Battlefield V DLSS Tested: Overpromised, Underdelivered

Nvidia'due south deep learning super sampling, or DLSS, is 1 of the highly predictable features nowadays on RTX graphics cards. We previewed it months ago, but until back up arrives on working titles, at that place was just and then much we could learn about information technology as well looking at demos and what Nvidia claims will be possible with it.

DLSS finally fabricated its way to both Battlefield V and Metro Exodus this month, and as is the usual case for us, we'll be going through a full visual and functioning breakdown in this article. For now we're sticking with Battlefield V in this investigation because according both Nvidia and developer 4A Games have said the DLSS implementation in Metro Exodus still needs some polish.

When we looked at DLSS back in September, there were only two demos Nvidia had for the launch of their RTX GPUs. The demos weren't particularly great every bit they were canned benchmarks, which we felt would give Nvidia's DLSS neural network an unrealistic advantage at optimizing the image quality, compared to a dynamic game surroundings. Notwithstanding, we still discovered back then that DLSS performed roughly the same as reducing the paradigm from 4K to 1800p, while providing roughly the same epitome quality as 1800p.

DLSS has as well been available in a real game for a little while, Terminal Fantasy Xv. But considering that game has a terrible anti-aliasing implementation, we decided it was not a good test bed and is not reflective of most other decent games. But both Battlefield V and Metro Exodus have good anti-aliasing, which provides an splendid comparison betwixt DLSS and a high quality native epitome.

In all DLSS games released so far, the feature is locked down, preventing you from merely enabling information technology with any combination of settings or resolutions. In the case of Battlefield V, you lot must have DXR reflections enabled to enable DLSS, so there is no option to use DLSS without ray tracing. That's already disappointing, considering nosotros feel well-nigh gamers should play with ray tracing switched off. So if y'all simply wanted to apply DLSS to boost functioning ray tracing bated, that is non possible.

But it's locked down further, on a GPU by GPU basis. If you're playing at 4K, all RTX cards can admission DLSS. Still if you're a 1440p gamer, the option is only available for the RTX 2080 and below. At 1080p, simply the RTX 2060 and 2070 tin use DLSS. And there are like limitations with Metro Exodus.

Co-ordinate to Nvidia, the reason for this restriction is that activating the neural network for DLSS takes a fixed corporeality of time for each frame. As your performance level increases, DLSS begins to occupy a proportionally higher percentage of the rendering time, up to a signal where for fast GPUs, it takes longer to process DLSS than information technology does for the native frame. So Nvidia has made the choice to block users from activating DLSS in situations where the performance uplift is negligible, or in some cases even worse than just native rendering.

The video version of this feature showcases scene differences in motility and greater detail in some instances:

This definitely prevents DLSS from being that one-click operation improving feature Nvidia advertised it as. Many pop configurations, peculiarly those that deliver high-framerate gameplay similar the RTX 2080 Ti at 1440p, can't benefit from DLSS.

Before nosotros bound into the prototype quality comparisons, we wanted to go over the performance results offset. Nosotros tested Battlefield V with a Cadre i9-9900K rig, and for this initial batch of 4K testing nosotros used a GeForce RTX 2080 Ti with all settings set to Ultra.

Testing at 4K with the RTX 2080 Ti

Above you have the performance at native 4K, so performance at 4K with DLSS enabled, and native 1440p. Crucially, we've also added in the functioning with the game set to 4K but with a 78% render calibration. As you can run across, the functioning of 4K at 78% resolution, which equates to a resolution of about 2995 10 1685, is roughly the same as 4K DLSS. This will come into play for the quality comparing. Then there's too functioning with ray tracing and DLSS disabled.

Compared to native 4K rendering, which with Ultra DXR reflections is only a 40 FPS experience, DLSS improves performance by 37 percent looking at boilerplate framerates. That'south definitely a sizable comeback, and again, it's the same as rendering at a 78% resolution scale. However it doesn't bring the game running at 4K with DXR back up to the operation level of the game running without Ultra ray tracing.

