Nvidia GeForce RTX 5080 Founders Edition Review

Just last week, NVIDIA dropped the RTX 5090 (review). The most powerful graphics card in the world donning the new Blackwell architecture, but arguably, along with being super powerful, means that it has a super price tag to match that is simply beyond the remit of most consumers wanting the latest technologies and the highest frame rates in games. So now, with the review embargo lifting for its little brother, the RTX 5080, coming in at a much more palatable $999, can we get performance that gives us enough of a generational uplift over the 40 series while reigning in on the value for money perspective? Well, that’s what we’re going to find out! Nvidia RTX 5080 In terms of the GPU, we have a die size of 378mm², so similar to the 4080 series and packs 45.6 billion transistors, which is actually less than both the 4080 and 4080 SUPER, which came in at 45.9 billion. Along with a bump from PCIe 4.0, we’re now sporting PCIe 5.0 and it’s something we’ll be having a look at in the near future to see if that’s a limiting factor in terms of performance due to the bandwidth being double that of PCIe 4.0, though as we know, most modern GPUs, even high-end ones like the new Blackwell based ones, rarely are able to saturate the bandwidth of PCIe 4.0 in most gaming scenarios, but we’ll look at it in some later content regardless. We’re also now pushing the memory, which while we still have 16GB, we’re now utilising GDDR7 on a 256-bit bus, which gives us 960GB/s of bandwidth, which is a step up from 716.8GB/s on the 4080 and 736.3GB/s on the 4080 SUPER. So, we have some somewhat strong uplifts in specs, but as we know, raw numbers like this, don’t exactly equate to raw performance. However, with the 4080 SUPER launching at the same $999 price tag, and the 4080 non-SUPER coming in more expensive at $1199, it’s definitely an interesting talking point to see where performance fits in based on price. Key Features Enhanced NVIDIA Broadcast App: Includes new AI-powered features like Studio Voice and Virtual Key Light for improved live streaming. Blackwell Architecture: The new Blackwell architecture boasts significant improvements in performance and efficiency. AI-Powered Rendering: Features like neural shaders and DLSS 4 leverage AI for groundbreaking graphics rendering. DLSS 4: This technology uses AI to generate more frames, leading to up to 8x performance boosts compared to traditional rendering. RTX Neural Shaders: Introduce AI networks into programmable shaders, enabling film-quality materials and lighting in games. RTX Neural Faces: Renders high-quality digital faces in real-time using AI. Reflex 2 with Frame Warp: Reduces latency in games by up to 75% for a more responsive experience. Autonomous Game Characters: New NVIDIA ACE technologies allow for AI-powered non-player characters (NPCs) that can perceive, plan, and act more like human players. AI Foundation Models: Support for NVIDIA NIM microservices enables users to build AI agents and assistants on their RTX AI PCs. FP4 Precision: Boosts AI image generation performance and allows for running generative AI models locally. Packaging Ok, so let's gloss over the packaging because it’s something that’s definitely important to consumers, especially when we think about added extras and goodies that come with a GPU, but sorry to disappoint, it’s exactly the same as the RTX 5090 Founders, which we did an unboxing on a little over a week ago now. That’s not to say that it’s bad, more just that if you’ve already seen it, then there’s nothing new here. Well I say nothing new, inside, you do get a 12V-2x6 adapter which takes it from 3 PCIe 8-pin connectors. Beyond that, the packaging is the same but obviously says 5080 debossed into the front, and you’ll be pleased to know, if you’ve seen my unboxing of the 5090, that I’ve finally figured out how to open the box now, so yeah, go me. A Closer Look From a first glance, it looks identical to the RTX 5090, but there is a small difference, and I’m not just talking about how it says RTX 5080 on it now, instead of 5090, but it’s actually a slightly different shade of colour, which in all honesty, I prefer the colour of the bigger brother, but I also wouldn’t be willing to drop double the price on it, just for aesthetic reasons, though of course, you get what is arguably a lot more power, though we will see what that means when we look at benchmarks shortly. Due to that being the other real difference between the two, we still have a dual-slot card, which you could claim is easier to cool on a less powerful card, though they’ve still kept the dimensions the same at 304mm in length and 137mm in width, so there’s still a lot of surface area to help spread the heat to dissipate. Much like the RTX 5090, NVIDIA with their Founders model are relying on their “fan pass-through” cooling system to help push airflow through the heatsink fins, and exhaust upwards, though there has been talk about what this means for CPU temperatures, and it’s something we’re definitely going to look at on both the 5080 and 5090 Founders cards in the very near future, because a lot of outlets do test outside of a chassis, and I honestly believe, that could be a huge difference in terms of performance, especially in terms of 1% low figures, which we saw on the 5090, were actually lower than expected. In terms of branding, there’s not really too much going on, short of just that really small NVIDIA logo once again, and I really like that. I know who makes the GPU. I know what I’ve bought. I don’t need reminding, so I’d definitely like to give NVIDIA some brownie points on that one for keeping it nice and clean. The card, like the 5090 has some understated LED lighting on the middle of the card and the GeForce RTX logo on the side, which sits next to the angled 12V-2x6 connector though the 5080 has considerably less total graphics power of just 360W which is just over 12% more than its predecessors, the RTX 4080 and 4080 SUPER, which both came in with a TGP of 320W and due to this, NVIDIA do recommend an 850W power supply with 3 available PCIe 8-pin connectors for the included adapter or a single PCIe Gen 5 cable capable of supplying at least 450W of power. In terms of display outputs, just like the 5090, it packs a single HDMI and 3 DisplayPort connectors, and while I’ve seen comments on the 5090 review we did with consumers upset that it only has a single HDMI, I’m personally ok with this as DisplayPort 2.1b offers much more bandwidth over HDMI 2.1b meaning higher resolutions and refresh rates are possible, so maybe that’s a nod to 8K gaming being possible in the near future. The PCB on the Founders 5080 is still extremely small and is classed to some as a work of art and design engineering, especially when you factor in that the GB203-400-A1 GPU is packing 10752 CUDA cores which is an 11% uplift over the RTX 4080 and a smaller 5% increase over the 4080 SUPER. Other specs include the same uplift of Texture Mapping Units, now at 336, the same for Streaming Multiprocessors, now at 84 along with the same for RT cores and the same uplift for Tensor cores, which now come in at 336. For Render Output Units, which are now at 128, we see 14% more when compared to both the RTX 4080 and SUPER variants and both the L1 and L2 cache remain unchanged from the Ada Lovelace predecessors. How We Test We continue to update our testing methods around once per year. As such, we re-test older hardware to reflect changes over time. These can be driver updates, Windows updates, game patches, and more; all of which have an impact on performance figures. Furthermore, we update our test benches to newer and more relevant hardware. This means that our new reviews aren’t always comparable to those of older reviews, so please compare the testing methodology on older reviews should you be trying to compare them with newer ones. Replicate Our Results When it comes to our benchmarks in our reviews, the benchmarks are pretty self-explanatory and are kept as simple as possible, although there are a few exceptions. Remember that your choice of a graphics card, CPU, the silicon lottery, and other factors can yield different numbers, and there’s always a margin for error when using any software. Links are provided below, as well as the settings we use. We encourage you to not just look at how one product compares to any other, but how it compares to your own. If you’re looking to build a new system, you should benchmark your current PC using the benchmarks available to you. You should then look at the percentage improvement from