RNG problems and idea to solve them

[Zilotz]

First things first - sometimes RNG gets beyond annoying, that is problem of any RNG in games in general.

But there is a solution to this problem, i will explain it as much as possible.

My idea is to add a counter for having guaranteed drop chance.

I think of it like this - player can buy a "device" that will have 5 slots initially(yeah, i know it will increase DB load like rivens did so it will cost 1000-2000plat for device and 100-200plat for every additional slot).

Every slot can track only one thing at any given time.

For example  - i need aura forma, eidolon lens, cetus wisp, arcane energize, arcane grace.

I put those chances into 'device' and it tracks the progress.

Initial data in device:

Slot1 - Aura Forma 0/67 (100/(1.5 which is lowest chance in drop tables for this item))

Slot2 - Eidolon Lens 0/21 (100/(4.73 which is lowest chance in drop tables for this item))

Slot3 - Cetus Wisp 0/10 (100/(9.39 which is lowest chance in drop tables for this item))

Slot4 - Arcane Energize 0/20 (100/5)

Slot5 - Arcane Grace 0/20 (100/5)

Every time i have chance to roll item in slot it adds 1 to the counter in that slot, once counter is filled i get the item and counter resets to 0.

You cannot put same item in other slots(like it is for mods), every time you change item in slot it will reset counter for the slot.

It is not pay to win, it is a way to pay for making results predictable.

For example if i want to get Legendary Core i can do it but it will require me to do (100/0.18)=556 sorties while keeping one of slots occupied with this task.

Yes, it can be used as resource booster but only for selected resources in slots so it will not be cost-effective for most of cases.

 

[Teridax68]

I very much agree with the core proposal: Warframe is in dire need of a pity mechanic for its RNG, and I feel there is strictly no reason to keep not having one when pity systems have now been a tried and true gameplay feature in RNG-based games for years, to their significant benefit. I do disagree with some of the specifics, though, as I think the system should be baseline, to begin with, but if we're going to make it an item, it should be obtainable without plat.

[Zilotz]

2 minutes ago, Teridax68 said:

it should be obtainable without plat.

But on the other side it increases server cost so it should be kind of investment to avoid DE lose profits if they implement it.

Sure i would like to get it for free as base mechanics but i understand that there are two sides - players and DE so it should benefit both sides.

[Teridax68]

3 minutes ago, Zilotz said:

But on the other side it increases server cost so it should be kind of investment to avoid DE lose profits if they implement it.

Increases server costs by how much? From what you're saying, you're tracking progress on five different, preexisting items -- a tiny server cost, particularly when considering all the items the game already keeps track of.

[Tsukinoki]

Not a bad idea behind adding something to serve as an RNG valve but a few questions:
-What happens if you get the drop "early" (e.g. it randomly drops say the 3rd time you run the mission).  Does this further add one to the counter (in which case it becomes a bit P2W because now you are guaranteed 1+ drops every x missions...especially for the high plat items that you can turn around quickly for a profit) or does it reset the counter to zero?

-What happens if you say put 2 corrupted mods into it (all with the same drop chance mind you)?  They count up at the same rate from the same missions...so would that mean that once you hit the limit you get multiple rewards from that mission?  Or does it pick one, give it to you, and then the next mission gives you the other one?

-How does this work with shared rewards?  E.G. everyone gets the same corrupted mod from the vault.  What if you were after say Transient Fortitude and someone else was after Blind Rage, and both of you had a full counter?  What would happen?  Who would get their reward?
This gets really messy for pug bounty missions, and doubly so for sorties if people have competing drop targets.

-Is this in addition to the drop from the mission?  Or does it replace the drop from the mission when the counter is full?

There needs to be a lot thought about...and probably quite a few adjustments in ow many runs it takes for some of the items.

Edited May 8, 2019 by Tsukinoki

[Zilotz]

@Tsukinoki Thanks for reading and asking!

