Reversing Morrowind's formulas

Everything about development and the OpenMW source code.
fragonard
Posts: 7
Joined: 09 Jun 2012, 21:49

Re: Reversing Morrowind's formulas

Post by fragonard »

Hi, I just recently learned about the development going on here and Signed up to maybe help out a little where I can.

I noticed that there is an error and a missing term in the enchanting formula on the wiki page. The INT and Luck multipliers are reversed and there is a enchantment chance GMST. In the last part (enchantment points *5) the 5 represents the results for a constant effect enchantment. The term should be:
enchantmentpoints (7.5/fEnchantmentChanceMult) for regular enchantments and:

enchantmentpoints (7.5/fEnchantmentChanceMult*fEnchantmentConstantChanceMult) for Constant ones.

This gives a result of 5 for CE and 2.5 for normal.
User avatar
scrawl
Posts: 2152
Joined: 18 Feb 2012, 11:51

Re: Reversing Morrowind's formulas

Post by scrawl »

Please add yourself to the wiki group ( http://openmw.org/forum/ucp.php?i=176 ) and correct the wiki.
Sadler
Posts: 24
Joined: 30 May 2012, 13:53

Re: Reversing Morrowind's formulas

Post by Sadler »

This may help with empirical formulas: http://formulize.nutonian.com/ . The best solution, of course - to reverseengineer all formulas, but it may be too slow. We may use this method to increase the speed of development. In future, of course, it may be replaced with strict formulas.
Tarius
Posts: 574
Joined: 24 Oct 2011, 19:29

Re: Reversing Morrowind's formulas

Post by Tarius »

Sadler wrote:This may help with empirical formulas: http://formulize.nutonian.com/ . The best solution, of course - to reverseengineer all formulas, but it may be too slow. We may use this method to increase the speed of development. In future, of course, it may be replaced with strict formulas.
Cool! This looks interesting, I must play with it immediately.

Edit: Aw, this is kickass. I wish I had this back in school.
It takes the data you enter, and will run through as many solutions as it can for as long as you want(or rather until its done going through the equation types you specified). It will then give you a list of the solutions(you can even watch the list while its running) and you can see how best they fit the line for the data you put in.

Have to look at this more to see how to specify for it to use certain values, but at first glace it looks like it would be great for finding formulas if you need them quickly. I think it might still take some work to get them to how MW has though.
Last edited by Tarius on 09 Jul 2012, 22:00, edited 1 time in total.
Epsilon
Posts: 40
Joined: 07 Jul 2012, 09:24

Re: Reversing Morrowind's formulas

Post by Epsilon »

Sadler wrote:This may help with empirical formulas: http://formulize.nutonian.com/ . The best solution, of course - to reverseengineer all formulas, but it may be too slow. We may use this method to increase the speed of development. In future, of course, it may be replaced with strict formulas.
You are awesome for posting this. I was just complaining to myself last night about how limited Wolfram Alpha is (and there's no way I can even afford the Student edition of Mathematica.)
wintry
Posts: 9
Joined: 10 Jul 2012, 19:08

Re: Reversing Morrowind's formulas

Post by wintry »

Hello.
I've noticed in one of the topics that some of you use MessageBox while reversing the formulas. You can improve your productivity with Morrowind Script Extender (MWSE). It allows you to use xLogMessage function in your scripts. This function "writes out formatted strings to the log file (MWScriptExtender.log)" as MWSE documentation states.
Epsilon
Posts: 40
Joined: 07 Jul 2012, 09:24

Re: Reversing Morrowind's formulas

Post by Epsilon »

wintry wrote:Hello.
I've noticed in one of the topics that some of you use MessageBox while reversing the formulas. You can improve your productivity with Morrowind Script Extender (MWSE). It allows you to use xLogMessage function in your scripts. This function "writes out formatted strings to the log file (MWScriptExtender.log)" as MWSE documentation states.
This was suggested to me, actually, but thanks!

Because of all the automation, all I really needed to do is punch some numbers into a spreadsheet once I'm done a round of tests. The pause actually gave me an opportunity to examine the results up to that point and see how the new piece of data fit in. Multitasking, in a way.

Depending on what I decide to test next, I very well may use MWSE to log results next time though.
User avatar
Zini
Posts: 5538
Joined: 06 Aug 2011, 15:16

Re: Reversing Morrowind's formulas

Post by Zini »

I would like to request a status report on the formulae research, particularly on the acrobatics/athletics stuff, alchemy and ingredients use. So far only the skill gain formulae made it into the wiki, which means the rest is still work in progress. How much progress? Any ETA?
Epsilon
Posts: 40
Joined: 07 Jul 2012, 09:24

Re: Reversing Morrowind's formulas

Post by Epsilon »

Zini wrote:I would like to request a status report on the formulae research, particularly on the acrobatics/athletics stuff, alchemy and ingredients use. So far only the skill gain formulae made it into the wiki, which means the rest is still work in progress. How much progress? Any ETA?
We've got a thread running under the Development forum that's got info on acrobatics here: http://forum.openmw.org/viewtopic.php?f ... 0&start=20.

Short version, I've got... 1/4th of the formula. We're taking a lot of data points, and it's definitely not a linear formula. Makes it a bit harder to figure out.

ETA is really unknown - I've got my processor crunching away at the data points, but when it'll stumble across the -right- formula to fit the data is anyone's guess. After we've got the second half of the acrobatics formula (minus encumbrance/damage,) it should be usable in-game as a stopgap.
User avatar
psi29a
Posts: 5362
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: Reversing Morrowind's formulas

Post by psi29a »

Epsilon wrote:ETA is really unknown - I've got my processor crunching away at the data points, but when it'll stumble across the -right- formula to fit the data is anyone's guess. After we've got the second half of the acrobatics formula (minus encumbrance/damage,) it should be usable in-game as a stopgap.
How much do you need and what kind of program is it? Racks of Xeons E5s should be good for something. 8-)
Post Reply