Nvidia’s new GeForce RTX 4080 Founders Edition hits shops in the present day, lightning-fast efficiency and a large new price ticket in tow. While final era’s awesome RTX 3080 value $700 at launch, delivering a monument 60 to 80 p.c efficiency uplift over its predecessor, this new mannequin is releasing with a staggering $1,200 price ticket. Yes, that’s proper: a $500 (or 71 p.c) improve over the price of its predecessor. For roughly 30 p.c extra efficiency.
Sheesh.
We’ve been placing the GeForce RTX 4080 via its paces. While Nvidia’s new GPU is astonishingly quick, blowing previous final gen’s RTX 3090 with ease whereas drawing a lot much less energy, that new price ticket casts a protracted shadow over its launch, to such an extent that we will’t even come shut to recommending this card at this aggressively insulting value—until, maybe, in the event you’re utilizing it for content creation below 8K resolution. Look for our complete RTX 4080 gaming efficiency evaluation to land within the coming days. Today, right here’s a high-level take a look at what you must know earlier than dropping $1,200 (or, ideally, not) on one in every of these beasts.
1: The GeForce RTX 4080 chews via video games
There’s no denying that the GeForce RTX 4080 is quick. Averaged throughout our 11-game testing suite—working quite a lot of graphics APIs, sport engines, and sport sorts—the GeForce RTX 4080 winds up about 7.5 p.c sooner than the vaunted RTX 3090, and 30.3 p.c sooner than the RTX 3080 (although actual efficiency uplift can range wildly relying on the sport). Unlike final era’s playing cards, it has no downside hitting the hallowed 60 frames per second mark at 4K decision with all the attention sweet cranked, even in robust titles like Cyberpunk 2077.
Brad Chacos/IDG
The identical holds true for ray-traced and DLSS-enhanced video games. As we noticed with the $1,600 GeForce RTX 4090, Nvidia’s new “Ada Lovelace” structure severely stepped up the standard of its devoted ray tracing and tensor cores to ship unparalleled efficiency in these cutting-edge video games—particularly within the handful of titles that help DLSS 3’s Frame Generation characteristic, which provides AI-generated visuals each different body to supercharge visible speeds (however not essentially responsiveness, as a result of these AI frames don’t reply to your inputs like a standard body would). At 1440p in F1 22, the RTX 4080 hits 118fps with customary DLSS 2 on and ray tracing energetic, and a whopping 234fps with DLSS 3 FG enabled!
2: The GeForce RTX 4080 chews via content material creation
If you utilize your graphics card for work in addition to play, you’ll like what’s on provide right here—perhaps. Adam “EposVox” Taylor, the web’s stream professor, already examined the RTX 4080’s workstation chops in a gauntlet of content material creation benchmarks. Bottom line? The large efficiency advantages he first witnessed in the RTX 4090 nonetheless maintain true right here, with the RTX 4080 (normally) smoking its last-gen predecessors in streaming, enhancing, AI artwork era, and extra.
But there’s a catch: The RTX 4080 solely comes with 16GB of onboard VRAM. That’s a lot for gaming, after all, however EposVox discovered it wanting for critical video editors. Check out his full eval for dozens upon dozens of benchmarks, however right here’s the important thing excerpt on why you could wish to spend up for a 4090’s 24GB capability or look ahead to a less expensive RTX 40-series choice as a substitute:
“Throwing the cardboard in my Threadripper Pro construct to textual content my intense 8K challenge filled with 8K RAW footage, Resolve SuperScaled 4K footage, plugins and results… sadly the 4080 couldn’t export this, identical to the earlier 80-tier playing cards I examined. 16GB of VRAM nonetheless isn’t sufficient, I suppose—plugins error out, shade graded footage takes longer than the 4090, and so forth. 8K is hard, however for $1,200 I might hope at this level to begin to see greater VRAM capacities when Intel is handing out $350 Arc graphics cards with 16GB.”
3: It’s large
One of the perks of Nvidia’s RTX 3080 Founders Edition was its dimension. Virtually all customized 3080 boards by the likes of EVGA, Asus, MSI, et cetera have been huge three- or four-slot playing cards that required numerous room in your PC case. The RTX 3080 Founders Edition stood out by being a normal two-slot card that match into nearly any system, and it seemed damned good doing it because of Nvidia’s gorgeous industrial design. Yay!
Brad Chacos/IDG
Not so with the RTX 4080 Founders Edition. Nvidia determined to present the RTX 4080 FE the identical humongous triple-slot cooler because the RTX 4090 this time round, which implies it’ll slip into far fewer methods. It’s a little bit of a perplexing design selection, because the extra environment friendly Lovelace-based RTX 4080 truly attracts much less energy than the RTX 3080 FE below full load in our assessments. Nvidia may’ve chosen a less expensive, extra manageable two-slot design but once more, I believe—although having an enormous, silent, and frigid three-slot cooler may assist make you are feeling higher about dropping $1,200 on this card I assume?
