• 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.

Poll: How do you game? Upscaling or native? (updated poll choices - 24/12, revote!)

How do you game?


  • Total voters
    237
Associate
Joined
1 Oct 2020
Posts
1,186
Again, explain to me, how the poll has been changed to increase dlss vote
He didn't say that. He said that the poll options were reassigned and still show the same thing.

My point is that if there is no change in the options, why do it? More granular data? Not really, all that introduces is unnecessary bias to the upscaling/native question. Adding further options which say the same thing in relation to the question dilutes the result. 1 native option vs 3 up scalers gave a set of results which showed the main upscaler used was DLSS. Then changing the options to 3 native options against the same up scalers (3 of which we know are unlikely to be selected) has little effect to the upscaler vote, but splits the native result into three, giving the impression that the dlss vote is, comparatively, holding its own. This is misleading when presented simply in bars, which is why in relation to this question gpu ownership is irrelevant.

What is the point of knowing someone's gpu for this question if not to start going on about nvidia superiority again? I agree, it is better, but not to the point where I think fsr (or xess, not tried it) should be discarded and derided.

I also think this thread is done now tbh. Maybe leave the poll up and lock replies, if that's possible? If it's purely info gathering, all the discussion is not required and can be left in its own thread.
 
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
He didn't say that. He said that the poll options were reassigned and still show the same thing.

My point is that if there is no change in the options, why do it? More granular data? Not really, all that introduces is unnecessary bias to the upscaling/native question. Adding further options which say the same thing in relation to the question dilutes the result. 1 native option vs 3 up scalers gave a set of results which showed the main upscaler used was DLSS. Then changing the options to 3 native options against the same up scalers (3 of which we know are unlikely to be selected) has little effect to the upscaler vote, but splits the native result into three, giving the impression that the dlss vote is, comparatively, holding its own. This is misleading when presented simply in bars, which is why in relation to this question gpu ownership is irrelevant.

What is the point of knowing someone's gpu for this question if not to start going on about nvidia superiority again? I agree, it is better, but not to the point where I think fsr (or xess, not tried it) should be discarded and derided.

I also think this thread is done now tbh. Maybe leave the poll up and lock replies, if that's possible? If it's purely info gathering, all the discussion is not required and can be left in its own thread.

Read the full post or you're missing the "context":

didn't like it and reassigned the poll and still got the same results as before

That's insinuating that I didn't like the results i.e. native coming out top thus changed the poll to try and change the outcome...... except the change to poll was never going to influence the vote since the change was not intended to change the outcome except provide a better insight where the votes for native where coming from. Like I said, I want to add these options to get more insight but alas the usual suspects will scream "don't like the results so changing the poll to see what I want to see" even though again, it will not change the overall outcome except providing more granual data i.e.

- DLDSR + DLSS
- Native Nvidia non rtx owners
- Native Nvidia rtx owners

Actually, some may prefer that as it may even result in less votes for "DLSS" :cry:

As for your points:

More granular data? Not really, all that introduces is unnecessary bias to the upscaling/native question

How? People who want to vote native still can, all they are doing is identifying themselves as owning nvidia or amd.

It doesn't dilute the results as the results are the "same", it only appears "diluted" because people don't like seeing dlss with a higher bar.

If people can't do simple maths, not really the problem of the thread/poll or a problem on my end tbh but when you have people who take MLID as gospel, I can't say I'm surprised.





TLDR: Thread/poll is good, just people need to keep to the discussion/topic and stop crying that "the poll was changed because didn't like the result"

EDIT:

Also a large part for splitting the poll into more options is because people didn't read the thread OP where they were to provide these answers along with their vote:

gpu - 3080
res - 3440x1440 and 4k
reason you use your chosen upscaling tech or native - dlss provides better IQ and temporal stability than native in most cases whilst providing a substantial performance boost, it also exhibits less ghosting than native TAA implementations. Also, in combination with DLDSR, it can provide much better IQ than even my 4k display for clarity yet still perform similarly or better than just using dlss or native

Before adding more granular options, I and others noticed it was mostly people posting with amd gpus than nvidia (and the poll was hidden too so no way to know who/what camp the votes were coming from except for it being mostly amd users based on the thread posts. I suspect you and perhaps purgatory haven't read the full thread to see these discussion points?
 
