Thursday, April 19, 2012

The character screen. How bad is it?

We all know that the charcater screen is an unreliable source of information. But how unreliable is it? In what condition does it display properly?

1. damage

2. attack rating/chance to hit

3. defence rating/chance to get hit

4. block

5. resist|||4 and afaik 3 is 100% correct. 2 is mostly correct but doesn't take into account any defense reducing mods on your gear or skills (like Eth runes with -25% target defense). The AR bonus you get from the amazon bow skills doesn't work even though it's accounted for in the LCS. Resists are correct, except for one case: When you die you lose the bonus you got from the Anya scrolls until you create/join a new game, but this isn't shown in the char screen. That might be all for 2-5.

Damage is very rarely 100% accurate but comes pretty close most of the time. I think it screws up poison damage. Straight +damage on weapons gets added to your displayed damage after enhanced damage instead of before. For example if you've got Grief in a weapon with 30 damage and 500% ED the LCS will show:

(30 + 500%) + 400 = 580

while your actual damage is:

(30 + 400) + 500% = 2580

I think +min and +max damage is also calculated wrongly but I can't recall how. Stuff like lightning mastery and +% poison damage gets applied twice only when a melee attack is used, but this is wrongly shown to be the case for ranged attacks as well.

That's all I can remember now...|||What about throwing and Wake of Inferno? Im also interested in arctic blast and normal inferno?|||Double Throw's displayed damage incorrectly applies the sum of +% Damage from Throwing Mastery and the Double Swing synergy bonus before +% Damage from all other sources, when it's simply added to other +% Damage: the result is that displayed damage can be up to three times higher than actual damage.

Inferno and Arctic Blast displayed damage is half actual damage. I haven't yet checked to see if this applies to Wake of Inferno as well.|||Is Inferno/Arctic Blast half or a third? Believe WoI is also bugged with the other two.|||When I tested it, both Inferno and Arctic Blast's actual damage per second was half displayed damage per second.|||Thx for the info, believe it was on here that I heard it was 1/3 of the displayed damage.|||Quote:








When I tested it, both Inferno and Arctic Blast's actual damage per second was half displayed damage per second.




I'm curious: How does exactly one test that? How can you know for sure that the listed damage is actually halved? What kind of tests did you do?|||From post #42 of the Inferno & Arctic Blast Bug topic posted in the Amazon Basin's Technical Discussion Archives (earlier posts report different tests with the same result):


Quote:








I've also confirmed that Arctic Blast and Inferno apply damage every other frame, but using a different testing method. I'd also like to point out that there isn't necessarilly a bug, at least as far as the mechanics are concerned: it could just be that, like many other skills, the displayed damage per second is calculated incorrectly.

When looking at Skills.txt, I noticed that the mana cost for level 1 AB was 0.375 mana, whereas 4 mana per second was the displayed cost. 481 mana is sufficient to regenerate 4 mana per second, but sustained casting still resulted in mana loss. 563 mana was required to ensure no mana was lost, which regenerates 4.6875 mana per second: this is consistent with the mana cost being applied an average of 12.5 times per second (0.375*12.5 = 4.6875).

I gave a Druid level 1 AB and 3,000 mana, regenerating 1 mana per frame, and edited Skills.txt to increase level 1 AB's cost to 4 mana (2 mana per frame) so that all mana would be consumed in 3,000 frames (2 minutes): I also increased startmana from 6 to 60, and ensured level 1 AB applied 1 cold damage.

I cast level 1 AB on Immune to Physical Zombies with Cold Resist 0%, no damage regeneration and 1,501 life. When mana ran out, it took two points of cold damage to kill them, indicating that 1,499 (1,501-2) cold damage had been applied by AB: this is consistent with AB applying cold damage in the second frame after casting, and every other frame thereafter. I performed a similar test on Inferno, and the results were the same.




From post #44:


Quote:








Incidentally, when I modified AB so that it applied poison damage for a single frame, the targets flickered green, indicating that poison damage and length were indeed being applied every other frame (flickering happened occasionally with cold length, but based on past experience poison length is a much more reliable indicator).




Finally, from post #45:


Quote:








Before I forget, there's something else suggesting that Inferno applies its damage correctly and it's displayed damage which is incorrect.

The DamageRate column of Missiles.txt affects how magic damage reduction applies to missiles applying incremental damage: since damage is applied in many small amounts, the effectiveness of MDR needs to be reduced so that it isn't too easy to counter incremental damage. Skills whose missiles apply damage every frame (like Immolation Arrow's immolationfire missiles) tend to have an entry of 41, which means MDR is multiplied by 41/1,024 (approximately 1/25). However, Inferno's missiles have an entry of 82, which means MDR is multiplied by 82/1,024... or approximately 2/25. Based on the other missiles with an entry in this column, there seems to be a correspondence between the entry and how often they apply their damage: see Tommi Gustafsson's MDR vs. Frame-Based Attacks for more details.

This isn't conclusive because Arctic Blast's missiles have an entry of 41, which could suggest that it was indeed meant to apply its damage every frame. However, since Arctic Blast was added later than Inferno this may be a mistake influenced by displayed damage.|||Has it been corrected that for #2 AR/chance to hit doesn't take into account the level difference between the player and the monsters for chance to hit? I did some rough testing a while ago which indicated it hadn't been fixed and I don't remember seeing anything that addressed that.

No comments:

Post a Comment