And if anything, I don't think the engine is the problem. He already shipped 4D toys, which I suspect is a way to monetize the devtools he made for Miegakure (nothing wrong with that). And even before that, the 4D was fine.
But while it is nice having good 4D, turning it into an entertaining game with good graphics, good puzzle design, a story that doesn't suck (even a simple one), etc... is hard. And we are not talking about a FPS or 2D platformer here, there are no game design books about making games like Miegakure, and there are many, many things that can go wrong. And to be honest, I hope for the best, but I don't expect much.
One big difficulty I see is that not only the designers have to understand 4D to make interesting worlds/puzzles, but the game has to teach that to players too, with the right difficulty curve.
