This generation is somewhat unusual because it (probably) represents the largest improvement in processing power between generations and the smallest improvement in meaningful usage of that processing power. I will attempt to explain what I mean by explaining the previous generation jumps.
Back in the dawn of the videogame console, most systems could only generate very simple 2D graphics which were not very expressive. The introduction of the NES changed everything because it was the first time the limitations of the system were less important than the capabilities of the system. The SNES and Genesis followed up with the perfection of the 2D gaming console; perfect not because further processing power was unusable (the NEO Geo was pretty impressive, and I'd love to see a sprite based game in 1080p on the PS3) but because further processing power was unimportant towards gameplay.
3D was the next step in games and although the SNES, Genesis, Sega CD and 32x were all capable of primitive 3D graphics it wasn't until the Playstation and Saturn were released that we had a viable 3D platform; once again the Dreamcast, PS2, XBox and Gamecube represent the perfection of 3D gaming. The PS3 and XBox 360 have demonstrated a very large improvement in graphical quality over the PS2 and XBox but have yet to show much of a meaningful improvement in gameplay.
The next generation will have greater processing power than this generation and some developers will choose to take advantage of that processing power to produce amazing graphics; what I think will be interesting is that (because of the success of the Wii) many developers will no longer feel the presure to take full advantage of the available processing power so many of the games' graphics may only be very similar in look to PS2/Gamecube/XBox/Wii games with higher resolution textures and somewhat greater polygonal detail.