4: It makes use of a controversial (and ugly) energy adapter
Similarly, the GeForce RTX 4080 additionally joins the RTX 4090 in utilizing a brand new energy adapter designed to attach its new-breed 12VHPWR energy port to a few customary 8-pin energy connectors. I feel these short-cabled adapters are fugly in your case, however they’re nothing new—last-gen RTX 30-series Founders Edition playing cards additionally shipped with barely totally different 12-pin adapters. What is new are dozens of experiences of melting 12VHPWR adapters on the RTX 4090. Yikes.
This isn’t something I’d personally fear about. I’ve had no issues with my 4080 or 4090 FE, and some dozen incidents over presumably tens (a whole bunch?) of 1000’s of RTX 4090 gross sales signifies a transparent downside, however not a pervasive one. And regardless of intensive testing, GamersNexus and Igor’s Labs have been unable to copy the melted cables seen by a handful of latest RTX 4090 house owners. Hopefully Nvidia figures out the foundation trigger, and it’s restricted to a foul batch of cables or connectors on 4090s alone.
But right here, it’s one other potential level of concern in a product filled with tiny considerations. And a $1,200 graphics playing cards actually shouldn’t stress you out. Which brings us to the primary level.
5: The GeForce RTX 4080 prices waaaaaaaay an excessive amount of
In no world do you have to pay $1,200 for the RTX 4080. And I believe Nvidia desires precisely that.
Our full benchmark outcomes gained’t be posted till our last RTX 4080 evaluation launches within the subsequent day or two—a burst pipe in the course of the night time foiled my release-day ambitions—however at a excessive stage, it’s about 30 p.c sooner than the RTX 3080 at 4K decision with settings maxed out and ray tracing/DLSS off. That’s a stable, if unspectacular generational soar (the RTX 4090’s 55 to 83 p.c leap over the 3090 was a lot extra spectacular, as was the 3080 vs. 2080), the kind you’d anticipate to see in the identical class shifting from one gen to the subsequent. But that’s on the identical value.
Again, Nvidia is asking for 71 p.c more money for that mere 30 p.c uplift—$699 versus $1,199. At the identical $1,199 value, the RTX 4080 is barely sooner than the RTX 3080 Ti by single-digit percentages. That’s completely insane and aggressively insulting to PC avid gamers. You ought to keep far, far-off from this graphics card at this value.
There are wider financial explanation why Nvidia most likely priced the RTX 4080 this fashion. The RTX 4090 stays an incredible choice even at $1,600; it could actually max out a high-refresh fee 4K monitor and chew via content material creation duties with out breaking a sweat, one thing no different GPU can declare, for simply $100 greater than the RTX 3090’s launch MSRP. It’s dear, however good.
The RTX 4080 is costly however does nothing to justify its downright ludicrous value. It does, nevertheless, handle to make each the RTX 4090 and the piles of older RTX 30-series playing cards look considerably extra affordable because of the psychology of price anchoring. Nvidia and its companions are sitting on a large stockpile of RTX 30-series GPUs after the cryptocurrency market crashed, and sunk demand for ultra-marked-up GPUs alongside it. Consider the RTX 3080: No one would anticipate to pay anyplace close to its MSRP of $700 greater than two years after its launch. But that’s the place we’re at proper now, and subsequent to the $1,200 RTX 4080, it’d appear like a cut price, as a result of the RTX 4080 is 71 p.c extra. Nvidia is conserving each these generations alive side-by-side for a purpose.
When Nvidia CEO Jensen Huang spoke to reporters after revealing Lovelace at GTC 2022, he stated, “The idea that a chip is going to go down in cost over time, unfortunately, is a story of the past.” But AMD simply revealed its next-gen “RDNA 3” Radeon GPUs and sure, they value lower than their predecessors. The Radeon RX 7900 XTX prices $999, in comparison with a $1,100 MSRP for the Radeon RX 6950 XT; the Radeon RX 7900 XT prices $899, in comparison with $999 for the Radeon RX 6900 XT (although the 7900 XT could have extra in widespread with last-gen’s 6800 XT).
We’ll must see the place efficiency falls when AMD’s playing cards launch on December 13, but when the Radeon RX 7900 XTX manages to hit the 1.5x to 1.7x uplift (by way of 6950 XT) that Radeon claims, Team Red’s providing may very effectively wind up simply behind the RTX 4090 and as quick or sooner than the RTX 4080 for a whole bunch of {dollars} much less.
Take all vendor-supplied benchmarks with an enormous handful of salt, but when the Radeon RX 7900 XTX is certainly greater than 1.5x sooner than the 6950 XT, it may wind up sooner than the 4080 in conventional video games for a whole bunch much less.
AMD
Don’t be a sucker and pay $1,200 for this card. Don’t be a sucker and purchase a 2-year-old last-gen mannequin at full value. Either purchase an RTX 4090 if you’d like no-compromises gaming and creation efficiency—or wait to see what Radeon affords this time round, or look ahead to Nvidia’s costs to probably fall after it sells via that mountain of older RTX 30-series playing cards. Because a 71 p.c (71 p.c) value improve for a 30 p.c efficiency improve ain’t it. That’s harsh, nevertheless it’s true.
Stay tuned for our full gaming evaluation quickly. In the meantime, take a look at Adam Taylor’s insanely detailed analysis of the GeForce RTX 4080’s content creation chops.