enchantment stacking: wiki contradiction (in General)
I don't believe I have ever had more than one minion with the same enchantment trained simultaneously, but I was planning on doing this for my new strategy because it was my understanding that it would be a more efficient usage of XP. This partial misconception was due to the following statement in the training section of the wiki:
"For example: 1 minion with a level 10,000 Guardian Angel (GA) and another one with a level 15,000 GA has the same result as 1 single minion having a level 25,000 GA. There is (at least) one drawback for having multiple minions training the same (defensive) enchantment, and that's the fact that Dispel Magic does act on each spell individually. Using the same example from above, a Dispel Magic spell with an effect of 20,000 will completely cancel out the 10,000 and 15,000 GA, but not the 25,000 GA (the result is a 5,000 GA)."
Directly above this it is stated that "All enchantments stack; that is, the effect of an enchantment trained on two minions will be the same as if the total exp had been trained by one minion." These statements are contradictory. The XP expenditures for these levels of GA are approximately:
10,000 - 86,985 XP
15,000 - 139,485 XP
25,000 - 239,940 XP
So using the XP combination method, having a level 10,000 and level 15,000 trained separately is the same as training 226,470 XP into GA on one minion. This is approximately a level 23,775 GA. The deviation from direct level combination obviously increases with the GA levels.
The XP combination method is definitely what is actually implemented due to the fact that Jon stated "Evasion skill and related items now stack differently; instead of evasion 30 + DB 30 = 60, it now stacks the way enchantments do: in this example, the total would be 39."
50,395 (30) - 509,787 XP
91,426 (39) - 1,002,159 XP
96,836 (40) - 1,067,079 XP
102,419(41) - 1,134,075 XP
509,787 x 2 = 1,019,574 XP, placing the combined effect at (39). According to my original misunderstanding, (30) stacked with (30) would result in the equivalent to a level 50,395 + 50,395 = 100,790 (40) evasion.
Basically, I'm a bit disappointed to discover that stacking enchantments is not as potentially beneficial as I originally thought. I thought that it was meant to be more efficient due to the lower levels being cheaper to train, but this benefit was countered by the fact that you would get completely wrecked by DM without an RoS. After thinking it through, it is still a bit more efficient to stack since Corn bonuses apply directly to the enchantment level instead of XP spent, but not nearly as efficient as the wiki lead me to believe.
Anyways, I'm sure most of you were already aware of this, but I guess all I'm trying to say is that we should update the wiki to ensure that others don't waste their time coming up with worthless strategies :) I suppose I'll do it tomorrow, but now I must rest.
August 17 2006 8:38 AM EDT
What is the contradiction then?
August 17 2006 8:40 AM EDT
i read it, a bit too complicated for my CB mind, even though i've been here a while.
explaining it to a child could work :)
August 17 2006 8:45 AM EDT
Just remember that XP has nothing to do with it. Enchantments stack by level, not by XP or by effect.
August 17 2006 9:12 AM EDT
I expect that Bartjan is correct in stating that when training the same enchantment on multiple minions the LEVEL stacks. I don't have the numbers to back it up, but he's been here longer than I have, and besides, if this wasn't the case the wiki probably would have been fixed by now.
The difference is that between MULTIPLE minions training the same enchantment, and a SINGLE minion training the same skill with the aid of items (i.e. Evasion + DBs). In the former case, levels stack, making it efficient XPwise but also vulnerable to DM. In the latter case, XP amounts stack, probably because of balance issues.
At least, that's how I interpret it, in which case the discrepancy to be corrected would be the allusion that Evasion stacks the way enchantments do. In any case, I do agree that it's unclear what the exact cases are and some clarifications are in order on the wiki; I'd do them myself, but I'd prefer someone confirm how things work before I make 'corrections' that are no more correct than they were before.
August 17 2006 9:16 AM EDT
The Evasion *effect* granted by a DB is converted to an equivalent Evasion *level* before it gets added to the Evasion *level* of the Evasion skill. The resulting Evasion *level* is then converted back to a certain Evasion *effect*.
