But…
There is no intersection between those two events…
Becuse the two events that we look at are
“1st yes, 2nd no”
and
“1st no, 2nd yes”.
I am not rude.
I am incredibly patient with you.
Yet you keep deflecting my valid criticism insted of seeing reason.
Yes you explained why you think that, but my calculation is not wrong. Your thinking is.
We were not both wrong. You were wrong.
Now that you accepted that your calculation was not correct, at least accept what is actually correct as well and do not divert the topic to something else that is irrelevant to this case.
There is no intersection.
I do not see a reason to calculate anything for you at this point.
You could not see the mistake in your own calculation, even after i made the effort to explain it to you in detail.
You falsely claimed that my calculation is wrong.
I do not expect you to be able to understand why (or admit that) my calculation of this would be correct.
Even if i wanted to calculate it for you, you did not even specify if it was supposed to be exactly 1, or at least 1.
If you want a T3+, you simply add the probabilities of getting a T3 and getting a T4.
If you were to calculate the chances for getting exactly 1, you would have to add the non intersecting probabilities of the two outcomes “1st yes, 2nd no” and “1st no, 2nd yes”.
If you wanted to calculate for getting at least 1, you would have to add the third outcome of “1st yes, 2nd yes”.
Or you could simply calculate the probability of “1st no, 2nd no” and substract it from 100%.
lol
The audacity to talk down to me like this, after all of my calculations have been right, and you have been unable to understand the mistake you made…
I gave you all the explanations and all the tools to calculate it yourself.
Do it if you care.
I will not waste any more time, trying to teach you statistics, since your responses to me correcting your mistakes were
“i am right, but you are wrong”
and
“i was wrong, but so were you”.
With your calculation, with a plasma gun you have 101.4% to get a wanted (so one of the 2 blessings you aimed for) on a gun that has 4 blessings + 1 unique (without even correcting the perks calculation like you suggest).
Like always, when you’re pushed into your contradictions, you cannot discuss.
I ask you something pretty easy… you have all numbers…
This example shows clearly that we cannot calculate that like that. This is exactly same situation.
We have a number of blessings in the pool (different per weapon), we have two events (crafting at blue, crafting at orange) and we want to find the odds to get a T3 or T4 blessing or at event 1 or at event 2, but we do not want that event 1 and event 2 give both the blessing we wanted.
If I follow your correct calculation we have:
32.35 + 9.56 + 21.5 + 38 = 101,4%
You see me as your enemy… but I am not. I can see the difficulties ou have to discuss with someone else, however here I just try to correct something you pointed.
What I say to you is, if I do like you suggest, I will find totally weird values…
Here what it gives when applied what you think is correct:
These stats are entirely wrong. You don’t have 46,57 chances to get an excellent flamer (crafted 3 and all of them are not what I wanted). you cannot have 101.4% chances to get the blessing you want on a plasma gun… same for power sword. You don’t have 25% chances to get a T4 blessing you want, or everybody would know it by now.
You did not follow my correct calculation. Because then the result would be correct.
You followed your own wrong calculation. That is why the result is wrong.
I won’t even look at it, because i expect that you did not do what i think is correct (which would be what is actually correct).
My last attempt.
Assuming that your values are correct:
probability to get correct blessing on blue : 32.35%+9.56%=41.91%
probability to get correct blessing on orange: 21.5%+38%=59.9%
On upgrade to blue, you can get a T3 OR a T4. Not both. These two outcomes are mutually exclusive. So their probabilities are added together, since we accept either of the two outcomes.
On upgrade to orange, again, you can get a T3 OR a T4. Not both. Again, these two outcomes are mutually exclusive. So, again, their probabilities are added together.
Now if you want to know the probability for getting a specific outcome on blue and also a specific outcome on orange, which is a chain of events, then you have to multiply the probabilities of these specific outcomes.
Probability to get ONLY correct blue? A chain of two independent events (yes blue + no orange ).
That is the probability of “yes on blue” multiplied with the probability of “no on orange”.
Probability to get ONLY correct orange? A chain of two independent events (no blue + yes orange).
That is the probability of “no on blue” multiplied with the probability of “yes on orange”.
Probability to get BOTH correct? A chain of two independent events. (yes blue +yes orange)
That is the probability of “yes on blue” multiplied with the probability of “yes on orange”.
-Probability to get NONE correct? A chain of two independent events. (no blue +no orange)
That is the probability of “no on blue” multiplied with the probability of “no on orange”.
If you want to know the probability to get AT LEAST 1 right, you have to add the probabilities of “only correct blue” and “only correct orange” and “both correct”.
This is NOT a chain of events. It is three independent and mutually exclusive events.
Alternatively you can simply substract the probabiltiy of not getting any from 100%.
Not if you treat this game like a second job! and know in your heart that the game owes you nothing, and will not save you against a string of bad luck rolls. I mean, the inventories on my most played characters are like a hoarders junkyards… with fps loss.
@Ralendil I don’t suppose there’s a way to contribute data to this sheet, is there? Doing probability math ain’t my forte, but I’ll gladly throw some data your way to flesh out the results further, even if it’s a relatively small addition.
