NVIDIA GeForce RTX 4080 Founders Edition
USD 1,199 USD and AUD 2,219
November 16, 2022
The arrival of the GeForce RTX 4090 last month set the scene for the next generation of NVIDIA GeForce graphics, built on the new Ada architecture. And it did so with all of the style and wonder of the ray-traced Night City found in Cyberpunk 2077. With settings turned up and the resolution set to 4K. In terms of performance and efficiency, it made the GeForce RTX 3090’s status as a beast for high-end PC gaming yesterday’s news. Quite literally, the day the RTX 4090 arrived with its 1.8X increase in performance over the RTX 3090 in 4K, well, it was like finding the warp whistle in Super Mario Bros. 3
and skipping a few generations. That sort of next-level.
The RTX 40 series debut also brought some brand-new rendering technology into the NVIDIA fold with DLSS 3. Which uses new specialised hardware to generate entire frames using GPU-rendered data, motion vectors, and all manner of technical wizardry to boost performance. An idea that’s hard to wrap your head around. Exclusive to the GeForce RTX 40 series, Frame Generation joins the existing DLSS Super Resolution (aka DLSS 2) and NVIDIA Reflex to form a TriForce of rendering power. Which is why it’s called DLSS 3. Okay, I’ll stop with the Nintendo references. For now.
Built on the same Ada architecture as the formidable GeForce RTX 4090, the GeForce RTX 4080 takes full advantage of DLSS 3. Its implementation in Microsoft Flight Simulator, as part of the new Anniversary update, is nothing short of a game changer. One of the most technically advanced and visually stunning PC experiences out there, it just wasn’t possible to play it in 4K with Ultra quality settings and maintain a solid 60 fps until, checks notes
, this week. Running on a GeForce RTX 4080, with Frame Generation, you’re looking at a solid 120 fps using the DLSS Super Resolution ‘Quality’ setting. And that can go even higher if you choose ‘Performance’.
Impressive stuff for sure, but the GeForce RTX 4080 is a step down from the GeForce RTX 4090 whilst still weirdly being the same physical size. The big question surrounding the GeForce RTX 4080 is how exactly it fits into the bigger picture. Now that the RTX 4090 set the bar so high, how does the next GPU in the new GeForce RTX 40 Series stack up? How does it compare to the GeForce RTX 3080, and the RTX 3090? The short answer is roughly somewhere there in the middle, but the main takeaway - like with the GeForce RTX 4090 - is that you’ve got another GPU from NVIDIA that is a true 4K crank-the-settings triple-digit-fps powerhouse.
Let’s dig in.
Introducing NVIDIA Ada
The below is a summary of the RTX 40 Series technology, applicable to all RTX 4080 models. An Ada Lovelace primer if you will.
Across PC and console, from indie games to competitive shooters to AAA blockbusters, we’re now in the age of visual fidelity plus
performance. This a transition that has happened somewhat gradually over the past decade, but we’re now firmly at a point where CG-like visuals need to be backed up by 60 frames-per-second performance at a minimum.
We’re also in the age of 4K resolutions and high-refresh-rate 1440p and ultrawide displays. With the GeForce RTX 30 series, we saw visual fidelity plus
performance at the heart of it all. The bonus of HDMI 2.1 support and a PCIe Gen 4 interface helped handle all the extra data and stuff needed as high-end gaming moved beyond 1080p. NVIDIA even took steps to improve system latency with the arrival of NVIDIA Reflex, which is now a key part of many competitive-focused games and a component of Ada’s DLSS 3 tech.
Built on the same Ada architecture as the formidable GeForce RTX 4090, the GeForce RTX 4080 takes full advantage of DLSS 3. Its implementation in Microsoft Flight Simulator, as part of the new Anniversary update, is nothing short of a game changer.
The new NVIDIA Ada Lovelace GPU architecture is named after mathematician Ada Lovelace, considered the world’s first computer programmer. Built on cutting-edge TSMC 4N process technology, the process shrinkage not only means more of everything in terms of NVIDIA CUDA cores and RT Cores but considerable improvements made to power efficiency.
From a pure numbers perspective, the full NVIDIA Ada GPU features 76.3 billion transistors, up to 18,432 CUDA Cores (70% more than the previous gen), and delivers clock speeds above 2500 MHz whilst maintaining the exact power requirements of the GeForce RTX 3090 Ti. Impressive. But that’s the fully unlocked version we probably won’t see in a desktop variant for a while. The GeForce RTX 4080 features 45.9 billion transistors, 9728 CUDA Cores, clock speeds of 2505 MHz, 16GB of GDDR6X memory, and a power rating similar to that of the GeForce RTX 3080. Thanks to Ada’s impressive efficiency though, it can perform 35% faster than the GeForce RTX 3090 in 4K whilst using 25% less power.
