Suggestion: Optimal Crest Usage

Sorry if this has been suggested before, I had a look and couldn’t see anything.

The Cost Efficiency Options added to the Upgrade Finder: Valorstones/Crest section is neat but I’ve often found myself wanting to know the most efficient way to spend the crests that I have right now, not ones that I might farm later. If I have 45 Runed Crests would it be better to upgrade my 610 item to 613 and a 616 item to 619 and another 616 to 619, or is it better to do the same 610 item to 619, or etc…

The Max Upgrade Level doesn’t really account for this decision, as if I set it to, say, 623 it assumes that I have the crests available for that upgrade on all items, when I don’t. I might have enough Gilded to upgrade 1 item to 623, but be short on Runed to get some others there right now.

I can accomplish the comparison myself by calculating my Max Upgrade Level for each item, setting the option, then Finding Upgrades, and repeat for each item, and then compare the amount of upgrade, but it would be really nice if I could enter in that I have 150 Weathered Crests, 200 Carved Crests, 90 Runed Crests, 30 Gilded Crests, and then have Mr Robot tell me which items to upgrade to which level to get the best value out of those crests right now.

We’re giving some thought to some changes and improvements to the cost efficiency feature.

People have asked for something similar to what you suggest, so we might add something like that. I think we’ll make it a bit more flexible though… and let people enter in any amount of currency so that they can plan ahead.

Given the relatively high cost of obtaining upgrade currencies, it’s usually a good idea to plan ahead at least a little bit. It also gives people a way to get a ranking even if they don’t have any currencies on them at the moment.

That said, as a player I still have concerns with this kind of ranking. An example helps to illustrate:

Say that I have 5 items that can be upgraded. They are all rank 1 right now, and they can all be upgraded to rank 8. Most upgrades switch from one currency to another going from rank 4 to 5, so you really have to break each item into two different upgrades:

1->4 (costs crest type 1 and valorstone amount X)
4->8 (costs crest type 2 and valorstone amount Y)

Thus you have 10 potential “upgrades”, 2 for each item.

Note that “partial” upgrades, e.g. only taking an item from 1->2 instead of all the way to 4, are going to be of essentially the same efficiency as the “full” upgrade. Thus if upgrading item A from 1->2 is your best option right now, your next best option will almost always be taking that same item from 2->3, then 3->4. In other words, when possible, you want to “finish” one item (to the point where the cost changes, at which point its efficiency changes and you re-evaluate).

Say that my most efficient upgrade is taking item A from 1->4. I have enough currency to do that, so I do it. Then my next best upgrade is taking item B from 1->4. I do that too. Then my next best upgrade is taking item A from 4->8, so I do that.

Now I have just a little bit of currency left. I don’t have enough to do my next-best upgrade, which is taking item B from 4->8. I don’t even have enough to take it from 4->5. But I do have enough to do a partial upgrade of my next best upgrade: I can take item C from 1->2. So I do that, and now I’m out of currency.

I would argue that last partial upgrade, taking item C from 1->2, is a waste. I would be better off saving that and using it next week (or whenever I get more currency) to continue upgrading item B instead.

Hopefully that example illustrates why I was hesitant to do this kind of ranking. By putting a hard cap on the amount of currency, you end up spreading your upgrades around to use up as much of the currency as possible… when often you would be better off saving it up to do higher efficiency upgrades next week or whenever you get new gear or more currency.