-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Include computed chances to get monster parts #22
Comments
I did this for 4U at https://mhapi.info (generated from the gathering hall studios sqlite db and python scripts at https://github.com/bd4/monster-hunter-scripts). The scripts are a totally hack job, but if someone wants to work on this, they may find something useful there. I may try to generate a compatible sqlite db from the mhrice.json and add it to mhapi.info, but not sure if/when I will have time. Note that I prefer using expected values over P(at least one). It's easier to calculate and to me more useful. |
So for chances of broken parts, carves, etc, calculating it is easy. |
My understanding of how quest rewards worked in 4U, is that you would have a certain min and max number of rewards in a quest line. Some of the rewards would be fixed for that quest, and others would come out of the pool based on probabilities. There would be a fixed chance for whether an extra reward in the line would be received, and rng would be tested each time. So if extend chance is p, chance of getting 3 or more extra rewards would be p^3. Chance of getting exactly three would be p^3(1-p)^n-3, where n is max number of extras possible. Based on my old code (https://github.com/bd4/monster-hunter-scripts/blob/master/mhapi/stats.py), default extend chance for quests in 4U without luck skill was 22/32, and so was chance of getting the 3rd cap reward. I have no idea if / how this has changed in world and rise. Expanding more on p(at least one) vs expected value, the idea of expected value is that it's the average number of a reward you would expect to get from doing an action many times (e.g. a quest taking into account carves, shiny drops, quest rewards, or cap rewards). I personally find that more useful than p(at least one), and I think it fits better with the normal human intuition of how things should work - if I do this quest twice - I will double my average expected value, whereas my p(at least one) will not double. The barrier to understanding for non-math people, is that you can have things like 0.08 expected value for Rathalos mantle from this quest - not immediately obvious what 0.08 mantles means to someone unfamiliar with it. Just means that on average, one would get 8 from doing the quest 100 times. In the end, using probability vs expected value comes down to what you are trying to understand / convey and somewhat to personal preference and familiarity. I think it's at least worth considering - and I never see it come up in monster hunter contexts for some reason. Edit: found the source for 22/32 number, ShadyFigure on reddit was a great source of info: |
Hello, I would like to share an idea that help people trying to use the website in order to know the chances of getting a monster part.
For example, it would be great to compute the chances of the zinogre claw and show the final chance depending if the player broke his arms, capture/carve it, etc.
The text was updated successfully, but these errors were encountered: