2014/09/21

The truth about crit rate

There were many speculations lately about how critical works in Wartune. Since we were able to reverse-engineer most other aspects of the game's combat system, this mystery needed to be resolved as well. In Doomlord, critical formulae is pretty simple, so I suspected it's not as complicated either here.

First of all, testing crit rate needs much more testing than figuring out a formulae. And even after a lot of test, the numbers won't be accurate. I did a LOT of testing with only 3 crit values: 6848, 7948 and 11383 vs many different opponents. And I still can't give an exact formulae, but now I know how it works.

We always knew that chance to crit has another component besides the value of crit. People said the higher their lvl / BR, the more crit they needed. We also knew that vs WB, you will crit a lot even with a low critical. I suspected that this mysterious second component is the opponent's level. After my experiments, it looks like this is true.



Look at the table. With the lower crit, I had still a crit rate of approx. 80% vs world boss. What is more important, I had similar crit rate vs. Iris in sylph atoll, who are lvl 50 like the World boss. Also it is obvious, that I crited less often vs higher level opponents. The third thing I noticed that although my crit increased, the crit rate did not go beyond 80% vs the lower-level monsters (I tested only with mage, so the cap for archers can be higher). 

So based on this data, I can say:

1.) Crit rate is capped at 80% for mages. As soon as you begin critting at 80% vs lvl 80 opponents (at crit 9k? or 10k?) there is no point in increasing your critical further, unless your opponent has a crit reduce effect (like guardian angel).

2.) Crit rate depends most likely on target's level. The higher their level, the lower the crit rate.

3.) Some additional modifiers can lower the crit rate. Like a knight's -5% crit bonus, which can apply to "knight class" monsters too (you see in the table, crit vs the first lych boss, which is lvl 72 like the rest, but knight class, is obviously lower), or the modifier in moonevil.

4.) If you have determination 10 and 7k+ crit, you should use crit instead of enchanted will destroyer, the extra damage compensates even the extra damage you take.

5.) Accept that random has a huge role in crits. (Notice I had lower average crit rate in Iris atoll with higher crit?) Sometimes you always crit, sometimes you crit much less. There is no bug. It's just luck :)

So I suspect that the formulae is somehing like this:
MIN (critical * X - target's level * Y - modifiers (knight -5%, guardian angel, etc) ) , 80)%

Please let me know if you collected any data which can cofirm or contradict what I have now :) Thanks!

12 comments:

  1. Another great analysis. Invaluable. Thanks a bunch for sharing this!

    ReplyDelete
  2. Thanks for sharing this! The only thing missing (or still to be tested?) is how/if the critrate would change with different lvl attacker with the same crit value. Btw, what was the mage lvl?

    ReplyDelete
    Replies
    1. I was already level 80 when I started all these testings, so yes, the question remains if the "level modifier" is dependent only on the target's level, or the level difference. But I'm pretty sure they did not overcomplicate this, having a higher level does not increase your chance to crit. So it's only your opponent's level. The higher your level, the stronger opponents you get, so you need to increase your crit anyway.

      Delete
  3. Has anyone figured out or already know what archer's max would be ? Thanks a bunch

    ReplyDelete
  4. is there any cap for archers? (assuming knights have the same cap as mages, according to ur calculation) or simply can get %100? after some critic

    ReplyDelete
    Replies
    1. I'm not very familiar with the archer class, sorry, someone must make tests.

      Delete
  5. Just found your post and I find it fascinating. A few things of note I would like to note/ask in the interest of being thorough.

    My previous calculations showed that the critical rate for PVE and PVP was different (when attacking players v monsters of the same level). This lead me to believe that the critical rate was based on BR; however my results did not match up with this again (though better results than when comparing vs level alone).

    A guild mate of mine proposed that the crit rate was based on the difference between the attackers attack power, and the defenders defense power, up to a maximum percent like you suggest. This idea definitely holds water for my eyes in several regards. The world boss supposedly has no defenses, which means that anyone with ANY crit rate has a very good chance of critting the world boss quite frequently. In addition I recall having difficulty landing successful crits on players who are much stronger than myself.

    Anyways, since you've done so much research on the subject I thought I would get a second opinion. Thanks for any info/insight you can give. Great post and thank you for reading.

    ReplyDelete
    Replies
    1. Thanks for your comment! Players often have modifiers vs crit (some equipment, skill, guardian angel astral) I think this is the reason why pvp and pve crit rate is different a bit at the same level. But I don't think ATK/DEF has any influence on this, I had the same crit rate versus 60k PDEF warrior and some monsters. Also, sky bosses have really high PATK (3-5 has some 360ATK, for example) but he very rarely crits on me. But we will never know the exact formulae, so just can make guesses :)

      Delete
  6. I was told about 3 or 4 months ago, by a friend who has another account on one of the Chinese servers, that an update note around 4 months ago mentioned crit being at best chances if as close as possible to 20% of attack stat. Have you done any testing using different attack values to compare? I have been going by this "20% rule" for a few months and seem to get good results from it, though I have no exact numbers or statistics available. Crit success seems to drop off when boosted/buffed from potions, scrolls or WB Inspiration, unless I retune my crit to match 20% of what my attack would be while buffed. I notice in your tests that you indicate different crit values, but I do not see any varying attack values listed.

    Also, in regard to the comment by Anon amous, RE: PVE vs PVP. Talents should also be taken in to consideration, as some will increase a percentile chance for crit, while others may decrease an oponent's chances, and not entirely certain which, if any, are specific to PVP.

    ReplyDelete
    Replies
    1. Yes, I mention in point 3.) that some additional modifierc can decrease crit rate. Replied your post in the other topic too, shortly, these are my tests, to argue these numbers, someone else have to do detailed testings as well.
      I recommend that you try to count your crit rate during WB (pve) and BG (pvp).

      Delete
  7. have 20% of matk/patk at least and u should have a better crit rate i also have mage ingame , at wb we all do crit but on pvp i do better crit than the rest (also if u have archer eud u have skill for 20% increase in rate ''so gurdian angel ca kiss my ass'' as well xD)

    ReplyDelete