Have you thought about doing an upgrade recommendation similar to Raidbots? Where it imports the amount of crests and stones that the character has and then does upgrades based on that? I’d find that SO helpful!!!
The Valorstones/Crests search will do a ranking based on cost efficiency. Start by reading the description in this post:
I have not looked at how raidbots does the currency search, but that post should explain some of the reasoning behind our approach.
Something worth mentioning that is not in that post: how much of each currency you have right now is usually not that interesting which is why we don’t use it. Of more interest is how much of each currency you are willing to spend time getting – usually you will get some of each currency every week.
That is another reason why we arrived at a system that tries to equate your currencies to effort/time and use that as the “unit” for an efficiency ranking, rather than simply looking at how many of each you currently own.
This makes sense, but I will say the way Raidbots does it is also nice. It’s more of hey if you spent this combination of crests/stones that you currently have across the gear pieces you “own” this would be the best option for you right now. I could see that being hard to add in and that your system is probably better after reading it. But it is hard for me to know how much of the currency I’m willing to gain when I honestly don’t keep track. Atm on my main I’m crest locked but not stone locked. So that kinda throws a wrench in the amount I’m willing to earn thing.
Yeah it really is a tough problem for players (and for websites like ours) to answer.
Our first idea was in fact to do it exactly like you said: import how much of each currency a person has, then tell them here: do these 3 upgrades with it. We could certainly do that.
But after thinking about that for a bit as a player… I realized that’s not what I really want. I’m getting more currency all the time… so I want to think about it more in terms of how hard it is for me to get the currency, not how much I happen to have right at this moment.
You can certainly just make up numbers for those settings that seem reasonable – just getting crests/stones in the right ratio is the most important thing, e.g. I tend to get about 1000 stones for every 100 crests that I obtain. The “hours of play” setting only changes the magnitude of the result, not the relative ranking (e.g. whether you get 1000 stones and 100 crests in 1 hour or 10 hours would make the result say it is a 1% or a 0.1% upgrade per hour spent, but the relative ranking of two items would not change).
Another note on that initial type of ranking (given X currency what are the best N upgrades that I could do with it): the way an optimization problem like that tends to work is this: first you find the most efficient upgrades (similar to what we show you already – a ranked list from biggest gain per currency spent to smallest).
Then, you do a second optimization that maximizes your total gain given your limited amount of currency. This won’t always take e.g. the top 3 most efficient upgrades (though often it will). Sometimes because of the total cost of upgrades, it might skip one here or there (e.g. upgrading this one item 7 times costs more total and might not fit under your currency limit, so it is skipped for an item that only needs to be upgraded 3 times).
And this brings us back to why I don’t like this type of optimization: if I go and play the game for another couple days, I could just do that more expensive (but more efficient) upgrade. By limiting the optimization to a very specific total amount of currency it will tend to make choices that I probably don’t want.
As a player, simply seeing the two ranked lists (biggest overall upgrades and most efficient upgrades) I can plan ahead a little better.