Jul
01
2009
10:10 pm

Damage Boost Calculations

Showcasing how terrible Cave’s design is.

There are various categories of damage boost:

Elemental/Affinity Boost

The most common and easiest form of damage boost to get.

Elemental tarots, percentage modification etc… all belongs to this category.

There’s a hardcap of 100% for this category of boost.

The toggle from control magic belongs to this category too, so for those using control magic, the cap is basically 90%

Racial Boost

Basically anything that adds damage against a race or family of devils.

Most of these boosts come from talismans and anti-racial soul enchants like Gyuuki.

Movement/Action Boost

Anything that adds damage to skills of a particular movement type.

The +10% shot/rapid/rush/spin rings and soul enchants like Nozuchi belongs to this category.

Stuff like Seiten Taisei’s (Sun Wukong) crystal and the 4 King bonus, which adds damage to skills that depends on weapon for damage type, falls in this category too.

General Boost

Probably the rarest kind of boost on Aeria now.

Anything that adds damage to all physical, gun or magic attacks.

Drill comp and fragrances belong to this category.

Crystals like Uriel which activates at a certain HP threshold belong to this category as well.

LNC Boost

Anything that adds damage to a particular alignment.

Stuff like Claymore, Harakiri Saber etc… belongs to this category.

Enhance

The effects from enhance belong to its own category.

Oddities

There are a LOT of things that are weirdly programmed, or programmed in such a way to get around Cave’s own limitation.

A good example is Kerberos/Orthos. Even though it adds +50% fire damage to all fire based skills, it’s actually considered as a movement boost rather than an elemental boost. Thus, it isn’t subjected to the 100% elemental cap.

Another example is the LNC glove, which add 4% damage to Chaos/Law target. It’s actually considered as a general boost instead of an LNC boost.

Lately, there are new necks that are added to Cave’s e-mail only lottery, and these necklace cast a buff that add +100% damage to the next fire/ice/lightning/zan skill you use. This is actually considered as a enhance boost and is once again, not subjected to the 100% elemental cap.

2 days ago, Cave added an “improved” rider set, which adds 20% fire/ice damage, and if the entire set is equipped, you gain an additional 10% fire/ice damage, and +1 to the activation stack of all fire/ice skills. The 10% additional fire/ice damage is considered an enhance boost.


Now, there are various theories on how damage boost stack, and none of them perfectly match the damage done in-game.

The closest theory is:

Elemental boost belong to its own category.
Racial and movement are counted as one, so they’re added together first.
General boost belong to its own category
Enhance and LNC boost are counted as one, so they’re added together first.

Each of these boost are then multiplied together:

tl;dr:

Elemental * (Racial+Movement) * General * (Enhance+LNC)


Improving Damage

The best way to improve your damage is to balance out your boost, since

1.1 * 1.1 * 1.1 * 1.1 => ~1.46 (10% boost for all category)
1.4 * 1.0 * 1.0 * 1.0 => 1.4 (40% boost to one category)

1.46 is definitely better than 1.4.

Before Sentinel/LNC China etc… appears, your general boost is typically low, because only fragrance and drill comp grants it.

Most will probably run around with something like this:

45% fire (20% weapon + 15% mod + 10% extra)
80% movement (50% Kerberos/Orthos, 10% ring + 10% (polt, nozuchi, mithra), 10% angel set, ignoring the wings for a nice round number)
40% racial (20% charm, 20% enchant)
10% general (Drill comp)
50% enhance

This will give:

(1+45/100) * (1+(80+40)/100) + (1+10/100) + (1+50/100) = 1.45 * 2.2 * 1.1 * 1.5 = ~5.26

Ignoring parts, if we switch the 10% from angel with 10% general damage from LNC china etc…

45% fire (20% weapon + 15% mod + 10% extra)
70% movement (50% Kerberos/Orthos, 10% ring + 10% (polt, nozuchi, mithra))
40% racial (20% charm, 20% enchant)
20% general (Drill comp, LNC China)
50% enhance

This will give:

(1+45/100) * (1+(70+40)/100) + (1+20/100) + (1+50/100) = 1.45 * 2.1 * 1.2 * 1.5 = ~5.48

The total percentage is the same, but the final calculation shows an actual 20% increase in damage.

This is one of the reason why damage is inflated on the JP servers compared to the EN/TW one.


Spreadsheet

Here is a VERY messy Excel spreadsheet that I personally use for damage simulations, as well as keep track of the boosts I have.

The damage is off by ~100-200 damage, but for most purposes, that’s accurate enough.

Use it at your own risk.

34 Comments »

  • Anna says:

    “This is one of the reason why damage is inflated on the JP servers compared to the EN/TW one.”

    Their turn will come soon, I’m still amazed by Angel Sets to Hakkas transition. But of course, there’s this super American player in JP server whose gears is soooo high end compared to us. We can never match him. tee-hee

  • toadGX says:

    For the magic control toggles how much % would you say they add?

    • 守護霊母Finella says:

      +10% damage, -20% cast time to skills that does that toggle’s damage.

    • Pearz says:

      “The toggle from control magic belongs to this category too, so for those using control magic, the cap is basically 90%”

      Read plox?

      • toadGX says:

        Thank you Finella for the nice response, and to you Pearz for pointing that out. I must have missed it while reading through.