your current hardware to the hardware tested here to give you a ballpark figure of how much of an upgrade this will provide you with. Processor Test Systems Test System (GPU) Motherboard Gigabyte B650E AORUS MASTER Buy Now Processor Ryzen 7 9800X3D Review Buy Now GPU Changes Per Review RAM 32 Corsair VENGEANCE RGB 6000MHz CL30 Review Buy Now CPU Cooler NZXT Kraken Z73 RGB Buy Now Power Supply NZXT C1000 Gold Review Buy Now Storage Seagate FIRECUDA 530 1TB Review Buy Now Chassis NZXT H9 Flow Review Buy Now OS Windows 11 Pro Version 21H2 Drivers Software We use a variety of software from different developers to assist in our testing and to give us the most consistent results possible. Some applications are specifically used for testing, while others are simply used for monitoring and to aid us in making sure that our results are as accurate as possible. Software Used 3DMark FireStrike & TimeSpy (download) Gaming To gauge performance levels which are easily reproduced time after time, by both ourselves and our readers, we now stick strictly to games default profiles. Given how powerful most modern cards are getting, we've began using more demanding profiles for most games, but will otherwise state on a per game basis if we use other profiles beyond High, Ultra, Maximum or their respective equivalent at all resolutions. V-Sync is always disabled in our testing, and all FPS caps that can be removed are. We also enable XMP and Resizable bar technology across all platforms to keep things as fair as possible. Games While we test a lot of games, it's not possible to display all of the data on the website or in our videos. Instead, we include access to all of our testing data for our Patreon members (link), where members get a ton of other benefits as well. What we do, however, is display charts for averages across all game titles to give a better picture of performance across all titles tested. A Plague Tale: Requiem Alan Wake II Baldur's Gate 3 Black Myth Wukong CS2 Cyberpunk 2077 F1 24 Fortnite Hogwarts Legacy Indiana Jones And The Great Circle Starfield War Thunder Resolutions 1080 (Full HD) 1440 4K Some tests will also include features such as NVIDIA DLSS and Ray Tracing. However, these will be clearly labelled and may only appear in reviews for graphics cards that support those features. Power Consumption Not being content with simply reading total system power draw, we now run F1 24 for an hour loop and measure power draw at the end of the loop for the most accurate measurement that represents real-world gameplay experience. Temperatures We take temperature readings after the same hour-long F1 24 loop with no other background programs running, again to simulate a real-world gaming experience. 3DMark TimeSpy Extreme 3DMark Time Spy Extreme is a DirectX 12 benchmark test that pushes even the most powerful gaming PCs to their limits. It renders at a demanding 4K resolution (3840 x 2160) and includes a CPU test designed to measure the performance of high-end processors with 8 or more cores. This makes it an ideal benchmark for evaluating the latest graphics cards and top-of-the-line CPUs, especially when considering demanding tasks like gaming at 4K resolution and beyond. Starting with a synthetic test and 3DMark Timespy and it’s here where we see the RTX 5080 nestled in below the RTX 4090. In terms of uplift, we do find a 13% increase in performance over the RTX 4080 and 8% over the RTX 4080 SUPER, which now puts the 5080 above the 7900 XTX from AMD by 6%, so a good start in terms of standings, but not exactly groundbreaking compared to other cards. A Plague Tale: Requiem A Plague Tale: Requiem is a stealth-survival horror adventure that serves as a direct sequel to A Plague Tale: Innocence. Players continue the harrowing journey of Amicia and Hugo de Rune, two young siblings navigating a plague-ridden France in the 14th century. The game features stunning visuals, intense action sequences, and a gripping narrative that explores themes of family, resilience, and the darkness within. For our first game, we ran A Plague Tale: Requiem and at 1440p, the RTX 5080 comes in packing 18% more performance than the RTX 4080 and 13% more than the 4080 SUPER, which while it doesn’t sound huge, considering that the 4080 SUPER came in 20% more expensive, this is quite a big win, though it does only put it 12% ahead of AMD which is currently retailing for $40 less at the moment. As we move up to 4K, there’s not much change with an 18% uplift over the RTX 4080, and 13% over the 4080 SUPER and again, a similar but somewhat smaller uplift, now at 10% over the 7900 XTX. One big point that I noticed, is that the 5080 does sport better performance increases in the 1% lows than what we saw with the 5090, though this looks likely to be down to the driver, which we looked at just a few days ago so who knows what future updates may bring. Alan Wake II Alan Wake II is a survival horror game and the long-awaited sequel to Remedy Entertainment's cult classic, Alan Wake. This time, the story takes a darker turn, delving deeper into the psychological and supernatural elements that made the original so compelling. Players can expect a chilling atmosphere, intense combat encounters, and a mind-bending narrative that will keep them on the edge of their seats. In Alan Wake II, the RTX 5080 packs another 17% more performance over the RTX 4080, and 11% over the 4080 SUPER. These numbers are fairly good, especially when factoring in the relevant price points, but more impressive again is the 1% low figures that sees the 5080 ahead by 19% over the 4080 and 16% over the 4080 SUPER. This also means, due to the uplift, that we find the 5080 leading the 7900 XTX by 10% in the averages, and 11% in the lows but does still put it 14% behind the RTX 4090 overall. As we move up to 4K, we see similar numbers, with 18% more performance over the RTX 4080 in the averages, and 13% more in the lows, while comparing to the 4080 SUPER, we see a smaller increase of 11% and 8% respectively. What this means for the competition is that NVIDIA now sits 13% ahead in the averages, and 8% in the lows for a price point that matches, at least in terms of MSRP launch price at the same $999. Baldur's Gate 3 Baldur's Gate 3 is a highly acclaimed role-playing game set in the Forgotten Realms universe. Developed by Larian Studios, it offers a deep and immersive experience with turn-based combat, intricate storylines, and a vast world to explore. Players can create their own custom characters or choose from a roster of origin characters, each with their own unique backstories and motivations. When looking at Baldur’s Gate 3, we don’t see as much of an increase with just a 15% rise in the averages, but we do see a drop of 7% in the lows when comparing to the RTX 4080. When we move up to the 4080 SUPER, the gains shrink, but still sees the 5080 packing 7% more performance overall, and again, the lows take a hit of 11%, though a part of me really wants to blame it on the drivers, as we’ve seen with the 5090 suddenly finding extra performance by updating them, and I feel that the 5080 may be in a similar boat in places. This does also put the 5080 11% behind the RTX 4090 in the averages, and only 8% ahead of AMD, though AMD really do come in much stronger in the lows. Moving up to 4K and performance is where you’d expect it to be based on what we’ve seen so far with a 19% lead over the RTX 4080, and 12% lead over the 4080 SUPER and 7900 XTX which came in with identical averages. Now where things get really interesting, is how our 4080 didn’t do great in terms of the 1% lows, and because of that, the 5080 comes in with 39% more performance there, though retests could see this balance out a bit more. We do however, see a 14% lead in the lows over the 4080 SUPER, and 20% over AMD, so there’s definitely some strong performance coming from the new Blackwell GPU in this title, though still not enough to get close to the 4090, as it comes in with 14% less performance in the averages, and 10% less in the lows. Black Myth Mukong Black Myth Wukong is an action role-playing game inspired by the classic Chinese novel "Journey to the West." Developed by Game Science, it features stunning graphics, challenging combat encounters, and a captivating story based on the legendary Monkey King. Players can expect a unique blend of Chinese mythology and action-packed gameplay. In Black Myth: Wukong at 1440p, and me being honest with you, the results are weird, mainly on the 4080, as the 5080 comes in with 13% more performance in the averages but a whopping 45% more in the lows. We did