i wonder what a linkup of all available caster types could do?
It is unclear if it is even possible. The linear proposal was something like
A - T - B - T - C
The question is if the 2 thinkamancers will be able to cooperate. Also, it means that A and C aren't directly connected to each other, the link has to pass through T-B-T.
If each mind has to be linked with each other mind, then the difficulty increases very fast as you add more casters.
With 2 casters you just need 1 link
A - T
With 3 casters it is harder, you need 3 links to connect all minds
. / \
The thinkamancer connects to both A and B, but also has to establish the A to B direct link.
With 4 casters, you need 6 links
| X |
The Thinkamancer connects to each of the 3 directly and then also needs to establish the A <-> B , A <-> C and B <-> C links.
This would require N*(N-1)/2 links, if N was the number of casters.
N=2: 1 link
N=3: 3 links
N=4: 6 links
N=5: 10 links
N=6: 15 links
N=7: 21 links
This could explain why 3+ casters in a link is not possible.
The linear version might just not allow a single pseudo-mind to function.
If it is assumed that each Thinkamancer can handle 3 links (based on N=3), then you might still be able to handle large linkups with more thinkamancers.
However, there are still problems.
N=4 would require 2 thinkamancers, so you end up with 2 thinkamancers and 2 other casters. This still only gives you a thinkamancy and 2 other types of caster.
| X |
This could still be a benefit, as the added thinkamancer could add raw power, even if not adding more abilities.
N=5 would require 3.33 Thinkamancers (so 4). This gets you 4 thinkamancers and 1 alternative caster.
N=6 would require 5 thinkamancers, so 5 thinkmancers and 1 alternative.
N=7 would require 7 thinkamancers, so no alternative casters would be possible.
The N=7 link could potentially grant massive thinkamancy abilities though.
N=8 would require 9.3 thinkamancers, so wouldn't be possible.