Haha.If you open the images in a new tab the first one is labelled 'native' and the second one which is sharper is FSR.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Fidelity Super Resolution in 2021
- Thread starter Grim5
- Start date
Maybe FSR is so good it's restoring lost detail like DLSSHaha.
.Which is FSR and which is native?
If we can't tell then FSR is doing it's job
I left the images labeled if you save them or open in a new tab. I deliberately left it for people to decide without any preconceived notions of what FSR "should" look like. Like I said earlier in the thread it is so easy to test out FSR by downloading the free RiftBreaker demo or even the complete DOTA 2 for comparison.
Now having tested FSR in both and I can see that the DOTA2 devs did a much better job than the RiftBreaker devs. So FSR, just like DLSS really has to be judged on a case by case basis.
Well your experiment definitely workedIf we can't tell then FSR is doing it's job
I left the images labeled if you save them or open in a new tab. I deliberately left it for people to decide without any preconceived notions of what FSR "should" look like.
.
The ground looks blurrier with FSR in Riftbreaker but for some reason in Dota 2 it's the opposite going by the screenshots unless the person who labelled them made a mistake
I must have made sure a dozen times because the FSR one looked subjectively better. More details on the textures but softer edges. But for a objective test you should download DOTA2 to check as it is free.
i just think dota 2 has less graphical elements, hence why fsr seems to be doing okayish job
it fails horribly in godfall and riftbreaker at 1080p. this can't be purely implementation difference. its gotta be that rifbreaker and godfall are much newer games and much more packed with detail, so FSR cannot do meaningful work on a more varied picture
it fails horribly in godfall and riftbreaker at 1080p. this can't be purely implementation difference. its gotta be that rifbreaker and godfall are much newer games and much more packed with detail, so FSR cannot do meaningful work on a more varied picture
Why do u think quake rtx upscales like it does as in dynamic scaling performance, its like dota and has less gfx details like polys and textures. If the taa adaptive upscaler that everyone praises works well in it do you think its cos its good or just cos it works ok on a ancient game with really low fidelity. I doubt ppl would praise its method if used in a modern title as bet it doesnt perform the same if it was say in a title that had assets like cyberpunk or the likes. Its only a guess mind that it wouldnt be praised as much as i think it wouldnt perform as good on a high asset game vs dead low one i.e quake.
Quake 2 RTX works very well.
Not bad for a ~25 year old FPS and yes, I'm a @Rroff fan -
From my test room - mirror on the left - if you look closely can see that the material lighting is correct for the mirror perspective as well which isn't the case in games like CP2077 and WDL, etc. that have used RT so far.
Game doesn't have bullet impacts as standard - I've not got around to adding bullet impacts that render correctly with RTX yet so it is a bit ugly but will suffice to demonstrate
One of the things that really stands out for me as well is the reflected scattered light
(Ignore the shell casings are floating and an old model with 8bit colour skins and not using RTX materials - I'm still very much prototyping)
The nice thing is you aren't limited to planes, etc. any object can have real time, accurate, reflections from all angles, etc. (though there is a limit currently on reflections of objects with refraction/reflections and/or multiple levels of transparency).
If you look at my test room screenshots (example below) it has large areas of glass on both sides, a large mirror and multiple glass, chrome and other reflective objects and is only 1.5% slower than the stock Quake 2 maps if that.
Typically the performance cost of rendering a scene in CP2077 with rasterisation maxed and RT maxed exceeds the impact it would have of rendering it via Quake 2's path tracer albeit the cost here comes at the constrained ray budget not being sufficient to remove noise from specular lighting to an ideal level.
EDIT: You can adjust the reflection/refraction bounces between 1 and 8, default 2, in Quake 2 RTX with only a moderate performance hit - obviously there is a point you don't get things rendered that would require more bounces to capture after a point hence the limited performance impact of increased scene complexity but there are few instances where that matters for a video game even though it would be nice to have higher.
It has its limits and there in some way to go yet to get ideal results with an ideal low level of visual noise from using denoising approaches to make up for limited ray budgets but the path tracer in Quake 2 RTX is much more than people give it credit for because they judge it from the stock Quake 2 maps.