retest this over, and over, and over, but there’s potentially a bug affecting the RTX 4080 specifically, as the 4080 SUPER didn’t have the same problems, and instead, the 5080 came in packing 10% more performance in the averages and 11% more in the lows. This also puts the new Blackwell-based GPU 19% ahead of AMD’s best offering, with a 24% disparity in the lows too, and now sees it only 9% behind NVIDIA’s RTX 4090. Then at 4K, the uplift over the 4080 increases to 20% in the averages, and a somewhat incredible 60% in the lows. Again, there’s clearly an issue with that particular card in this particular game, and could be for a multitude of reasons, of which we will investigate further as time goes on. That aside, we see 15% more performance over the 4080 SUPER in the averages, and 12% more in the lows, while compared to AMD, we see 15% more performance on the 5080 in both the averages and the lows, which also puts it around the same level behind the RTX 4090. CS2 CS2 (Counter-Strike 2) is the highly anticipated sequel to the iconic tactical first-person shooter, Counter-Strike: Global Offensive. Developed by Valve, it promises to revolutionise the competitive FPS scene with updated graphics, refined gameplay mechanics, and new features. Players can expect the same intense tactical gameplay that has made the franchise so popular, but with a fresh coat of paint and exciting new additions. Counter-Strike 2 at 1440p clearly has a CPU bottleneck issue, and due to that, there really isn’t much information to go off of, as cards are simply all over the place. The only data I feel that we could potentially discuss is over the 4080 SUPER, where we see 20% more performance on the 5080 averages, and 9% more in the lows, but even then, I’d be inclined to take these results, at this resolution, with a pinch of salt. As we move up to 4K, things balance out a lot better, though we don’t see huge differences between each GPU, and instead see single-digit increases over the 4080 and 4080 SUPER of 2.5% and 2% respectively. What’s more interesting here is how the 5080 takes a bit of a hit in the lows and comes in with 10% less performance than our 4080, and 2% less than the 4080 SUPER, though it is nice to see that the 5080 only falls 1% lower than the RTX 4090, but again, the lows do make it suffer a little, but it’s enough to best AMD in the averages, but they fair extremely well in the lows. Cyberpunk 2077 Cyberpunk 2077 is an open-world action role-playing game set in the dystopian metropolis of Night City. Developed by CD Projekt Red, it offers a vast and immersive world filled with diverse characters, complex storylines, and a wide range of customization options. Players can create their own unique cyberpunk character and explore the city's underbelly, completing missions, engaging in cybernetic enhancements, and making choices that shape their destiny. Moving over to Cyberpunk at 1440p, the uplifts are some of the strongest we’ve seen with 23% more performance over the 4080, and 21% more over the 4080 SUPER. Due to the uplift we seen, we now bridge the gap to the 4090, which only comes in with 7% more performance, and this also means that we get 10% more frames over the 7900 XTX. In terms of the lows, these are also very impressive with 27% more than the 4080, 23% more than the 4080 SUPER and 1% more than the 4090, though AMD come in the stronger on this one, putting the 5080 6% behind. As we move up to 4K, we continue to get strong performance from the 5080 that pushes past the 4080 by 23% in the averages, and 24% in the lows, with similarly strong performance over the 4080 SUPER of 20% and 24% respectively. This does also mean that we pack 7% more frames than AMD, though they still push ahead in the lows, putting the 5080 2% behind, but performance is strong enough to now only be sitting 8% behind the 4090, which is a card that retails for, in some cases, three times the MSRP cost of the 5080. F1 24 F1 24 is the official video game of the 2024 Formula 1 World Championship. Developed by Codemasters, it offers a realistic and immersive racing experience with all the official teams, drivers, and circuits. Players can compete in the full F1 season, participate in online multiplayer races, or create their own custom championships. Moving over to F1 24 the increases in frame rates at 1440 aren’t as strong as what we just saw. There are still some welcomed increases of over 10% in both the averages and the lows over the 4080, and 6% and 5% respectively on the 4080 SUPER, but I think consumers are just expecting more, though from our figures, at 1440p, it’s pretty clear to see that we’re being held back by the rest of the system across all cards. Moving to 4K, the RTX 5080 showcases strong performance gains across the board. Compared to the RTX 4080, it delivers a substantial 20% improvement in the averages and 16% in lows. The gains over the 4080 SUPER are slightly more modest but still significant, with 16% in the averages and 13% in lows. However, when stacked against the RTX 4090, we see a dip of 8% in both averages and lows, but for a card that’s cheaper overall, I’m fine with this. Then against AMD’s 7900 XTX, the 5080 holds a solid lead with 10% better averages and an 8% advantage in the lows, reaffirming its dominance at this resolution. Fortnite Fortnite is a massively popular online multiplayer game that combines elements of battle royale, building, and crafting. Developed by Epic Games, it features a constantly evolving world, a wide range of cosmetic items, and a variety of game modes. Players can team up with friends or compete against others in fast-paced matches where the last one standing wins. In Fortnite at 1440p, the RTX 5080 delivers a mixed bag of results, with notable gains in the averages but struggles in the lows. Compared to the RTX 4080, it comes in 8% ahead in the averages but falls 6% behind in the lows, showing inconsistent performance in more demanding moments. Against the 4080 SUPER, the gap narrows slightly, though still leads by 5% in the averages but lags by 9% in the lows. When compared to the RTX 4090, the 5080 trails by 7% in the averages and 8% in the lows, which isn’t too bad considering the price disparity. For a similar price, there’s the 7900 XTX, but here the 5080 leads by 21% in the averages but essentially breaks even in the lows, with just a 0.3% difference though 21% extra performance for the same price is hard to ignore. At 4K, the RTX 5080 continues to show solid improvements over the last generation but does only pull ahead of the RTX 4080 by 11% in both the averages and the lows, offering consistent gains but I think consumers will expect more. The margins remain the same against the 4080 SUPER, where the 5080 also leads by 11% in the averages and 10% in the lows and when compared to the RTX 4090, the 5080 falls behind by 14% in the averages and 2% in the lows which is actually a lot better than expected. Against the 7900 XTX, the 5080 comes out 29% ahead in both the averages and the lows, offering a compelling argument for gamers looking for balanced performance at this resolution without spending over $1000 though that is just the MSRP price and retail price may be worse. Indiana Jones and the Great Circle Indiana Jones And The Great Circle is an upcoming action-adventure game based on the iconic film franchise. Developed by Bethesda Game Studios, it promises to deliver an epic and cinematic experience with a globe-trotting adventure, thrilling action sequences, and a captivating story. Players will step into the shoes of Indiana Jones and embark on a quest to uncover the secrets of a mysterious artefact. In Indiana Jones And The Great Circle, the performance at 1440p is a mixed bag. The RTX 5080 sees an 8% uplift in the averages and a modest 1% improvement in the lows over the RTX 4080, while the 4080 SUPER offers slightly smaller gains where the 5080 comes in ahead by 5% in the averages and with no change in the lows. When compared to the RTX 4090, the 5080 struggles to keep up, falling 14% behind in the averages and suffering a sharp 26% drop in lows though with such a big price disparity, that’s not unexpected. Meanwhile, the NVIDIA Vs AMD battle heats up as the 5080 comes in 15% more powerful than the 7900 XTX in the averages, and 5% better in the 1% lows. At 4K, the 5080 continues to deliver some small uplifts. Compared to the RTX 4080, it shows a 12% improvement in the averages but struggles to make any big blows in the lows, falling 8% behind. The margin narrows slightly when compared to the 4080 SUPER, with the 5080 leading still, but only