Behind the scenes, well, inside the silicon, there’s the new Shader Execution Reordering which means that rendering tasks happen more efficiently and without bottlenecks. Ray-tracing is the realistic rendering of light, covering shadows, reflections, and how the sun interacts with objects. Traditionally games bake (pre-render) lighting because the effect is costly to do in real-time, and more prominently found in the realm of film and Pixar animation. With NVIDIA being a pioneer when it comes to real-time ray-tracing hardware, Ada RT represents the next evolution in the stuff that makes games like Cyberpunk 2077, Dying Light 2, Metro Exodus, Ghostwire: Tokyo, DOOM Eternal, and more look as impressive as they do.
Ada’s new AV1 encoder is 40% more efficient than the H.264 video encoder used in the GeForce RTX 30 Series. For streamers and content creators, the GeForce RTX 40 Series is also a game changer, delivering better quality video using the same bitrate.
Ada RT Cores are capable of doing all of the complex calculations to deliver real-time ray-tracing two times faster than the GeForce RTX 30 Series’ Ampere architecture. So much so CD Projekt Red will be releasing an update to Cyberpunk 2077 to significantly upgrade the level of ray-tracing in the game (not available at the time of this review). This ‘Overdrive’ update will make full use of DLSS 3, the cutting-edge form of neural graphics exclusive to the GeForce RTX 40 series.
DLSS 3 takes advantage of new Optical Flow Acceleration to generate entire frames. It’s all thanks to the Ada Fourth-Generation Tensor Cores, hardware that turns the GeForce RTX 4080 into an AI supercomputer of sorts. On top of this, there are also advances in video encoding. Ada’s new AV1 encoder is 40% more efficient than the H.264 video encoder used in the GeForce RTX 30 Series. For streamers and content creators, the GeForce RTX 40 Series is also a game changer, delivering better quality video using the same bitrate.
Ada, and the GeForce RTX 4080, represent a considerable leap forward in graphics technology.
Of Numbers and Heat Sinks
The NVIDIA GeForce RTX 4080 Founders Edition look and visual design is an extension of the stunning Founders Edition that debuted with the GeForce RTX 30 series. Albeit in a redesigned form to allow for better airflow. Even though spec-wise there’s a pretty big difference between the RTX 4080 and the RTX 4090, the physical size of the two cards, and housing, are the same. Meaning you’ll need quite a large case, with a potential bracket for support, to properly install the GeForce RTX 4080.
The sheer size of the GPU is the first thing you see when opening the box, and it kind of sets an expectation for some impressive performance. Which the GeForce RTX 4080 delivers, more on that in a bit.
NVIDIA GeForce RTX 4080 Founders Edition
- Architecture (GPU): Ada
- CUDA Cores: 9728
- RT Cores: 76 (3rd Gen)
- Tensor Cores: 304 (4th Gen)
- GPU Boost Clock: 2505 MHz
- Memory Capacity: 16GB GDDR6X
- Memory Interface: 256-bit
- Memory Bandwidth: 716.8 GB/s
- TDP: 320W
The good news is that the large physical size does mean that in terms of staying cool and keeping quiet, the NVIDIA GeForce RTX 4080 Founders Edition is more than sorted. Across several benchmarks, temperatures remained well within the 60-70 degree range, sometimes sitting firmly in the region of 50 degrees, even when stressed. Plus, it’s basically silent.
Although the 320W power draw is similar to the GeForce RTX 3080, the GeForce RTX 40 series presents a major improvement in terms of performance per watt over the Ampere generation. In testing several titles it wasn’t uncommon for the power draw to hover at around 250W when gaming in 4K, and around 200W when gaming in 1440p. Comparatively the RTX 3080 pretty much always hovers around its 320W limit when pushed, so when the RTX 4080 easily trumps the performance of the RTX 3080 - it does so using less power.
Before we dive into benchmarks and performance though, which will feature so many numbers you might just end up having a few numerical-based nightmares, it’s worth keeping the above comparison in mind. Not only in terms of the big gap between the GeForce RTX 4080 and RTX 4090 specs but how it compares to the two top models in the previous Ampere generation. This is a testament to just how well Ada scales, maintaining incredible performance.
The GeForce RTX 4080 only features 10% more CUDA Cores that the GeForce RTX 3080, 8% fewer than the RTX 3090, and a sizable 40% less than the GeForce RTX 4090. And yet it still handily outperforms both RTX 30 Series GPUs. So what does that mean for performance, exactly? Let’s find out.
With the GeForce RTX 4090, we began to see a situation where gaming in 1440p kind of hit a ceiling in terms of what was possible, pointing to that flagship GPU being more suitable for high-end 4K gaming using a high refresh-rate display. To the point where it doesn’t make a lot of sense otherwise. Or at least pair it with an ultrawide that sits somewhere between 1440p and 4K when it comes to raw pixel counts. In the console space, 4K TVs rule the day, but even so - you need considerable hardware to maintain fast frame rates.