Only Quake 2 RTX has a proper path tracing implementation of any sort which is fully used for all rendering (albeit even that isn't without some cheats*) - all the other games use a combination of some or all of using ray tracing only or partially in screen space, using simplified re-renders of the scene often missing a lot of details and often captured at a far lower resolution. In CP2077 for instance a small amount of ray tracing is used as a reference for scattered light and only from a limited number of sources (primarily the sun) and then the rest faked up using traditional techniques referenced against the smaller number of ray traced samples (I'm overly simplifying the system they use). A lot of the shadows are generic imposters projected into the scene rather than using proper ray tracing as well again with a limited amount of ray tracing used to guide the results. The sad thing is in many cases just getting a lite version of those features running still has a good slice of the performance overhead of a more complete solution for ray tracing the whole scene but is the minimum implementation required just to get some small level of the features in use.
Quake 2 RTX still has many compromises - especially the ray budget right now just doesn't allow for denoising results to an optimal level - but it is far far more capable than people who haven't spent time experimenting with it allow for and can significantly exceed the visual results of traditional techniques even though it can't hit the heights of offline ray tracers yet and is no way limited to the Quake 2 engine for results. We still need ideally 2-4x the ray budget to be able to satisfactorily remove noise from specular lighting, etc. despite the claims of some there is no reason why it can't be used to render a game like CP2077 with viable performance.
* For instance specular lighting is only fully path traced out to medium distance - the specular lighting on far objects uses a fast approximate cheat which is mostly indistinguishable to the real thing though not 100%. Caustics are only partially implemented (for time rather than performance as the stock maps don't really have features that would use it) and use fast approximate simulation which can look quite fake in certain scenarios (but again the developer never spent any time on it).
EDIT:
Turned off the sun light to show all the windows, etc. are doing reflections, etc.:
With lots of reflections and refraction, etc. as well as other features going on there isn't a huge performance difference to a standard map this holds true for even more complex scenes as well - for LOLs ran it at 4k on my 3070FE - 1440p is more where playable performance is at:
Old video from when I first played around with Quake 2 RTX but RT water can be really nice - this is only scratching the surface of it:
EDIT: Can do some decent water effects in general though it needs something better than Quake 2 to show it off really:
This is pretty cool (if you know what to look for):
Not only is the blaster effect reflected in the various surfaces but they are also reflecting the reflection of it in other nearby surfaces! (i.e. the chrome on the chairs is both reflecting it from the world but also the mirror and vice versa)
I believed the same until I tried FSR in DOTA 2 at 1080p and we have a developer who has experience implementing FSR saying it's not an inherent flaw with FSR. He specifically said that he had to change the TAA used to get better results and that it's down to how a developer implements it with their existing AA routines.
I have also tested 1080p in Terminator Resistance and the FSR looks subjectively better there as well. So it could well be lower resolution game textures are getting a bump in sharpness but it is still better than native.
So right now the evidence suggests it will be an implementation issue, but time will tell.
Last edited:
While I can't say for sure how the implementation in Quake 2, as quoted above, would work in a modern game I've pushed it quite far and it has stood up well with far more complex detail on screen than the stock maps/textures. It should be able to handle something like CP2077.
fsr looks good in terminator because it's a dark game
sorry to present a rebuttal every time, but this is what it is. fire up cyberpunk, use 1080p, use dlss performance mode, skip the time to night. you will practically see that dlss works wonderful, crazy, amazingly good. skip time to noon, and all of a sudden that magic quality breaks down in brighter pixels
its a matter of composition. this is why the medium developers got away with 648p/900p on series s/x. they could've done further optimizations to make the game run at higher resolutions, but having "dark" scenes and "dark" compositions allows even lower resolutions to work somewhat okay
Your problem is that you aren't presenting any rebuttals backed by evidence, only opinion. A quick Google search for images of Terminator Resistance would show your claim that "it's a dark game" to be innacurate. There are plenty of daylight areas within Terminator Resistance that allow proper FSR testing.
You seem like a reasonable person who at least tries to engage honestly and I don't mean this come across as confrontational. But people have been presenting actual evidence for pages now but each time you "rebut" that evidence, you tend do so without any objective or factual data. Your posts come across like you simply reject evidence in an almost flippant "I refuse to believe it" way. Maybe I am being harsh and you are just cautiously optimistic.