Last edited:
Associate
Joined
1 Oct 2020
Posts
1,186
No, I suspect we've read it and can see the issues. Gonna leave this thread now - i normally like your insights into a lot of this forum as even if I don't always agree, I get the point of view you're putting across. There is just too much scrabbling for re-interpretation here.
 
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
No, I suspect we've read it and can see the issues. Gonna leave this thread now - i normally like your insights into a lot of this forum as even if I don't always agree, I get the point of view you're putting across. There is just too much scrabbling for re-interpretation here.

Well wouldn't be the first time you and him have stated yous didn't read the thread :p

I get where you and others are coming from and you would be correct if I had for example removed native entirely but as explained many times, that isn't the case.

Btw I'm curious what your thoughts are on the CP 2077 comparison I posted. Would you still say that dlss would not provide a better than native experience here? If so, why not?
 

mrk

mrk

Man of Honour
Joined
18 Oct 2002
Posts
101,063
Location
South Coast
All I can say is in my subjective (but fairly extensive) experience is that when a game doesn't use DLSS properly (half baked implementation), then DLAA can give better AA, and only AA since DLAA is purely just a modern more efficient AA method than SMAA/FXAA/TAA etc. At all other times DLSS (1440p) or DLDSR+DLSS Performance offers the better image quality (AI image reconstruction), and with that comes free AA due to the nature of upscaling.

I've previously demonstrated that there's no major difference in games like Cyberpunk between DLSS Quality @ 3440x1440 output, DLDSR @ 5160x2160 output with DLSS Performance, and native 3440x1440 DLAA - There is however rather a big difference in framerate - Which is where DLSS Quality 3440x1440 wins every time. Cyberpunk is a good example as it implements all these tech really well and the engine is optimised, so is a good baseline to compare other engines against.
 
Last edited:
Soldato
Joined
28 May 2007
Posts
10,102
Oooph. Where to begin with this post..

Best go make a cup of tea for this one. For others maybe grab some popcorn because I over did eating last night and woke up and can't get back to sleep :cry:




So no input lag. Now that we got that out of the way.

Let's go back to what you said before. You said "ain't about how nice the game looks it's more about high fps and input lag".

So therefore in your own words it is not about native. What does native even mean to you otherwise?

As you said, other competitive gamers would not even be on 1440p. As I understand these guys use 1080p and even use lowest graphics settings.

So clearly you do care about how nice a game looks :)




I see the problem. You are conflating things here and letting emotions cloud your judgement.

If you separate things, you will find people loving DLSS like me doing so for the opposite reasons.

How does rejecting using upscaling aid anything? The situation is what it is. Upscaling is here to stay. Not using it is simply cutting ones nose to spite one's face if that is the primary reason for not using it.

Even if upscaling was not here, these companies would be finding a way to charge you ever increasing amounts. That's just what they do. That is why I have no loyalty to neither of them.

So yeah, not using tools available on this basis just makes no sense imo. It changes nothing.




I understand what you are saying. But yet again being objective, inflation absolutely makes a difference. But it does not end there. Those jumps used to happen because back then both architectural and going to smaller and smaller manufacturing processes was easier and cheaper.

Living in those days in one's mind and having those expectations will do you no good. Those days are gone. It is now much more expensive jumping on to smaller manufacturers processes. It also takes a lot more r&d to develop better architectures.

I accept that we won't be getting 70-100% jumps each year for the same money. That would not be sustainable even if these companies aimed to break even on gaming cards. If you think otherwise I am afraid it is you that is living on the moon.

I know what your thinking now. This guy is defending them and their huge profits. No mate. I know full well they could be charging less and still making a healthy profit.

Do you see me with a 4090 in my sig? Did I pay a huge premium on a 3090? Nope. So no, I don't think the current situation is good.

But I also live in the real world. It is what it is. Ignoring upscaling because of the reasons you have given is only shooting yourself in the leg.

If your expectations are 70-100% improvement per year at the same price and think that was sustainable forever then you are not even just living on another planet, solar system or galaxy. Nope, you sir are living in another universe... :cry:




Ah so you do have an inkling that it is not as easy no more. We are getting shafted because these companies keep looking for ways to keep increasing profits as much as they can.

Why not be objective though. Why worry about what the noobs or others do or don't? The tech is here. When I use DLDSR and combine it with DLSS, there is no ifs or buts about it. I see a big difference in image quality. That is me being objective to what I see. I am not going to cut of my nose to spite my face by not using it.