The GeForce RTX 4080 is similar to the RTX 4090, it is a 4K gaming GPU so the following benchmarks cover 1440p and 4K. A sizable leap over the RTX 3080 and the RTX 3090.
Before we jump in though, here’s a quick look at our new 2022 testing rig.
- GPU: NVIDIA GeForce RTX 4090 Founders Edition
- CPU: AMD Ryzen 9 7900X
- Motherboard: MSI MAG X670E Carbon WiFi
- Memory: Corsair DOMINATOR PLATINUM RGB DDR5 DRAM 5200MHz
Here’s the first batch of games testest in 1440p and 4K resolutions with all detail settings set to Ultra-equivalent levels and DLSS and ray-tracing enabled where possible. For DLSS the Quality mode was selected to ensure image quality was on par with native rendering, though we’re of the firm belief that DLSS Super Resolution ‘Performance’ is a great option for 4K, offering even more performance than what’s seen below.
One of the surprise game updates this past week was the addition of in-race ray-tracing to the stunning Forza Horizon 5 with DLSS Super Resolution support. The above benchmark figures represent the latest Forza with Extreme quality settings and ray-tracing turned on. The visual bump in fidelity is awesome and the GeForce RTX 4080 handles it all without breaking a sweat (the fans mostly take care of that) pushing triple-digit performance in 1440p and 4K.
Some of the biggest performance increases over the previous Ampere generation were found in titles that don’t feature ray tracing, with Assassin’s Creed Valhalla and Shadow of the Tomb Raider presenting a 1.5X increase.
In terms of 1440p performance covering 21 games with and without ray-tracing or DLSS support, the difference between the GeForce RTX 4080 and the GeForce RTX 3090 is roughly a 25% or 1.2X increase in favour of the new RTX 4080. This is impressive for a card that uses less power and isn’t the high-end flagship model, but also a reminder that cards like the RTX 3090 and now the RTX 4080 push this resolution as far as it can go. Death Stranding has a frame-rate cap of 240 so getting to see performance on the RTX 4080 in that range is awesome.
And, DOOM Eternal with RT and DLSS pushing higher than 300 fps in 1440p, whilst testing in action-packed sections, is eye-opening. That said some of the biggest performance increases over the previous Ampere generation were found in titles that don’t feature ray tracing, with Assassin’s Creed Valhalla and Shadow of the Tomb Raider presenting a 1.5X increase.
Naturally, there is a drop off in performance when going from 1440p to 4K, and for the GeForce RTX 4080 that sits at around 30%. In 4K you’re looking at performance that regularly hits triple digits or above 60 fps. Compared to the GeForce RTX 3090 the increase jumps to 35% which is roughly what you would expect for a generational jump for a flagship model like the RTX 4090. But, the GeForce RTX 4090 we got exists, which has smashed pretty much all performance records out there. It’s here where we learn that the RTX 4080 is roughly 30% slower than the RTX 4090 when it comes to fully maxed-out 4K gaming. Which says a lot about the top-end, and costly, Ada offering.
It’s here where we learn that the RTX 4080 is roughly 30% slower than the RTX 4090 when it comes to fully maxed-out 4K gaming. Which says a lot about the top-end, and costly, Ada offering.
It’s a pretty sizable gap, more than what we saw in the Ampere generation. This means that the RTX 4090 sits in its own space, which is reflected by the pricing difference between the two. So yeah, 4K performance is a notable step down from the RTX 4090 but it’s also a notable step up from the RTX 3090. This is why that “somewhere there in the middle” line from the intro to this review was a vague but apt way to sum it up.
Still, Overwatch 2 using Epic quality settings pushes above 200 fps in 4K, and can still hit playable frame rates even in 8K. The GeForce RTX 4080 is a 4K beast.
To provide a clearer picture of the 3D Mark results, we’ve included both the GeForce RTX 4090 and the GeForce RTX 3090 results captured on the same system. The key to understanding 3D Mark results is to look at the various sets of tests targeting different resolutions and different APIs.
Fire Strike is a DirectX 11 test, with the three benchmarks covering 1080p, 1440p, and 4K resolutions. Comparing the RTX 4080 numbers to the RTX 3090 the increase across all tests sits at around 1.3X. Port Royal - which tests ray-tracing, sees a 1.4X increase over the RTX 3090. These results line up with the game performance seen above and are a testament to the power of the GeForce RTX 4080.
But there’s one more component to the performance story, DLSS 3.