by 8% in the averages and the gap increases as it sits behind by 12% in the lows. Against the RTX 4090, the 5080 faces an uphill battle, falling 6% behind in the averages and suffering a steep 31% deficit in lows—further showing to us that there are some clear issues with 1% low performance, much like we saw on the RTX 5090. On the AMD front, the 5080 takes the crown with 26% more performance over the 7900 XTX in the averages, but due to its lacklustre performance in the lows, it comes in evenly matched. Starfield Starfield is a highly anticipated space exploration role-playing game developed by Bethesda Game Studios. It offers a vast and immersive galaxy to explore, with countless planets, star systems, and alien civilizations to discover. Players can create their own custom character and embark on a journey to uncover the secrets of the universe, completing missions, engaging in space combat, and making choices that shape their destiny. In Starfield at 1440p, the RTX 5080 offers modest improvements over the previous generation but fails to deliver any groundbreaking leaps. Compared to the RTX 4080, it provides a slight 5% gain in the averages, though it makes up ground with a 27.5% improvement in the lows, offering smoother gameplay in demanding moments. Against the 4080 SUPER, the 5080 edges ahead by just 2% in the averages but maintains a stronger 34% lead in the lows. Then when stacked against the RTX 4090, the 5080 struggles, falling 16% behind in the averages and trailing by 1% in the lows, reaffirming the 4090’s position as the dominant high-end card. Meanwhile, in the NVIDIA vs AMD battle, the 5080 outperforms the 7900 XTX with a modest 4% improvement in the averages and 5% in the lows, though the overall gains here feel underwhelming for what consumers might expect from this new release but what’s more interesting is how the 5080 now increases its performance in the lows, whereas other games, it suffered. At 4K, the RTX 5080 continues to show mixed performance, offering modest improvements over the previous generation. Compared to the RTX 4080, it delivers an 8% improvement in the averages and a stronger 23% boost in the lows, once again emphasizing its ability to smooth out demanding moments in certain games. Against the 4080 SUPER, the 5080 inches ahead with a 2% lead in the averages and a more notable 17% gain in the lows. However, when compared to the RTX 4090, the 5080 struggles in the averages, again falling 16% behind, though it manages to slightly outperform in the lows with a 1% lead, showcasing improved consistency in this title. In the NVIDIA vs AMD matchup, the 5080 edges out the 7900 XTX with an 8% lead in the averages and a 10% improvement in the lows, solidifying its position as a reliable option for gamers looking to balance performance and smooth gameplay. Interestingly, much like at 1440p, the 5080 excels in the lows in Starfield, but overall averages will tell us more. War Thunder War Thunder is a free-to-play vehicular combat multiplayer game that focuses on historical battles across World War II and the Cold War. Developed by Gaijin Entertainment, it features a wide range of aircraft, tanks, and naval vessels from various nations. Players can participate in large-scale battles with other players, completing objectives and striving for victory. Then lastly War Thunder at 1440p gives us a 13% uplift over the RTX 4080 while the move from the 4080 SUPER comes in much less at just 5%, though the 1% lows are much healthier on the new Blackwell-based GPU. This doesn’t sound overly impressive but does widen the gap over AMD as the 5080 now wides the gap as it sits 26% ahead of AMD, so unless the red team can pull it out of the bag with the new RDNA4 cards, it does seem like NVIDIA have the market sewn up. At 4K, we see the 5080 coming in with 8% more performance than the RTX 4080 and a pretty measly 2% over the 4080 SUPER, though again the 1% lows do come in stronger with a 17% uplift, which does again widen that gap over AMD, though I think consumers may have just been expecting a little bit more. Overall Average FPS Results are added together, then averaged out for each graphics card at each resolution to give you an overall picture of their relative performance. As we look at the overall averages, starting at 1080p, we see the 5080 sat pretty much exactly where we expected based on the rest of the charts that we looked at, sitting right below the 4090, but above the 4080 SUPER. This resolution has the tamest differences we saw, as a result of the CPU bottleneck we experience, so we only see a 5% increase in performance over the last generation, but equally, it doesn’t fall that far behind with only 3% less performance than the 4090. So nothing too extraordinary here, but if you’re buying a 5080 then realistically you aren’t playing at 1080p anyway. Something a bit more realistic is 1440p, where we see some of the gaps start to widen somewhat, with the 5080 now sitting 9% ahead of the 4080 and 4080 SUPER, which came in with an identical average FPS. Now I do want to put a caveat on this, as the only reason they come in identically is due to Counter-Strike 2 running pretty abysmally on the 4080 SUPER and therefore bringing both 4080 cards closer together again, though omitting this result would see more of a difference, but we’re focussing on the 5080 here, and not so much the 4080. Now, the gaps do get wider on both sides of the card however, and whilst we did see a bit of a boost with the older 80 series cards, we also see the 4090 take a bigger lead ahead of the 5080, now with 9% more performance than the newer card. This and 4K is really where you’ll be playing with a card like this so it’s nice to see a slight improvement over the same tier of card from last generation as I think gone are the days where a newer XX80 series would try to replace the older XX90 series. 4K is more than playable with an average frame rate of 163 FPS, which ends up being 9% higher than the 4080 SUPER, much like we saw at 1440p while compared to the non-SUPER variant, the newer 5080 comes in around 12% ahead. Compared to the 4090 we see that the 5080 is outperformed by 11%, making this the biggest gap between the two we’ve seen. For its pricing, the 5080 delivers solid results. It's not groundbreaking but still competes well. Against AMD's RX 7900 XTX, which is priced similarly, the RTX 5080 shows a 16% lead in average FPS—a notable win for NVIDIA. The lows do look a lot better in the overall averages, but I do still feel that while some games had a strong 1% low performance, others still suffered, and this could see a lot of change through driver refinements. Performance here is fine, but I do feel that the card needs to mature a little and performance could increase as time goes on. 3DMark Port Royal 3DMark Port Royal is a benchmarking tool specifically designed to measure the ray tracing performance of graphics cards. As the world's first dedicated real-time ray tracing benchmark for gamers, it provides a realistic and practical example of what to expect from ray tracing in games. In Port Royal, we see some notable performance gains over the previous generation, with the RTX 5080 scoring 18% higher than the RTX 4080 SUPER and delivering a 24% improvement over the non-SUPER variant. However, the 5080 still sits below the RTX 4090, trailing by 15%, which highlights the continued dominance of NVIDIA's flagship. Since this benchmark focuses on ray tracing, the results also reflect NVIDIA's advanced ray tracing technologies, which maintain a clear edge over the competition. That said, AMD's RX 7900 XTX performed surprisingly well, coming in just behind the RTX 4080. Despite this, the RTX 5080 outperforms the 7900 XTX by a substantial 37%, a result that gives us a strong indication of what to expect in ray tracing performance during game testing. Overall, the RTX 5080 showcases impressive progress but this is only a synthetic test and we need to see what that means in terms of raytracing in games. Ray Tracing Alan Wake 2 Alan Wake II pushes the boundaries of visual fidelity with its ambitious implementation of "full ray tracing," also known as path tracing. This means ray tracing handles all lighting, reflections, and shadows, creating a more realistic and immersive environment. The game features ray-traced direct lighting for accurate and dynamic shadows, ray-traced indirect lighting for realistic bounced light, ray-traced reflections on surfaces like water and glass, and ray-traced transparency for convincing interaction with transparent objects. In Alan Wake II with ray tracing