nesdev.com

I don't believe at all that the PS4 will have an equivalent kind of nostalgia in 30 years that NES does now. I think the curve of game hardware capabilities is kind of asymptotic and we're already into some pretty marginal differences between generations.

Not to mention there isn't anything in particular that defines a game as a ps3 or xbox 360 game. They all look the same accross all systems. It's the age of multiplatform games that look identical to each other, meaning the "platform" has lost its meaning/character.

I would understand it if you made a game for kinect or wii remote, but then again not really.

_________________
https://twitter.com/bitinkstudios <- Follow me on twitter! Thanks!

Kinect and Wiimote can both be used with a PC, so you don't really need the console they came from to play with their possibilities. Same with PS Move, and there are other various VR control schemes that are floating around at the moment.

The Wii U's screen gamepad seemed like it could be used for interesting asymmetrical views, but I can't think of any Wii U game where it actually paid off. With Starfox Zero it actually made the game much harder to get the hang of in a way that was not really fun to learn IMO. I've seen several recent games get paired with a smartphone app to allow other kinds of asymmetrical view like that too, though, so the idea isn't exactly limited to the Wii U. (...and of course the Gamecube + GBA, or Dreamcast controller offered similar things on various scales.)

In most cases the controller itself can be divorced from the console, if there's something interesting about that controller that you want to exploit. (...and there's an almost endless supply of obscure novelty peripherals out there if you wanna go spelunking.)

But to be fair, you can tell a huge difference between PS4 graphics and PS2 graphics, so in a sense they have their own style. It seems kinda dismissive to say modern systems don’t have any differences.

I have hardly ever played games on any system more recent than PS2, but to my understanding the only real difference is a much larger resolution and maybe better framrate, which is due to progress being made on flat-screen HD TV technology. But even there we reached a point where more resolution would be completely useless (i.e. not visible to the eye). So for the next generation they'll probably try to sell some "3D virtual reality" stuff again even though it has been proved multiple times this was causing major headaches along other problems.

Its odd though the perception has changed, I don't think a Xbox 360 game is that different to a Xbox One game, but then I play one and its "WOW was it really this bad?". But we are hitting the limits were the problem is not adding more capabilities to the machines its adding more dev time to be able to use them. Content generation is the far larger and harder to solve problem.

I think people will get nostalgic for a PS4 in 30 years, ones first console is always special.


The only fun thing to do with a PS3 I can think of is to make Farbrausch style demos on the CELL.

I'm not the only one to notice that physics and AI haven't improved as much as graphics, if at all. A lot of AAA games nowadays look stunning until you try to interact with something, at which point it all falls apart.

Some of this is just because developers aren't doing what they could be - running in place in front of walls instead of reacting to the wall is a good example - but a lot of the physics may be waiting on more powerful CPUs. Fluid dynamics in particular tend to look really bad; a body of water in a modern game can look completely real until you jump in and start splashing around, producing splash effects that would have been unremarkable on the Nintendo 64 and totally failing to alter the lovely realistic wave pattern that looked so impressive a moment ago. I've seen semi-realistic interactive water in a modern game, but it was an indie game where you play as a swimming frog, and the Reynolds number was super low. (Also, I'm pretty sure the refractive effect was wrong, as usual. I don't think I've ever seen correct refraction in a video game, and it's not like it'd be hard.)

Video games took a wrong turn somewhere. Nobody seems to care about world interaction any more; it's honestly embarrassing that the decent physics in Breath of the Wild were such a revelation (and showed up the physics in Horizon Zero Dawn so thoroughly). Racing games of all things can fail to credibly simulate a three-car pileup. AI has been mostly stagnant for a decade. Graphics have settled into a standard way of faking things, and if you go outside the range of stuff that looks good in demos the seams start to show. Real-time shadows can look worse nowadays than they did on the N64. The last generation with a standard 60 fps expectation was the PS2/GC/Xbox gen, so all of this photorealistic stuff held together with spit and baling wire doesn't even move smoothly...