Cyberpunk 2077 wasn’t included in the tests above, and that’s because NVIDIA provided reviewers with a special branch to be able to play and test the game using the new DLSS 3 rendering - amongst other titles already out and updated with DLSS 3 like F1 22, A Plague Tale: Requiem, and an impressive Unity tech demo called Enemies. DLSS 3 is another game changer in the making, with NIVIDA confirming that over 30 titles will see DLSS 3 integrated in the coming months.
One of the best things about it, apart from the fact that it works (and looks great), is how it goes beyond brute force to deliver performance. A 2X increase in performance is kind of the baseline, going all the way up to 3.5X and beyond.
As mentioned earlier, DLSS 3 isn’t simply the next version of the impressive AI-rendering tech we call DLSS - the 3 is about the combination of DLSS Super Resolution (which is what DLSS 2 is today), DLSS Frame Generation (the new magic), and NVIDIA Reflex. DLSS Frame Generation is limited to Ada-based GPUs like the RTX 4080 because it uses brand-new AI hardware to generate frames. Specifically the new Optical Flow Accelerator inside Ada architecture.
One of the best things about it, apart from the fact that it works (and looks great), is how it goes beyond brute force to deliver performance. A 2X increase in performance is kind of the baseline, going all the way up to 3.5X and beyond. As per the benchmarks above DLSS Super Resolution was set to Quality mode to maintain the highest visual fidelity, and to push the GeForce RTX 4080.
The 2.7X increase in performance over native 4K rendering in Cyberpunk 2077 is a sign that the days of brute-force rendering being the be-all-end-all are slowly coming to an end. It’s, yeah, amazing tech. The impressive adoption and implementation of DLSS Super Resolution into hundreds of games is a testament to NVIDIA’s investment in AI, and Frame Generation is the exciting next step.
It’s not perfect, and as a first version, there are issues. There have been a few in-depth visual breakdowns that have showcased some clearly messed up frames with objects looking very wrong or UI elements taking on artifacts. But here’s the thing, once you push 100 or 200 frames-per-second it’s virtually impossible to notice these issues. And for titles like Microsoft Flight Simulator and A Plague Tale: Requiem, where slow and measured camera moves are the norms - DLSS 3 is, well, the ideal way to play on an RTX 4080.
The impact on PC latency, compared to native rendering, is often reduced too, or you see a relatively small increase. Analysing frames, calculating motion, and then creating brand-new frames and putting those on-screen does come as a cost but thankfully NVIDIA Reflex manages to keep that well within playable levels. Now enabling NVIDIA Reflex without Frame Generation and the difference is a lot more pronounced, which does point to the technology not being suitable for competitive games. But then again, competitive titles are the ones optimised to run fast (just take a look at the Overwatch 2 benchmarks) so this isn’t a concern.
A 4K Powerhouse
There was no GeForce RTX 2090, on account of the RTX 2080 and x80 cards historically being the flagship for NVIDIA GeForce graphics. With the arrival of the GeForce RTX 3090, NVIDIA took the idea of a Titan-style over-the-top no expense spared graphics card and slotted it into the line-up. For many it became the flagship GPU for the GeForce RTX 30 series, pushing the architecture as far as it could go while still being something you could consider for day-to-day gaming.
Two years ago Ampere and the GeForce RTX 3080 launched and it was one of the most impressive hardware debuts to date. The fact that Ada and the RTX 4080 push things even further, in such a short timeframe, is impressive to see.
It also pushed the price of graphics cards as far as they could go. Comparing the RTX 3080 to the RTX 3090, in terms of in-game performance, and the difference between the two wasn’t all that massive. The RTX 3090 was a beast, but the gap was such that the RTX 3080 felt mostly the same. This isn’t the case for the GeForce RTX 4080 and the GeForce RTX 4090, where you’re looking at around a 30% bump in 4K performance when jumping up to the RTX 4090. But, when you look at the pricing these two new NVIDIA GPUs begin to make a lot more sense. Yes, the price point for the GeForce RTX 4080 is high but it’s priced according to the performance in relation to the GeForce RTX 4090.
Two years ago Ampere and the GeForce RTX 3080 launched and it was one of the most impressive hardware debuts to date. The fact that Ada and the RTX 4080 push things even further, in such a short timeframe, is impressive to see. This is high-end 4K gaming that feels a step or even two steps ahead of where gaming is right now. DLSS 3 and its use of Frame Generations is exciting, and we can’t wait to see the technology evolve and mature in the coming months and years.
Although only briefly touched on, the content creation chops for the GeForce RTX 4080 are next level too. Everything from streaming to video editing to 3D modeling, NVIDIA has delivered. The GeForce RTX 4080 might not hit the same heights as the RTX 4090, but that’s a GPU that feels out of this world. Which makes it something of a more down-to-earth graphics card that delivers uncompromised 4K gaming.