• 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.

RDNA3 Navi 31 update

Its similar to the pixel peepers that can tell DLSS is superior than FSR. Like people that have better than 20/20 vision, yes its a thing but the general population cannot tell the difference. I think this topic had a thread years ago discussing dividing the factions. :)
 
Its similar to the pixel peepers that can tell DLSS is superior than FSR. Like people that have better than 20/20 vision, yes its a thing but the general population cannot tell the difference. I think this topic had a thread years ago discussing dividing the factions. :)
I had quite a side by side with DLSS and FSR (FSR 1.0 in fairness) and did find DLSS to be superior. Not a huge amount in it at the quality settings, but as you went lower and lower in settings the difference was quite noticeable, and I'm no pixel peeper.

FSR 2.0 sounds like a vast improvement however, and I haven't had chance to do a side by side with that yet.
 
I had quite a side by side with DLSS and FSR (FSR 1.0 in fairness) and did find DLSS to be superior. Not a huge amount in it at the quality settings, but as you went lower and lower in settings the difference was quite noticeable, and I'm no pixel peeper.

FSR 2.0 sounds like a vast improvement however, and I haven't had chance to do a side by side with that yet.

Depends which iteration you use. As it was a generalisation I imagine you get the point. If you look at say DLSS 1.0 and FSR1.0 let me know how noticeable you find that! ;)
 
wasn't it usually the opposite and Nvidia cards are more vibrant
When LCDs were first coming in people used to enable digital vibrance on Nvidia cards to make colours pop a bit more. Nowadays the opposite problem is more likely - most monitors are set up too vividly. But left to default I've always found it was monitor dependant, not card - I now use a callibrator and the adjustment was pretty identical between AMD and Nvidia back when I tried both.
 
Depends which iteration you use. As it was a generalisation I imagine you get the point. If you look at say DLSS 1.0 and FSR1.0 let me know how noticeable you find that! ;)
Oh, I expect so. I don't think I've ever used DLSS 1.0, it's why I wanted to be clear that I'd only tried FSR 1.0. I can't just pit DLSS 2.whatever to FSR 1.0 and take that forward as a fair comparison for the rest of my days.
 
Human nature and your brain adjusts to differences imho, unless you go full on OCD and start looking for defects but then you'll also find them in the game asset themselves.

What is far more useful imho is being able to apply these up sampling technologies to legacy titles and games where the developer has no intention of making further changes. I want this to work fully in VR all the time not some select games that are still in development.
 
There was this ancient article (in a paper magazine, think it was Computer Shopper), where they did a group test of sound cards. And they had proper equipment.
Turned out that Creative did something to the sound which scored them low in the technical test, but in the subjective tests they got good scores.

Might the default monitor colours thing be something similar? Technical not correct vs what people like?
 
When LCDs were first coming in people used to enable digital vibrance on Nvidia cards to make colours pop a bit more. Nowadays the opposite problem is more likely - most monitors are set up too vividly. But left to default I've always found it was monitor dependant, not card - I now use a callibrator and the adjustment was pretty identical between AMD and Nvidia back when I tried both.
Maybe as I'm using a VA panel a little extra helps. I'm not bothering to be scientific about it, just tweak until it looks good to me.
 
There was this ancient article (in a paper magazine, think it was Computer Shopper), where they did a group test of sound cards. And they had proper equipment.
Turned out that Creative did something to the sound which scored them low in the technical test, but in the subjective tests they got good scores.

Might the default monitor colours thing be something similar? Technical not correct vs what people like?
I think they find people usually prefer vibrant to accurate, just look at Samsung :cool:
 
I think they find people usually prefer vibrant to accurate, just look at Samsung :cool:
Older Samsung (TVs) used to be worse iirc. Remember looking at all the TV's at Oxford Street and all the Samsungs stood out. Obviously everyone wanted one. We've got one of their last plasma TVs I think, and that's much more normal.
 
I'm always skeptical of these claims from both sides - What metrics are "2x the performance" based on, do we know or is it a vague promise?
You're right to be sceptical, without specifics those claims don't mean much. Double compute, or rasterization or ray tracing or some or all of them? I'd say most likely is in that order, which matters to you most depends on your workload.
 
That sounds sensible - compute seems to be the easiest thing to put numbers into as well as the other methods can vary for reasons outside of the user's control or setup. Don't get me wrong, if both sides are able to double performance, then fantastic, but lets wait and see the details. Not some bloke waving a prototype promising the moon on a 4k stick.
 
7900XT said to be over 2x the 6900XT. War is heating up!

Maybe in RT? I have no doubt AMD know how important it is to get their RT performance up and i think it would be a mistake to under-estimate AMD's ability, they have proven themselves extremely capable over the last few years, lets wait and see, it does seem like a very tall order even if we are just talking about RT performance.
 
Maybe in RT? I have no doubt AMD know how important it is to get their RT performance up and i think it would be a mistake to under-estimate AMD's ability, they have proven themselves extremely capable over the last few years, lets wait and see, it does seem like a very tall order even if we are just talking about RT performance.
That's the interesting thing now. Because they can't just up the clocks or make chips larger they have to be more innovative, makes it much harder to predict the improvements especially if it's a novel approach.
 
Back
Top Bottom