When the Wii U titles were unveiled at E3, we finally get a chance to look at what it was capable of in terms of graphics. Of course, this is only the first generation of Wii U titles, so you can expect some improvements over its life time. However, we were wondering how the Wii U would compare to Sony and Microsoft’s next-generation consoles. Typically, the leap between generations is quite high, so there are lofty expectations out there for Microsoft’s and Sony’s next hardware.
The Nintendo CEO knows that people may hold out on the Nintendo Wii U if they believe that something “radical” is around the corner, so he came out in defense of his product and said “we can at least assure you that the Wii U will not have such a big difference as the Wii had in comparison to how, on other platforms, developers could expect very different graphic capabilities of generating HD-applicable high-resolution graphics”. I have underlined “capabilities” because wording is important here. Many people assume that Mr Iwata is talking about graphics performance, but he is not.
Capabilities != Performance
Of course, many media were wondering: how would Mr. Satoru Iwata (Nintendo’s CEO) know what he’s going to be up against in terms of graphics speed/power? There are rumors of course, but most likely he doesn’t know with any level of certainty (or someone from the competition should be fired). So, how can he make this claim?
Well, here’s the thing: he is not talking about hardware *performance*, but about hardware *capabilities*. And when talking about capabilities, he can be absolutely certain of one thing: if they come out next year, both PS4 and Xbox8 will more or less have graphics capabilities found in DirectX 11. Today’s consoles mostly have features found in DirectX 9.
If you look at today’s Wii U games, they basically look like slightly more polished than PS3 and Xbox 360 games, but overall, they aren’t radically different mainly because developers are still using DX9-like techniques. In time, this will change when they will use GPU computing and geometry tessellation, just to cite the big ones.
Graphics performance is about number of cores, frequency and memory bandwidth
But capabilities don’t tell much of a story, really. It used to be that graphics processors (GPUs) could easily be differentiated with new hardware feature like texture mapping, then bump-mapping, or stencil-shadows.
But nowadays, GPUs are mostly programmable and there are only a few elements (like frame-buffer blending, or texture filtering) that are still using fixed-functions. Modern GPUs tend to differentiate only by raw power. To do that, most GPU vendors add hundreds or thousands of compute cores, which are the fundamental building clocks of GPUs. Memory bandwidth is also a critical factor in overall graphics performance, and again, there can be very exotic solutions out there.
Conclusion – it’s all about the wording
Mr. Satoru Iwata can easily say that “capabilities” will be very close -it’s true- but what you should actually understand from his statement is that all those consoles will be able to use “similar rendering techniques”, that’s it. As for raw performance, there is simply not enough information to guess, even for him. This is a big deal because even when using the same rendering techniques, a game can look and feel very different if you can throw 4X or 10X the number of objects in the scene. Now, let’s wait and see how Wii U will do when the PS4 and Xbox 8 / Xbox 720 land, because this is still an open, and important question.