The uncertainty principle is NOT tied to our abilities to observe.
The uncertainty principle is
\sigma_x \sigma_p \geq \hbar / 2.
What this says is if you have an bunch of identically prepared states of a particle, and you go through and measure the momentum of half of them, and the position of the other half, you will get a distribution for both quantities. Even if your measuring device is perfect, you will still get distributions -- granted, you will know the distribution very well, but you won't change it's shape.
\sigma_x is the standard deviation of the distribution in position and \sigma_p is the standard deviation of the distribution in momentum. Thus, the uncertainty principle implies that both distributions cannot be arbitrarily sharply peaked -- if one is sharply peaked, the other must be broad. Of course it is possible that both are broad because it is an inequality rather than an equality.
This much is experimentally verifiable, and hence doesn't depend on the interpretation of quantum mechanics. The orthodox interpretation of quantum mechanics, in fact, asserts that a particle does not even have definite position and momentum. The Bohmian interpretation of quantum mechanics, on the other hand, DOES assert that particles have definite position and momentum. Both interpretations give rise to the same predictions about the outcomes of experiments. Therefore, whether or not a particle has a definite position and momentum that "god" could measure is not an experimental question.
Finally, it is worth noting that the uncertainty relation is not a postulate of quantum mechanics -- it is derived from the postulates of quantum mechanics. Hence, it's not really coherent to ask "What if we could change the uncertainty relation?" because to change the uncertainty relation, you'd have to throw out quantum mechanics.