What to expect from next-gen graphics [XBSX, PS5]

I find repetitive content really reality-breaking, as part of the uncanny-valley problem. If we had photorealistic rendering, but everything made out of the same repeating textures and repeating interior decor, it'd create the sense of a fake reality, because the real world isn't like that. I think repetitious content is an area that needs improving the same as animation and everything else.

 

Yeah. And having no need to repeat stuff on purpose, level editing work could be much easier, and procedural generation too.

But at this point i don't know how to solve the storage space problem. Current compression methods are not nearly good enough.

Recently i have implemented Voxel DAG compression, which is said to do 'wonders'. But it does wonders only at really high resolutions where any geometry becomes pretty flat at finest voxel levels. At lower resolution it is not much better than regular octree.
Still, the idea is interesting: It works by searching branches of the tree that are equal. So if you have two equal houses bath aligned to the voxel grid, you only need the space of one. You get automatic 'instancing' at all levels of the tree, from a single 4^3 cube of voxels up to a full house.
Limitation here is: Grid alignment must match, and rotation is not supported at all. (Thinking of Bruce Dell again )

But could this idea be used for lossy texture compression? Searching for similar 8*8 quads of texture, make a dictionary of quads, and compose a huge virtual texture from a small dictionary? Maybe even hierarchical, so the result is a sum of multiple mip levels, each level compressed on its own this way?
I think this could work. There is the problem of discontinuity between quads, but this may be solvable or acceptable.
However, if you want rotation too, then the compression algo would likely take years to complete.

No idea. That's why i think storage space is the main future limitation for games.

 

Ooof. Just had nightmares of Dragon Age 2 again.

 

The official ninja theory twitter accoun reweeted this tweet.

 

Just apply a mirror operation to the map and/or reverse the top/bottom of the cave for additional variations.

Also, have you considered changing the colour of snow to make, for example, red sand?

 

Ray traced hybrid (ground up) games are going to be something else. I think a lot of people struggled with the difference with this crop of games (bolt ons), but there will be no question about the difference it will add for next gen.

I feel like this is a super critical milestone WRT next gen consoles. If this is developed in mind for next gen - aside from the software requirements, the hardware must be capable of supporting this. I think I'd focus more on how this is achieved rather than a raw TF discussion. As long as there is enough TF to support graphics like this, then that should be enough to answer the question of a generational difference. Also hopefully this will put to bed the discussion that you could invest more in traditional compute based lighting and beat out the dog slog of hybrid ray tracing.

The question is how the engines are being coded to support this level of fidelity to run in real time. What techniques are they using, what bottlenecks, what limitations, restrictions, what the performance is like etc.

 

I think after this gen we will probably reach a diminishing return but not this time. It will probably be a mix of better than ever rasterization, dynamic light raytracing and static Global illumination.

 

Even with all that, we'll still see the likes of "IQ in 2027 is Unacceptable".

 

Forza could use proper shadowing.

 

Next gen is going to be really nice looking 30 FPS slide shows.

 

Yup, even at Extreme PC settings, it looks horrendous with lots of low res shadows everywhere.
Come to think of it, grass LOD needs to be extended too, even standing still I can notice those distant hills being devoid of any ground details.

 

And this should be an easier problem to solve to creating believable looking distinct faces. You could make 20 or 30 basic chair, table (and other furniture) models and have them procedurally adjusted in dimensions, colour and materials to hugely increase variety. Procedural textures (not necessarily realtime) could also help. GTA does an amazing job of keeping exteriors looking very distinct but it's generally interiors where things really stick out when everybody has the case couch, coffee-machine and TV.

It's a wonder that we're reaching the point of rendering great-looking people but it's largely inanimate environment that is breaking emersion.

 

There's enough power for impressive improvements in hair and clothes on player characters and 'hero' NPC's. I'm curious as to where the quality of unimportant/background NPC's will land. Like the rest of next gen graphics, there are technical and production challenges to upping their fidelity.

If companies like Bethesda only upgrade their existing character tech, we'll end up with lovely hair floating above a potato face.

Edit: with more power and SSDs I'm expecting really great things for next gen crowds.

 

technically identical no, but by the overall looks we should be close.

 

Ray traced stick figures......

 

West of Loathing has that.