In motion when playing the game it is difficult to tell the difference with Ultra Quality, as confirmed by multiple reviewers and users alike. There is a slight drop in quality which no one disputes, but it is minor enough that it is totally worth the 40-50% performance improvemen - if you need the extra performance.
-
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
Even with UQ I can see the loss of detail on the grass, certain rock surfaces, in some games fluids have a significant quality loss, to be fair many areas of the image are practically if not indistinguishable unless you zoom in 400% which is very good for spatial based techniques - however given the performance/quality delta vs temporal techniques I find the lauding of it rather hard to take seriously especially given the comments from some posters about DLSS which is inconsistent with their stance on FSR.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 31,179
I think we can all agree that both implementations aren't perfect, but both are at that 'good enough' stage to be usable by the majority and enjoyed.
Its such a weird argument. FSR doesn't make your game look worse, it gives you the option to make your game look a tiny bit worse in exchange for a big jump in performance.
I could understand not liking it if it forced you to use it, but its just giving people more ways of tweaking their game settings, y'know that thing that has been used as a major argument about why PC gaming is better since the early days of gaming
I could understand not liking it if it forced you to use it, but its just giving people more ways of tweaking their game settings, y'know that thing that has been used as a major argument about why PC gaming is better since the early days of gaming
Do you think the people using FSR will care about minor image quality loss that is difficult to spot while playing? Nope they won’t care at all which explains the glowing press and user feedback, except from one outlet who produce sponsored content for a competitor.
I’m not sure why you can’t understand the lauding of it from press and gamers using FSR (from both vendors - check the AMD and Nvidia subreddits). The reasons why have been explained to the naysayers multiple times already just look back through the last dozen pages.
FSR may not be for you Rroff, or DP, but neither of you would ever use an AMD GPU anyway so FSR is not aimed at either of you for that and also for people that cannot accept any loss in image quality, or people that will comb through images, zoomed in at 400%, to notice a distant tree in the background has some missing detail on one of the outer branches. Exaggerating a little, but you get the jist.
There are two factors for me here:
I just find it a bit odd to see it lauded so much when despite being quite good for spatial reconstruction it falls short of what AMD could have pulled off with a bit more effort and incorporating other techniques including temporal sampling even without a deep learning model. Having spent a lot of time playing with Quake 2 RTX's upscaling it is very noticeable to me.
Then there are those people whose position is inconsistent between when nVidia does it and when AMD does it.
This is pure insanity it is well beyond fanboyism. So AMD deliberately stops anyone that bought a 6000 card or wants to buy a 6000 card, to get double FPS in games like Metro or Cyberpunk. Because they are afraid the games won't look as good as it does with DLSS.
Keep your clients with low FPS and they will be happy. But wait, their games already look worse because DLSS is better than native.
Another ...
Any engine developer will want as many features as he can get in his engine. No matter if he thinks some are better than the others. It's like saying they will force you to use Nanite or Lumen because they spent a lot of resources on these things. And about an inferior tech ... they already added DLSS.
Well this is the FSR 1.0, things have to start somewhere. Who knows what will happen in the future.
A bit more effort? You are downplaying the effort that was put into FSR to launch with support in multiple games, with support for another 12 coming shortly, and 45 developers on board already. Is that not enough effort? It’s better than launching with two games and then nothing for a while.
One thing we can all agree on, it’s a very good start with no major negatives only positives. Maybe that’s me being optimistic.
What you will find though is that the negatives will be largely coming from people that have no intention of using AMD GPUs or FSR.
TBH Quake 2 does not have the most complex geometry out there. Even if the temporal upscaler looks ok'ish in that game, it may not look the same in other games. Plus the processing time is also a factor, any temporal solution will take more time to process. And a lot of time to be optimized and fine tuned for each game. Next thing, FSR will become open source, someone will make a bad temporal implementation and everybody will think it is AMD fault because their temporal upscaler sucks since it has no AI like DLSS has. :
One I've posted before and while you can tell the source is low res this quite impressed me as the input resolution is barely 360p - lower than many people first played these kind of games at back in the 90s. Which is why I find it hard to accept the praise FSR is getting as it would have its work cut out to match this.
Like with the path tracing it doesn't really benefit from the relatively low complexity of the game - I've posted some screenshots from custom maps, etc. with much higher detail than the stock maps where it still holds up even if you stick like 4 million polys worth of animated models in the scene, etc.
TBH Quake 2 does not have the most complex geometry out there. Even if the temporal upscaler looks ok'ish in that game, it may not look the same in other games. Plus the processing time is also a factor, any temporal solution will take more time to process. And a lot of time to be optimized and fine tuned for each game. Next thing, FSR will become open source, someone will make a bad temporal implementation and everybody will think it is AMD fault because their temporal upscaler sucks since it has no AI like DLSS has. :
Like with the path tracing it doesn't really benefit from the relatively low complexity of the game - I've posted some screenshots from custom maps, etc. with much higher detail than the stock maps where it still holds up even if you stick like 4 million polys worth of animated models in the scene, etc.
What games? I thought they were unplayable demos, the games moved on rails and DLSS still looked like crap.
But this is how the game looked back then when playing in software mode.One I've posted before and while you can tell the source is low res this quite impressed me as the input resolution is barely 360p - lower than many people first played these kind of games at back in the 90s. Which is why I find it hard to accept the praise FSR is getting as it would have its work cut out to match this.
But this is how the game looked back then when playing in software mode.
I probably should have uploaded the low res source image as well - it is quite impressive comparing them.
Well this is the FSR 1.0, things have to start somewhere. Who knows what will happen in the future.
A bit more effort? You are downplaying the effort that was put into FSR to launch with support in multiple games, with support for another 12 coming shortly, and 45 developers on board already. Is that not enough effort? It’s better than launching with two games and then nothing for a while.
One thing we can all agree on, it’s a very good start with no major negatives only positives. Maybe that’s me being optimistic.
What you will find though is that the negatives will be largely coming from people that have no intention of using AMD GPUs or FSR.
It is a lot easier to support a lot of titles when fundamentally "all" you are doing is running a glorified post processing filter on a buffer.
I was joking but again there are some drawbacks in any temporal upscaler and it is no guarantee that it will do the same job every time. If they were hiding FSR behind Big Navi paywall, maybe you could have asked them for a better upscaler but for an open source one, they picked the easiest/simplest/ route. And it does an ok job.
You are still massively downplaying the effort required, but I can see I am not going to change your mind so I will leave it there.
Permabanned
- Joined
- 12 Sep 2013
- Posts
- 9,221
- Location
- Knowhere
Soldato
- Joined
- 25 Nov 2011
- Posts
- 20,675
- Location
- The KOP
200%??
Did you watch any of the release video for DLSS 1.0 on battlefield and Metro? it was easy to see just looking at video of the gameplay no need for screenshots or zoomed in.
Soldato
- Joined
- 25 Nov 2011
- Posts
- 20,675
- Location
- The KOP
But comparisons show from reviews that ultra quality very little difference between native? So what the game is originally showing at native is kept intacked so by your logic the original native game lack good texture quality.
Simmering pixels?
Mind showing some examples?