16 minutes ago, Tsukinoki said:

-What happens if you get the drop "early" (e.g. it randomly drops say the 3rd time you run the mission).  Does this further add one to the counter (in which case it becomes a bit P2W because now you are guaranteed 1+ drops every x missions...especially for the high plat items that you can turn around quickly for a profit) or does it reset the counter to zero?

I would like it to keep counter.

Yeah, it is a bit p2w but will still require effort like boosters do.

18 minutes ago, Tsukinoki said:

-What happens if you say put 2 corrupted mods into it (all with the same drop chance mind you)?  They count up at the same rate from the same missions...so would that mean that once you hit the limit you get multiple rewards from that mission?  Or does it pick one, give it to you, and then the next mission gives you the other one?

That's a good question, didn't think of it at start.

For corrupted mods it can be solved easily enough(by making them incompatible with each other like Umbral Vitality and Vitality do for warfame slots).

21 minutes ago, Tsukinoki said:

-How does this work with shared rewards?

It does not. As every device is individual it will only work individually by counting chances. 

And still everyone gets normal drop like it is now, device can add to owner additional thing(s) when counter(s) is(are) full.

It is a way to individually reduce grind by investing plat AND still doing missions.

 

 

 

[Zilotz]

2 hours ago, Teridax68 said:

Increases server costs by how much? From what you're saying, you're tracking progress on five different, preexisting items -- a tiny server cost, particularly when considering all the items the game already keeps track of.

I cannot tell, all i know is that DE mentioned server cost increase from rivens. I don't know how much but it was important enough to mention.

[Marine027]

vor 2 Stunden schrieb Teridax68:

Increases server costs by how much? From what you're saying, you're tracking progress on five different, preexisting items -- a tiny server cost, particularly when considering all the items the game already keeps track of.

Rivens are a big problem simply due they not have consistent stats, each existing riven is a uniqunes in itself, that is simply way different then having a copy of lets say vitality where its stats nevere change and will always rank the same up.

Riven simply take alot of server space due the stats and descriptions onto them, same for the challanges to unlock and on top able to reroll them at any time.

Same way we have limited Loadout slots, because each Loadout is individual. And havign the ability to make your own loadout, simply needs alot of server space, alot of data to transfer to the servers and to you as player, onto the trade chat, into missions, etc.

Warframe is no MMO with several servers or region servers simply to store things like housing etc.

[HexOmega111x]

Il y a 3 heures, Zilotz a dit :

Yeah, it is a bit p2w but will still require effort like boosters do.

Then it is a bad idea. The mechanic you propose is cool but any little bit of p2w element should be avoided. Because if you start to legitimate any p2w elements, there will soon be no limit to what is acceptable. Don't even start with that. 

Edit: the idea itself is cool, not how you want to make it work. 

Edited May 8, 2019 by HexOmega111x

[Zilotz]

50 minutes ago, HexOmega111x said:

The mechanic you propose is cool but any little bit of p2w element should be avoided.

Are boosters avoided?

As i told before - i would like it to be free but i'm not sure if DE will wish to add it as core mechanics instead of paid item(you can read reasons in this topic).

 

[peterc3]

Why would I ever buy anything with Plat if I knew, right there in the UI, how many times I'd have to get something guaranteed?

[HexOmega111x]

il y a une heure, Zilotz a dit :

Are boosters avoided?

As i told before - i would like it to be free but i'm not sure if DE will wish to add it as core mechanics instead of paid item(you can read reasons in this topic).

 

Boosters are not p2w. There is no p2w elements in this game at all. But I guess this is again an ever ending debate. 

Your idea can work without involving plat at all. DE never sell or lock any part of its content for real money or plat. You always have full access to every content this game has to offer. 

We could get the device via a quest and then have to loot tokens to activate a slot. I know that the idea is to get rid of the weight of the rng mechanic but this token could be available in every rotation with a drop chance of 2% in arbitration for exemple. 

