Again, you are mistaken. I assumed that you could execute any finite number of instructions in an instant. Computing Solomonoff probabilities requires executing an infinite number of instructions, since it implies assigning probabilities to all possible hypotheses that result in the appearances.
In other words, if you assume the ability to execute an infinite number of instructions (as opposed to simply the instantaneous execution of any finite number), you will indeed be able to "compute" the incomputable. But you will also be able to solve the halting problem, by running a program for an infinite number of steps and checking whether it halts during that process or not. As you said earlier, this is not what is typically meant by computable.
(If that is not clear enough for you, consider the fact that a Turing machine is allowed an infinite amount of "memory" by definition, and the amount of time it takes to execute a program is no part of the formalism. So "computable" and "incomputable" in standard terminology do indeed apply to computers with infinite resources in the sense that I specified.)
Solomonoff induction is not in fact infinite due to the Occam prior, because a minimax branch pruning algorithm eventually trims high-complexity possibilities.
If it's worth saying, but not worth its own post, then it goes here.
Notes for future OT posters:
1. Please add the 'open_thread' tag.
2. Check if there is an active Open Thread before posting a new one. (Immediately before; refresh the list-of-threads page before posting.)
3. Open Threads should start on Monday, and end on Sunday.
4. Unflag the two options "Notify me of new top level comments on this article" and "