TLDR? No! Go get some popcorn and enjoy the damned post. My name is not Purgatory, I don't do walls of text at every given opportunity! :p

No gonna lie man i am to hungover to make to much sense of your post. Was also probably to drunk when i made mine. In short if i am getting close to my max refresh of 240hz i am already where i need to be input lag wise. I am 45 so my relexes ain't gonna need any more fps and the game is fluid enough. With fsr and dlss i might get more frames but most likely adding in ghosting and a softer image. There is just no need for me to use the tech. I don't do what the pro's do and go for low image quality to get even more fps as i can get 170 to 240fps depending on the map at max with all blur effects off. If i favoured high visual games like yourself i would maybe take these techs more serious but no need as the speed i run about maps leaves no time to enjoy the visuals. A nice clean image and high fps is all i need to shoot people in the head. Cod Mw3 looks pretty clean in native although i am not looking for issues that may be there with taa.

Also at 1440p i here fsr is not really what you want to use and is mainly good for 4k. When my head clears i will turn it on and see if it makes a difference but probs after new year.

P.s i really need to stop posting 10 beers in haha
 
Last edited:

TNA

TNA

Caporegime
Joined
13 Mar 2008
Posts
28,343
Location
Greater London
No gonna lie man i am to hungover to make to much sense of your post. Was also probably to drunk when i made mine. In short if i am getting close to my max refresh of 240hz i am already where i need to be input lag wise. I am 45 so my relexes ain't gonna need any more fps and the game is fluid enough. With fsr and dlss i might get more frames but most likely adding in ghosting and a softer image. There is just no need for me to use the tech. I don't do what the pro's do and go for low image quality to get even more fps as i can get 170 to 240fps depending on the map at max with all blur effects off. If i favoured high visual games like yourself i would maybe take these techs more serious but no need as the speed i run about maps leaves no time to enjoy the visuals. A nice clean image and high fps is all i need to shoot people in the head. Cod Mw3 looks pretty clean in native although i am not looking for issues that may be there with taa.

Also at 1440p i here fsr is not really what you want to use and is mainly good for 4k. When my head clears i will turn it on and see if it makes a difference but probs after new year.

P.s i really need to stop posting 10 beers in haha

It's cool man. Have a read of it when your head clears. But no biggie. I had fun typing it :cry:

Merry Christmas and a Happy New Year by the way :D
 
Associate
Joined
15 Oct 2018
Posts
1,294
I also wonder about the point of this poll. I think we all accept DLSS is vastly superior to FSR.

The real question to me is what are you happier with - native Cyberpunk at ultra-non RT settings at 80-100+ fps >1440p on a 7900 XT - or RT (?quality) plus DLSS to make it viable on an RTX 4070 Ti (whether that's 30 fps or 70 fps only nVidia owners who have opted for this combo of features can tell)

nVidia cards have the RT advantage + DLSS, but lack the sheer raster horsepower of the AMD cards in the same price brackets (and now VRAM).

Apples to oranges analogy comes to mind.
 
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
I also wonder about the point of this poll. I think we all accept DLSS is vastly superior to FSR.

The real question to me is what are you happier with - native Cyberpunk at ultra-non RT settings at 80-100+ fps >1440p on a 7900 XT - or RT (?quality) plus DLSS to make it viable on an RTX 4070 Ti (whether that's 30 fps or 70 fps only nVidia owners who have opted for this combo of features can tell)

nVidia cards have the RT advantage + DLSS, but lack the sheer raster horsepower of the AMD cards in the same price brackets (and now VRAM).

Apples to oranges analogy comes to mind.

As explained many times throughout the thread, point was to see if people do actually "use" upscaling tech or native for their average day to day gaming. It was never about preference as that is a pointless poll as everyone would prefer higher resolution, higher fps, higher bit depth, path tracing and so on.

Personally one of the main points I take away from the poll is that whilst FSR is free/open and for all, this poll somewhat shows that even though it has all that going for it, it's a bit pointless if there are very few people actually using it.

And yes I agree, that's why I stated in my post/vote, upscaling allows the impossible to happen i.e. being able to use higher graphical settings which impact the IQ more than running lesser settings at a native res. Whilst this has frame gen on too, it showcases the mindset I have from a upscaling pov (and FG):

 

Klo

Klo

Soldato
Joined
20 Nov 2005
Posts
4,111
Location
South East
As explained many times throughout the thread, point was to see if people do actually "use" upscaling tech or native for their average day to day gaming. It was never about preference as that is a pointless poll as everyone would prefer higher resolution, higher fps, higher bit depth, path tracing and so on.

