UESP Forums

The discussion page was getting a bit difficult to follow, so here's what's going on. I've found a formula for the duration/strength of a potion/poison, what we're lacking is how to calculate the septium value of a mixture. Here's what we currently have for information:

Links to discussion pages where this information can be viewed in a bit more detail:
Alchemy - http://www.uesp.net/wiki/Skyrim_talk:Alchemy
Alchemy Effects - http://www.uesp.net/wiki/Skyrim_talk:Al ... n_value.3F


Terms: These are the terms I've come to use as it helps to keep confusion down.

• Magnitude - This is the value from the EFIT information in the INGR section of Skyrim.esm, typically values range 0-5.
• Strength - This is the strength of an effect as it is on a potion.
• Duration - The duration of an effect as it is on a potion.
• DoI - The duration specified in the INGR records, in the EFIT values.
• TSV - A Tab Separated Values list of the EFIT values directly dumped from Skyrim.esm. Link is on this discussion page http://www.uesp.net/wiki/Skyrim_talk:Alchemy (can't paste it on the forums)
• max ingredient - The ingredient chosen to be used to calculate strength/duration/septium value.

What we currently know (or feel strongly that we do know, but could be needing tweaking):

• The EFIT information in INGR/ALCH records are: float (magnitude), uint32 (area - always 0), uint32 (duration)
• When calculating magnitude and duration only one ingredient is looked at per effect. That ingredient typically has the highest magnitude for that effect of those selected.
• When two ingredients have identical magnitudes, the duration is then checked and the higher is the one used for calculations.
• When the max ingredient has a magnitude of 0 or 100 the formula is used on the duration, otherwise the formula is used on the magnitude. ** Note: I haven't come across anything that shows both strength and duration increases, so this may need to be tweaked.
• (SPECULATION) If both magnitude and duration is the same on all ingredients chosen, a number which determines the septium value of the potion is then reviewed.
• One formula is used to determine strength and duration. (formula posted and explained below)
• When two (possibly more) ingredients have a duration of zero, the magnitude is reduced by 1 in calculating the strength and septium value. - Incorrect (Fluff)
• We know that the magnitude and duration from the max ingredient are taken into account along with alchemy level when determining the septium value of the mixture.
• The septium values have a distinct pattern of increasing every 10 levels starting with level 5, however it skips levels 25, 65, 105, ...

Formula for calculating strength and duration:

Code:

(floor(A/5)*.1+4)*B

Where:

• A = Alchemy level
• B = The magnitude or DoI from the max ingredient.
• C = Bonus from enchanted armor
• D = Number of points in Alchemist perk multiplied by .2
• E = Number of points in Poisoner perk multiplied by .25 *
• F = Number of points in Benefactor perk multiplied by .25 **
• G = Number of points in Physician perk multiplied by .25 ***

* - Only applies to poisons.
** - Only applies to potions.
*** - Only applies to health and stamina effects.
C,D,E,F and G will be applied to the formula when more is known on them



Final notes:

• If you're going to post information that you feel might be beneficial to finding the formula for the septium values of a certain mixture, please be sure to include: Alchemy level, perks, names of the ingredients mixed, your results. Also supplying the EFIT information if you have it would save us some time while reviewing your information.
• I have written a program that can quickly show you the magnitude and DoI based on the TSV, if you're interested in it send me a message as I can't link it or attach it here.
• If there is anything I forgot to mention from the discussion page please let me know and I'll add it to this post.

I didn't feel this thread belonged in the "Mod-Skyrim" forum because we're not really modding anything. If I was incorrect in this assumption I will recreate this thread if it can't be moved.

I didn't even know uesp had a forum.

Anyway, here's some more discussion. Note that this is with Alchemy 100, 5 levels of Alchemist, Physician, and Benefactor.
I think it's more likely that the following statement is true: "When calculating the value and strength of a particular effect, only one ingredient is looked at. That ingredient typically has the highest magnitude of those selected for that effect." For example, consider Crimson Nirnroot + Nirnroot. The Damage Health effect is the typical Nirnroot Damage Health, but the Damage Stamina is Crimson Nirnroot's Damage Stamina.
Nope. Nirnroot + River Betty is typical River Betty Damage Health, but both have duration of zero. Further, neither Deathbell + Nirnroot or Beehive Husk + Thistle Branch is an example of both ingredients having 0 duration; Deathbell has a normal duration of 1 for Damage Health and Beehive Husk has a normal duration of 60 for Resist Poison.