enabled at 1440p, the RTX 5080 showcases its strengths in handling demanding graphical features but doesn’t entirely dominate across the board. Compared to the RTX 4080, it delivers a solid 14% improvement in the averages and a 12% boost in the lows, offering a noticeable uplift in performance. Against the 4080 SUPER, the gains are more modest, with the 5080 coming in 5% ahead in the averages and 6% in the lows, showing only a slight advantage over this refreshed card. However, when stacked against the RTX 4090, the 5080 falls short, trailing by 15% in the averages and 13% in the lows, reinforcing the 4090's position as the go-to for ray tracing workloads though cost is obviously a big factor. The most striking comparison comes in the NVIDIA vs. AMD matchup, where the 5080 outshines the 7900 XTX by a staggering 70% in the averages and 62% in the lows, demonstrating how far AMD still has to go in the raytracing space. At 4K with ray tracing enabled, the RTX 5080 continues to deliver respectable improvements over the previous generation but still shows some limitations. Compared to the RTX 4080, it offers a 13% improvement in the averages and an identical 13% boost in the lows, showcasing at least, consistent gains. The 4080 SUPER sees smaller margins, with the 5080 coming in just 3% ahead in the averages and 5% in the lows, indicating a narrower performance gap. When compared to the RTX 4090, the 5080 lags behind, trailing by 10% in the averages and a more pronounced 20% in the lows, which highlights its struggle to compete at the top end of the market but considering it’s a lot cheaper, this could be a game changer in terms of the cost per frame. When looking at how AMD do, the 5080 absolutely dominates the 7900 XTX, leading by a jaw-dropping 106% in the averages and 84% in the lows, making it clear that AMD's ray tracing capabilities remain significantly behind NVIDIA’s latest offerings. Ray Tracing Black Myth Wukong Black Myth Wukong is set to be a visual tour-de-force, thanks to its implementation of full ray tracing, also known as path tracing. This means ray tracing is used to render all lighting, reflections, and shadows within its fantastical world. Ray-traced reflections enhance the serene beauty of water features and the shine of polished surfaces, while ray-traced shadows add depth and realism to every scene. Ray-traced indirect lighting allows light to bounce naturally, creating a more believable environment. Even the game's particle effects, often a challenge for ray tracing, are rendered with impressive accuracy. In Black Myth: Wukong at 1440p with ray tracing enabled, the RTX 5080 showcases substantial gains over previous-generation cards. Compared to the RTX 4080, it delivers a 20% improvement in the averages and an equally impressive 20% boost in the lows, offering a smoother and more consistent experience overall. The margins narrow slightly against the 4080 SUPER, where the 5080 still comes out ahead with a 15% improvement in both the averages and the lows. However, the RTX 4090 still retains its high-end dominance, with the 5080 trailing by 10% in both the averages and the lows. The most striking result, however, is in the NVIDIA vs. AMD comparison. Here, the 5080 continues to utterly obliterate the 7900 XTX, boasting a staggering 233% lead in the averages and an even more incredible 307% improvement in the lows. This cements the 5080 as the clear winner for ray tracing in this title, especially at its given price point. At 4K, the RTX 5080 continues to deliver solid improvements over the previous generation. Compared to the RTX 4080, it offers a 24% boost in the averages and a 22% improvement in the lows, maintaining a strong lead. Against the 4080 SUPER, the margins are slightly smaller, with the 5080 ahead by 19% in the averages and 12% in the lows while again, the RTX 4090 remains competitive, at least in terms of performance with the 5080 falling just 3% behind in the averages and 15% in the lows, showing that NVIDIA’s flagship of last generation still holds its ground at 4K. Once again, the AMD matchup is striking, with the 5080 outpacing the 7900 XTX by an incredible 287% in the averages and an even larger 350% in the lows, further cementing NVIDIA’s advantage in ray tracing performance, though who knows what the RDNA4 lineup will hold. Ray Tracing Cyberpunk 2077 Cyberpunk 2077 was already a visually impressive game, but the implementation of "Ray Tracing: Overdrive Mode" takes things to a whole new level. This mode utilises path tracing, a form of full ray tracing, to render all lighting, reflections, and shadows with exceptional accuracy. This means every neon sign, streetlamp, headlight, and screen in Night City contributes to the illumination of the environment, creating a truly breathtaking and immersive experience. Ray-traced reflections bring the city's rain-slicked streets and glass skyscrapers to life, while ray-traced shadows add depth and realism to every scene. In Cyberpunk at 1440p with ray tracing enabled, the RTX 5080 shows respectable performance gains but doesn't push the boundaries significantly. Compared to the RTX 4080, it achieves an 11% improvement in the averages and an 8% boost in the lows, providing a smoother experience in demanding scenes like what we tested. Against the 4080 SUPER, the performance difference narrows further, with the 5080 coming in just 4% ahead in the averages and 2.5% in the lows. Then when compared to the RTX 4090, the 5080 lags behind by 22% in the averages and 21% in the lows, which I feel some may be a little disappointed in. In the battle with the 7900 XTX, the 5080 pulls far ahead again, leading by 85% in the averages and 86% in the lows. At 4K, the RTX 5080 continues to demonstrate its performance gains over the previous generation. Compared to the RTX 4080, it delivers an 18% improvement in the averages and a 17% boost in the lows, maintaining a strong advantage. Against the 4080 SUPER, the 5080 pulls ahead by 8% in the averages and 11% in the lows, showcasing a noticeable, if not groundbreaking, lead. However, when compared to the RTX 4090, the 5080 trails by 24% in the averages and 22% in the lows, falling short of the Ada flagship’s raw power. The 7900 XTX, on the other hand, lags well behind, with the 5080 outperforming it by 86% in the averages and 90% in the lows, providing a much more stable and enjoyable experience in this graphically demanding title. Ray Tracing F1 24 F1 24 brings the thrill of Formula 1 racing to life with its implementation of ray tracing, adding a new layer of realism to the on-track action. Ray-traced reflections accurately capture the gleaming surfaces of the cars and the vibrant colours of the race tracks, while ray-traced shadows add depth and detail to every scene. The introduction of ray-traced ambient occlusion further enhances the visual fidelity, providing more realistic and nuanced shading. Even the crowds in the grandstands benefit from ray tracing, with more detailed and dynamic lighting. In F1 24 at 1440p, the RTX 5080 puts in a strong performance, starting with a 17% improvement in the averages and an even more impressive 27% boost in the lows compared to the RTX 4080, highlighting smoother gameplay in critical moments. Moving to the 4080 SUPER, the gap narrows slightly, with the 5080 leading by 10% in the averages and 7% in the lows, still retaining a notable advantage. Against the RTX 4090, the 5080 falls behind by 15% in the averages, but it manages to come out ahead in the lows with a 10% improvement, offering a more consistent experience in some scenarios. Finally, when facing off against the 7900 XTX, the 5080 solidifies its position with a 28% lead in the averages and a 14% advantage in the lows. Then at 4K, the RTX 5080 continues to deliver notable gains. Compared to the RTX 4080, it achieves a 21% improvement in the averages and an 18% boost in the lows, maintaining its lead as the better option for higher resolutions. Against the 4080 SUPER, the margins are smaller, with the 5080 ahead by 14% in the averages and just 5% in the lows, showing a diminishing but still present advantage. When stacked against the RTX 4090, the 5080 falls behind by 18% in the averages and 12% in the lows, demonstrating that NVIDIA's former flagship still holds strong in raw raytracing-based performance. Finally, when compared to the 7900 XTX, the 5080 comes out ahead once more, with a 38% lead in the averages and a 23% advantage in the low. Ray Tracing Overall Average FPS In terms of the averages during ray tracing, when we look at 1080p the RTX 5080 shows itself to be a strong