I don't think we have heard anything from the developers as far as implementation, but should be interesting. Are developers using UnReal or Unity going to focus and spend precious development time on upscaling techniques that may fall short of internal upscaling techniques provided by the engine; TSR will be available on PC's and consoles.
D
Deleted member 13524
Guest
Can FSR be used to boost quality beyond a monitor's resolution through downscaling?
For example, on a 1440p monitor to render at 4K FSR Quality but then downscale to 1440p using Virtual Super Resolution. There could be some use for it as a replacement to antialiasing perhaps?
One problem could be the smaller UI elements.
Agreed.
I don't think most gamers understand exactly how much of a waste of resources native 4K rendering actually is, even on very large 4K TVs. Unless they're getting up from the couch to look at the TV from a distance of 2 palms, or doing 5x zoom-ins from desktop monitors, very few people are going to effectively notice the difference between 4K FSR Ultra and native 4K. Especially in motion and not stopping at every other frame to analyse it.
Same thing for Series X and PS5, but we'd already know this from the scarce amount of new games on the new-gens being rendered at native 4K.
Probably not UE5 developers because those can just use TSR which is better than FSR (spatial + temporal). Unity and Frostbite are engines already appearing in the list of devs supporting FSR, so most probably yes.
Valve an Ubisoft are in that list as well, so I'd expect Source (Half Life), Snowdrop (Division, Avatar), Anvil (Assassin's Creed) and Disrupt (Watchdogs) to support it too. Dunia is already confirmed to use it, considering Far Cry 6.
For example, on a 1440p monitor to render at 4K FSR Quality but then downscale to 1440p using Virtual Super Resolution. There could be some use for it as a replacement to antialiasing perhaps?
One problem could be the smaller UI elements.
Agreed.
I don't think most gamers understand exactly how much of a waste of resources native 4K rendering actually is, even on very large 4K TVs. Unless they're getting up from the couch to look at the TV from a distance of 2 palms, or doing 5x zoom-ins from desktop monitors, very few people are going to effectively notice the difference between 4K FSR Ultra and native 4K. Especially in motion and not stopping at every other frame to analyse it.
Same thing for Series X and PS5, but we'd already know this from the scarce amount of new games on the new-gens being rendered at native 4K.
Probably not UE5 developers because those can just use TSR which is better than FSR (spatial + temporal). Unity and Frostbite are engines already appearing in the list of devs supporting FSR, so most probably yes.
Valve an Ubisoft are in that list as well, so I'd expect Source (Half Life), Snowdrop (Division, Avatar), Anvil (Assassin's Creed) and Disrupt (Watchdogs) to support it too. Dunia is already confirmed to use it, considering Far Cry 6.
ToTTenTranz said:
looking at Alex's more capable analysis that the rest of these amateurish websites there is a clear and significant downgrade in details. Its not little and hard to notice. Its a visibly softer image with compromised details in multiple areas. Alex also points how the aliasing of the base resolution remains in the upscaled image. So pixel crawl, shimmering and all that stuff remains. Image stability was one of the highlights of dlss in a game like Death Stranding for example
Good question. You can do this with a combination of DLSS and DSR today. On a 1080p monitor upscale to 4K using DLSS and downscale back to 1080p using DSR.
Including it in games as it is now or waiting till it has matured enough to include as an option? The big tell will be which studios over the next few weeks announce what games will be using FSR as an option.
FSR is not an AA method. So, no, it doesnt improve image quality over the native resolution.
Flappy Pannus
Veteran
Digital Foundry's video here:
Watching that first, then coming into this thread was somewhat surprising when seeing some of comments from people that were generally impressed. If anything I've been generally upbeat on this before release, I've argued that I don't think this needs to 'match' DLSS to be worthwhile at all as some have, simply by nature of it being open source and supported across a wide variety of hardware is a feature unto itself - I definitely don't want it to replace DLSS as the gold-standard, but having another 'decent' reconstruction tech that's easily integrated as another option for non-RTX cards is a good thing, especially as on the PC we see a far sparser use of checkerboarding/TAA upsampling across games.
However, the results as Alex showed are less than impressive, and bear in mind Alex doesn't even make a direct comparison to DLSS in the video (DLSS isn't even mentioned until the very end). Rather just comparing it against regular bilinear and TAA upscaling, it falls down hard against the latter imo.
UE4 TAA looks significantly superior here. I can't speak to how good it is compared to say, Insomniac's very impressive reconstruction tech or basic checkerboarding, but I don't expect it would far well against those either. So that this is at 'least' a worthwhile method for the new consoles to squeeze out extra performance is dubious when it likely can't compete with already established reconstruction methods they've been using for years.