Personally one of the main points I take away from the poll is that whilst FSR is free/open and for all, this poll somewhat shows that even though it has all that going for it, it's a bit pointless if there are very few people actually using it.

And yes I agree, that's why I stated in my post/vote, upscaling allows the impossible to happen i.e. being able to use higher graphical settings which impact the IQ more than running lesser settings at a native res. Whilst this has frame gen on too, it showcases the mindset I have from a upscaling pov (and FG):

For all you know, it could just mean that Nvidia users desperately need every little bit of help with performance they can get with all their 'AI' crutches, while AMD cards are just better and don't.

This entire thread is just an Nvidia fanboy circular self pleasuring.
 
Last edited:
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
For all you know, it could just mean that Nvidia users desperately need every little bit of help with performance they can get with all their 'AI' crutches, while AMD cards are just better and don't.

This entire thread is just an Nvidia fanboy circular self pleasuring.

No offence but it's comments like this which are adding nothing to the thread and dragging it into the fanboy crap territory and also just comes accross as defensive over amds position with silly comments like this:

every little bit of help with performance they can get with all their 'AI' crutches, while AMD cards are just better and don't.

- we know equilvalent nvidia and amd cards compete neck in neck on raster, benchmarks show and prove this
- we know nvidia are way ahead in RT, benchmarks show and prove this
- we know nvidia are ahead in their software features as evidenced many times

The only way your post would be factual is if the amd voters/owners are using 7900xt and above gpu compared to dlss nvidia users using older/weaker gpus than a 7900xt/x. Given that there are at least 3 4090 users voting dlss, they either prefer the dlss IQ/benefits or/and the games they are playing on a daily basis are demanding e.g. path tracing at 4k.

Again, this is why there is this guideline in the OP:

gpu - 3080
res - 3440x1440 and 4k
reason you use your chosen upscaling tech or native - dlss provides better IQ and temporal stability than native in most cases whilst providing a substantial performance boost, it also exhibits less ghosting than native TAA implementations. Also, in combination with DLDSR, it can provide much better IQ than even my 4k display for clarity yet still perform similarly or better than just using dlss or native
 

Klo

Klo

Soldato
Joined
20 Nov 2005
Posts
4,111
Location
South East
No offence but it's comments like this which are adding nothing to the thread and dragging it into the fanboy crap territory and also just comes accross as defensive over amds position with silly comments like this:
This entire thread comes across as 'fanboy crap', it serves no purpose and doesn't contribute anything to the forum. Everyone knows Nvidia's DLSS is better than FSR, you and others never stop mentioning it. You didn't get the results you wanted the first time round, so you edited the poll just so you can claim "look, even AMD card owners don't use FSR" for some reason. Even if its true, so what? You own an Nvidia card, it doesn't really affect you. The only reason I can think is so that you can 'win' internet arguments defending poor little Nvidia's honour.
 
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
This entire thread comes across as 'fanboy crap', it serves no purpose and doesn't contribute anything to the forum. Everyone knows Nvidia's DLSS is better than FSR, you and others never stop mentioning it. You didn't get the results you wanted the first time round, so you edited the poll just so you can claim "look, even AMD card owners don't use FSR" for some reason. Even if its true, so what? You own an Nvidia card, it doesn't really affect you. The only reason I can think is so that you can 'win' internet arguments defending poor little Nvidia's honour.

If that is how you read the thread then again, this is a problem with how you are viewing the thread with your mindset. It's a gpu sub forum where we have threads on all kinds of discussion points, why is this thread (created for the purpose to see how gamers actually game) any different? Is there something with the poll results that you don't like perhaps?

Again, this thread is nothing to do with dlss being better than fsr etc. You can take from the poll that not many people use/value FSR but that's it and the reason I find that insightful is because some people bang on about how amd create things for "everyone" to use yet, as shown, very few are actually using it therefore is this really such a big win for amd/fsr? Same way you can take from the poll that despite nvidia owners having dlss (and arguably "paying" for it, they don't use dlss), why are they then paying more for such features? I would love to include the rtx and non rtx options but alas, people will have a fit because "zOMG changing the poll cause didn't like the results!!!!" even though it would be a good insight to see if the reason is because they perhaps don't have access to a feature that can only be used on RTX gpus.