Still hard to get but not an infinite grind and you can alter the drop chance of everything in the game so tokens should not be easy to get and you should still  have to work to be able to do so. 

[Teridax68]

7 hours ago, Marine027 said:

Rivens are a big problem simply due they not have consistent stats, each existing riven is a uniqunes in itself, that is simply way different then having a copy of lets say vitality where its stats nevere change and will always rank the same up.

Okay, but then preexisting items aren't procedurally generated: there would only be a very limited number of items to remember, and the game remembers these items already. The only added memory cost would simply be which of those items the player would be tracking, along with whichever increment, so in this respect this would be far less of a server storage cost than most other things in the game, e.g. mods (especially Rivens), relics, loadouts, cosmetic setups, etc.

[Arcira]

Here is a way to make things more predictable:

Instead of a 1% drop for an item you need to farm 10 equal parts (not like warframe parts) with a 10% drop chance each which you have to combine in the end.

Why? Because math. If you know how probability works you will see extrem cases are far more unlikely if there are more parts with a higher drop chance than a single part with a small drop chance. Even though the average remains the same.

Edited May 9, 2019 by Arcira

[AshenHaze]

17 hours ago, Zilotz said:

But on the other side it increases server cost so it should be kind of investment to avoid DE lose profits if they implement it.

Sure i would like to get it for free as base mechanics but i understand that there are two sides - players and DE so it should benefit both sides.

Do none of you understand the term "opportunity cost?" DE would probably lose more income as the above mentioned program is essentially p2w. Plus, does this game honestly need yet another needless platinum sink? If y'all jsut picked up the market once or twice you can easily trade stuff to get plat and then trade for the item you want... Unless you're going rivens then you gotta pay protection to the mafia.

I still personally think that warframe should have some form of ingame bartering system that works better than meroo's bazaar but uses ducats. That way you can distance the idea that people are paying to win and distancing the issues that come with cash for ingame currency. Plus, having the different currency gives  people a reason to use that new barter system rather than just trading for plat alone via warframe.market.

Also, there are in the works a few bills that are hammering on P2W based mechanics in games (similar to lootboxes) as it is seen as a form of predatory market that targets children so DE would have a lot to lose by adding "plat for everything" and will probably start implementing mechanics that go against that here in the mid future. I've read a few articles on such bills in the US, EU, AUS, and a few other places.

[Zilotz]

3 hours ago, AshenHaze said:

Do none of you understand the term "opportunity cost?" DE would probably lose more income as the above mentioned program is essentially p2w. Plus, does this game honestly need yet another needless platinum sink? If y'all jsut picked up the market once or twice you can easily trade stuff to get plat and then trade for the item you want... Unless you're going rivens then you gotta pay protection to the mafia.

I still personally think that warframe should have some form of ingame bartering system that works better than meroo's bazaar but uses ducats. That way you can distance the idea that people are paying to win and distancing the issues that come with cash for ingame currency. Plus, having the different currency gives  people a reason to use that new barter system rather than just trading for plat alone via warframe.market.

Also, there are in the works a few bills that are hammering on P2W based mechanics in games (similar to lootboxes) as it is seen as a form of predatory market that targets children so DE would have a lot to lose by adding "plat for everything" and will probably start implementing mechanics that go against that here in the mid future. I've read a few articles on such bills in the US, EU, AUS, and a few other places.

Just think of it like a permanent booster - it is a bit similar in mechanics(but doubles chance for specific items only and has no time limit).

Personally i would pay for it easily(even if it would cost 10 000plat) because RNG is really annoying for me(having bad luck with it since years makes me work 2-10 times more than expected by drop chance to get an item).

For trading system - that is other topic to discuss and is not related to this one.

[Leyers_of_facade]

6 hours ago, Arcira said:

Here is a way to make things more predictable:

Instead of a 1% drop for an item you need to farm 10 equal parts (not like warframe parts) with a 10% drop chance each which you have to combine in the end.

