While reaching a point in my new game, where I can start to make some godly powerful spells by boosting my own intelligence, willpower and spell casting skills, I have found that the calculation success formula for spell casting doesn't work as advertised in the wiki, and how I assumed it was. (http://www.uesp.net/wiki/Morrowind:Spells#Spellcasting_Formulas)

I tried to make some very high powered spells that required me to predict a needed restoration score + willpower very exactly, but when I cast all the support buffs, I was still at 0 percent for the target spell I actually wanted to cast. (http://abload.de/img/mgescreenshot002o9kwy.jpg)

If you follow the formula, any point in restoration skill should provide you with 2.5% chance, so alone those 5293 you see in the screen should enable me to cast more than 12500 magicka expensive spells with the restoration skill alone. Then also about 1025 from willpower. Yet, the spell in the list there, "God Apotheose Jin 2", which consists only of 4x500-Intelligence + 4x500-Willpower effects (so only restoration skill is required), is still not castable despite only costing 12120.

If you look even more closely, you will see two other spells in the list of about 900 cost, which should be fully castable with the high willpower alone, never mind that I also got over 150% from unbuffed skills there, but it also shows me sub 100 total percent success.

...So I tested a bit, and it turns out the formula doesn't really add this "fatigue overdrive", meaning the extra 25% you get to all your skills and attributes. Normally your skill level varies from 75%-125% depending fluently on your fatigue level between completely exhausted and full. Instead, the maximum is actually 100%, and probably (have not tested this) goes down to 50% at total exhaustion. Every skill point thus only provides 2%, and every 5th instead of 4th willpower point also provides 1%.

..It is difficult to test this with other skills or attacking, because ofc. you don't see your chances written down so clearly, but maybe even there the formula is not correct, as it should have the same 75%-125% structure.

Now my first question is: Is the Wiki just out of date? I would be inclined to say no, since it even often mentions post-last-patch progress from mods, like the morrowind code patch.

And if not: Could a mod have been responsible for those changes? I am pretty proficient with all creation kits, but I could not find a game variable so far that clearly influences these hard game mechanics at the core, so I am a little doubtful of that option too. However, it wouldn't happen without precedent, as I was already a couple of times forced to custom rework some unwanted mod influences; most namely from the great "Morrowind Rebirth" which went way beyond its original mission of graphic and content improvement.(like sneaking in a rebalance of most weapons away from original lore -as if they had the authority to decide what is fair and people should want-; or taking out spell effects from the enchantment list, like sanctuary, restore health or restore fatigue! Yepp, the most classic player artifact, the deadric tower shield that bestowed near health and fatigue invulnerability, Morrowind Rebirth says "we don't like you having that". "some balancing tweaks and improvements" )

Is that another of those cases? And if, then where is the function that governs these things that they manipulated. (already searched through the scripts too btw., but this seems to be more fundamental as expected)

The https://wiki.openmw.org/index.php?title=Research:Magic is usually a more reliable source for game mechanic details in my experience. I haven't looked into spellcasting in particular though, so I don't know how accurate the UESP article is or isn't.

Thanks for the link, very interesting. The fatigueTerm there indeed provides that the casting system uses the exact same fatigue function as any other skill, so this explains my still unreliable to hit chances even up into the 60s of skill levels. (wouldn't be mad with that though, because it means weapon skill still does something later on)

I can now also see which variable does what. Previously I suspected FatigueSpellMult and -Base entries to mean there could be fatigue loss for spellcasting, but it was left out later by the designers, as these were at 0. However, it was really the only element in the list of game-variables that linked fatigue and "magic-something", so I wasn't sure. Now I can conclude they must have tampered with fFatigueBase and -Mult, which should be at 1.25 and 0.5 if I see that correctly. I will test to overwrite this and see my chances and report back.

///Edit:

Well, from all I see in the code there, those two variables were the only thing that could have explained this influence. I set my custom -end of load order- patch esp to override these into 1.25 and 0.5, but I still see the exact same chances.

I am still pretty sure that the error lies in this function, as the numbers otherwise add up perfectly to towards 1.00 for FatigueBase, which would be too much coincidence.

Could it be a load order issue here for some reason, or does it take the game time to apply some changed core values here and there? (didn't for others that I changed so far)

Are you using MCP? Often when things don't turn out as expected from the code it turns out to be because MCP changes them.

Otherwise you could post the exact stats for the spell in question and I could take a look on my end.