There’s some crossover point in network size where an attack becomes unfeasible.
A larger network means a) more attacking vaults are required so the cost is higher and b) less safecoin can be stolen per section so the benefit is less.
The exact crossover point depends on exchange rates and costs and average vaults per section etc, but the result is that a larger network is generally more secure because it’s less desirable to attack. It’d be cool to have a tool that could pull in all this live data and report what a secure size of network would be.
As an example, using a safecoin price (SP) of $0.50 USD and a lifetime vault cost (VC) of $1.00, a secure network size would be at least 464K vaults.
VPS = vaults per section = 50 (assumed)
NS = network size (we want to calculate this)
Q = Quorum = 0.5
TC = total safecoin = 2^32
TS = total sections = NS / VPS
SCPS = safecoin per section = TC / TS
TAV = total attacking vaults required = NS * Q
B = benefit of attacking = SCPS * SP
C = cost of attacking = TAV * VC
The section is on the fringe of secure when B = C
Substituting in
SCPS * SP = TAV * VC
(TC / TS) * SP = (NS * Q) * VC
(TC / (NS / VPS)) * SP = (NS * Q) * VC
Solving for NS
NS = ((TC * VPS * SP) / (Q * VC))^0.5
Using the assumed values
NS = ((2^32*50*0.5)/(0.5*1))^0.5
= 463,410 vaults as a minimum to make an attack unviable.
There are some intuitive results that come out of this:
Increasing exchange rate but same cost means the attack is more beneficial, so there must be a larger network to make the attack more expensive. Doubling only the exchange rate results in a network of 655K vaults being the new minimum secure size.
Increasing vault cost but same exchange rate means the attack is less beneficial, so there can be a smaller network. Doubling only the vault cost results in a network of 328K vaults being the new minimum secure size.
It’s a little scary to think the exchange rate can be used as a dial for security, and that higher exchange rate means a larger network is required, but the results are pretty fascinating.
Some rambling observations…
Ageing is a security feature because it increases lifetime vault cost.
Datacentres have a lower lifetime vault cost than a small scale farmer due to economies of scale.
Home users have a theoretically zero lifetime vault cost since it’s included as part of their other internet / computing costs which they’d pay even if they weren’t farming.
Adding the idea of a distribution of various vault costs across the network makes this calculation a bit more challenging.
Total attacking vaults required may be less than NS * Q, especially for small networks.
This is a pretty complex thing to model accurately, and the formula derived above is the most basic starting point. It’d be cool to dive a bit deeper but it escalates in complexity fairly rapidly.