In general, the highest magnitude (or duration if magnitude is tied) does determine the strength of the potion. The effect that is most wonky is Damage Health. Here, the ingredient priority is [magnitude,duration]: Jarrin Root [200.0,1], River Betty [5.0,0], Nirnroot [2.0,0], Crimson Nirnroot [6.0,1], Deathbell [3.0,1], and everything else [2.0,1]. I think it's important to note that the value (which I think we're most interested in) steadily decreases down the ingredient priority, but both the magnitude and the strength do not. An obvious question: why is the Crimson Nirnroot Damage Health effect worth less than the Nirnroot or the River Betty Damage Health effect, despite having a higher magnitude?

The Beehive Husk + Thistle Branch is also an anomaly. Instead of using Thistle Branch Resist Poison [3.0,0], it uses Beehive Husk Resist Poison [2.0,60]. The normal Resist Poison is [4.0,60]. Whenever either of those two ingredients are combined with a normal Resist Poison, the normal Resist Poison effect is used. This makes more sense than the Damage Health effect, since I could see the duration being a trait that could cause an ingredient's effect to be dominant. With the Damage Health, it makes no sense that the Nirnroot Damage Health [2.0,0] would be dominant over the normal Damage Health effect [2.0,1].

I believe the Ravage Stamina effect is also interesting. As expected, the Deathbell Ravage Stamina effect is dominant when combined with other ingredients (due to having a tied magnitude and a larger duration). Perhaps not as expected, the strength of the potion is exactly the same, with or without Deathbell. Only the value is greater.

By the way, here's the formula with the perks added in.

For positive effects:
(floor(A/5)*.1+4) * B * (1 + Alchemist value) * (1 + Benefactor value)
For positive Health/Magicka/Stamina effects:
(floor(A/5)*.1+4) * B * (1 + Alchemist value) * (1 + Benefactor value) * (1 + Physician value)
For negative effects:
(floor(A/5)*.1+4) * B * (1 + Alchemist value) * (1 + Poisoner value)

The Alchemist value is the number of Alchemist perks you have multiplied by 0.2.
The Benefactor, Physician, and Poisoner value is 0.25 if you have the perk, 0 if you don't.
The final magnitude is rounded to the closest integer.

I'm going to bed so much of this I'll have to look into tomorrow, but for a quick reply:


This may be a fluke with the Damage Health effect, have any other instances of this happening?


I'll revise my speculation in the main post. The Beehive Husk/Thistle Branch was supposed to be a general advisory because any other combination of ingredients would result in a more potent poison, I must've gotten backwards somewhere with it.


That's the question we're trying to answer :) How do the septium values get calculated? I suspect the answer lies in the ENIT information of either the INGR records or the ALCH records, but currently I haven't been able to look too deeply into this suspicion. It could very well be that some calculation is made and form that calculation it pulls a specific ENIT in the ALCH information that we're unaware of too. Many possibilities at this point.


I'm not so sure on this, a quick glance at the tool I wrote for magnitudes/DoI display shows that they're all a magnitude of 2, which means to me that no matter what ingredients you mix the strength will be the same, but if you have Deathbell the duration (thus septium value) will increase. Am I misunderstanding you here?

Yes. Crimson Nirnroot + Jarrin Root produces a poison with Jarrin Root's Damage Health and Crimson Nirnroot's Damage Stamina. I know it still uses the Damage Health effect, but as far as I can tell, those are the only two combinations of two ingredients that will result in a potion where the effects are split between two ingredients rather than all coming from one dominant ingredient. Really though, it shouldn't be surprising that it's per effect. After all, you could match up two ingredients for one effect, and then add one more to match up for a second effect. If the dominant ingredient for one effect doesn't have the other effect, of course the data about that effect must come from a different ingredient.