contender, coming in just behind the RTX 4090 with an average FPS of 137. This places it 5% ahead of the RTX 4080 SUPER, which delivered 127 FPS on average, and 13% ahead of the standard RTX 4080 at 121 FPS. Compared to AMD’s RX 7900 XTX, the gap is even more noticeable, with the 5080 boasting a 56% advantage. This performance makes it clear that NVIDIA’s architectural improvements are paying off in ray tracing-heavy scenarios, even at lower resolutions. While the 1% lows are also solid at 101 FPS, the RTX 4090’s consistent performance keeps it firmly ahead in this resolution. At 1440p, the RTX 5080 continues to shine, with an average FPS of 98, placing it 7% ahead of the RTX 4080 SUPER and 17% ahead of the RTX 4080. While these gaps widen slightly compared to 1080p, the 5080 still trails the RTX 4090 by 19%, showing that the flagship card maintains a clear lead in this resolution. When compared to the RX 7900 XTX, the 5080’s performance advantage shrinks somewhat, now sitting at 38%, but it’s still a substantial lead that demonstrates NVIDIA’s dominance in ray tracing workloads. The 1% lows remain consistent at 79 FPS, ensuring a smooth gameplay experience for demanding titles. Moving up to 4K, the RTX 5080 manages an average FPS of 54, which, while playable in many cases, starts to push the limits of this card in ray tracing-heavy games. It sits 7% ahead of the RTX 4080 SUPER at 49 FPS and 17% ahead of the standard RTX 4080 at 46 FPS, maintaining the same relative performance gap we saw at 1440p. The RTX 4090, however, pulls even further ahead, delivering a 24% higher average FPS but we do have to remind ourselves of the clear price disparity. Against the RX 7900 XTX, the 5080 remains dominant, outperforming it by 74%, so if raytracing is what you’re after, there really is no choice other than NVIDIA. The 1% lows drop to 44 FPS here, indicating that while the performance is solid, the card may struggle in more demanding ray-traced scenes. Upscaling Alan Wake II Alan Wake II is a survival horror game and the long-awaited sequel to Remedy Entertainment’s cult classic, Alan Wake. This time, the story takes a darker turn, delving deeper into the psychological and supernatural elements that made the original so compelling. Players can expect a chilling atmosphere, intense combat encounters, and a mind-bending narrative that will keep them on the edge of their seats. With Alan Wake II, we weren’t able to manually choose between different DLSS models, so our focus is on the performance improvements when enabling each level of upscaling. Starting with DLSS, which was set to balanced across all our tests, the results were immediately impressive. The average FPS more than doubled compared to native rendering, with a substantial 102% increase. Latency also saw a noticeable reduction, dropping from 48.41ms at native to just 25.27ms, cutting it nearly in half. While that alone is a strong result, the real gains came with frame generation, which boosted performance by 210% over native, adding around 60 extra frames per second, though latency crept back up slightly to 34.04ms. Building on this, enabling Frame Generation X3 pushed performance even higher, delivering a 320% increase compared to native rendering which added another 60 FPS to the total, making a noticeable difference, although latency continued to rise, landing at 37.35ms. Then when we enabled Frame Generation X4, the gains were slightly smaller but still significant, with a 49 FPS jump over the previous setting and this brought the total performance increase to an incredible 403% over native, with the frame rate landing just shy of 300 FPS, while latency increased to 40.99ms. Even so, X4’s latency remained lower than native, showing that the performance improvements come with minimal sacrifices in responsiveness. Turning on raytracing provided similar trends, but with smaller overall performance gains compared to non-ray-traced settings. From native rendering, enabling DLSS delivered a 138% performance uplift, equivalent to a 36 FPS improvement, while reducing latency from 100.97ms down to 45.41ms which is a significant improvement. Adding frame generation made an even bigger difference, pushing performance up by 308% over native and adding another 80 FPS, though latency did climb a bit to 57.09ms. The jump from just DLSS to frame generation wasn’t as dramatic, with an increase of 44 FPS, and the trend continued with Frame Generation X3 and X4, where the final performance increase hit an astonishing 627% over native rendering. At this point, latency rose to 63.36ms, but it still stayed well below the 100.97ms we got with native rendering. The performance gains are impressive, but the way latency scales across these settings is equally worth noting. In both ray-traced and non-ray-traced scenarios, DLSS alone halved the latency compared to native, offering a substantial improvement in responsiveness. As we moved into multi-frame generation settings like X3 and X4, latency began to climb back up slightly, but even at its highest, it never exceeded native rendering levels. This demonstrates that while the advanced settings add complexity, they still manage to deliver both performance and responsiveness in a balanced manner. Overall, the upscaling and frame generation technologies in Alan Wake II provide more than just higher frame rates—they deliver smoother gameplay and improved latency, even under the most demanding settings. This showcases how features like DLSS and Frame Generation X4 can redefine the gaming experience, ensuring better performance without sacrificing responsiveness. Upscaling Cyberpunk Cyberpunk 2077 is an open-world, action-adventure RPG set in the dark future of Night City — a dangerous megalopolis obsessed with power, glamour, and ceaseless body modification. You play as V, a mercenary outlaw going after a one-of-a-kind implant that is the key to immortality. Cyberpunk 2077 supports up to DLSS 4 scaling and the latest Frame Gen technologies. Cyberpunk 2077 follows a similar trend to what we saw in Alan Wake II, with each level of upscaling delivering improved performance and reduced latency compared to native rendering. Enabling DLSS on both the Transformer and CNN models provided substantial gains, with latency nearly halved at the base DLSS setting before climbing slightly with each level of frame generation. Interestingly, while Alan Wake II showed consistent performance scaling with advanced settings, Cyberpunk adds a twist—CNN consistently outperforms the Transformer model in both frame rate and latency, with around 6% more frames and slightly faster response times. This deviation highlights a potential focus of the Transformer model on better visuals rather than raw performance, something that we’ll explore further when comparing image quality. Nevertheless, both models maintain the pattern we’ve come to expect from NVIDIA’s upscaling technologies, delivering massive performance gains across all settings, all while keeping latency improvements below native rendering. The choice here ultimately comes down to whether gamers prioritize responsiveness or visual fidelity, but the underlying trend aligns closely with what we observed in Alan Wake II. With ray tracing enabled, Cyberpunk 2077 shows the same consistent trend we’ve seen before, with each upscaling and frame generation setting delivering higher frame rates and gradually increasing latency. The CNN model continues to outperform the Transformer model, offering both better frame rates and lower latency, likely due to its focus on raw performance over visual enhancements. Regardless of the model used, the gains are substantial—DLSS alone cuts latency by more than half compared to native rendering, and each frame generation level pushes performance significantly higher. Whether you prioritize the responsiveness of CNN or the potential visual fidelity of the Transformer model, enabling upscaling provides a massive boost, ensuring a smoother experience even with ray tracing turned on. Path tracing in Cyberpunk 2077 shows a significant uplift across all upscaling and frame generation settings, making an otherwise unplayable experience viable. Interestingly, while the CNN model generally performs well in terms of latency, the Transformer model pulls ahead in overall performance at the highest settings. For example, DLSS CNN cuts latency dramatically, from 153.23ms at native to 51.98ms, while boosting performance by 300%. However, at Frame Generation X4, the Transformer model delivers the best result, achieving 132 FPS compared to CNN’s 128 FPS, with slightly