Correct, but he's not trying to do that the way previous INT 27 casters would have in earlier centuries, by finding current magical anomalies and poking at them until he figures out what's going on.

He's building a thing that can simulate possible things that could be going on, so that he can, for example, feed in 30 different possibilities, and see if any of them reproduce the behavior of reality.

He's doing that by building a fully general simulator toolset, which he can then adapt to simulate possible laws of magic as a special case.  In principle, he could calculate out the possibilities himself by hand, given infinite time, but he doesn't have infinite time, so he's making a thing that does the calculations for him.

To do that, he needs pieces that interact in precise ways, so that they can model precise hypotheses; and they need to go on interacting in that precise way even when there's a lot of them, which is the hard part with magic.

Ideally, he needs small parts like that which can be spun off by other stable magical structures, so that he can spin off a hundred thousand of them that will last for ten minutes apiece in a regular pattern whose resulting forces will allow it to be stable.

Correct, but he's not trying to do that the way previous INT 27 casters would have in earlier centuries, by finding current magical anomalies and poking at them until he figures out what's going on.

He's building a thing that can simulate possible things that could be going on, so that he can, for example, feed in 30 different possibilities, and see if any of them reproduce the behavior of reality.  Ideally, he will be able to search for them automatically, in small spaces.

He's doing that by building a fully general simulator toolset, which he can then adapt to simulate possible laws of magic as a special case.  In principle, he could calculate out the possibilities himself by hand, given infinite time, but he doesn't have infinite time, so he's making a thing that does the calculations for him.

To do that, he needs pieces that interact in precise ways, so that they can model precise hypotheses; and they need to go on interacting in that precise way even when there's a lot of them, which is the hard part with magic.

Ideally, he needs small parts like that which can be spun off by other stable magical structures, so that he can spin off a hundred thousand of them that will last for ten minutes apiece in a regular pattern whose resulting forces will allow it to be stable.

Correct, but he's not trying to do that the way previous INT 27 casters would have in earlier centuries, by finding current magical anomalies and poking at them until he figures out what's going on.

He's building a thing that can simulate possible things that could be going on, so that he can, for example, feed in 30 different possibilities, and see if any of them reproduce the behavior of reality, and think analytically about how they fell short

He's doing that by building a fully general simulator toolset, which he can then adapt to simulate possible laws of magic as a special case.  In principle, he could calculate out the possibilities himself by hand, given infinite time, but he doesn't have infinite time, so he's making a thing that does the calculations for him.

To do that, he needs pieces that interact in precise ways, so that they can model precise hypotheses; and they need to go on interacting in that precise way even when there's a lot of them, which is the hard part with magic.

Ideally, he needs small parts like that which can be spun off by other stable magical structures, so that he can spin off a hundred thousand of them that will last for ten minutes apiece in a regular pattern whose resulting forces will allow it to be stable.

Correct, but he's not trying to do that the way previous INT 27 casters would have in earlier centuries, by finding current magical anomalies and poking at them until he figures out what's going on.

He's building a thing that can simulate possible things that could be going on, so that he can, for example, feed in 30 different possibilities, and see if any of them reproduce the behavior of reality, and think analytically about how they fell short

He's doing that by building a general toolset for simulating continuous processes, which he can then adapt to simulate possible laws of magic as a special case.  In principle, he could calculate out the possibilities himself by hand, given infinite time, but he doesn't have infinite time, so he's making a thing that does the calculations for him.

To do that, he needs pieces that interact in precise ways, so that they can model precise hypotheses; and they need to go on interacting in that precise way even when there's a lot of them, which is the hard part with magic.

Ideally, he needs small parts like that which can be spun off by other stable magical structures, so that he can spin off a hundred thousand of them that will last for ten minutes apiece in a regular pattern whose resulting forces will allow it to be stable.

Correct, but he's not trying to do that the way previous INT 27 casters would have in earlier centuries, by finding current magical anomalies and poking at them until he figures out what's going on.

He's building a thing that can simulate possible things that could be going on, so that he can, for example, feed in 30 different possibilities, and see if any of them reproduce the behavior of reality, and think analytically about how they fell short

He's doing that by building a general toolset for simulating continuous processes, which he can then adapt to simulate possible laws of magic as a special case.  In principle, he could calculate out the possibilities himself by hand, given infinite time, but he doesn't have infinite time, so he's making a thing that does the calculations for him.

To do that, he needs pieces that interact in precise ways, so that they can model precise hypotheses; and they need to go on interacting in that precise way even when there's a lot of them, which is the hard part with magic.

Ideally, he needs small regular tiling parts which can be spun off by other stable magical structures, so that he can spin off a hundred thousand of them that will last for ten minutes apiece in a regular pattern whose resulting internal forces will allow it to be stable.

Correct, but he's not trying to do that the way previous INT 27 casters would have in earlier centuries, by finding current magical anomalies and poking at them until he figures out what's going on.

He's building a thing that can simulate possible things that could be going on, so that he can, for example, feed in 30 different possibilities, and see if any of them reproduce the behavior of reality, and think analytically about how they fell short.

He's doing that by building a general toolset for simulating continuous processes, which he can then adapt to simulate possible laws of magic as a special case.  In principle, he could calculate out the possibilities himself by hand, given infinite time, but he doesn't have infinite time, so he's making a thing that does the calculations for him.

To do that, he needs pieces that interact in precise ways, so that they can model precise hypotheses; and they need to go on interacting in that precise way even when there's a lot of them, which is the hard part with magic.

Ideally, he needs small regular tiling parts which can be spun off by other stable magical structures, so that he can spin off a hundred thousand of them that will last for ten minutes apiece in a regular pattern whose resulting internal forces will allow it to be stable.