Switching off DXR led to an 88% performance improvement. This is Ultra DXR non Low, only our previous testing has shown the functioning gain going from Low to Off to withal be around fifty%.

Screenshots and image comparisons below are all 4K. You lot can click to see full-sized versions.

But the real kicker is looking at the visual quality comparisons. We'll commencement with native 4K versus 4K DLSS. Across all the scenes we tested, there is a astringent loss of detail when switching on DLSS. Merely look at the trees in this scene, it's exactly what y'all'd await from a 4K presentation – sharp, clean, high item on both the foliage and torso textures.

DLSS is similar a stiff blur filter has been applied. Texture detail is completely wiped out, in some cases information technology's like you lot've loaded a low texture manner, while some of the fine co-operative particular has been blurred away or even thickened in some cases.

Of course, this is to be expected. DLSS was never going to provide the same image quality as native 4K, while providing a 37% performance uplift. That would exist black magic. Merely the quality deviation comparing the two is almost laughable, in how far away DLSS is from the native presentation in these stressful areas.

It gets worse though, and we'll switch to a dissimilar scene for this ane. Here is a comparison betwixt DLSS and our 78 percent resolution calibration, roughly 1685p, which we found to perform exactly the same equally DLSS. It'south a complete non-contest. The 1685p epitome destroys the DLSS image in terms of sharpness, texture detail, clarity, basically everything.

But look at the quality difference betwixt these two areas, when zoomed in. The 78% scaled image preserves the fine detail on the rocks, the sign, the sandbags, the fabric, the gun, everywhere! With DLSS, everything is blurred to the point that this detail is lost. And nosotros're not talking near a state of affairs where DLSS is noticeably improve at anti-aliasing, the 1685p prototype is already using Battlefield's TAA implementation which is quite good.

There are some instances where DLSS is smoother, looking at extremely fine particular when zoomed reveals less aliasing in the thin tree branches, but this has come at the complete loss of particular in the leafage and everywhere else. And nosotros only spotted this when zoomed in. Looking at the full image on a regular 4K PC monitor, it'southward hard to tell the departure in aliasing because the pixel count is already very high, and the non-DLSS presentation with TAA removes most of the key artifacts yous'd spot without anti-aliasing. This leaves DLSS looking like Vaseline has been smeared on the brandish.

That final scene was especially bad for DLSS, but there's a common theme throughout the areas we tested. And we're not even talking near DLSS versus native 4K hither, nosotros're talking about DLSS versus a 1685p epitome upscaled to 4K, both of which evangelize the same performance. 1685p is a footling behind native 4K in terms of sharpness as you'd expect from upscaling, simply DLSS is miles behind either of them.

In some situations we wouldn't fifty-fifty say DLSS is superior to a 1440p image upscaled to 4K. 1440p sees a further loss of fine detail to the paradigm, and in some environments DLSS can restore that detail and smooth out any jagged artefacts. However the DLSS image is still very soft and blurred, often with lower texture detail. It depends on the environment and it'southward definitely a lot closer than some of our previous comparisons, but a lot of the aforementioned problems remain. It gets closer again if you downscale the 4K DLSS image to native 1440p and do the comparison at that resolution, rather than upscaling both to 4K, simply DLSS isn't clearly ameliorate.

Battlefield Five'southward regular TAA implementation is very good, too, and doesn't blur out the paradigm similar TAA tin do in other games. Concluding Fantasy Fifteen for example, is quite blurry with TAA enabled. Compare TAA to DLSS in that game, and it'due south like. Only Battlefield Five is sharper overall, and DLSS only tin't match upwards. In fact, the native and scaled images completely obliterate the blurry mess that is DLSS.

Testing with the RTX 2070 at 1440p

We ran some cursory tests at 1440p also, this time with an RTX 2070. Here DLSS performs more in line with an 84% resolution scale, and then roughly 2150 x 1210, delivering a modest 18% performance improvement over native 1440p.

