The Crescent Rises Again
A series of posts on the return of Crescent, what happened, and what will happen next.
Part 4: Taking Inventory
A series of posts on the return of Crescent, what happened, and what will happen next.
Part 4: Taking Inventory
Hello hello, Crescent. Welcome back!
We're in the final steps of our pre-production work, and as you already saw, our open beta will be arriving this weekend. Along with the many improvements across the board, one of the main projects we had is to redesign and recreate the server's economy. Unfortunately, a combination of poor design choices and a bunch of unexpected bugs ended up wounding the economy up pretty badly. Truth to be told, we didn't evaluate things properly last time, but this was also a golden opportunity to learn how to do it better.
Now that we had plenty of time to work on this side of Crescent, we took the time to come up with a new game plan for our economy and inventory system - and this is what we will discuss in this final chapter of the relaunch series. Let's go!
How Not To Run A Market
The first step in our economy redesign, before writing any line of code, was to analyze and understand what went wrong last time. It's not too difficult to figure this one out, but here is what we came up with:
- Our drop rates and chances were way off. Not only by design, but also due to a code error that actually bumped them far higher than we intended. The result was an absolute overflow of valuable items that were meant to be very rare, immediately devaluing almost everything.
- Crates were poorly designed. The attempt to combine both weapons and cosmetics into a single crate turned out to be a poor idea, and the result was a bunch of bloated and meaningless crates that very rarely were even able to actually give you what you want.
- Scrap values were both low and inconsistent. Instead of properly scaling up with quality to make rare items actually valuable, we went with an insignificant increase for each quality tier. Add in the lack of proportions between these values and the crafting values, and you get a big mess.
- Items had no clear 'initial value' to anchor their value in the long run. How much is a legendary worth? What about a rare? How does one even know? With scrap values failing to represent the actual value of an item, this quickly became a mess.
- The lack of proper currency sinks made it possible for players to hoard lots and lots of it without a way to steadily get rid of it as well. This is in part due to the fact we never got around to finish our currency sink items last time.
Indeed, there were quite many pitfalls in our old design. To be brutally honest, most of it was plain wrong. But hey, getting almost everything wrong is great way to learn what not to do next time, right?
Now that we know what went wrong, it's time to go over what we did to fix those problems.
Items, Drops and Boxes
First, let's tackle the way players get items. The old design was a strange mishmash of various ideas that are each good on their own, but don't quite work together.
Items were obtained either through a drop, or from caches/crates (we call them boxes for simplicity). This, however, turned out to be unnecessary complex: we had to maintain two different chance systems (one for direct drops, and one for loot from boxes), and calculating their overall impact became more difficult. Then we had task crates, login reward crate, and a few other varieties, all working with different mechanics. Whew.
The first step, therefore, was to combine all these and eliminate the unnecessary duality. We got rid of direct item drops and of most of the non-standard boxes, leaving us with a single, controlled way of obtaining items: boxes. This allows us to only have to worry about a single 'income' source and balance it accordingly, and also prevent the confusion that our multiple methods made before. What this means in practice is that all rewards will also be provided in the form of a box, either a standard one or a special one.
With just one variable in the picture now, we redid the chances of each quality and item type. Very rare items (e.g. Legendary and Exotic) are now rarer and more valuable, while the lower tiers are more common.
Next, we had to tackle the boxes themselves. One of the main complaints we heard in old Crescent is on how weapons and cosmetics were mixed together, making it too difficult to actually get what you want. With soundbits just arriving last time, this immediately made boxes insanely cluttered with them, making the possibility of earning a high tier weapon or valuable item in general even lower. Not only that, but the boxes themselves were a bit of a mess, without a clear indication of where to get a specific item and what exactly is inside it.
Our next step, therefore, was to sort correct this mess. We started by separating the boxes into categories: weapon boxes, a.k.a. Caches; and cosmetic boxes, a.k.a. Cases. This way, we can separate the gameplay-related content (i.e. weapon, WepMods and such) from the cosmetics/customizations (i.e. Soundbits, player models, etc). This eliminates one of the biggest frustrations from old Crescent, and also allows us to be more flexible with how we create and add more content: it's now clear what goes where.
