPUs actually shine.
In 2016 the GTX 1080 launched at $599. In 2018 the RTX 2080 launched at $699. In 2020 the RTX 3080 had a price ticket of $699. We are able to all see one thing of a sample right here. Nonetheless, cue the “file scratch” noise for the RTX 4080 launch at $1,199. That’s an outlandish value leap, and it has turned lots of people off. What’s attention-grabbing is the value is smart in contrast with the $1,599 RTX 4090. But when it had stored up the pricing scheme from earlier than and bought it for $899 or so, it could be an on the spot basic. Clearly, Nvidia believes it could cost extra for its GPUs than it used to. Not less than, it could accomplish that for the high-end stuff. It’s most likely proper, because the RTX 4090 has been bought out since launch. We’re positive the identical destiny awaits the RTX 4080.
In a vacuum, Nvidia’s pricing is smart when it comes to its efficiency. Nonetheless, again within the olden days, we’d see a brand new technology of GPUs supply extra efficiency than their predecessor on the identical value. That’s clearly not the case anymore. After the RTX 4090 launch, Jensen mentioned these days have been over since Moore’s Regulation is useless. The whole lot prices extra now, he mentioned (we’re paraphrasing). Nonetheless, it’ll take AMD GPUs for us to place Nvidia’s newest effort into correct context.
On the one hand, the RTX 4080 is a blisteringly quick GPU. It could simply deal with each 4K and 1440p at excessive fps, and with DLSS 3 it’ll outpace Ampere playing cards even additional, finally. And it runs cool and quiet too. It’s simply so costly, particularly when the RTX 4090 is proper there. No one who buys it will likely be dissatisfied by its efficiency. However in addition they could be kicking themselves when the RTX 4080 Ti comes alongside, particularly if Nvidia provides it on the identical value.