Alchemy

[Myckel]

what about player->setAlchemy 100 and then player->setAlchemy desired value?

It seems that the value of the progress is kept after changing the alchemy level and translated to the progress where you are at for the new level. E.g.

Level 60 -> 29
Level 61 -> 26
Level 99 -> 18
Level 50 -> 35
(The progresses above are just made up, don't reflect the real game).

However, first levelling (to reset the progress counter to 0) seems to work.

[Zini]

That is useful information. Should go into the wiki.

[Myckel]

That is useful information. Should go into the wiki.

Any specific place (I guess it doesn't really fit in the formulae page)?

[Zini]

It does. Right besides the skill gain formula.

[Zini]

It seems we have missed an edge case for the skill gain analysis.

When you are partially progressed through a skill and then change your skill level via a script instruction, the progress is adjusted accordingly. But what happens, if you reduce your level so much that the current skill gain progress would result in gaining another level at the new smaller level?

[Myckel]

It seems we have missed an edge case for the skill gain analysis.

When you are partially progressed through a skill and then change your skill level via a script instruction, the progress is adjusted accordingly. But what happens, if you reduce your level so much that the current skill gain progress would result in gaining another level at the new smaller level?

The progress is maxed out to 100, next new generated potion generates a new level, if I remember correctly.

[Myckel]

Ok, here is a small update from my side.

I've found a way to somewhat accurately measure the effect of the change of attributes to the chance. Problem is that you need about 200 samples for each combo. Then, the spread is still big, but you can see the effect.
I worked on the influence of Luck on the chance and higher Luck gives a higher chance on success (which is what can also be seen in Lazaroth's result).

On my personal side, my holiday is currently finished (in two weeks I have another week), so that leaves me little time to work on this. I also joined the Liberated Pixel Cup contest (already before I started on this), so a lot of spare time goes into that. I'll try to work a bit on the chance formula in the weekends, but the speed will drop a bit.

[Lazaroth]

I worked on the influence of Luck on the chance and higher Luck gives a higher chance on success (which is what can also be seen in Lazaroth's result).

Have you seen if intelligence change any different than luck? As far as I could tell by the tests I did, luck is equivalent to intelligence.

Problem is that you need about 200 samples for each combo. Then, the spread is still big, but you can see the effect.

I've been using a program called GS Auto Clicker to get large sample pools fast. 200 potions would take 10 s + some stopping to correct for the alchemy skill gain.


I will have more time for OpenMW in the next couple of days again, so expect some more work on negative effects and chance. Is there something you need Myckel for your work, or do you have anything I can continue?

[Myckel]

I worked on the influence of Luck on the chance and higher Luck gives a higher chance on success (which is what can also be seen in Lazaroth's result).

Have you seen if intelligence change any different than luck? As far as I could tell by the tests I did, luck is equivalent to intelligence.

I've only been focussing on Luck at the moment. I first wanted to know how to take measurements that are somewhat accurate to use, although not still sure how these values relate to each other.

Problem is that you need about 200 samples for each combo. Then, the spread is still big, but you can see the effect.

I've been using a program called GS Auto Clicker to get large sample pools fast. 200 potions would take 10 s + some stopping to correct for the alchemy skill gain.

I will have more time for OpenMW in the next couple of days again, so expect some more work on negative effects and chance. Is there something you need Myckel for your work, or do you have anything I can continue?

The problem with my measurements is not the clicking for potion creation, but continuously switching between morrowind and the spreadsheet to record the data. I measure the amount of failed potions before I get a successful one, for every progression within a level. This is laborious and generates maybe to much data, but rather to much then too less.

But please continue in the way you want to do it yourself, at the moment I think different viewpoints can be good to compare the results later (if the formula is good it should fit in both cases) or even help to come to a formula. I tested the formula you gave earlier with my data and that didn't fit (went off at higher Luck levels, if I remember correctly).

One thing you could test with that auto-clicker is if the interval between clicks has an influence (I really don't hope so, but sometimes I get the feeling that it does (most likely my mind is playing a trick on me)).

[Zini]

I have run into a new alchemy related problem. When eating an ingredient the first effect is applied to the actor. But I can't see anything n the data structures about things like magnitude and duration. Any input on that?