So I don't mind presenting evidence and I have repeatedly said I wish to see more proof that FSR can do much better at 1080p because at 4K the results are excellent. Though while we wait for that evidence, here is an FSR on vs off at 1080p in a daylight level from Terminator Resistance.
While Terminator is not pushing any graphical boundaries it is certainly decent enough looking and the FSR implementation is very good even at 1080p.
Last edited:
sorry but the FSR image looks very oversharpened
this effect was less pronounced on dota 2 so that got a pass from me. but this one looks overly done
i wish there were a way to tweak the sharpening level, a little bit less would be better? would look more natural like native image maybe
is this FSR ultra quality mode?
i have no bad intentions, i really want fsr and dlss to succeed at 1080p, but sadly i wouldn'T have enjoyed that oversharpened image, that's a personal choice so everyone's mileage may vary
this effect was less pronounced on dota 2 so that got a pass from me. but this one looks overly done
i wish there were a way to tweak the sharpening level, a little bit less would be better? would look more natural like native image maybe
is this FSR ultra quality mode?
i have no bad intentions, i really want fsr and dlss to succeed at 1080p, but sadly i wouldn'T have enjoyed that oversharpened image, that's a personal choice so everyone's mileage may vary
Last edited:
Them terminator shots in the day the gfx look bad texture wise, flat and generic in a lot of the scene like the grey cement and stairs areas look so dull. The ground is boring to look at. Lol i dont think anyone would mind a little fsr in that scene as nothing worth pixel hunting at.
Im presuming top image is fsr as its a bit softer and bottom is native as sharper and more detail?
Im presuming top image is fsr as its a bit softer and bottom is native as sharper and more detail?
no, lollll, fsr image is bottom, top is nativeThem terminator shots in the day the gfx look bad texture wise, flat and generic in a lot of the scene like the grey cement and stairs areas look so dull. The ground is boring to look at. Lol i dont think anyone would mind a little fsr in that scene as nothing worth pixel hunting at.
Im presuming top image is fsr as its a bit softer and bottom is native as sharper and more detail?
its gotta be, i never seen any "native" game looking oversharpened like that. FSR is known to employ sharpening so thats gotta be it
Yes the softer one is native and the sharpened one is FSR. I was very careful not to pick a favourite becase it is going to be subjective. Your reactions from both of you demonstrates my point that how FSR looks even at 1080p is going to be subjective and will depend on how the developer implements it and tweaks it.
I would say that given the evidence of the various implementations of FSR, we will see some better or worse examples depending on the game. So it won't be a case of FSR always sucks at 1080p, or FSR will always cause blur etc.
If I were to base my conclussions on nothing but how FSR looks in The RifBreaker at 1080p I would call it a failure. Though given further examples of TR and DOTA2 I can see that the results vary by game and developer.
I would say that given the evidence of the various implementations of FSR, we will see some better or worse examples depending on the game. So it won't be a case of FSR always sucks at 1080p, or FSR will always cause blur etc.
If I were to base my conclussions on nothing but how FSR looks in The RifBreaker at 1080p I would call it a failure. Though given further examples of TR and DOTA2 I can see that the results vary by game and developer.
Last edited:
Maybe in the future fsr can work in a lot of games by the way of using it in wrappers/injectors like dgvoodoo or the way gta5 got it modded in or something, may even work on older titles that might work.
About the sharpening with fsr, im guessing if ppl complain to devs they can adjust the cas sharpening of it later in a patch or open up a setting to let ppl decide themselves how much to sharpen in say fsr 1.1 code etc.
About the sharpening with fsr, im guessing if ppl complain to devs they can adjust the cas sharpening of it later in a patch or open up a setting to let ppl decide themselves how much to sharpen in say fsr 1.1 code etc.
Soldato
- Joined
- 25 Nov 2011
- Posts
- 20,672
- Location
- The KOP
@Grim5 He isn't wrong though. Just play the game if you noticed the effect of these tech like DLSS 1.0 just playing then its not doing a good enough job.
FSR works really well. I had a few of my friends over on the weekend and I did some blind test with them on Godfall not one of them could notice the difference.
FSR works really well. I had a few of my friends over on the weekend and I did some blind test with them on Godfall not one of them could notice the difference.