For most of my life, real time ray tracing was considered to be the holy grail of GPU tech. Now that we're getting the first generation of cards that can do it, what's the next holy grail or is there one?
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
What is the next step for graphics tech after ray tracing?
- Thread starter Deleted member 1659
- Start date
good ai has been possible for awhile but devs dont want to alienate players
Bullshit. Where have the advancements been? Nemesis system in Middle Earth? Not much else
The next step would be real ray tracing, not this rasterisation guided by extremely rough ray tracing that we currently have.
Also, physics. I want hair with thousands of strands on every head. I want every piece of clothing to be hung on the model with full cloth physics. I want fluids to be fully simulated. I want it all at once. We have such a long way to go.
Also, physics. I want hair with thousands of strands on every head. I want every piece of clothing to be hung on the model with full cloth physics. I want fluids to be fully simulated. I want it all at once. We have such a long way to go.
The dual stick controller limits you. What AI do you want when games need to employ an army of automations for every game because the dual stick pad is and always has been uther shit?
Magnetism, safe windows, contextual actions, aim assist and so on. Any AI is just smoke when you need all this
Smell-o-vision.
Real answer: this isn't graphics related but how about better artificial sound synthesis? It's kind of funny that we have near human looking real time graphics but still have to rely on voice actors. Think of the amount of branching interactive dialogue you could fit into a game if all the information needed to be stored was just the text, and the game engine would render it in the appropriate voice with all its inflections and nuances, according to some predefined configuration.
I've been waiting years for someone to move us in that direction.
Real answer: this isn't graphics related but how about better artificial sound synthesis? It's kind of funny that we have near human looking real time graphics but still have to rely on voice actors. Think of the amount of branching interactive dialogue you could fit into a game if all the information needed to be stored was just the text, and the game engine would render it in the appropriate voice with all its inflections and nuances, according to some predefined configuration.
I've been waiting years for someone to move us in that direction.
It's gonna be a decade before all graphics are able to be raytraced, rather than just some lighting and reflection effects against normal polygons. So we got a while to wait to see what's next.
But, VR is next.
But, VR is next.
Heh. Most of the stuff in FEAR was fake to make the players think the AI is behaving in a certain way. But I guess as long as you don't notice it, it's fantastic. :p
Oh, and everything being a corridor helps script things a lot better.
A lot more. We already have self learning AI agents that behave like an actual AI and not the "game AI we have known for the last 50 years".
They are not used anywhere because simply playing against a self learning AI is not fun.
Pancake panel gaming has already reached diminishing returns, Hence why they want you now to buy 8K tv's. So many resources wasted on trying to get native 4K. So when we reach 4K/HDR/RT levels that's it, Well done game over.
VR/AR is where it all should shift next. Far more impressive/immersive tech that actually does change the way we play games.
VR/AR is where it all should shift next. Far more impressive/immersive tech that actually does change the way we play games.
indeed. there are only a specific set of scenarios where making enemy AI super smart sounds fun, gameplay wise. for a lot of people, they just wanna shoot some dumb henchmen :p that said, some truly smart NPCs would be cool.
slow down, real time ray tracing is still not a mature technology.
anyway, i expect with the new SSDs in the console that we will have a leap in terms of loading and asset streaming, "from pressing play into gameplay in x seconds" and also to be able to have "world shifting" mechanics with that, being able to switch dimensions or some rad stuff like that in a few seconds into a completely new area, i guess its not percisely graphics, but its enough of a "woah" moment for me to consider it one.
anyway, i expect with the new SSDs in the console that we will have a leap in terms of loading and asset streaming, "from pressing play into gameplay in x seconds" and also to be able to have "world shifting" mechanics with that, being able to switch dimensions or some rad stuff like that in a few seconds into a completely new area, i guess its not percisely graphics, but its enough of a "woah" moment for me to consider it one.
It's arguable that this justifies the "not much else" comment, though. Any development in AI which isn't useful for creating fun gameplay, isn't obviously relevant for someone looking for advancement in game design.
But I would tend to agree that... issues in video game AI design today typically aren't due to the AI systems not being complex enough, or to the AI not being smart enough, and in some cases what get perceived as "AI issues" don't really source from the AI systems themselves.
Raytracing is the pinnacle in CGI right now imho
the next steps should be creating the hardware capable enough to do toy story in real-time
Where we are lacking now is in animations which are hard and expensive
There are also a lot of advancements left in display hardware though
like peak Dolby vision level HDR, more resolution, higher refresh rate, glassless 3D, OLED without burn-in, VRR
the next steps should be creating the hardware capable enough to do toy story in real-time
Where we are lacking now is in animations which are hard and expensive
There are also a lot of advancements left in display hardware though
like peak Dolby vision level HDR, more resolution, higher refresh rate, glassless 3D, OLED without burn-in, VRR
I'm hoping on bigger denser worlds with large NPC crowds/enemies and fully destructible environments as a standard.