...what was I talking about? Right, CPU power. Maybe the next generation will have enough CPU power to do decent water physics. I bet you could make a cool jet-ski game or something...

Maybe I'm being a bit unfair...

... Really? It's almost free when you're raytracing (although the caustics aren't), but I don't see how dealing with the very complex lens that is the water-air junction would be easy when rasterizing triangles...

That said, my knowledge about opengl stuff is thin, and my knowledge about directx is nonexistant, so I could easily be wrong.

Seconded. In fact I'd argue the AI is getting worse, as people waste more money on graphics to the exclusion of all else.


Actually, correct refraction requires raytracing. There's no way to render water using only primary rays (rays passing through the eye, which is the only kind rasterization can handle) that isn't fake.

EDIT: where'd that ninja come from?


...seriously? Other games are worse? I honestly thought the physics puzzles in BotW were game-breakingly bad. It's the first game I've played with major physics content, and I thought it was just another worthless gimmick. If this is the state of the art, you're right, it's an embarassment.

Okay, maybe I didn't think that through carefully enough. Strict exactitude would indeed require the ability to render each pixel at an arbitrary point on an arbitrary polygon based on the refraction angle result for that pixel. Same with reflection.

But there's a large gap between strict exactitude and what actually shows up in games. Ever notice how in real life, water tends to look shallower than it is? Something as simple as distorting the underwater terrain mesh based on viewing angle would have been reasonably cheap on the N64, and I don't recall ever seeing anything like it.

I vaguely remember the water in Prince Of Persia on Xbox/GCN looked pretty persuasive? But the camera never got particularly close, and it was only 480i. Also not in the sequels, where they evidently just targeted the PS2's capabilities and didn't do much to improve the appearance of the GCN/Xbox ports.

In Raytracing, refraction is basically "free" (just casting another ray). In games though, raytracing is basically never an option except for a few novelty games.


The hard part about refraction in games is how you render/determine what's "under" the surface of the water. If the surface of the water was flat and still, you could render the entire scene upside down under the water, and use that are your reflected version, or an offset lookup to that for your refracted version.

The problem is, refraction and water requires a continuous variation of the surface, i.e. every different angle requires a different viewpoint on that reflection. You can't get all of that from one upside-down viewpoint, you'd need a different view from each point on the curved surface. No-go. In general, the technique is to render the upside-down scene once (and save to a texture) and then use the angle of refraction to warp the lookup to that texture -- this tends to look very bad at the edges of the water, where it will probably be displaced up or down from where it is being told to refract from, but you can try to do some clamping at these edges to restrict the refraction there, or hide it with "sea foam" or something.

With an emphasis on the "maybe". I think there are less 60fps games from the 7th generation than the 6th to be honest, as open world games became really popular.


The base PS4 already has a 1.6GHz x86 64 8 core CPU; people (who aren't game developers) will bitch and moan about how it's not powerful enough or whatever, but unless it has the world's worst IPC, or most developers are very bad at coding for multiple cores, it doesn't seem possible that it isn't up to a level of processing video games would require. I think mediocre physics have more to do with a lack of effort on developers parts. It's like AI in video games; I never really seem to notice any dip in performance with computer opponents, so unless the developers left just enough headroom for the AI, I don't think it's a CPU issue.


You'd love these physics https://youtu.be/8lrBwSgCovE I really love how photorealistic they're trying to make racing games look, when you can slam directly into a wall at 100mph and only the body paneling will dent... Ever heard of BeamNG.Drive? It's probably the best display of physics in a video game that I can think of. It's much more of a car crash simulator than an actual game though.

You're a programmer, if you think the problem requires that few resources, have a go I can tell you it doesn't, but you'll probably only believe it once you've done something similar. Essay on the requirements of physics and AI omitted for brevity, as well as that describing Bobcat performance.

Technical reasons aside(as I have no knowledge on how stuff could be done or whether they're easy to achieve) I think sometimes this might be done on purpose for gameplay reasons(or just lazy programming). If done too realistically, the player may not be able to judge easily what's inside and how deep it is, etc., before jumping straight into a pool, so this may be an acceptable break from reality.