Also, it does affect me as I've explained many times, I'm locked to nvidia because amd can't provide an upscaler as good as dlss (which I use 99% of the time and would do regardless of the gpu power), which means I'm limited in my choice as to what I can buy and then for the few games where DLSS is not included, I have to sacrifice graphical settings in order to meet my fps requirement because FSR is not at a standard I consider to be usable.

And as explained again for the hundreth time, please do explain to me how the change to the poll, which only makes it more granular is effecting the results this second time round? So far, I've asked this a few times and no one has been able to explain this..... I wonder why.

If you can't add to the discussion without making it out to be a nvidia vs amd thing then don't post, as simple as that.








EDIT:



BTW, as a heads up, any more posts which can't keep to the relevant discussion or/and thread topic (since these posts in the last 2-3 pages are now just going around in circles) will be RTM to be deleted.
 
Last edited:
Soldato
Joined
7 Dec 2010
Posts
8,299
Location
Leeds
28/12/23 Combined the results for anyone interested.

70.3% of OcUk users participating in the poll are running Native.

29.7% of OcUk users running Upscaling.

Also to add :-


z38tJ5u.jpg
 
Last edited:
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
FSR3 Frame Gen is getting used by everyone inc yourself?

Why is it an issue for you that AMD open source their tech?

Bold statement, should we create a poll to see how many are actually using frame gen? ;) So far, I don't think many people are actually using FSR 3/FG due to the mentioned issues surrounding, "noticeable" artifacts, broken frame pacing with VRR, HDR issues but alas it's seems to be hit and miss depending on the end users setup as evidenced in the fsr 3 thread and in depth analytical videos by the likes of DF/Alex.

As of now, I'm using it in one game I am playing right now, cp 2077 with PT (which we all know, needs US and FG).

Where I have said amd making open source stuff is an issue for me?

However, if amd were to take ownership of their solutions where they are solely responsible for the improvements and uptake, they could possibly be in a better situation and their solutions could be on par with nvidias solutions especially if they were to tailor it in such a way which would utilise RDNAs strengths but of course, this would mean them having to dedicate engineering time to such solutions and working with developers more to have better uptake. Working in the development industry and having used both open and closed source solutions, each has their strengths and weaknesses but as customers wanting a solution which "just works" and doesn't require huge time from our engineering side to "figure out", closed source generally wins, you aren't just paying for a solution which has been QC thoroughly but also paying for the customer service.

Intel arguably have the best approach right now with xess, it's open source and out there for people to tinker but as shown, they are very much taking ownership of making improvements to it (where as amd are throwing it to the community/developers to run free with it, which is exactly what amd have said they like about open source approach themselves in interviews with Alex and way back in the day with Roy) but alas, their partnership/outreach capabilities is obviously poor given the lack of uptake for xess.

Do you not want to see FSR improve to match DLSS? How do you propose that is done? So that the likes of myself who favour upscaling can have an option to move away from nvidia? (saying just buy gpu with the grunt/vram is not a valid option, as shown, 4090/7900xtx **** the bed for high res and high refresh rate gaming especially RT/PT)
 
Caporegime
OP
Joined
4 Jun 2009
Posts
31,348
Also to add :-


z38tJ5u.jpg


And that's why I created this thread as explained, this threads poll is about what you use, not what you prefer:

As per poll:

- native
- fsr
- dlss
- xess

This is only for upscaling, not frame gen.

And in your post, post your config so for me:

gpu - 3080
res - 3440x1440 and 4k
reason you use your chosen upscaling tech or native - dlss provides better IQ and temporal stability than native in most cases whilst providing a substantial performance boost, it also exhibits less ghosting than native TAA implementations. Also, in combination with DLDSR, it can provide much better IQ than even my 4k display for clarity yet still perform similarly or better than just using dlss or native

As per usual, keep the fanboy crap out and in another thread.

EDIT:

This is purely for how and what you use in your day to day gaming (as in most of the time), not "preference". If you prefer native but use dlss, fsr, xess, you vote for whatever one it is that you use and don't vote for native. In your post, you can explain your preference to use native or upscaling for whatever reason.
As explained many times throughout the thread, point was to see if people do actually "use" upscaling tech or native for their average day to day gaming. It was never about preference as that is a pointless poll as everyone would prefer higher resolution, higher fps, higher bit depth, path tracing and so on.
 
Last edited:
Back
Top Bottom