Next, we tackled the actual contents of each box. We took all of our existing content and organized it into equally-sized boxes, with an equal amount of each quality in each of them. In doing this, we also had an opportunity to re-organize some of our item's qualities, moving things between tiers to better fit their nature. Finally, we made some improvements to the descriptions of the boxes and the unboxing UI, including the handy and long requested "Open Another" button.
With our item obtaining methods and boxes finally sorted and organized, we ticked the first two tasks in the to-do list. Let's move on, then...
Next in line is our main currency: shards. Shards, as you remember, are obtained mainly through scrapping unwanted items (though they can also be awarded from various other places), with the item quality dictating the amount of scrap received. This means that a high quality item, say Exotic, is worth more shards than a low quality item, say Uncommon. This in turn meant that higher value items can be sacrificed for greater reward, making them a worthy trade-off.
...or so it was in theory. In reality, however, we made yet another poor decision when deciding the actual values of each tier. We ended up giving only a little more shards for higher quality items, an insignificant amount that made it really unworthy overall. This was done in an attempt to avoid generating too many shards by scrapped items, but our calculations were flawed. With box prices far higher than scrap values as well (more on this in a bit), scrapping items ended up barely giving anything to the player. Players had to scrap dozens of items for a single box, which like I mentioned above were already not that great. Whoops.
Another problem was that the bad scrap values made it impossible for them to become the 'base values' of items. Our goal with scrapping was not only to make them a source of income, but also to serve as a price guide for them. In other words: the scrap value of an item would represent its initial worth, and a base to start trading with. Unfortunately, the values failed to represent the actual worth and rarity of items due to how low they were, and the market as a whole was thrown into a lawless state where players made up prices without anything to guide them.
The solution here is obvious: to increase the scrap values, and properly scale them to fit the rarity and quality of an item. A legendary item, that most rare of its kin, shouldn't be just 5% more than its common counter part. It should be far, far more, especially now that we re-balanced the chances of each quality.
And so, we did just that. We gave a nice big boost to scrap values across the board, and then proportionally scaled the value of each quality accordingly. For example: Legendary weapons' scrap value saw a whoppin' 25,000% increase. Yes, actually this much. This number seems exaggerated or even ridiculous, but it just shows how off were our initial values. The rest of the qualities also got similar treatments, according to their place on the scale. Now, finally, scrapping is worth it.
The new values also make much more sense when it comes to pricing items. We did some number crunching and came up with a formula that gave us those values based on the quality chances I mentioned before. This way, we can be sure that the rarity is properly reflected in the item's value, and in turn will serve as a starting point for market pricing.
But that's not the only value we fixed.
Sinking Their Ships
There is one more problem left, one that also inherits from the previous ones: currency hoarding. Players constantly generate shards when they play CTTT, but how do they spend it? This is an important aspect of any economy that does not rely on a finite amount of currency. What it means is: if currency comes in, it must also go out. Otherwise, you'll find yourself dealing with inflation problems - and this is a serious headache.
We have multiple currency sinks planned for Crescent, most of them in the form of one-time consumable items that offer various benefits. Those of you who follow our dev showcase channel in Discord probably remember the sneak peek into a new item: Fun Round Token. This is a good example of what a currency sink is like: something that takes out/destroys currency, thus acting as a counter-force against the generation of additional currency. Yet despite this design, we had a simple problem with it in old Crescent: we never got to finish those sinks. The time constraints and everything else that plagued old Crescent did not pass on this; we simply never made it to implement most of those sinks. This was technically somehow countered by the low scrap values, but that's not how it's suppose to work, so...
Anyway, now that new Crescent is coming, we made sure to implement those new items and ideas we had and have them ready from the get go. And indeed, there are no less than 5 of them, each giving its own special benefit. I will leave the discovery of what those actually are to you. ;-)
As we were working on these sinks, an idea came to our minds: maybe we can hit two birds with one stone here. We need a way to encourage players to spend more currency, but we also need a way for new players to kick-start their inventory, which is yet another complaint we received a lot last time, on the steep starting curve our economy forced. So why not do both?