Already there. See Ratchet & Clank on PS4 Pro.
And it blows my mind that people are still surprised by this and apparently have no clue how marketing works. How many times do we need to explain that graphics sell games and not the speed at which these graphics are shown? The mainstream doesn't give a fuck about 30 or 60 frames per second. You can't show framerate on a screenshot in a review or on the back of a box. You can't let a casual player "feel" framerate (if they notice the difference at all) via a trailer.
Already there. See Ratchet & Clank on PS4 Pro.
And it blows my mind that people are still surprised by this and apparently have no clue how marketing works. How many times do we need to explain that graphics sell games and not the speed at which these graphics are shown? The mainstream doesn't give a fuck about 30 or 60 frames per second. You can't show framerate on a screenshot in a review or on the back of a box. You can't let a casual player "feel" framerate (if they notice the difference at all) via a trailer.
Hell, see ACTUAL Toy Story on Kingdom Hearts 3. I think Digital Foundry did a comparison video for the game's implementation vs the movies.I'm hoping on bigger denser worlds with large NPC crowds/enemies and fully destructible environments as a standard.
Already there. See Ratchet & Clank on PS4 Pro.
And it blows my mind that people are still surprised by this and apparently have no clue how marketing works. How many times do we need to explain that graphics sell games and not the speed at which these graphics are shown? The mainstream doesn't give a fuck about 30 or 60 frames per second. You can't show framerate on a screenshot in a review or on the back of a box. You can't let a casual player "feel" framerate (if they notice the difference at all) via a trailer.
Next gen, all graphic techs + methods will be used more and more sophisticated versions will be made.
I hope devs try improve A.I and npc/world realism, A.I is not just harder enemies, its also having npcs which behave realisticly.
As morpheus once said "the matrix is made of rules but those rules can be broken"
Well, I want devs try to break the rules and break free of common videogame restrictions.
I hope devs try improve A.I and npc/world realism, A.I is not just harder enemies, its also having npcs which behave realisticly.
As morpheus once said "the matrix is made of rules but those rules can be broken"
Well, I want devs try to break the rules and break free of common videogame restrictions.
Last edited:
I presume simulations can be scaled up further and further for another long while. Not sure what the next new big thing will be though.
That's the one silver lining to the dark streaming-only future. All sorts of funky netcode shenanigans will drastically change. No more jumpy roll-backs or seeing different things on your screen than the other players. If we can get rid of the "I definitely hit you, I have video evidence" "no you didn't, I also have video evidence" arguments, that'd be nice.
You should tell this to the staff that worked on The Hobbit.
Motion blur, clarity of image, details perceived, all things equal (which is obviously not possible outside of CGI trailers since at 60fps you have half the time to render the frame) 60fps vs 30fps will always look drastically different.
Given enough computing power and with diminishing returns of other graphic improvements I have 0 doubt that games will move to higher framerates sooner or later (and I'm not even talking about responsiveness).
Pretty much what I was going to say. It would be great to see better hair and cloth physics, physics-based deformation, etc. And once ray-tracing takes off, it'll enable devs to make their environments more dynamic without sacrificing lighting fidelity. More destruction, more moveables, etc.
I think the field of animation is also going to make great strides during the next gen. More fluid movement, more believable and contextual movement, etc.
Call of Duty, Fortnite, Rocket League and Apex wouldn't be anywhere near as big without 60fps. Normal people do care.
on the contrary for the streaming future, sure no more 'funky' rollback...in replacement you will see graphical lossy pixel artifacts on your crosshairs and control input latency.........that you have no video evidence of to back yourself with.
I do side work for a company that handles excess work for studios. One company we are dealing with is working with sparse voxel octrees and some of the results in the amount of detail they can achieve is pretty staggering. One of their tech demos went from literal particles of dirt to a massive building with all detail being real world geometry instead of normal/parallax mapping. If it takes off, it could be highly impressive.
Because AI evidently isn't graphics and I don't think people even understand what they mean when they say they want better AI anyway...
Physics.
Physics, physics, physics.
Physics driven everything: in lighting, animation, materials, physical collision (down to a molecular level ideally).
Physics.
Physics, physics, physics.
Physics driven everything: in lighting, animation, materials, physical collision (down to a molecular level ideally).
This is actually very interesting.
I was thinking about this a couple days ago , like, we have some nice AI voices nowadays, voices that doesn't sound like Loquendo anymore.
Is anything actually coming out of this, or is there a goal/outcome in mind? I recall seeing this a year or two ago and haven't heard much since.
erm...these games are not as big as they are because of 60fps....what are you smoking....they were 60fps due to the nature of the games themselves, not the actual selling point.
GTAV and RDR2 are not anywhere near 60fps (on consoles) so what's your reason for it?