Regarding teleportation and pruning - got it. Makes more sense to have complete destruction there for a clean start.

Thinking out loud:

What if, after a disaster block, the Fee for a hunter is very expensive. 100 HUCs (random number), this tapers down after X blocks (~~10k?~~ 3k?) to 1 HUC.

Obviously wouldn't solve the problem, but maybe can~~ limit the competition~~ make harder for people to create thousands of bots after disaster block

Also, these hunters would have to drop 100 HUCs if they are killed - I'm not sure if the code automatically relates price of team with general death, or if they are both just set as 1... so not sure how much work it is to do.

edit: may have negative effect on new players if it's 100 huc

Here's the problem distilled into a single word:

Distribution.

I believe that what we're fumbling around with is, "How can the coins get distributed as broadly as possible?"

Higher distribution is the key. It leads to the network effect and exponential growth. (What's the technical term for that? There's an equation for it - something like N^2.)

If it's 100 coins per hunter, and there are X coins on the map, I might think to myself, "Heck, I'll just write a bot to wait for a disaster, then spawn hunters to go collect. What is the optimal number of hunters if there are X coins?"

But 100 coins is likely to put a damper on new or small players.

But, I think the idea has merit. Perhaps something like a curve that curves over time back to 1 HUC per team?

For example... (these are just numbers I'm pulling out of my butt for an example)

Team #1 costs 1 HUC.

Team #2 costs 2 HUC.

Team #3 costs 3 HUC.

Team #4 costs 4 HUC.

Team #5 costs 5 HUC.

Team #6 costs 10 HUC.

Team #7 costs 20 HUC.

Team #8 costs 30 HUC.

Team #9 costs 40 HUC.

Team #10 costs 50 HUC.

Team #11 costs 100 HUC.

Team #12 costs 200 HUC.

Team #13 costs 300 HUC.

Team #14 costs 400 HUC.

Team #15 costs 500 HUC.

Team #16 costs 1000 HUC.

Team #17 costs 2000 HUC.

...

Team #21 costs 10,000 HUC.

Team #22 costs 11,000 HUC.

...

etc.

Then for each block over the next 4 hours (or some time - maybe 12 hours or 2 hours?) the costs shrink proportionately (along some curve or linearly) down to 1 HUC per team.

That will not address players with multiple computers or VMs, but at some point they really need a LOT of computers/VMs.

It is unlikely that many players will have many computers to "get around" those costs. And with disasters being infrequent, it's not going to be cost effective to buy more hardware. (But renting cloud computing might be practical.)

The question then is about the time advantage. Hunters can only go up to the mid-way point where the entrance to the middle of the map is before they must turn back and be guaranteed of cashing in their coins. So, that's a major consideration in doing the math.

The aim should be to prevent botters from creating thousands of bots and to give human players enough time to "get in on the action" so that coins are more widely distributed (so that there is a larger network effect).

A possible twist to that scenario is for a disaster to take place in 2 parts:

1) Disaster "pending" notification - it will happen in X blocks.

2) Until X blocks have passed and the disaster takes place, all coins dropped go back to a coin spawning area.

That would give everyone notice of the disaster, which would benefit humans most of all as they can't pay attention 100% of the time where a bot can. The X blocks there give the human players a "grace" period that is (almost) useless to the bots.