Rune of Removal

Feedback and Suggestions
81

I still don’t know that I agree on it being “random”. I think the disconnect is actually on where people are looking at it. If you are looking at the roll, it is random. If you are looking at the outcome, it isn’t so random. If the outcome is mostly deterministic, its not really random.

One could say that vaccination effectiveness is random because it has a 5% chance of not working for someone for example. If the advertising of the vaccine was then “It’s random if it works or not”, nobody would get it.

There is an expectation of “random” that the rune doesn’t really meet.

There seems to be a lot of things in this game that “feel bad”.

83

I don’t understand your argument. Having various outcomes distributed by a certain spread of probabilities has no relationship as to whether something is random.

Random /= Luck and Random /= situations (like vaccines) that include unknown variables (genetics, undiagnosed health conditions, vitamin D intake, etc.).

I want to recommend a book “Fooled by Randomness” to everyone that thinks random is used incorrectly in the description of RoR. The problem with the description is not that it says “random,” the problem is that it does not provide enough information about what is being randomly determined.

1 Like
84

There is a bag of 10 marbles, 5 red and 5 black. You reach in and take a marble, what is the probability that you take a red marble? a black marble?

Now there is a bag with 10 marbles, 3 red, 2 yellow, and 5 black. What is the probability that you get take a red marble? a black marble? a yellow marble?

Now, lets say these marbles are put in a row in front of you and numbered 0 to 9.

It would looks something like this
[r, r, r, y, y, b, b, b, b, b]

Lets say that the example above are the 0th, 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, 8th, and 9th marbles in the row.

Now roll a 10 sided dice, and do modular division on the result. 10 % 10 (modular division) = 0
Therefore,
1 % 10 = 1
2 % 10 = 2
3 % 10 = 3
4 % 10 = 4
5 % 10 = 5
6 % 10 = 6
7 % 10 = 7
8 % 10 = 8
9 % 10 = 9
10 % 10 = 0

Whatever the result of rolling the 10 sided dice (which is random) modularly divided by 10 is now the marble you choose to remove.

This happens to be a (simplified) way of how this is done in general for any number of N items that you need to randomly select from.

This is also (without looking at source code) how this is done for Rune of Removal (more or less). The only difference being, once you roll, say a 2 on the 10 sided dice, you now have to remove all red marbles from the row. Because Rune of Removal randomly (dice roll) selects an affix to remove, all it does is weight (different number of red or yellow or black marbles in the row = different probabilities to be selected) based on the number of tiers that affix has.

3 Likes
85

There is no such thing as “mostly deterministic” . You are either deterministic or you are non-deterministic. There is absolutely no determinism in the way Rune of Removal selects an affix to remove. It is random, you cannot know what happens 100% of the time when you use it. You only know the probabilities. However, it is deterministic in the sense that when you use it, you know that you will remove one affix.

The reason the crafting system in LE is said to be deterministic is due to the fact that when you craft you know that IF it succeeds you know what the result will be, there is no randomness in what is applied (affix wise, instability wise – not so much) to the item that you’re crafting.

2 Likes
86

That is totally fine, i just specifically were refering to the suggestion of Jakira, which was a pretty long sentence.

1 Like
87

My bad, I didn’t realize how quoting you there seemed like I was disagreeing with you. I think we agree on pretty much everything in this thread.

1 Like
88

i honestly think they should allow to use both sides of the shards: both modifier and support item side.
like if you want to remove certain roll, you use rune of removal and then using glyph of stability would increase chance removal of selected roll by say 50% and the glyph guardian would guarrantee to remove that affix or prefix. i mean it adds instability already, so why not?

i’d even say similarly with rune of refinment, glyphs could “cement” or protect selected roll in a similar fashion. It’s not a big thing and would not really break anything.

89

Dude, I’ve been brainstorming suggestions for the crafting system and I come at it from a different angle but your suggestions are aiming at exactly the same goal. Nice to know I’m not the only one.

90

You all need to stop. It has nothing to do with the dictionary definition of random. The problem is what people perceive random to be. Stop talking down to me like I’m stupid and you don’t have to sit there with your marble novel of a post like I am in 4th grade. You completely missed the point I was making.