  • Tanarin says:

    Not completely related to this, (OK, so it isn’t related at all really,) but are enemy weaknesses figured with the same basic concept, or is that handled completely differently? Also, what about damage decreasers, (For example the Lord of Armor toggle,) where are those figured in the calculations? What about if your opponent can decrease damage they take, is that figured later or before?

    • 守護霊母Finella says:

      I’ll come back to this when I’m more awake (4am now, just finish my work)

      Resistance/weakness = flat multiplicative modifier.

      It doesn’t matter if you multiply it at the end or the beginning, since it’s multiplicative.

      Damage reduction are calculated after resistance etc…

      By right, defense are multiplicative, but Cave screwed up somewhere along the line and fragrances/regal 4-0 shared the same category, aka making them stack additively.

      Instead of (1-0.5) x (1-0.5) x (1-0.5) = 0.125, you get
      (1-0.5) x 1-(0.5+0.5) = 0

      • Tanarin says:

        Hey, thanks for the info. Now what about the other end of spectrum, the -100% damage penalty that 4-0 RP adds? I assume it have to be something like a general “boost.”

        • 守護霊母Finella says:

          Two theories, general or enhance.

          Ignoring the pun, the more generally accepted theory is that it’s a general boost penalty.

  • Pearz says:

    I was under the impresion they stacked because it worked same as the Damage formula. Thought they both belonged to general Modifier Category. At least thats the assumption I made when I was testing how the damage formula worked a couple weeks ago…..

  • Goro says:

    Hi Finella,

    You’re really knowledgeable about megaten. I was wondering if you enlighten me about critical rate.

    Does having more CRT and Luck affect Critical hit rate? Does it matter much if you have 140 CRT and 70 luck?

    Also for an Avalon, is it worth it to add CRT to Slot 2 or Penetrate?

    Thanks,
    Goro

    • 守護霊母Finella says:

      At this point, no one is really confident in how luck/crit rate/crit prevention is calculated.

      The general accepted theory is that luck adds a certain amount of crit rate and your crit rate vs your target’s crit prevention is used to calculate the final crit chance.

      Luck seems to affect crit more than raw crit rate.

      Result of some random tests:

      Against a pixie (25 crit prevention)
      1 Luck : 2%
      60 Luck : 11%

      Against cait sith (45 crit prevention)
      1 Luck : 1%
      40 luck : 2%
      70 luck : 5%
      100 luck: 7%

      Against a pixie (25 crit prevention)
      60 luck: 6%
      60 luck + 10 crit: 6%
      60 luck + 1-0 attack’s toggle (+300): 15%
      60 luck + 1-0 attack’s toggle (+300) + 10 crit: 15%
      60 luck + 9-0 weapon knowledge’s passive (+20), 1-0 regal presence’s toggle (+20) and 1-0 attack’s toggle (+300): 21%
      60 luck + 9-0 weapon knowledge’s passive (+20), 1-0 regal presence’s toggle (+20) and 1-0 attack’s toggle (+300) + 10 crit: 21%
      100 luck + 9-0 weapon knowledge’s passive (+20), 1-0 regal presence’s toggle (+20) and 1-0 attack’s toggle (+300) + 10 crit: 40%

      The final crit chance is very streaky and it’s really hard to determine the exact calculation.

      Luck and crit rate does matter, but emphasis shouldn’t really be focused on them, at an expense of p/rng/m.attack.

      As for Avalon, penetrate, which I assume is the affinity used by rifle, would be better, unless you enchant an element to your Avalon, and changed it to do elemental damage.

      • Pearz says:

        Question: About WK, the 9 classes would give a passive bonus too, and was the + amount for WK and all the toggle add amounts pulled from client files?

        • 守護霊母Finella says:

          Values for the weapon knowledge passives/attack toggles are pulled from the client files.

          As for increase based on rank/class of weapon knowledge, it, along with the rest of how crit works is just a pain in the arse to test. Look at the first example and last example:

          60 Luck : 11%

          60 luck: 6%

          This is across 500 hits. First example with 0 weapon knowledge, second example with 9-0, yet, 5% less crit.

          Like I mentioned before, crit is very streaky and quite hard to test, partially due to how the random numbers are generated in Imagine.

          I need a significantly larger sample size (1k-2k hits) before anything worthwhile can be concluded.

          • Pearz says:

            Just wondering where you got base numbers for added crit, since rather take sample and try and figure it out myself

      • MaxCherry says:

        “60 luck: 6%
        60 luck + 10 crit: 6%”

        this +10 crit added, do you remember what it was from? e.g kali tarot, weapon with 10 crit stat?

        thanks

  • Goro says:

    Thanks Finella!

  • anoymous says:

    hehe sorry for too much question again, so basically which is better +1 crit or +1 luk? since you said luk seems increase more? oh and if its used with high level demon of course.

  • whoopie says:

    Is there a ‘movement’ category percent cap?

  • whoopie says:

    150% movement cap~ for sure it’s logged! But I am sure it’s different than my home IP as you noticed.

  • whoopie says:

    Really the caps are interesting~ I actually learned something new here. Who would have known!

  • MaxCherry says:

    “Crystals like Uriel which activates at a certain HP threshold belong to this category as well.”

    i know old post and possibly outdated…
    dunno how it is for jp server but for aeria, dying lights are coming affinity not general

  • SK says:

    Is this information still relevant? or have the standards for each boost changed?

  • joshirosama says:

    i got a question which would be better for shotmages?
    useing sirus SS on face and loseing combo or keep the the combo?

RSS feed for comments on this post. TrackBack URL


Leave a Reply

Submit Comment