Quote (laztheripper @ Sep 10 2019 05:05am)
Code
The +X bonus is set by a RND[100] (this will give a range of values from 0 to 99) and if the item is an imbue it will add ilvl/2 (drop fractions).
If 90 or above, +3
Between 60 and 89: +2
Lower than 60: +1
The total skills on the item will have been set by a RND[100] and if an imbue add ilvl to that random number.
If 91 or above: 3 skills
Between 71 and 90: 2 skills
70 or lower: 1 skill
If we take what you claim,
81% of 1/5 = 1 / (5 / 0.81)
and then the chance of getting +1-3 on that specific skill is as stated right above.
that's one skill, and what you said about the selected skills assumes one or another skill is picked first
and that all of them are on the same tier, which can't be assumed.
No, that's not what I said. There's still 81% for each other skill of being from tier 1.
Hence (81%)^3 = 0.81 * 0.81 * 0.81 for all 3 skills being from tier 1.
Quote (laztheripper @ Sep 10 2019 05:05am)
so ~1/10 of whatever 1 / (5 / 0.81) is. multiplied with the chance of getting whatever other skill you wanted. It's very, very far from being what you said it would be.
There's no 6 tries, that's just not how it works.
And yet... How do explain that "forbidden" staffmods may appear (even if rarely) ?
Your post #49 : what's the mechanism behind this, according to you ?
Actually, after skill A has been picked up, a roll occurs for the next skill (assuming here we're talking about a multi skills item).
Saying that the next skill has 1/5 chance of being picked up means if the roll ends with skill A, that skill is chosen twice, and the pick is lost. That's not how it works.
In this situation, the roll is done again (up to 6 times), until a
new skill is picked up.
Quote (laztheripper @ Sep 10 2019 05:05am)
You ask how imbue affects staff mod calcs, it's a big part of this and whatever info you're using as reference is incomplete or wrong
No, I'm not talking about imbue here. Just regular drops.
It appears that I have the same info about imbuing than what is explained in your code (just above). That's a good news :-)
Quote (laztheripper @ Sep 10 2019 05:05am)
I wrote a little brute forcer to get an idea of what I should be seeing, and my values are right. Plug in the limits and the rules, loop it 9 billion times, print out the odds, compare with the values affixcalc says and what you got, you'll see
That's a very good way to proceed. But how did you wrote your program ? A simulation should follow the same mechanism as the game itself...
Quote (laztheripper @ Sep 10 2019 05:05am)
As for the calculator, yeah I know about it. It doesn't do nearly as much
This tells you throw, h2h, 2h, etc damage with the mods of the item, defense calcs, max sockets, min ilvl, and loads more
Yes of course. And that's why I said it was an impressive work.
I'd very like to see such calculator achieved, including all the features you put in it.
And that's why I'm trying to help to get accurate values on the chances.
Again, good luck !