Graphics card generation is over and it was mostly trash
techradar.com
techradar.com
Personally I dislike DLSS so much:
If I enable it in Cyberpunk 2077, there are some visible noise patterns that pop in and out of view especially near shadows on flat textures all the time, it looks horrible and distracting. And if I disable DLSS, if the game updates sometimes it enables it back so I see the horrible things again!
Why do you need AI to upscale anyway if it looks this bad, why not just use Lanczos/bicubic/... upscaling? They are not only simpler and faster, but look better too, because they don't create the above mentioned horrible visible noise patterns.
Then they have the inter-frame thing that causes input lag, another bad degradation.
Then they 'solve' input lag by touting a vendor lock-in-sounding feature (Nvidia Reflex), while the real solution to input lag is not add vendor lock-in feature but ensure you compute 1 frame at the time and do so fast (which is of course not simple, but is a matter of optimization and not of brand-named features, and perhaps not generating and inserting additional frames would help).
Then there is the fact that DLSS is something that needs to be supported by games themselves: why? This again makes it sounds vendor-lock-in-ish because they can have it "exclusively" in some games. But nothing about upscaling, nor frame generation, sounds like something that requires game engine support to me. The graphics card should have enough info to do that by having all the pixels. I mean, it was always possible to play games at different resolutions (DLSS is basically also like playing the game at lower resolution), and there always has been some upscaling required for that (whether it's the monitor or the graphics card doing it). This never required special engine support.
Because DLSS is a AI-based solution with customized settings and inputs for each game. (The “Deep Learning SuperSampling” is right in the title.) This is also why it requires NVIDIA’s bespoke Tensor cores to function. DLSS 2.0 removed most per-game optimization but there’s still some support work required.
AMD’s FSR is basically a traditional upscaling solution, just with low overhead. FSR runs fine on NVIDIA and even the Nintendo Switch - but almost all reviewers agree DLSS is superior in most games.
Except, perhaps, Cyberpunk 2077. But that’s not a fair metric to measure DLSS by because everything about that game has been a technical disaster.
Any idea why this needs bespoke cores? GPU's were already used for AI and crypto mining before this, without bespoke cores for 1 specific application
> The “Deep Learning SuperSampling” is right in the title
With that name, it could have been based off of a single pre-training on many games, to work on games in general
> With that name, it could have been based off of a single pre-training on many games, to work on games in general
Well, I’m not NVIDIA, so I don’t know if you could. I just know AMD tried and the results were almost universally considered inferior.
I disagree. It's not even close how much better DLSS looks than traditional upscaling. I've run into very few noticeable artifacts from DLSS that are distracting, and this seems to be the general consensus among most consumers.
> DLSS is something that needs to be supported by games themselves: why?
It requires motion vectors and depth buffers in addition to the low res aliased images.
DLSS looks poor to me and I have yet to actually keep it enabled.
This is the same reason 60FPS in modern games isn’t “enough” while it felt buttery smooth on old console fighting games. Consistent FPS is vastly more important than average FPS.
Here I was thinking the goal was entertainment.
The reality is that we're dealing with trade-offs, and one of them is reducing resolution to improve FPS. If I can't play at native resolution with adequate FPS, the options are to reduce quality or resolution. I personally prefer to keep quality as high as possible, even if that means a lower resolution.
If you reduce resolution, the first thing you'll notice is that the result looks blurry, or otherwise just less sharp thanks to whatever is doing the default upscaling. DLSS lets me reduce resolution while retaining a sharp image that looks like native resolution. That's a win-win if you don't enjoy 4k@30fps.
> This is the same reason 60FPS in modern games isn’t “enough”
There are hundreds of millions of console users that disagree. It's great to get a billion FPS, but the reality is that outside of a few people most gamers just don't notice the difference after a certain point.
Entertainment doesn’t need to be responsive. I can watch a movie at 24 FPS and it’s fine hell most YouTube videos are fine at 360p. First person games however need to be immersive to be entertaining rather than annoying.
> It’s great to get a billion FPS, but the reality is that outside of a few people most gamers just don’t notice the difference after a certain point.
You’re talking about something different. I’m saying a monitor set to 60 FPS is fine as long as the game is rock solid at 60 FPS. However, frame rates vary so wildly a game can easily average 120 FPS and suddenly then drop 4 frames which people notice.
