(English is my second language. If you see any mistakes, please, don't hurt me too much.)
Hello. I have recently posted a rough sketch of this idea on some forum thread but it was left unnoticed. I decided to show it to people here in more detail to see if the idea is any good at all. Basically, it's about reducing the amount of "random" factor in BP/resource/mod farming:
Instead of adjusting the drop tables or making a trade/token/whatever system I suggest to implement pseudo-random distribution for item/resource drops. As far as I know, at least few competitive games utilize PRD when calculating anything related to chance and luck.
As of now, the game seems to drop stuff only based upon a static drop rate of a certain item. With 10% drop chance you'd think that a certain item will drop at least once per 10 runs. However, what 10% actually means is that the item will drop, in average, once per 10 runs on a scale of total amount of runs reaching infinity, and this is exactly what we see when people can't get a certain item/blueprint/mod to drop in over 50 runs when lucky ones get it in their first try. As in, streaks of success and streaks of failures can occur just as often as evenly distributed occurrences.
PRD, however, uses non-static percentage, as in, it compares numbers generated by the RNG to a value calculated as [base_chance]*[iterations]. Each time the item hasn't dropped, the chance for success increases. Oncce the desired item has dropped, the amount of iterations resets to 1, and the total chance, in our case, to 10%.
E.g.: Let's say, we want to get Mag Prime Systems Blueprint, which drops in T3 Survival with, as datamining shows, about 11.06% chance, let's round it up down to 10%. With true random distribution (which appears to be used now), every consecutive run will result with this BP awarded with 10% chance. Like I said before and as proven by numerous players across the board, it does not equal to 1 BP per 10 runs and the chance is actually significantly lower.
With PRD, chances will be as follows:
Run №/Drop chance
- 1 ---------- 10%
- 2 ---------- 20%
- 3 ---------- 30%
- 4 ---------- 40%
- 10 ---------- 100%
- 11 ---------- 90%
- 20 ---------- 10%
- 21 ---------- 20%
and so on
For each run not awarded with the BP, the next one will have a higher chance of desired outcome. On a large scale, it will drop at least once per ten runs, the drop won't occur several times in a row, and situations of absolutely no drop for more than tens of runs will either be very rare or cease to exist at all.
Now, you see that I listed chances for further runs past 10th, although it would have a 100% chance. This is for cases when the 10th run is done in co-op session, where chances will differ, but I'll get to that a bit later.
An alternative would be for the chance to stop at 100% until the item drops, thus resetting the counter. This might also used for boss drops, which are no more than 1 BP at a time out of 3 in total. With each BP drop chance counted individually and have reached 100%, it would be very likely that the three of them will drop for next three subsequent runs.
Now about co-op. As I mentioned before, chances in co-op may be different from individual ones, since every player has their own PRD iteration, yet they still must have the same chance of drop. I can think of three ways of implementing this:
1) take only host's iterations into account and calculate chances based on his current PRD iteration. 2) take the highest/lowest PRD iteration among the party. 3) take the average of all players' PRD iteration in the party.
The first two may drastically bias players towards either solo or online mode, while the third is sort of the equalizer.
So, on the one hand, one would prefer playing solo to have a solid chance of receiving the BP in 10 runs without PUGs entering your game and ruining your chances from 100% to 32,5%. On the other hand, playing online would be a better option as you may end up in a game with average chance higher than your own, so that you don't have to do ten runs in a row and still have some random element in the game.
Also, this system can be implemented for things like environmental hazards, faction swaps (some people say that those occur rarely), etc.
This is just a rough example and I think the PRD behaves a bit differently (lower base chances, different iterations, for example, the relation between number of consecutive attempts and chance is represented as a sine graph instead of a linear one, etc.)
Drop chance of a certain item increases after each failure to receive said item, eventually reaching 100% after N attempts, number of which depends on item's base drop chance. Counter resets to base chance after the item has dropped.
Pros and cons:
- +Steadier and more consistent chance of desired drops without ridiculously long streaks of failures
- +With proper implementation still maintains the "random" element, as such, prevents players from running out of content to acquire before new content is added
- +It might promote either co-op or solo modes with different chances for rewards
- -Without proper implementation players will run out of content way too fast.
- -It might promote either co-op or solo modes with different chances for rewards
- -Some may still prefer true random as this is not a competitive game
- -Probably hard to implement, may or may not require an entire overhaul of drop system. Dunno, not a dev.
Of course, I might completely misunderstand how the random works in this game and then this post is complete garbage.