I think we're saying the same thing. What I'm saying is odd about it is that both potions appear to have exactly the same effects, but one is worth more, just like Nirnroot Damage Health. I know this is due to the DoI, but this information is hidden from the user at this point. From a user's point of view, you can make two identical potions, one of which is inexplicably worth more. In addition, this means that the 'value' formula has to take both magnitude and DoI into account, not just whichever one is the determining factor of the dominant effect.

To me, that means that there must be more data on the effects somewhere. I can't think of a single mathematical way to arrive at a value such that the following two things are true: 1) Nirnroot Damage Health is worth more than Ectoplasm Damage Health, and 2) Giant's Toe Fortify Health is worth more than Wheat's Fortify Health. In both cases, only the DoI is different, yet they go in opposite directions. Nirnroot's Damage Health has a smaller DoI than Ectoplasm's, and is worth more. Giant's Toe Fortify Health has a larger DoI than Wheat's, and is worth more. Does not compute.

After waking up I re-read what you said, and I agree. I tested it in-game too, and you're correct. First post now reflects that



Yep, that's why I keep mentioning extra values "hidden" in the .esm.

Something I just noted while doing normal play, there seems to be a certain synergy with my restore magicka potions. I'm tossing this up to "coincidence" for now, but I thought it was noteworthy. These potions were crafted at various levels while playing normally through the game, unfortunately it doesn't seem that there's such a synergy with my health pots, here's my numbers:

Strength = 104, Value = 99, Difference = 5
Strength = 103, Value = 98, Difference = 5
Strength = 73, Value = 68, Difference = 5
Strength = 32, Value = 27, Difference = 5

Like I said, probably coincidence because removing the 'bonus' from alchemy level alone would no longer make this true, but interesting none the less. I take this as the game taunting me :(

More minor quibbles.

Code:

(floor(A/5)*.1+4)*B*(1+C+D+E+F)

I'm not sure about how enchants modify the formula, but I do know that you have to multiply each perk separately to arrive at the correct value. Note that (1+D) * (1+E) * (1+F) where the user knows 5 ranks of Alchemist, Benefactor, and Physician is equal to 3.125, where as (1+D+E+F) is only 2.5. Also, Physician applies to positive effects. It's Poisoner that only applies to poisons.

Removed from the formula until we have numbers backing up any claim.

Take a look at the Armor Rating section on the Armor page. The perks are combined in that formula in exactly the same way (which is why I tried it that way first). In addition, the effect strengths check out when I use the formula with the perks multiplied together. The final strength is rounded to the nearest integer, but it hasn't failed on any effect that I've checked.

this stuff is gold! keep up the good work guys

Here are a couple of value formulas for different effects:

Waterbreathing: value = floor(duration * 7/2 - 6)
Alteration: value = floor(magnitude * 13/6 - 14/3)

These were derived from a character with zero perks starting with zero Alchemy and gaining one level of Alchemy at a time. Once I noticed a pattern, I skipped ahead in "safe" increments, verifying no change before a change. To derive Alteration's 'b', I had to go up to 144 Alchemy. It's possible that both 'b's are still wrong, but accurate enough to not cause a septim difference for Alchemy values between 0-100. At higher values, values may be off by a septim.

What I took from this exercise:
1) Obviously, each effect has a different value equation. While I expect each of them to have a form of floor(ax + b), I also expect 'a' and 'b' to vary between differing effects. For example, Fortify Destruction likely has a very similar (or the same) value equation as Fortify Alteration. However, I would not expect Damage Health to have a similar equation at all.
2) Once again, there has to be more data in there somewhere. I used these two effects as starting points because all of the ingredients that have these effects have the same EFIT data and because they are two of the rarer effects. Deriving a value equation with an effect that has ingredients with varying EFIT data would be far more challenging.

Neither of those are right. I tried testing with full Alchemy perks at various levels of Alchemy skill. It appears the value equations are non-linear. With the creation kit scheduled to come out in January, I have no problem just waiting for it.

no new information yet?