Why? Because math. If you know how probability works you will see extrem cases are far more unlikely if there are more parts with a higher drop chance than a single part with a small drop chance. Even though the average remains the same.

Interesting proposal. Did the calculations and here is the graph.

While the probability of completing the goal is converges to 1 at a much faster speed under your suggestion, it is notable that for that it requires ~109 - 110 runs till the probability of 10% 10 times overcome the 1% once. Interestingly, this point is actually higher than the expected number of runs. In addition, for much lower values (eg, 50 runs), we can see that under your suggestion, the probability that someone is done within 50 runs is minimal (2.45%) while with the current system it is at least reasonably probable (39.50%).

While extreme cases of number of runs are less likely, it is at the cost of significantly lowering the chance of everyone at obtaining the desired item in earlier runs. I am not too convinced that this is better.

Edited May 9, 2019 by Leyers_of_facade

[Arcira]

5 hours ago, Leyers_of_facade said:

snip

My numbers arn´t accurate and just there to explain the idea. If you want an intersection point at .5 you´d actually need ~12-13% but it doesn´t really matter here.

So yea you are right the probability of getting an item early is quite a bit lower. Nonetheless most people only see there "chances" but forget to take a closer look at the graph.

You can either get lucky and get the item early (<50%) or screwed and you have to do more runs than intended (>50%). Both cases are equally likely with the difference that by the nature of rng you will never have a 100% chance to get what you want. So getting unlucky simply has more "potential" (in the negative sense).

Keep in mind rng systems always work against and not for you.

Edited May 9, 2019 by Arcira

[SordidDreams]

On 2019-05-08 at 2:24 PM, Zilotz said:

sometimes RNG gets beyond annoying, that is problem

That's the flaw in your reasoning right there. From DE's perspective, that's not a problem, it's the whole point. You're playing a free-to-play game that, like all f2p games, makes a profit by annoying you into saying "f*ck this" and just buying the item with money. If farming was fun and satisfying, nobody would ever pay to skip or accelerate it. Nobody would pay to have less fun.

Edited May 9, 2019 by SordidDreams

[Zilotz]

1 hour ago, SordidDreams said:

annoying you into saying "f*ck this" and just buying the item with money.

Nope, because annoying to "f* this" means i get annoyed and just stop playing for some time.

Also not all items are trade-able at all.

Playing is fun and playing if we can know we will be rewarded will be even better, right?

[Birdframe_Prime]

See, the fun thing about the RNG in Warframe is this:

Every single drop table is limited, and every single drop table is known.

You run for the Ephemera or the Aura Forma, for example, the place to go farm them is entirely known, and the drop chance is entirely known, plus the other items in that drop list are all known with their own documented chances of occurance. Yes, its not exactly as fun that the initial drop chance is 1.5% for the one reward, and only ever 2.5% at max (although this is actually being adjusted for the Jovian Concord update), but because of this it's as managed as any RNG system is ever going to get.

If, for example, the rewards for that same game mode included four hundred other possible things that dropped in other locations as well? Then you might, just might, have an argument for this system. As it's not... then RNG has already got management on it.

Which leads me on to this:

On 2019-05-08 at 1:24 PM, Zilotz said:

It is not pay to win, it is a way to pay for making results predictable.

Are the players that pay the thousands of Plat to attain this item, and the hundreds of plat to slot a tracker device into it, gaining a way to circumvent the RNG stated drop tables by increasing their chance at a specific item?

Yes. It literally changes up the drop chances of a reward (not a resource, not an in-mission drop, but an end reward) to make them better than the base game's chances without it.

Are players that do not pay for this item still going to have the exact same problems as always with RNG?

Yes.

Does this constitute the ability to pay for a clear advantage over other players?

Yes.

Then this is pay to win.

