Ok...
An engine is just an engine. That's it. It doesn't mean squat. How a developer utilises an engine is what counts. Most of what the Crytec guys are doing can be reproduced elsewhere (though I dare say not as impressively showcased as their efforts

). Some systems use 'y' (3dsmax etc), everyone else uses 'z'(Directx, the world, etc.). There's no real difference.
From what Anet (Jeff Strain) have said, their engine is scalable. Therefore graphical upgrades will happen as time goes on. As they have already been happening (Yay! GW has normal maps. For all the good it did the game). You'll see nothing like Crysis however. That's an FPS. Those game-types are almost made to show-off hardware and software performance. Doesn't mean the visuals could be used successfully (or even be beneficial) in other game types. Theres no reason to suggest that GW couldn't benefit from some of the effects within Crytek's new toy however (The lighting model for instance.)
In all honesty, a GW 2 is far too much work for too little gain. Anet either update everything they have already done, or they create a brand new game. Both ridiculous prospects.
They either spend the next 2-3 years updating 3 games worth of content, or they start from scratch and lose more or less everything they have already established.
Who would care to start all thier accounts again? Not enough people to justify the cost.
Implementation for just the Pvp guys wouldn't be too difficult, but then competition would degrade due to the inevitable plague of hardware, and thus performance issues, while all the Pve guys bitch endlessly.
Plus the business model would take a swift kick in the face (Unless they charge 300 bucks for the game and force existing players to sign contracts to ensure they buy the thing when it arrives somwhere in the next 3-20 years).
Only fix things when they (are about to) break...