lower latency at 69.25ms versus CNN’s 71.54ms. This highlights how the Transformer model can balance higher performance with improved latency, particularly at the top settings, while CNN maintains a strong showing overall. Regardless of the chosen model, the impact of upscaling on path tracing performance can turn Cyberpunk into a playable and visually stunning experience. Upscaling Visual Comparison We also wanted to show visuals, because numbers with latency are one thing, but side by side comparisons are always going to be a key point, because no one wants super high performance if it results in the game looking like garbage, and we will be doing some more content on this soon in a lot more titles, comparing performance and visuals. For the visuals in Alan Wake II we see at native the game is expectedly amazing looking, and when we enable upscaling we see some loss of quality, with things being just a little more fuzzy looking, and with a little more noise, but nothing too bad and definitely nothing you'll notice when gaming. When we enable frame generation we really don’t see anything that stands out visually, and I wasn't able to pick up on anything out of place or any artifacts, so all in all, this is a pretty good way of playing the game. With raytracing turned on we don’t see too much changing, but it does provide a more realistic look. This is probably because raytracing is kind of always enabled in alan wake, and the raytracing settings are more about path tracing, so that’s expected as we aren’t getting the same sort of jump in visuals as we would going from no ray tracing at all to ray tracing enabled. When we enable upscaling we don’t really see much difference at all, and almost nothing is noticeable visually. This is mostly the case when we enable multi-frame generation too, but one very small difference is that when the light flashes on the wall, we see the graffiti’s reflection flicker a bit, so it’s really not a whole lot of obvious differences, and overall i think it’s a worthwhile sacrifice to make for much better frame rates. Moving over to Cyberpunk, and if you haven't seen any of our 5090 content then may want to check that out as we have already gone over the visual differences between the different settings, so for the sake of brevity, I won't be going into as much detail here and since we’ve learned that there really isn't any noticeable difference in quality so for this we’ll only be comparing native to frame gen. Comparing the game regularly with multi-frame generation enabled doesn’t really give much of a visual difference, and in this scene in particular, takes almost nothing away. You could argue that there is more noise but it’s so negligible that you can’t really tell but what if we compare the models though? Well, when we put our frame generation clips together we see very few differences once again, but if you look closer than you find that the CNN model is a little bit less clear, with the most noticeable point being underneath the light where you can see the smoke has more visual noise on the older model, whereas the transformer model provides a slightly clearer visual. Raytracing then improves our visuals, with better lighting and reflections as you would expect, but it does add some visual noise; and when we take a look at how multi-frame generation affects things we can see, again, barely any difference. One very small thing I noticed was that the reflection of the neon sign in the puddle had a very slight delay in its movement when compared to the idle camera movement and the sign on the wall. This was present without upscaling enabled but it did get a very slight exaggeration with upscaling. When we look at CNN though we find that not much changes, but if you’re paying close attention, you can see that it does introduce a more obvious noise into the visuals, albeit only slightly Path tracing is last and yet again, visuals are improved over ray tracing, even if it does vastly change the way the scene looks. When we compare native to multi-frame gen we see some minor differences like we saw with the visual noise like before, but we also see the entire scene brighten up, and this scene takes place in a room that has no outside lighting so it remains consistent throughout, meaning that this isn’t a change in game time causing the sun to light the room up more. This is a pretty interesting difference and I'm not really sure why it happened, as we didn’t see this kind of behaviour with the other tests so the fact that it shows up with path tracing is a bit odd. Then when we look at CNN we see that it once again only really introduces a bit of extra visual noise Power & Temperatures So gaming performance and upscaling to one side. We always want to take a look at the cooler and see how that performs along with power requirements as these figures play a pivotal part in picking which brand of GPU and which model, so we booted up F1 24 and stuck it on an hour-long loop to see how good, or bad the Founders card actually is. It’s here where the 5080 delivers solid thermal and power efficiency, maintaining a GPU temperature between 64°C and 68°C, averaging around 66°C. The GPU Memory Junction Temperature follows closely, peaking at 67°C and holding an average of 68°C, indicating effective cooling performance overall. Performance-wise, the GPU clock hovers around 2700 MHz, while the memory clock remains locked in at 15,001 MHz. GPU usage stays high, averaging 98%, with memory usage at a steady 86% though during this test, our fan speed remained quiet, fluctuating slightly but averaging just under 1500 RPM. For power consumption, it’s pretty well-managed, with the total board power ranging from 310W to 325W which is where we expected things to be given its 360W max power rating. Cost Per Frame MSRP Before we dive into the cost-per-frame analysis, it’s important to consider the context of MSRP pricing. While it provides a baseline for comparisons, real-world availability and market conditions often mean prices can fluctuate, sometimes significantly. That said, MSRP values are still a good reference point for understanding the relative value of different GPUs in NVIDIA’s Blackwell lineup and beyond. Looking specifically at 1440p performance, the RTX 5080 comes in at an MSRP cost per frame of $3.98. This positions it as one of the most efficient high-end cards in terms of value, outperforming the RTX 4080 SUPER, which costs $4.35 per frame, and even undercutting AMD’s RX 7900 XTX at $4.44. This shows that NVIDIA has managed to deliver a more competitive price-to-performance ratio for gamers who want strong performance without venturing into flagship territory. Compared to the RTX 5090, which comes in at $6.61 per frame, the 5080 is significantly more cost-effective, offering a 40% lower cost per frame while maintaining respectable performance levels. Even the RTX 4090, at $5.78 per frame, can’t compete with the 5080’s value, highlighting the 5080 as the card to choose for gamers who care about getting the most performance per dollar in NVIDIA’s higher-tier GPUs. Ultimately, the RTX 5080 strikes an excellent balance between cost and performance at MSRP. While not the absolute cheapest in terms of cost per frame, it offers an appealing option for gamers who want high-end capabilities without stepping up to the RTX 5090 or compromising on features with a lower-tier card. Of course, MSRP is just part of the equation, and the real-world value of the RTX 5080 becomes even more interesting when we consider typical selling prices, which we’ll look at soon. Looking at 4K performance, the RTX 5080 comes in with a cost per frame of $6.13. This again puts it in a really good position when compared to other cards, like the RTX 4080 SUPER at $6.67 per frame and the RTX 4080 at $8.29 per frame. Even AMD’s RX 7900 XTX, priced at $7.08 per frame, falls behind the RTX 5080, which again shows that NVIDIA have managed to get things right here. Cost Per Frame Selling Price When compared to NVIDIA’s flagship RTX 5090, which comes in at $8.60 per frame, the 5080 offers a 29% reduction per frame. While the 5090 is undoubtedly the performance leader, the 5080 is far more efficient in delivering value for money, making it a smarter choice for most gamers who don’t want to stretch their budget to extreme levels. Even the RTX 4090, which costs $8.76 per frame, struggles to match the efficiency of the RTX 5080 in this context, but this is all about MSRP, and we all know that the 4090 in reality is much more expensive. Speaking of reality, at 1440p in terms of selling price, the RTX 5080, as