A little late, but not selling a self-produced resource which you need or want to consume is not the same has having purchased it for consumption. This is poor economic analysis as well as poor physical analysis. It's not even using "opportunity cost" correctly, to be honest.
You can't really refer to the value of the materials to build a relic as an opportunity cost if the desired goal is to build a relic because the alternative (not building a relic) is mutually exclusive with our desired outcome. We can talk about that value as the opportunity cost of building the relic (we could have bought other things, we could have not spent the time making the money to buy the relic, etc.)
But since we
are making a relic, the assumption is that we need 17700 currency (or whatever the cost is) and we have to procure that somehow. Assuming liquid capital is available, and simplifying the system to the two extremes "produce it all" or "purchase it all" - although in most cases both processes will occur - we can now talk about opportunity costs in terms of time or the risk incurred by reducing available capital, among other potential "costs."
The example where purchasing the relic now increases our efficiency at gaining capital is a good example of opportunity cost used correctly - as the difference between pre-relic farming and post-relic farming - is a cost we incur if we instead choose self-production at a slower rate. The nice thing about this sort of cost is that it is easily quantified and thus can be analyzed objectively. Another example of opportunity cost used correctly was the sacrifice of potential purchases (of course this risk diminishes to zero as the ratio of required target capital for a particular outcome (in this case, building a relic) to our initial available capital increases but for most players this risk is not diminished enough to ignore).
When we start to talk about opportunity cost in terms of time we have to consult leisure-consumption models and consider marginal rates of substitution, and our analysis stops being objective and becomes subjective. How much can we consider dynamis farming "work" versus "leisure?" If the means of production are enjoyable enough, self-produced consumables will have a "cost" less than their market value if purchased with already accrued capital (one might then argue that the the value of selling is greater because of this, but this is irrelevant because we assumed we need to consume some amount of currency and selling it therefore does not accomplish our goal except in the case where we expect a massive downward market shift).
This mistake could have been avoided by paying careful attention to the first sentence of the Wikipedia article for opportunity cost: "Opportunity cost is the cost of any activity measured in terms of the value of the
next best alternative forgone (that is not chosen)." If the assumption is that we are building a relic by some means, then we cannot consider the market value of self-produced currency to be an opportunity cost because we are either going to self-produce or we are going to purchase. The correct comparison is the marginal rate of substitution between time/leisure (if any) sacrificed to produce the currency and the current market value - along with any other costs or risks associated with either self-production or capital reduction as previously mentioned.
http://sighroll.com/wp-content/uploa...oreyouknow.jpg