It's such a subjective evaluation. I was plenty immersed in Elder Scrolls Daggerfall, despite constantly falling through the floor. I was plenty immersed with Cyberpunk 2077, despite minor car accidents rocketing me to the moon.
> frame rates vary so wildly a game
Which is a problem that DLSS helps solve, by improving FPS while minimizing the sacrifice to quality. I'm not sure what's the problem? If a few wonky shadows or other minor graphical glitches ruin the immersion for you, it sounds like you've set the bar too high and have unrealistic expectations.
I much prefer turning shadows off than have them do something crazy. Graphics don’t need to be awesome just consistent. Minecraft is in many ways worse than Daggerfall, but you quickly get used to the world. Skyrim is fine for a more realistic world even without mods, until a texture messes up and looks crazy. That’s the moment when you go from playing at game to looking at a screen.
IMO Cyberpunk would be strictly better off with fewer effects and worst textures in exchange for fewer glitches. Voluntarily selecting more glitches please aka DLSS is just a terrible trade off from my perspective.
DLSS requires motion vectors, depth buffer, and jitter in addition to the rendered frame. DLSS 2.0 is effectively Temporal Anti-aliasing (TAA) with deep learning replacing the heuristic approach of TAA implementations and thus needs as much integration as TAA does.
https://old.reddit.com/r/hardware/comments/fvgf7d/how_dlss_2...
Modern upscaling solutions (eg TAA) require not just the pixel outputs but various inter-frame/metadata such as depth, normals, motion vectors, last frame, etc. DLSS is more or less just the spiritual successor to TAA with machine learning sprinkled in. Integration with the game engine is needed because the non-color metadata needs to be provided to the DLSS and how each engine computes those things is a bit different.
No, Nvidia Reflex is a step in the right direction. You want to sample input at the latest possible time and then create a frame for that input. Sampling input at the beginning of a frame is the worst time you can choose as that will guarantee at least a frame of latency. If you are able to render in 20% of the time of a frame you want to sample the input and then start rendering 80% of the way through the current frame that way you are only add 20% of a frame of latency at a minimum in this part of the system.
This is how things are done in VR, but the location of input devices are also exterpolated to predict where they should be located at time they will be displayed.
That's impossible, if you change your movement it will mispredict and thus cause more lag, while exactly when you change is when it should feel the best. Consistency feels better.
Just like no magic gizmo can reduce audio lag other than have actual proper low latency standards (for bluetooth audio): it can never predict when you will hit the button to fire your weapon, etc..., in games
It will not cause more lag, it will just not be accurate to where your cursor actually should be when the frame gets displayed. For people with strong computers you would only need to predict a few ms forward in time which should be possible to get really accurate.
>it can never predict when you will hit the button to fire your weapon
That is not what is happening. What is happening is that time between registering the click and the frame being displayed. The prediction I'm talking about would be predicting subframe mouse movements. This behaviour is more for smooth mouse look. For shooting you want to have a precise timestamp of when the click happened so that you can see if at that moment in time the shot would have hit. The position where you are looking when shooting and the position where you are looking on screen are different things.
Old news, but in a similar vein -- games that incorporated PhysX back in the day required nVidia hardware to make it work. I imagine its a similar issue for PhysX and DLSS, in that the game engine has to know to expect PhysX calculations and have something to run those calculations, and then use those calculations for debris or ragdoll physics. It could be turned on or off (so AMD cards could still play the game with a different physics backend).
It sounds like another case of nVidia making a cool technology and game devs choosing to use that cool tech in their game (or not, looking at you Starfield).
It probably could run on other GPU but Nvidia isn’t the type to just give tech to their competitors.
Modern game engines aren't credibly capable of 4k gaming, given the market's desire for high refresh rates. The fine print will show that they are doing 4k@30Hz which is like going back in time to the PlayStation 2 era but with really fancy geometry.
The bar to clear is 4k@120Hz. However getting there requires a full stack refresh. A high end panel made in the last two years that not only can handle high refresh 4k, but has the required state of the art HDMI or DisplayPort tech to stream that many bits.
Since nobody else is in place yet -- the monitors, the game engines -- there's no incentive for the graphics card to be early to the party. RTX 40 series was for people with more money than market sense.
