Forgive me for not going into the math on this. I just dont feel like it. As such I have to say my opinion is theoretical at this point.
Im going to throw this out here by example. Keep in mind the designs aren't optimal. So for sake of argument, there may come a certain point in eu/t vs efficiency that its more productive to increase our chamber size. But for the resources used and the optimal output im going to argue that multiple single reactors is the way to go. The designs I will be throwing out are going to be pretty crap mark 2's.
Take this low efficiency single chamber design for instance:
now take this 6 chamber design for instance:
Its an exact duplicate of the first, except this one cant keep up with the heat at all without filling the other 3 chambers. We could fill those chambers, and make it a little more stable:
For 9 extra coolant cells and 3 extra hd's.
At this point i would like to point out that I understand that the design of these reactors, nor the cooling is optimal. MUCH LESS the efficiency of the uranium.
lets drop a chamber:
add a few coolants, its still relatively stable.
instead of 3 chambers, we can save a chamber and use the 2 others plus some easy resources to make a whole new reactor, encase it in water, and end up with a reactor that still doubles our eu/t and has FAR less heat.
Here's the issue, the 33 external cooling of the water is only good for one reactor, so we still end up with a larger cooldown time as we add chambers and uranium. The runtime drops, the cycles drop, and there's no good reason for it as far as I can see.
There are some really awesome designs out there right now, but uran efficiency isn't really that big of a deal, especially with a separate breeder. Whats the point of chambers vs single reactors?