Warframe has and still does endorse the idea of 'pay to go faster', where you can gain access to the base items (although not fully-levelled and definitely not any different to a player that built the blueprint) and to boosters that double the gains for your time investment. Boosters are borderline, to be fair, because there's an actual 2x drop chance booster which flat doubles the chance of an item being dropped from an enemy, but at the same time this does not actually increase the chances of getting a rare item, just the chances of getting a drop in the first place. Boosters also do not, in any way, affect mission reward items, such as the Aura Forma.

But Warframe has consistently removed anything that's pay-to-win, such as the slot-machine style of visual randomisation on Kubrows, and even is still removing things that have a loot-box style (hence the complete removal of the Mod Packs from the market).

It genuinely doesn't matter what price you put on it, if you're putting any price at all on it then you're making it pay to win.

However...

There is something that I would support being implemented in Warframe.

The grand slam of three changes to how drop tables and loot for the mission rewards are managed that would comprehensively allow players to grind-to-win in the most reliable way possible while retaining RNG. 

Spoiler

 

1. The 'personal drop table'
2. The 'scarcity' bias.
3. The 'repeat rewards' bias.

The first step would be to make mission rewards all conform to the Relic system style, where each player gets an individual reward and there's a short window where you can choose to have a copy of another player's reward instead. This means that if you're searching for a specific item from the drop table, playing in a group means you are literally 4x more likely to gain that reward, and also that players who already have the specific reward themselves can choose another that they have not received yet.

The 'scarcity bias' is then a refinement of this system. If you have items from the drop table in your inventory, such as parts for a Warframe, or a specific mod, you are less likely to get repeats of it as a mission reward (normal drop tables for killing enemies still apply). If you have multiples of that item, copies of the reward, this then proportionally lowers the chances further.

A further shift would be the 'repeat rewards' bias, where if you're playing for multiple waves of endless missions, your chance to get a repeat reward goes down.

Imagine if you're on Arbitrations, and getting an Endo reward reduced the chances of getting it again, increasing the chances of getting any of the others, hmm? On Rotation C in particular, where the Ephemera drops and only one Ayatan Sculpture drops (the combined Endo/Ayatan drop chance is currently 81.5% of the drop table), you could lower the chances of getting Endo/Ayatan for the next Rotation C, and with Scarcity Bias it then lowers the chances of getting repeats of mods you own for Rotation C as well. Meaning that it could be possible to play the game enough to skew the rewards to the point where you have greater chance of getting Aura Forma and Ephemera in that mission than you do of getting anything else combined. In theory. In practical use, you could probably raise the chance of the Ephemera and Aura Forma up to around 10% if you were lucky with the drops on Rotation A and B (getting Endo twice would lower the chances of Endo in Rotation C quite a bit...).

Without using a pay-to-win system.

I can also see the response some people are ready to type on here as well; why not just buy or trade the mods you don't want so that you have a higher chance of getting the mods you do? Why not just skew your own drop table in that way?

On that, if you're looking for mods, why would you buy three different ones to skew the table rather than just buy the one you want? And on the Ephemera, you still have to compete with the massive proportional chance to still get Ayatans or Endo. The base chance of those is 90.5% on Rotation A, then 88% on Rotation B, and then it's 81.5% on Rotation C. While you can skew the Mods, you've still got those in the rotation that can't be artificially skewed by buying them so you'd be left with 80-90% chance every time to get something other than your desired reward. And only getting that reward in game can lower the chances of it happening again in the same game, meaning you play to win, not pay to win.

Basically with you able to choose from 4 rolls at the loot every time, bias the drop table by playing the game, and even biasing the table further by playing the particular mission, a good team or clan could farm up practically anything they set out to get as long as they stayed in the mission. In fact, a really good team could practically guarantee an Aura Forma or Ephemera drop by the 9th or 10th rotation. Return of the Long Run meta!

(As a note; for actual Relics the bias would not apply, due to them having their own adjustable drop tables via Trace Refinement.)