So what is this for? Based on what I've seen so far, if I see this - and only this - in a game as the only reconstruction method option, that will basically be its purpose - 'better than nothing' - and perhaps compared to bilinear upscaling + sharpening, even that meagre improvement might be largely negated. Of course goes without saying if this can't compete with regular TAA upscaling it's not remotely in the same league as DLSS.
Maybe it will improve, but as others have postulated I don't see that happening until it becomes more than a spatial upscaling method. From what I've seen it's less FSR and more BBB (Barely Better Bilinear).
Flappy Pannus
Veteran
And has massive shimmering, destroying the image quality. It's not a replacement for any AA method, if anything it increases pixel shimmering over no AA at all.
Rootax
Veteran
I'm one of the guy that, in a lot of games, prefer no taa at all vs taa. I prefer shimmering and very crisp details, that no shimmering but bluriness. But as time pass, not a lot of game are allowing that without artefacts in some effects...
D
Deleted member 13524
Guest
Watching that first, then coming into this thread was somewhat surprising when seeing some of comments from people that were generally impressed. If anything I've been generally upbeat on this before release, I've argued that I don't think this needs to 'match' DLSS to be worthwhile at all as some have, simply by nature of it being open source and supported across a wide variety of hardware is a feature unto itself - I definitely don't want it to replace DLSS as the gold-standard, but having another 'decent' reconstruction tech that's easily integrated as another option for non-RTX cards is a good thing, especially as on the PC we see a far sparser use of checkerboarding/TAA upsampling across games.
However, the results as Alex showed are less than impressive, and bear in mind Alex doesn't even make a direct comparison to DLSS in the video (DLSS isn't even mentioned until the very end). Rather just comparing it against regular bilinear and TAA upscaling, it falls down hard against the latter imo.
UE4 TAA looks significantly superior here. I can't speak to how good it is compared to say, Insomniac's very impressive reconstruction tech or basic checkerboarding, but I don't expect it would far well against those either. So that this is at 'least' a worthwhile method for the new consoles to squeeze out extra performance is dubious when it likely can't compete with already established reconstruction methods they've been using for years.
So what is this for? Based on what I've seen so far, if I see this - and only this - in a game as the only reconstruction method option, that will basically be its purpose - 'better than nothing' - and perhaps compared to bilinear upscaling + sharpening, even that meagre improvement might be largely negated. Of course goes without saying if this can't compete with regular TAA upscaling it's not remotely in the same league as DLSS.
Maybe it will improve, but as others have postulated I don't see that happening until it becomes more than a spatial upscaling method. From what I've seen it's less FSR and more BBB (Barely Better Bilinear).
Why are you presenting a whole thesis based on the false assumption that only performance mode exists? All the positive comments in this thread refer to the Ultra mode.
The supported games on the short-medium term are already announced in the FSR page.
I think this answers @Dictator 's question in his video.
It's not my case, but some people do hate temporal artifacts.
I guess after this less than impressive showing the question is: was this already being researched or in the pipeline for AMD or was it a knee jerk reaction quickly put together just to have something, anything, to appease fans asking for a DLSS competitor? Sounds like a terrible waste of time..
FSR will have the same temporal artefacts - FSR will use TAA in a game that uses TAA. It should not be seen as an alternative to TAA U or any image reconstruction technique as I view it - rather as a superior way to upscale an image in the case where that is the only option. But ideally, that should never be the only option. No one should really just want a game to FSR - that would be a bad outcome for a game's potential image and performance quality on PC.
Subtlesnake
Regular
The patent was from 2019. Perhaps future versions of the technology will have a temporal component, or perhaps it can be integrated with TAAU.
Sounds like extra third of framerate to me. On any card one can reasonably dare to play current games. So the point is to diminish the DLSS advantage of Nvidia.
Flappy Pannus
Veteran
The singular comparison image is from the Performance mode, as the TAA upscaling from UE4 is from the same starting resolution - 1080p, with very similar performance uplifts. My thoughts on it are based from the entire video, most of which focused on the 'Ultra Quality' mode, where the image degradation was still very noticeable.
Yes, FSR Balanced/Ultra will produce better output than that image - but TAA upscaling that starts with a native 1440p (as most implementations do on this gen of consoles), or checkerboarding at 1920x2160 will also produce improvements over TAA upscaling from 1080p.
I can't seem how it's diminishing anything when it's inferior to TAA...
