next »

[SolarLune]

I expect better graphics and a more solid networking infrastructure (even from Nintendo), but they'll all pretty much be the same consoles as ever. Every console's trying to do the same thing - be a one-stop-shop for entertainment in your living room. At least the Wii U's trying something different with the gamepad. That's something I discussed with my brother - consoles keep trying to have better graphics, but in my opinion, it's getting unnecessary.

Graphics can't or won't improve more than they are currently without the consoles being expensive. Even if the graphics did improve drastically, would you care that much? Do you really stop to see the sun beams breaking past the rocky crags of a canyon in the middle of a giant robot fight? Do you really notice just how high-resolution everything is when you're trying to focus on inputting a correct combo? And does being able to push a trillion polygons a second matter if you're playing Limbo 2? No, not really. Having good graphics is nice, but you quickly forget about them when you're playing the game.

When you turn on a great-looking game, you might say, "Wow," but if you spend an hour being bored to tears just looking around with nothing to do, you'll probably never pick up the game again. Having a good underlying art style, clean and nice graphics (not cutting-edge) and solid and fun gameplay is much more interesting and engaging than just pushing more polygons and better textures, and that can be done on the current systems.

I'll probably end up buying a Wii U, and perhaps nothing more, since it's probably going to have enough power to handle all ports (even if they look worse) and seems like the most unique console out of the three.

[J-Snake]

I am not a fanboy of graphics but I think if you see amazing high-res visuals at 60fps some will change their opinion about how graphics can matter. The time just hasn't come yet (only in some trailers). But yeah, the mechanics, the game-system and art is most relevant to me aswell.

But the thing is that graphic-cards have been pushed a lot, of course keeping power-consumption in borders. Graphics cards are parallel-computing monsters. What would be cool if there would be something like OpenCl for consoles, which means if the gpu isn't fully used it might spend its ressources to compute parallel game-logic parts.

[JMStark]

720p, 30fps, more online stuff like free-to-play and streaming demos, $200-$300 consoles with monthly subscriptions to subsidize cost (subscription provides cable-like content and other such features), visually there will be improvements in dynamic lighting and shadows (most important thing to improve at the moment IMO), some other post-processing effects that look cool in screenshots but are not noticeable when playing, the same types of games as last generation, at least in the beginning, that take advantage of additional CPU and RAM to provide better physics and more dynamic worlds, and connectivity with phones and tablets. As always, there are some neat things and some unfortunate things. Games will become more expense to make, but not by too much since the assets are already being made at such a high quality. There will probably be more of a reliance on engines like UE4 and CE3 (good tools such as these will help reduce the cost of making games). I am really interested to see where independent devs will fit in; the range of products will be wider than ever. Visuals are no longer going to be the focus; I might be wrong on this, but diminishing returns are going to start happening.

I will remain a PC gamer.

[SolarLune]

That sounds very logical and sensible. I wonder if the console market will truly crash if even the best-selling games take more money to create than they make back.

[PompiPompi]

Probably DX11 and OpenGL equivalent features. Which means things like Tessellation which currently don't exist in consoles since they have an older feature set.
As someone said, also CUDA and OpenCL stuff.
This might also be a boon for PC, since if console game developers will start fiddling with DX11 stuff it will also trickle more to the PC.

However, I am not even sure when all this will happen, since everyone seems to be focused on mobile ad tablets these days.

[J-Snake]

DirectX11 on next gen? Nah, still feels like an artificial add.

[PompiPompi]

I don't understand what you mean by that?

[J-Snake]

I mean it is just a feature. It doesn't automatically enhance visual fidelity and performance. Shader units do.

[PompiPompi]

Well it lets you do more with less.
A wimpy DX11 GPU might do more than a beefy GPU with only DX9 support. Because the DX11 is more EFFICIENT.
For instance, have you watched the iPad compared to Android tablet playing Shadowgun? It didn't look much better on the iPad, and on the Android it had fabric phsysics which didn't exist on the iPad version.
Features are also important for performance.
(The iPad GPU has more power btw)

[JMStark]

With certain publishers claiming that they need to sell 5-7 million copies of a title in order to just break even, the absurd idea of there even being another crash is a little less absurd. Fortunately, it would not be like the crash of 1983, with factors such as digital distribution making the situation, well, different. Really, the only thing that would crash would be the majority high end AAA development, but as I said before, it is unlikely. There will just be more studio closures and what not as more middle of the road developers who make AA games are unable to keep up. Indies, PC developers, and anyone who focuses on the causal phone/tablet market would be fine, since for them the situation is quite the opposite (they can easily make more money than they spent on development since they do not have movie sized budgets to worry about).

[J-Snake]

Tesselation is very expensive, I think. So a good dx9 card is a better solution, Crysis1 prooved it already. Comon, it is hard to take that shadow-gun stuff seriously in this perspective. It looks like 2 Watt version of Gears of War, achieved with certain optimization-decisions and trickery. I don't see what those features have to have to do with dx11.

[PompiPompi]

So you are saying that all the features developments in GPU in the past 7 years is just a hoax and it's no useful for anyone because you can do the same stuff with DX9?
You sir, obviously don't know what you are talking about.
The only thing I "worry" about the gaming industry is that mobiles and tablets are becoming a bigger share of the pie and companies are going to focus on mobile and tablets.
If there is any bubble to worry about it's things like facebook.

[JMStark]

Tessellation shouldn't be the main focus anyways, since it doesn't really contribute much of anything to the quality of the image. Like I have said previously, the main focus should be on better dynamic lighting and shadows; these two elements are vital to the visually experience and just are not quite there yet in games.

And yes, Facebook is probably a bubble. As I previously said, the idea of a crash is absurd. However, the unsustainable budgets of major studios are going to cause some problems if a solution to them is not found.

[PompiPompi]

Tessellation is not the only thing that they added since DX9...
And Tessellation is not only for improving image quality but also improving performance.
And about dynamic lighting, compute shader which was introduced in DX11 is the only first step taken torward ray tracing, and thst is what I suppose you mean by improving dynamic lighting, because eventually there is only much you can do while avoiding ray tracing in order to improve lighting quality.

[J-Snake]

So you are saying that all the features developments in GPU in the past 7 years is just a hoax and it's no useful for anyone because you can do the same stuff with DX9?

It was not a hoax but proved to be not the right way how to go about it. The way is just to move away from fixed-function features and do everything on your own with shaders. Vertex and Fragment-shaders are the most important ones. If you sum up the graphics-card, it is just a large amount of stupid independent shader-units. The more you can put on a card, the more it can process since traditional rendering methods are perfect for parallel processing.

Like mentioned in the above post, more of dynamic lighting would be cool. A dynamic lighting - program is nothing but a shader executed in real time. The more shader-units, the more you can do in a frame. That's what it is all about and there is no sign of change.