 

 

Honestly... I could go on and on with this topic... I already have, so I'll stop ^^

[MJ12]

On 2019-05-08 at 3:02 AM, Teridax68 said:

I very much agree with the core proposal: Warframe is in dire need of a pity mechanic for its RNG, and I feel there is strictly no reason to keep not having one when pity systems have now been a tried and true gameplay feature in RNG-based games for years, to their significant benefit. I do disagree with some of the specifics, though, as I think the system should be baseline, to begin with, but if we're going to make it an item, it should be obtainable without plat.

The core idea of having some kind of pity mechanic for extremely rare rewards isn't bad but the implementation in the OP needs work. You've already touched on the problem of making the pity mechanic an unlockable paid feature. But there's also the math issue.

A pity mechanic generally only kicks in if you suffer from extremely bad luck, rather than letting you obtain a reward immediately upon reaching the number of runs you'd need to have a mean 1.0 rewards (assuming an infinitely large sample size). This would actually significantly increase effective droprates. Most of the time pity mechanics are actually pretty hard to activate without a string of continued and extremely bad luck. For example, the 100% 'pity' gacha rate in Fire Emblem Heroes, as a random example I looked up, requires 120 rolls without a single 5-star pull. That's a 1-in-3,000,000 occurrence.

This also means that if droprates are designed around an expectation of how fast a player can acquire something, your pity rate mechanic is going to need to be obfuscated in a way the OP's doesn't, because otherwise it's going to be incredibly intimidating. If a player logs on and sees "Guaranteed Aura Forma Reward: 0/300 rewards obtained from Arbitrations" that's not exactly going to reassure them. Because of various laws, said mechanic and its inner workings will have to be signposted somewhere, but the OP suggestion has the issue that there needs to be some friction regarding checking it, otherwise it implies a much more intimidating grind than actually exists for the majority of players, and that has its own issues.

Edited May 9, 2019 by MJ12

[Zilotz]

42 minutes ago, MJ12 said:

If a player logs on and sees "Guaranteed Aura Forma Reward: 0/300 rewards obtained from Arbitrations" that's not exactly going to reassure them.

For old players it will be big progress. Why? Because some of them did 1000+ runs(if claim 300 as designed chance which actually isn't, it is 67 actually with 1.5% chance) and still getting nothing.

Also my proposal is not to replace chance of getting but to add a guaranteed one with exactly designed rate(which is calculated from drop tables).

I propose to get lowest chance(from drop rates) of additional thing and still it does not stop you from "possibly"(really not for me except one time in few years) getting item before that.

 

[MJ12]

5 minutes ago, Zilotz said:

For old players it will be big progress. Why? Because some of them did 1000+ runs(if claim 300 as designed chance which actually isn't, it is 67 actually with 1.5% chance) and still getting nothing.

Also my proposal is not to replace chance of getting but to add a guaranteed one with exactly designed rate(which is calculated from drop tables).

I propose to get lowest chance(from drop rates) of additional thing and still it does not stop you from "possibly"(really not for me except one time in few years) getting item before that.

 

Yes, on average you need 67 runs to get an Aura Forma. This is why any RNG bypassing pity mechanic isn't going to give you a guaranteed Aura Forma at 67 runs, but rather it's going to take 200 or 300 runs before the mechanic kicks in with a guarantee. Because pity mechanics aren't implemented to guarantee that no player will do more than the average number of runs. They're implemented to guarantee that a player won't waste time doing far more than the average number of runs, or pulls, or whatever RNG reward mechanic is used, without getting what they want.

[SordidDreams]

3 hours ago, Zilotz said:

Nope, because annoying to "f* this" means i get annoyed and just stop playing for some time.

That would be why f2p games tend to have a very high player turnover and millions upon millions of dead accounts. Take a look at your friendslist lately?

Edited May 9, 2019 by SordidDreams