Office of this is due to 1440p performing meliorate on the RTX 2070 than 4K does on the 2080 Ti with Ultra DXR. The other function is the RTX 2070's more limited RT and Tensor core resources.

Nosotros're seeing virtually of the same issues at 1440p equally we did at 4K. The DLSS image is softer and blurrier compared to upscaling a 1210p image to 1440p. In some areas there is less aliasing, simply that reward is non worth it when the overall image is so blurry. DLSS is an improvement in some areas over native 1080p, but information technology'south not a definitive victory.

There's no other mode to put it: DLSS sucks correct now.

From what we can see in Battlefield V, as well equally Metro Exodus which is nevertheless an early implementation, DLSS is complete garbage and a huge waste of time. A lot of people hated on ray tracing and the implementations we've seen so far, just to me, DLSS is by far the worst of the two key RTX features.

Correct now, DLSS provides much worse paradigm quality than upscaling from a resolution that provides equivalent performance (Nvidia has even acknowledged information technology's not adept correct now, though with less straightforward adjectives). For 4K gamers with a 2080 Ti, our upscaled 1685p image looked far improve than DLSS, while providing an equivalent performance uplift over native 4K. At 1440p, upscaling from 1210p also looked better than DLSS.

This is a significantly worse outcome than our initial DLSS investigation. Previously, in an optimal canned criterion, DLSS looked and performed similar an 1800p image. And so information technology provided no benefit. Here, DLSS looks worse than a 1685p image, at the same performance. And then enabling DLSS in a real-earth game like Battleground 5 is actually worse than using simple resolution upscaling that'southward been available in games for decades.

And that'southward why DLSS is complete rubbish. At least with ray tracing we're getting a superior image than what we had before, at an admittedly high performance cost. With DLSS, nosotros're not gaining annihilation in either the performance or visuals section compared to traditional techniques.

This is what we feared when DLSS was commencement demonstrated. While not all that impressive in canned demos either, we assumed having a repeatable benchmark run with the exact aforementioned frames in each run would nowadays the absolute best case scenario for the AI-based reconstruction DLSS uses. The AI could just railroad train itself on loftier resolution samples of the frames it will then reconstruct afterward.

But games are far more than dynamic than this. It's incommunicable at this stage to train a neural network for every single frame a game could output. And then when DLSS is paired with a existent world game, not just a demo, it's completely falling apart trying to reconstruct frames it's never seen before. Considering this is a neural network that is constantly learning, yes, DLSS could improve over time. But the gap between DLSS and our equivalent upscaled images is and then large, that who knows how long it will take for DLSS to catch upward, if it'due south even possible.

There are times where the comparison betwixt DLSS on and off volition be more favorable, even in dynamic games. Titles that have bad anti-aliasing implementations, such as Last Fantasy XV or more recent games like Resident Evil 2 with TAA, are already quite blurry. But many mod titles accept been optimizing TAA to expect meliorate than ever, preserving more particular and there DLSS can't go on up. Battlefield V being a prime example.

And this is without mentioning all the limitations that come with DLSS, including limited resolution support depending on the GPU you have, and for Battlefield Five, the inability to apply DLSS without ray tracing. On top of that, Nvidia admits it'due south useless for loftier frame rate gaming, different resolution upscaling which works beyond everything.

Every bit it stands today, DLSS in Battlefield V is so bad we think information technology should be removed from the game. Having the setting there might tempt gamers into using it when at best information technology provides no benefit, and at worst information technology's degrading the experience. Only when DLSS is at least equivalent to resolution scaling should the feature be reintroduced into the game.

But let'southward be clear, gamers and RTX owners were promised something entirely dissimilar: ray tracing with near no performance hit cheers to DLSS, which could even improve graphics quality. That is not what we accept received in the few titles that support either feature.

Unfortunately, DLSS was a feature that people were thinking would be the best of the RTX features, judging by the way Nvidia was selling it. The claims amounted to a costless performance uplift thanks to the ability of AI. But the reality is so far from this, information technology'south laughable.