Nvidia is going to pay gaming lip service and bide its time until there's an actual growth vector in that space. All their attention is, and should be, on monetizing the 'AI' hype curve.
I think, as you say, AI will be the next paradigm for gamers, and having specialised chips on these GPUs that can turn a scripted NPC story-based interaction into a do-what-you-want-under-these-contraints a la "AI town"[1] might keep them in business a little longer before they have to start upping the visuals again.
No one (at least that I've heard) is saying, for instance, that Undertale or Deltarune are "visual masterpieces". They don't need to be. They're well-made games that do and say some interesting things.
https://store.steampowered.com/app/1388770/Cruelty_Squad/
(Haven't had the chance to play yet, but I've heard great things)
If I want to play a games these days I normally go to emulators and back to old games. Old arcade games and computer games from the 80s and 90s era may look dated but some of them are more fun.
Also, the current trend is to make games more accessible and actually better for casual gamers. Most games now have customisable difficulty and quality of life features. When I think of 80s and 90s games I think of dying a lot and having to replay, not being able to figure things out without buying a walkthrough, no quicksave etc.
Whaaaat?? Nothing wrong with emulators, but there's tons of great, short, casual games out there! Here's a list of some of my favorites to get started, you can find a ton more of you search for lists containing some of these titles: Gris, Journey, A Short Hike, The Witness, Oxenfree, Celeste. And if you really want to dig deep into the short/indie games scene, check out https://itch.io (though it can be kinda hard to find the gems there).
2. Art Style
?. Sound Design (unsure if under gameplay or on its own).
3. Graphical Fidelity
This is what people look for in a game and what sells them, atleast in my opinion.
Some of biggest mainstream games of the last decade or so are Minecraft, and Fortnite.
StarCraft II is usually played competitively on low graphics settings, to improve visual clarity. The game is 13 years old, and on high graphics still looks fine when compared to any modern title.
For more twitchy competitive stuff I can see the need for FPS but that’s just a single corner of gaming.
GPU vendors keep pushing new pretty tech that their cards can't quite keep up with and even a 18ms frame time means 30 Hz with no dynamic refresh.
Bugs that cause the game to fall below the minimum specs will definitely get fixed. Whereas bugs that cause the game to fall below recommended specs will be deferred to at best a day 1 patch, but also maybe never.
So if you want the minimum specs to enforce 4k@60Hz and really mean it, we need to push the recommended specs higher still.
1080p is fine for min spec.
120Hz is more than twice as hard of a performance number to hit compared to 60Hz.
It’s also the fact to some degree that the safety factor on the 12vhpwr connector is less than the old 8-pin, so smaller issues lead to melting. I know that I am slightly stressed about my 4090 even though I’ve thoroughly verified that it’s fully plugged in.
[1] https://www.techspot.com/news/71172-nvidia-boss-moore-law-de...
As someone who is targeting Raspberry 4 and Vision Five 2 where the low end is still 2GB shared RAM there is zero point to go higher.
Embrace the limitation!
I can still play all games, the only games that struggle are the ones that hit their head in efficiency roof because of engines or pushing the boundaries too far.
Most titles are CPU bound too!
I'd like to see GPUs that are closer to dumb frame buffers again, with their need for fast RAM access served by the main system RAM and that capability benefiting the whole system.
Compare it to the 20XX series where the top of the line card only has 11GB of VRAM, less than the 1080Ti... and the flagship feature "raytracing" just isn't viable at the power of any of those 20XX cards.
Pretty sure we have at least a year before the 50xx series will be released, assuming it's called that.
But, didn’t mention that because that isn’t the topic of desktop rendering.
That makes things really hard when your absolute minimum input latency (that nobody gets in practice) is already 24ms at 60Hz.
Maybe once we start talking about higher frequencies it can make sense. VR is trying hard for higher refresh rates since we are more sensitive to input latency for things we are experiencing and is hoping that tech helps get there.
Also the performance benefits aren't as enormous as you would think. For instance unless you are pixel bound (which to be fair most things are in 4k games) it actually hurts your performance by slowing down the CPU.
My fear is it will end up being yet another tool that gets abused to where everyone associates it with a lack of quality.
After all looking to your left and seeing and old image would be very much against the goal of increasing performance.