Loot luck is bugged (with proof)

Just gonna add my 2 cents in here.

First, I am going to reiterate what my mentor, who likely helped write one of the first CS courses, and what I have stated to students over the decades:

You cannot, and NEVER will be able to generate, a completely random string of variables/results. (In regards to computational output.)

This is a technology limitation that has existed since it’s inception of the DIDO system, and still to present day it is unlikely that we will overcome it. I know that everyone thinks AI is amazing and all that, but still at it’s core, any AI CANNOT generate a random output.

Now CS majors/students/armchair fanatics understand the reference of this. For everyone else, it means that no matter how “RNJesus” the result/variable seems to your perception, it is all based in code and came from “somewhere”.

That all being said, I can confirm that luck loot is indeed broken, and furthermore, they broke it intentionally for the Winter Convergence event. Why? Well I don’t need fancy calculations and source script to know first off, it’s much easier than that. Because of the way that guaranteed loot and object interaction works in a script environment, it essentially “breaks” the script of generation.

Everyone talks about loot tables like that means something. It does not. Loot “tables” haven’t been used in modern gaming for nearly a decade now. This isn’t old school D&D. They are entries in a database, which yes is still technically a table, but also not as well.

When you have guaranteed drops, out of special containers, the script that Mathrandil obtained and calculated/tested goes out the window. Although many people would believe that it only applies to certain containers, and I agree that is how it should work too, the coders/devs that work on this haven’t the foggiest idea what they are doing, because it’s not all one person/company’s code. AG outsources a lot of this development, which is why there are gapingly obvious holes in consistency. It’s really that bad.

Because of where it has to reference the character, an interrupt of sorts has to be thrown into the mix on the character side script, that doesn’t exactly jive too well with the existing “luck” calculations. Because of this, a character’s “luck” is not being considered at all after daily allocations of event loot are met through w/e “event objectives”.

If you would like to prove this, go on an elite chest run at least 30 hours after the last time you did all of your daily convergence/tree runs, as well as the last time you obtained any gypsum.

You will find that your loot is SIGNIFICANTLY better than as if you did it the other way around, but go ahead and then do it the other way around just to be thorough.

So the big question you will have after almost 3 days is WHY? Well, I don’t exactly know why either, but back to the confirmed way that AG uses contracted coders, I would be willing to bet that piecemeal way of doing things has led to luck scripts being ignored after a character’s daily allocation of “guaranteed loot” has been filled, and is no longer running the scripts after that point.

The biggest tell in this is prior to the WC patch, you could easily get 5 or 10 chests in a 3-4 hour gaming session that would seem to “bug” and not issue loot. Relogging, porting, and even redownloading did not fix this issue. This is likely because the “loot scripts” were receiving too many simultaneous requests for processing, and your particular “request” (when you looted a box) was stuck in queue, and dismissed because there is no queue hold time for these requests.
(You can also tell because upon relogging, the chest appears looted for your character, when you clearly did not get any loot)

Although at the end, this is all conjecture, it is logical conjecture based on knowing enough about how scripts work, system limitations, database limitations, and being able to see the bigger picture.

AG knows about all the problems, but they don’t know where, or probably even how, to start fixing them. Eventually they will be forced to retain a company whose script and assets rule the majority of the game, and there will likely be a “major overhaul” to how loot works sometime in the next 3 or 4 months. Which really is just one company coming in and “unfucking” the loot adjustments by completely deleting the mess we have now.

Until then, we are all stuck with what it is, just make sure you do your tree/sack/present looting for the event AFTER you do everything else in the “normal” game for now if you want your luck to have a heavier consideration in your chest runs. Doesn’t mean you will always get better stuff, but at least this way you can still have the “dice thrown” for your character’s luck, whereas I have found the other way around the “dice” are not being thrown at all.

19 Likes

Please do not say you are just unlucky. The sample size is scientifically proven big enough. You don’t need a sample size of a million chests. I don’t mind that it is broken. I just want to know what is broken. If devs would tell us f.e. “hey the pristine pearls are overwriting the luck currently on each piece. Do not use it until we fix it” that would be totally fine for me. I just think they could give us any response.

1 Like

so what should we do? what gear, accessories, and weapons should we use ?

This really does depend how the % is translated.

My understanding is that trophies and food are not infact percentages but fixed values.

Your gear percentage luck is applied ontop of that so the best overall luck source will always be trophies as this will only improve the yield of your percentages.

It would be nice to get some confirmation on luck and how it works exactly.

I’m only clueless when you piss me off and i don’t want to fix something.

And you dont understand probability calculation. Out of 100 Chest you can except to get 2 Trophys, if your Chance is 2%.
It is possible that u dont get anything but its unlikely. And after 900 Chest, you can say: There is something wrong if he not even get close to 9 trophies.
probability calculation is really accurate to think he is one of the rar cases is just stupid, more data needed at the end to make sure, but you dont understand probability if you think, he should NOT get 2 trophies.

LOL. And when you dont know why that super cool web app you wrote wont work when its actually deployed in a real 3 tier environment but works perfectly when the web, app and db are all installed on your workstation which you made 14 different config changes to and didnt bother to include them in the install notes. “It works fine on my workstation!!” Heard it all before.

Im guessing you stay pissed off.