we know, comes in at a cost per frame of $3.98. The RTX 4080 and 4080 SUPER now lag behind at $6.90 and $6.96 per frame, respectively, showing just how much value the 5080 provides in NVIDIA’s stack or how much the 4080 and 4080 SUPER don’t because of their inflated retail pricing. The worst one is the flagship RTX 4090 from last generation, sitting at $10.12 per frame, which can’t come close to matching the 5080’s pricing efficiency, further cementing it as a standout option for high-performance 1440p gaming. Looking at AMD’s offerings, the RX 7900 XTX, priced at $4.26 per frame, and the RX 7900 XT at $3.52 per frame, does provide some stiff competition. However, the 5080 narrows the gap significantly when we factor in performance. Ultimately, at the selling price, the RTX 5080 looks good at 1440p, but we don’t know if retailers are going to try and profiteer from the 5080 and try to sell it for higher than MSRP from the get-go. When we move up to 4K, the RTX 5080 continues to hold its own in terms of value. The RTX 4080 SUPER and 4080 now fall short, priced at $10.68 and $10.96 per frame, respectively, showing that the 5080 is the far better value proposition for NVIDIA enthusiasts, especially if wanting the latest technology. The RTX 4090, unsurprisingly, performs well but at a staggering $15.33 per frame, which is just beyond what anyone should be paying. For AMD, the competition is closer, as the RX 7900 XTX comes in at $6.79 per frame, while the RX 7900 XT offers slightly better value at $5.91 per frame. Both are strong contenders in the value space, but the 5080 bridges the gap when we factor in other elements not shown in this chart, like ray tracing and upscaling performance. Budget options like the RX 7900 GRE and RTX 4070 SUPER deliver the best cost per frame, coming in at $5.53 and $6.15, respectively. However, these cards are targeted at a different market segment and don’t offer the same level of performance for 4K gaming, which is where the 5080 excels, and you won’t get the benefits of the Blackwell architecture, including DLSS with multi-frame generation. How Much Does it Cost? The Nvidia GeForce RTX 5080 Founders Edition launches with an MSRP of £979 on January 30th 2025. However, there will be custom cards from partner brands too, and while some may be MSRP cards, many are likely to be above that with custom cooling and overclocking configurations. You can check for stock on the Nvidia website here, and for custom cards at Scan.co.uk here. Overview Let's break this down because the RTX 5080 sits in an interesting position within NVIDIA’s Blackwell lineup. It’s not the flagship 5090, but it’s far from a mid-range option. Instead, it aims to deliver high-end gaming performance with fewer compromises and a more palatable price point. While the 5090 is designed for enthusiasts and professionals looking for no-compromise performance, the 5080 caters to those who want a balance between cutting-edge features and a reasonable investment. Rasterisation Performance Starting with rasterisation, the RTX 5080 delivers a strong uplift over its predecessor, the RTX 4080. In many cases, we see a 10-15% performance increase, which is notable for this tier. However, as with other cards in the series, the 1% lows still feel like they need some attention. It’s not to say they’re bad – they’re perfectly acceptable for gaming – but they occasionally lag behind the card’s average performance, which can create moments of inconsistency in demanding titles. This could likely be addressed through driver optimisations, which tend to smooth out over time. Interestingly, while rasterisation performance is solid, it’s clear that NVIDIA is no longer focusing purely on this metric for generational improvements. Instead, they’re looking to leverage new technologies like ray tracing and upscaling to deliver meaningful performance gains and more immersive visuals. While there’s still room for improvement in rasterisation alone, it’s clear that NVIDIA sees the future of gaming as a blend of traditional rendering and newer technologies, and the RTX 5080 is a great example of this direction. Ray Tracing Performance Ray tracing has been one of NVIDIA’s major selling points for several generations now, and the RTX 5080 continues to push this technology forward. Compared to the RTX 4080, the 5080 handles ray tracing significantly better, offering smoother gameplay and higher frame rates in titles that use advanced lighting, shadows, reflections, and global illumination. What’s particularly impressive is how the 5080 compares to AMD’s offerings, with NVIDIA still leading the way in ray tracing workloads. The performance gap becomes even more evident at higher resolutions like 1440p and 4K, where the 5080 delivers playable frame rates even without upscaling. As games continue to adopt more ray-traced elements and developers aim for fully ray-traced environments, the RTX 5080 shows that it’s ready for this shift. Its architecture is well-suited for next-gen visuals, offering a blend of raw power and efficiency that ensures games not only look stunning but run smoothly as well. That said, the future of ray tracing isn’t just about making it playable – it’s about creating fully immersive environments. Developers are moving towards real-time global illumination, reflections, shadows, and ambient occlusion being calculated simultaneously, which represents the next level of realism. The RTX 5080 may not achieve the same performance levels as the 5090 in these scenarios, but it handles them well enough to be a great option for those looking to enjoy cutting-edge graphics without jumping to the flagship model. And if it doesn’t, then that’s where upscaling comes in. Upscaling and Multi-Frame Generation (MFG) Upscaling and multi-frame generation on the RTX 5080 delivers impressive performance improvements, with significant gains over native rendering across all settings. These features not only boost frame rates dramatically but also manage to maintain lower latency, even as more advanced settings are enabled. The balance between performance and responsiveness makes these technologies a standout piece of tech, and something that I’m keen to see progress as the hardware and software mature. Who Is the RTX 5080 For? The RTX 5080 is clearly targeted at gamers and enthusiasts who want high-end performance without stepping into the extreme pricing and power requirements of the RTX 5090. It’s a card that thrives at 1440p and 4K gaming, offering a great mix of features, performance, and price. While it doesn’t aim to be the king of the hill, it provides a more approachable option that still showcases much of what the Blackwell architecture can deliver without going into 5090 territory. Unlike the 5090, the 5080 feels less like a showcase of what’s possible and more like a workhorse for gamers who want a reliable, high-performance GPU that can handle the demands of modern and next-gen games. It’s not about pushing every boundary; it’s about delivering a well-rounded experience that balances power, efficiency, and price. I feel it’s only going to get better as drivers and game optimisations come along. With it even easier for developers to incorporate upscaling and MFG into their games, we should start to see more and more titles adopt this latest technology, though 75 games at launch is still pretty impressive too. Should I Buy One? So, wrapping things up. The RTX 5080 is a card that feels like it was designed with gamers in mind. It takes the strengths of the 5090 – its cutting-edge features, improved architecture, and raw power – and puts it all into a product that’s more accessible while still delivering a fantastic gaming experience. It may not have the absolute best performance in every category, but it doesn’t need to. Instead, it strikes a balance that will appeal to a wider audience, offering next-gen capabilities without the need for flagship pricing or hardware requirements. While it’s not for everyone, the RTX 5080 is a great example of NVIDIA’s ability to scale its technologies across multiple tiers, ensuring that even non-flagship products deliver an impressive experience. It’s a card that shows what’s possible when you balance innovation with practicality, and it’s exciting to see how it sets the tone for the rest of the Blackwell lineup. To be completely honest with you, aside from the 5090, as it's the top dog, I’m excited more about the RTX 5070 and what that can deliver, especially given its price point. Maybe in the near future, we’ll do a re-review of this card as those all-important driver updates come in.