EDIT: and is there a specific weapon to provide data for? I don’t know if/by how much differing numbers of blessings per tier alters the odds
Have someone you know read both of your posts back to you, even after all this, if you don’t see what the problem is you need to do some soul-searching. Sorry for coming out swinging at you, but even walking by Hadron pisses me off these days. The crafting system is ass, feels like ass, and needs to be reworked. And not just does it suck, but it’s psychologically predatory BS too. This kind of sh!t game design should be illegal by international treaty, and individual country’s laws. FS should by fined by the Swedish govt (exponentially) everyday they keep this system on. This sh!t is unacceptable as it is even after recent changes. F#ck me man, if I wanted to deal with this type of sh!t I’d just go back to playing gacha games and League of Legends. I have way too many friends stuck on what is a lifetime suicide watch because they can’t get off these garbage games and game design like this.
In conclusion the odds to get what you want are low. Amazing. People already figured this out in beta, back when it was an rng roll to even get the weapon you want to play, never mind the stats.
This one was too personal for me. Playing League of Legends on Ranked solo was the most toxic yet addicting relationship I’ve ever had with a video game. I tried coming back to it a few years ago. Then I realized why this game was unhealthy to me and why it nearly gave me a heart attack. So I grinded to Gold then never played again. It’s why I lament the insidious design of Darktide, because I know the negative effects it has on people, design like that isn’t healthy for people.
Yeah there’s nothing redeemable about the design, from the fact that their adjusted costs are bubkus (all it means is upgrading trash gear is like 20% cheaper on average, but you still have the non existent material drops of low difficulty) to this most recent lock change meaning absolutely nothing. Brunts should seriously have a bad luck protection built in where each purchase raises the floor of item ratings until it hits max. You could still sit there and have to spit out over a dozen of a given weapon considering you have to go take it to Hadron afterwards, so I don’t get how this would break their item system. It just means that at some point you could possibly brute force their luck system with your play time, instead of it being meaningless besides whatever enjoyment you have with the game (I’m more addicted to this game’s movement system than anything else, so I still play).
Wow, some actual research about the blessing/perk rates! Thanks @Ralendil!
Haven’t read everything yet, but the probabilities here seem off.
Based on my (quick) calculations getting a t4 blessing out of 2 specific desired ones on a 6 blessing weapon (like braced autogun) is around ~24.5%. Hitting 1 specific is around ~12.6%, based on the data provided.
Sorry, this is nonsense. I’ve read this thread over, and I see so much wrong with your claims. I’ve seen that @flawless hasn’t gotten through so I’m not going to bother.
If you find the time to learn enough about statistics to do this accurately, you’ll realize your sample size is way too small to draw conclusions anywhere close to what you’re claiming.
I think the idea was just to provide enough data for an estimation, not for exact values.
But small probabilities come with large variance, and getting close to exact values for an rng system with such low probabilities, would take a lot of data indeed.
Anyway, independent of the intention being estimations or exact values…
Assuming that OP has collected the data honestly, it would be a shame to waste the time and effort that went into it. The calculations really should be done correctly for sure, so that the data can be evaluated properly.
Ok… after some time, I understand what you mean.
However, the calcul doesn’t take in account the perk wanted but not T4 in your exclusions (so a T1, T2, T3 perk).
Considering we have chances 2/15 to find a perk we want (13.33%) and that there is 26.1% chances to get a T4, we have 3.48% chances to get what we want at green.
Considering we have
2 chances over 14 to find the wanted perk if we have not got it at green and 25.5% chances to get a T4, this means the odds are 3.64% to find wanted perk at purple in this hypothesis
1 chances over 14 to find the wanted perk if we have got it at green and 25.5% chances to get a T4, this means the odds are 1.82% to find wanted perk at purple in this hypothesis
This means, if I am correct, that the odds to get a T4 perk either at green or either at purple are:
Chances to get right perk at green and not at purple:
chances to get what we want at green: 3.48%
chances to NOT get what we want at purple with first event as a success: 1-0.0182 = 0.9818
odds: 0.9818 x 0.0348 = 0.03416664 ≈ 3,42%
Chances to get right perk at purple when we did not get it at green:
chances to get what we want at purple when we were not lucky at green: 1.82%
chances to NOT get what we want at green: 1-0.0348 = 0.9652
odds: 0.9652 x 0.0364 = 0.03513328 ≈ 3,51%
Odds to get T4 perk OR at green OR at purple: 0.03416664 + 0.03513328 = 0.0692992 ≈ 6.93%
This is the same thing that i calculated in the citation below.
Except
i used 25% probability for T4 (both green and purple)
you used 26.1% and 25.5% probabilities for T4 (green and purple respectively)
i showed the calculation for each probability via fractions (based on number of perks) and the probability to get a T4 perk
you used a decimal for the “complete” probabilities instead
26.1 is the actual value… that’s why i used it.
But there is a difference between the 2 ways to calculate it.
Here your calcul with actual values I used:
(2/15) x 0.261 x (13/14) = 0.323142857142857
(13/15) x (2/14) x 0.255 = 0.315714285714286
= 0.638857142857143 (instead of 0.0692992)
Small difference. Cause when you take 13/14 (or 13/15) you forget the 14th with T1, T2, or T3
True. I actually forgot to include the case that one of the two good perks rolls at T1-3 on the “not successful” upgrade.
But aside from that, it should all be correct.
Aside from the fact that i forgot to include the aforementioned case, and you forgot a 0 behind the decimal, our calculations are now in agreement.