Clarification on how the rune actually works may be enough to correct the perception people have of it.

If people were really going to math the randomness of it, if they wanted to remove an affix that was only 1 tier on an 11 tier item, they should probably upgrade the tier 1 to a tier 2 first to approximately double the chance that that is the affix removed.

“Mostly deterministic” = you have a 99.99999% chance that something succeeds, you assume it will succeed. The ones that don’t are then considered anomalies.

91

In addition to that, it would be nice if people stopped saying it’s not random when it is. There’s more than just 1 group of people involved in this argument & it behooves everybody to stop. Then maybe @Sarno would be happy & might give us some more sneak peeks of 0.8.2 without trolling us.

92

Yes, that is why I put “random” in quotes. Because people are talking past each other. It drives me nuts that people don’t give others the benefit of the doubt and they take everything so literally. There is nuance and people are trying to get a point across even if it isn’t easy because they don’t know how to word exactly what they mean. They may not even realize the real issue they are having.

I would prefer this thread just be closed. It is not productive at all at this point.

1 Like
93

I apologize if you think that I was talking down to you in any way. That was not my intent.

That “marble novel of a post like you’re in 4th grade” is exactly the way professors teach this very subject to people of any age. No one has really disagreed that clarification would be best on how it works.

I must disagree, there is no nuance to this at all. As @Llama8 suggested, there is a definition for randomness, and deterministic – when someone makes a claim that the definition of these words are different from what they are, I personally believe its better to address the claim than not. This reminds me of the many times I didn’t understand something in school and my professors would sit there and tell me how I was wrong until I understood (and it irritated me a lot). They knew the real issue I was having, and continued to argue/teach me. I realize that may not be the same contract that people have on this forum with one another, but we are here to ask questions, help the dev’s, and help each other.

2 Likes
94

As I’ve said earlier, the problem here is that if you ask people what is the probability to randomly choose 1 item out of 4 most of them (who can calculate percentages) would say it’s 25% because that is the most logical answer without the additional information of the choice being weighted.

Yes, if you went to college and were introduced to the wonders of statistics and discrete mathematics then you might ask if it’s weighted before answering the question but I’m not even sure of that. It is totally justified to assume that a random choice of 1 out of 4 will yield to 25% chance to choose each of the items.

Yes, you and Llama8 are right, it’s rolling on different things, it’s not rolling on the 4 affixes, it rolling on each tier of all affixes. Yes, it is random just like 1 in a Googol is random but it’s not the “same random” as an ordinary person usually perceives “random”.

There are at least a couple of people asking on chat each day if the rune of removal is broken because every single player who have not checked it on discord or knows that it’s weighted from other sources assumes that the outcome is evenly distributed.

95

This is not about smart people figuring out if the system is weighted or not. Nobody knew that before asking the devs. They gave us the answer that the systems weights towards tiers.

People literally stated that it’s not random if something is weighted. This is wrong and got explained several times. The linking and education started when people kept on arguing.

I agree with @darkdeal that this doesn’t add anything to the topic. The point is that some people don’t like that it’s weighted and or that the weighing is not mentioned on the tooltip.

But there would also have been a way to get out of this by just admitting that one is wrong and somebody else is right. shrugs

96

That’s not true.

97

Can you explain this?

There are two statements in that post. Which do you think is not true?

Shit… I forgot about rule #8: Dont feed the Llama

:raised_hands::carrot::llama:

3 Likes
98

Have you met the internet? :stuck_out_tongue_closed_eyes:

2 Likes
99

If rune of removal is weighted to remove good tier mods first which it is and add instability that’s like slapping your face and calling your Mama ugly this is ridiculous

100

Really? Do you not like the current implementation? You should have mentioned!

1 Like
101

Woot? Somebody slapped my ugly mama??
:worried:

Just in addition to the topic:

Besides the issue that the rune removes the affixes with higher tiers more likely (and this doesn’t work well with optimising exalted items), why not show the exact chances of removing an affix? If it is solely based on tier, it could be made visible, just like the system shows the chances to craft an affix.

If you put in a RoR into the rune spot it could display % numbers next to the affixes on the item that shows the chances of removing the affix.

2 Likes