It depends. A proven conjecture is IMO better than an unproven one. But usually, a new proof would shed new lights or build bridges between concepts that were

It depends. A proven conjecture is IMO better than an unproven one. But usually, a new proof would shed new lights or build bridges between concepts that were before unrelated. And in that sense, a proof not understandable by humans is disappointing, because it doesn't really fullfil the need to understand the reason behind why it's true.