A friend of mine looted 1 trophy in 800h playtime, full luck gear but no pearls. I found like 14-15 by that time.
Then he got like 1-2 jewelries with all in all 1.5-2% luck more on it (cause higher gs and pearl).
He then looted a Stacked Deck and 2x Forgemasters Notes in like 30 mins!!!
Felt very weird to me.

Im not really into math and statistics but even with maxed luck bonus, your 2% chance of getting trophy seems oddly high.

My bet would be that there was in fact a drop chance change during pvp flag luck bonus patch and such artifacts ended in separate batch roll.

I’m not sure I understand your questions. If you read the post, it just means you have to do your chest runs BEFORE you do your WC event loots/trees/sacks etc. The luck gear you wear is up to you.

I also reconfirmed this again last night. I was out running stuff to new towns and noticed my harvesting was close to a prestige point, so once I got it there I received a chest, the first one I had ever gotten.

I pulled a topaz gypsum and the legendary blisterweave x 6 out of the chest. After that, my normal chest loots were amazingly bad.

So effectively, if you do anything that gives you a “guaranteed” higher loot… It appears to nullify the future luck calculations for your gear that you wear, as if you were not wearing any at all.

Hm… I can’t really confirm this.
I am still getting the same loot as before.
Can we get an answer on this whole luck topic from a @Developer?

OP, first, awesome post. Thanks for all of the links explaining the loot tables and math.

I looked into the loot buckets a bit in the specific link below and I think you might have misread them slightly. Also I am not an expert so it’s entirely possible I am incorrect.

If I read through the JSON file and search on “Item145”, representing all of the items in the ContainerTrophyArtifacts bucket, there are three items that also appear to have equal weighting to your trophies that you overlooked: Ember Stone, Dummy Frame, and Fluffy Stuffing. Essentially, if you got a few of these during your chest data set, but did not count them as “trophy” drops, you may need to reassess your conclusions.

Thoughts? Does this help the bug be less broken, if you happened to track drops of these items too?

tl;dr - It looks like any time an Ember Stone, Dummy Frame or Fluffy Stuffing drops, if the randomness had been ever so slightly different, it would have been a trophy, since they come from the same loot bucket.

1 Like

Hey @Weevilily thanks for that. Yeah, you are right on that I think. I just checked with my friends and we dropped 2 of those items overall in our whole new world gametime. I guess 1 of those came from one of those runs. So we can up the dropped amount to 2 instead of 1 total in my data. But even if it was three something really feels off (for me at least).

If the @devs cant be bothered to reply after all the requests for clarity on this subject then i think ill take a break from ‘testing’ the game for them for a while…poor imo

2 Likes

Dont feel bad about taking a permanent break. Please do.

bump… ive been doing chest runs non stop in MD/WF/RW, for about 5-6 days straight. like NONSTOP. i have not found one trophy mat in that time frame. however I used to do these same chest runs… before I got my voidbent, pearls, trophies, etc (before update) and on one run I got 2 storage chests and an armorers journal. Same day on a different run got quartermaster notes. Next day got armorers journal again.

I’m not good at coding/math at all… but I can easily tell something is off. It’s as if the more luck I have the worst drops I get, now that I’m basically maxed on luck, I have literally not gotten a single trophy mat. Pls @Developer figure this out.

2 Likes

between monday and tuesday of last week… i got amazingly 14 trophies that earned me 100k.

But before that I had heard a rumor that the lucky trophy was in trouble after merging the servers. They said to remove and put the trophies again … after that I got the following trophies :

  • 4 Stacked Deck
  • 3 Journal of Aeternum Flora
  • 3 Armorer’s Journal
  • 2 Surveyor’s Tools
  • 1 Pure Resin
  • 1 Philosopher’s Stone

At the time I did this I only owned 3 trophies of cards. After I bought load dice and incredibly after Wednesday my luck turned negative… I didn’t get 1 trophy at all. I will be performing the same process now.

Another observation was that I apparently spent 3 days doing any chest race… I don’t know if this has any interference, I’ll test it now with a friend who spent the same amount of time inactive

@TheLawRich can u help us ?

2 Likes

This is one of the best informed posts on the entire forum. This guy knows what he is talking about. That post has opened my eyes somewhat. I knew that computers could not technically produce pure “random” numbers however I have no technical background to understand it. Thanks!

I started farming chests about 2 weeks ago. I was getting solid drops with only luck bags, jewelry and voidbent. I was usually doing 2 max 3 runs on reek and shattered. Some day I got 4 drops. Since then I gradually upgraded trophies and equipped pearls. I also got max luck jewel. I am doing about 10 runs a day and maybe get 1 and I am not getting any legendary pieces either (green 600gs all the time). What I noticed is that I am getting more variety items, like rarer schematics. My guess with this is that the increased luck the loot pool is increased and you roll from more items making your chances lower overall.

i work in statistic i study Math advance statistic course and i can said that the number is good. you cant just said is random. In statistic we can make calculation to prove that something is not right if you flip a coin 10 time and you get head 9 time on 10 it can be random but not if it happens lot of time. You can check the % of chance that you get 2 mat trophy on a 900 try with a binominal law. a function that you can use in excel and the prob is 2.38E-06

for reference
https://stattrek.com/online-calculator/binomial.aspx

6 Likes