"So using the XP combination method, having a level 10,000 and level 15,000 trained separately is the same as training 226,470 XP into GA on one minion. This is approximately a level 23,775 GA. The deviation from direct level combination obviously increases with the GA levels."
This is the confusion.
25K on a single minion equals 239,940 XP.
10K + 15K on two minions equals 226,570 XP.
So for the same effect (25K GA) the two minions have saved 13,470 over the single minion.
Not a great deal, but still a saving.
Thanks for the responses everyone.
Shade: There are actually two contradictions. First, the wiki states that enchantment stacking is based on XP, but then the example claims that stacking is based on levels. I did a little test that clarified this discrepancy for me. The second contradiction concerns Jon's statement that evasion item-skill stacking is synonymous with enchantment-enchantment stacking. I'll elaborate on my test and my thoughts on item-skill stacking later in the post.
Flamey: I attempted to explain this to a small child this morning. She called me a nerd and then proceeded to give me a swift kick to the nuts.
Bartjan: Are you sure of this? According to Jon's example of 30 + 30 = 39, it seems to me that the effects are converted to levels, which are then converted to XP amounts. The XP amounts are summed and converted back into an equivalent level. Granted, I obtained 50,395 (30) from my regression analysis, so it is an approximation. The percent error would have to be quite significant to cause my calculation to be incorrect though. Looking at people's exact reported numbers, 50,395 is most likely an underestimate, so adding 30 to 30 level-wise might even put it at an effect of (41). The way I see it, the only way for 30 + 30 to yield 39 is XP stacking.
Vector: After doing some testing, I agree with you. Enchantment-enchantment stacking is calculated level-wise, and item-skill stacking is calculate XP-wise.
Here's a summary of my E-E stacking test:
Death: 957,345/784,719 (478,692) AS
Nymandus: 151,893 HP
Death cast Ablative Shield on all friendly Minions (347623)
Nymandus took no damage and had 499,516 HP afterwards. This is of course a 347,623 increase.
Death: 957,345/784,719 (478,692) AS
Nymandus: 49/50 (34) AS - 91 XP trained
Nymandus: 151,893 HP
Nymandus cast Ablative Shield on all friendly Minions (27)
Death cast Ablative Shield on all friendly Minions (347640)
Nymandus took no damage and had 499,533 HP afterwards. This is a 347,640 increase. In order for Death's AS to cast for 17 additional points, he would need 17/.3625 = 47 more levels, which corresponds to 658 XP. Therefore, this supports level-wise E-E stacking. Also, it appears that stacked enchantments combined as each minion casts, and the total stacked amount is displayed on the final caster.
I see that the wiki has been changed concerning E-E stacking, but now Jon's statement that E-E and item-skill stacking are performed the same way seems to be misleading. I suppose I may be wrong, but someone will have to explain to me how >50,000 x 2 yields less than 100,000.
August 17 2006 1:40 PM EDT
Stacking has nothing to do with XP.
AS stacks by level. AS is a difficult spell to use for an example, as it also has a multi-minion penalty. Try with another enchatment, if possible.
E-E and item stacking are performed the same but remember its the level not the number in brackets which is added. Same for protection.
I agree that enchantments stack by level. I took the multi minion penalty into account, and my test supports level-wise AS stacking. But if you look at the DB-skill evasion stacking analysis in my first post, it basically proves that this type of stacking is XP based, unless my calculations are significantly inaccurate. Please explain anything you find faulty in my DB-skill evasion stacking analysis.
August 17 2006 1:56 PM EDT
Let's respond with another question: is Evasion 96,835 (39) effectively in battle almost the same as Evasion 91,426 (39) or as Evasion 96,836 (40) ??
Evasion 96,835 (39) gives more defensive dex than Evasion 91,426 (39) while offering the same pth reduction. :P
It should also be noted that the DB portion doesn't add defensive dexterity.
August 17 2006 2:13 PM EDT
Is Pth (and similar stats, like Protection) internally an integer, or a float?
This thread is closed to new posts.
However, you are welcome to reference it
from a new thread; link this with the html
<a href="/bboard/q-and-a-fetch-msg.tcl?msg_id=001sgb">enchantment stacking: wiki contradiction</a>