This means that number 1 wallet would be able to run ~68 service nodes ( without worry about being dropped by the spread. Interesting stuff.
68 full service/bitcoin nodes. That would be a serious commitment to bandwidth bills
Interesting, what is the calculation that you made to come up with 68?
I was going by mental arithmetic so I'll show you my rough working from memory. I apologise for the mental back of fag packet calculations!
Number 1 wallet has ~200k SPR
Current coin supply is ~2,500,000
Max number of snodes (coin supply/2880)= 868
Based on the false premise that every coin will be used in a servicenode (I have no data to guess how many coins will be in snodes):
Max coin supply / Max snodes = 2880SPR
Meaning that if I was the owner of #1 wallet and I wanted to guarantee that all of my service nodes were in the spread I'd load each node with 2800SPR which means that I'd be able to afford to fund roughly 68 snodes in this way.
EDIT: Correcting price error
A quick check on AWS estimates that it would cost $3000pm to run the following:
68 x T2.micro
68 x 60GB magnetic volumes
13,600GB outbound traffic
1360GB inbound traffic (free anyway)
This means the #1 wallet holder would need to see about 130,000SPR per month (at current price) to break even)
goodness gracious, that's certainly cost prohibitive. I was thinking a remote server in the range of $ 5 month/node would be borderline feasible but would need the price of spr to reach beyond .000205 to be profitable.
Since the primary objective is decentralization, why not tie cpu mining to spr's first reward service? It would strengthen the network, be something anyone could jump into with little cost, wouldn't be too difficult to implement, and provide preliminary insight into how the competition dynamics will play out.