Remember when I said our scrapping prices were low? Why was that, actually? What was the scale we compared them to? You guessed it: boxes. The main use of shards, after all, is to craft additional boxes in order to gain more loot. It's a bit like recycling: get rid of the items you don't need anymore by turning them into a different set of items, and along the way they also serve as wonderful currency sinks. Everybody win here! And indeed, that was the perfect plan...
...if players could afford those boxes. But this is where our final flawed decision stood: box prices were completely unproportional to the income from scrapping and other sources. One had to get rid of dozens of items, sometimes of high quality as well, just to afford a single box - which is worth a single item. I think you can see where this is going.
It only made sense, therefore, to reduce this price to something more proportional to the value of items. And indeed, along with our scrapping value changes, we also gave a new, more sane price to our boxes. Sounds good!
But now we face a different problem: boxes being affordable are great for new players who are just starting out, but they become a big problem in the hands of already established players with a lot of currency and valuables. To those players, this is a cheap and easy way to hoard large amounts of currency with very little risk - exactly the problem we are trying to combat. But if we make them more expensive, newer and less established players will suffer.
And so, we came up with an idea: scaling prices. The gist of it is simple: boxes start at a low, affordable cost, but with each purchase, their price increases by a little bit. This means that initially, players can get their first few items very quickly - but the more they buy, the less they get in return. This is what's known as the "Diminishing returns" concept, in which the reward for a certain amount of offer declines every time.
In the case of our economy, this is a brilliant solution: new players can quickly earn new items to start out their journey, but powerful players cannot abuse this by hoarding the cheap loot in big piles, as the more they spend, the more expensive it becomes. Two birds with one stone.
Of course, to make sure this doesn't go completely out of proportions and that prices are still fair, we placed numerous restrictions on this scaling system. First, there is an upper cap for the price, to ensure prices don't sky rocket; and second, this mechanic will reset once a week, resetting all box prices to their original value, to avoid getting people locked in a limbo where they are forced to spend a lot of shards just to get something.
Lastly, there was another offender in terms of currency control: the lottery. And boy, was it an ugly one. Despite not being our intention, a bunch of players quickly figured they can utilize the lottery to generate huge amounts of shards with very small investments. We realized this eventually and took measures to stop it - but it was already too late at that point.
Obviously the lottery needs a fundamental change to properly work. This is something we feel like is only going to make things more difficult early on, and so we decided to disable it completely until a later date, when we're wiser and better prepared to give it another shot. Therefore, the lottery will not be included in our relaunch of CTTT, and will only come at a later date (as part of a bigger system, if all goes to plan).
A Joint Effort
This is it for the big changes. With all this work, we were able to negate and patch a lot of the problems we faced in our old economy design, while still keeping the good and enjoyable bits of it. No doubt you guys will immediately feel this big improvement and find our economy a lot more stable and well thought than last time.
However, if there's one thing we learned, it's the importance of feedback. And when it comes to a sensitive topic such as the economy, we could use it more than ever. This is I encourage you all to tell me what you think about these changes, and offer any ideas or improvements you might have. These are some drastic changes after all, and they will have a great affect on Crescent's economy. We did our research, but we will always appreciate more help. This is a joint effort, and you can help us do it right.
And that's it! Sorry for the lack of pretty pictures this time - I am actually not at my PC as I type this, so the best I can provide is some good old text this time. You'll have to excuse me!
This is it for our relaunch series. I got to cover most of what I wanted, thankfully, and I hope this answered some of your questions, or just gave you an interesting insight into how the development of Crescent works behind the scenes. As always, the team and I are open and welcoming to any question and thoughts you have to share, especially on a topic such as this (see above).
Don't forget: we will be running our special relaunch live stream later this week, in which we will go over some things I didn't cover yet in this series (yes, there are such things) and also answer your questions directly. Keep an eye on the forums and Discord for more information.
And, more importantly: I hope to see you at the open beta this weekend! As promised, more details and exact times will be posted ASAP. Stay tuned for this as well!
As always, thank you very much for being here, for reading, and for supporting Crescent to this day, through the various ups and downs. It is always a pleasure to hear kind and encouraging words, and I appreciate it a lot. I hope all of us will be able to enjoy Crescent for a long time, both by playing it and creating it. If you have something you want to say to me, my DMs are always open on Discord - go ahead!