PDA

View Full Version : Effective Damage per Second (eDPS) as function of enemy armor class



EightFeathers
04-06-2011, 08:51 AM
Since I´ve only used bows so far, I often wonder why so many birds carry heavy X-bows, and why pink X-bows are sold so quickly at respectable prices (the X-bow used in the present example was sold after 5 min after I found in the shop it for 7999). Using the excellent data from Physiologic (http://www.spacetimestudios.com/showthread.php?16817-Guide-to-Advanced-Mechanics-in-PL-DPS-Crits-etc.), I therefore compared effective DPS (eDPS) of a lvl. 30 X-bow and bow, as a function of enemy armor class. I assumed a hit-rate of 85,6 % with both weapons (see Physiologics post), and also considered critical hits since armor-substracted damage is less relative to total damage when scoring a critical hit.

Edit: I was wrong on this point. Armor is actually subtracted twice from critical hits, crits does therefore not help to overcome high enemy armor.

Also, in this example Focus skill is active increasing crit to 35 % (approximate number!).

Bows comapred are Jewel Thiefs Heavy Crossbow (pink, lvl. 30, damage 40-54, speed 1.4, DPS of weapon only: 33,6) and Jewel Thiefs Recon Bow (pink, lvl. 30, damage 25-34, speed 0.7, DPS of weapon only: 42,1). To show how eDPS was calculated relative to enemy armor, I exemplify with enemy armor = 10:

http://i56.tinypic.com/2ik8xv7.jpg

Using this "algorithm", I plotted eDPS against enemy armor for the two weapons:

http://i52.tinypic.com/2q3rspz.jpg

I finally get it. Despite higher DPS for the bow, the two weapons perform equally in terms of eDPS already at enemy armor = 20. After that, the X-bow performs better. Therefore, the X-bow is the logical choice for low-lvl archers in high-level dungeons!

Ellyidol
04-06-2011, 08:55 AM
Excellent post proving why base-damage is better than DPS :)

Major
04-06-2011, 08:59 AM
Wow neat and organized and i can actually understand it. Good work, btw welcome to the forums.

drewcapu
04-06-2011, 09:20 AM
Cool.

Can't wait to see more L30s in BS!!!

;)

FluffNStuff
04-06-2011, 10:04 AM
Now add 40 damage to each item for a bear rage (which is 60 2/3 of the time). Note it is added to damage, not dps, which makes it benefit faster weapons.

WhoIsThis
04-06-2011, 07:36 PM
Precisely why slower recharging, but heavier weapons are preferred.

The optimal recharge time is probably 2 seconds for dealing with mobs and slower for dealing with bosses (as they have very high armor). You don't want slower than 2 seconds against mobs because there are many targets and this leads to overkill. By contrast, against bosses, you want a weapon that deals as much damage as possible per hit for maximum effective damage.

Physiologic
04-07-2011, 02:47 AM
A caveat to a caveat! - Effective DPS works clean on paper but take into account armor debuffs that your party can do. On paper, you can assume there are absolutely no enemy being armor debuffed, so damage seems much more favorable to look at than effective DPS.

This is hardly the case on the field, where you have bears, archers, and mages constantly debuffing armor from mobs of enemies at a time. At many times, enemy armor may go way below the armor threshold (in your example, 20) so that the Recon bow reigns supreme, and effective DPS is actually better than damage. Of course, enemies die way too fast for us to see any discernible difference between the two weapons, but this can be seen a bit better when you're at bosses.

In any case, this is a very awesome graph that easily exemplifies the significance of effective DPS. I may steal it later...with your permission (and credit you) of course :)

EightFeathers
04-07-2011, 03:11 AM
In any case, this is a very awesome graph that easily exemplifies the significance of effective DPS. I may steal it later...with your permission (and credit you) of course :)

Thanks, you have all permissions! However, reading one of your posts this morning, I realize that armor is substracted twice from critical hits. This changes the graph a bit:

http://i56.tinypic.com/29oj11f.jpg

With the adjusted calculations both weapons perform similar already at enemy armor = 10, X-bow being more powerful after this point. Although I get how important armor debuffs are in practical play, I do not think enemy armor in high-level dungeons are debuffed much below 10. Therefore, I think my point is still valid for low-level characters in high-level dungeons (assuming calculations are correct :) )

If anyone is interested in the whole set of data (in case I made more mistakes ;) ), look here (http://i52.tinypic.com/qnu1hd.jpg). I really would appreciate if someone has the time to verify the calculations, send me a message if you want the Excel source-file by email.

WhoIsThis
04-07-2011, 06:40 PM
This is hardly the case on the field, where you have bears, archers, and mages constantly debuffing armor from mobs of enemies at a time. At many times, enemy armor may go way below the armor threshold (in your example, 20) so that the Recon bow reigns supreme, and effective DPS is actually better than damage. Of course, enemies die way too fast for us to see any discernible difference between the two weapons, but this can be seen a bit better when you're at bosses.

Not all PUGs will debuff (As an example, it may surprise you but there are some level 50 birds with as few as 1 point in break armor. I have even met a level 55 bird who claimed that they had 0 points in break armor, having maxed out a lot of other skills.). Scary, but true.

That said, even debuffed, targets are still going to have armor - how much will depend on each target. Remember, mage and bear debuffs don't stack, and only break armor stacks for birds.

Kossi
04-07-2011, 06:56 PM
lv30 recon bow?

EightFeathers
04-08-2011, 03:15 AM
lv30 recon bow?

Your point being you want me to compare some high-level bows, more relevant for high-level dungeons? I found two comparable weapons in my inventory, a ballista and bow of the nimble (greens). This is what I found:

http://i56.tinypic.com/2e1c6s7.jpg

The two weapons perform similarly at enemy armor = 70. After that, the ballista is better. In AO2 Physiologic has data indicating that mobs have armor of 70 or less, making bow the better choice. Bosses might be another matter entirely, though. Even after de-buffing. I will try to find some de-buffed armor-values of bosses in TotGO tonight :p

Ellyidol, Major: Thanks for welcoming me!


Now add 40 damage to each item for a bear rage (which is 60 2/3 of the time). Note it is added to damage, not dps, which makes it benefit faster weapons.

That would really seem to favor weapon speed over damage!


The optimal recharge time is probably 2 seconds for dealing with mobs and slower for dealing with bosses (as they have very high armor).

If you want me to compare a weapon speed 2 sec weapon with another, just give me the stats.

shmoo
04-08-2011, 03:38 AM
Is there a thread that has the details of the enemies? Like their armor value? For mobs and bosses. That would be another nice project if such a thread does not exist.

Some sort of Pokedex but for enemy monsters lol.

EDIT: That way we can have an idea of what armor we are up against when testing what you have found. :D

Does crit play a factor?

EightFeathers
04-08-2011, 04:22 AM
Does crit play a factor?

In the beginning I thought so, since I believed armor was only subtracted once from critical hits. This favored high DPS / low damage weapons, since a crit would deal damage anyway to mobs with high armor. But in a Physiologic thread I saw data that armor is subtracted twice from crits. For this reason, what weapon is most effective in therms of eDPS seems actually not to be affected by critical hits at all.

WhoIsThis
04-08-2011, 04:56 PM
If you want me to compare a weapon speed 2 sec weapon with another, just give me the stats.

I was talking about a hypothetical weapon, although I'm sure something exists. It may also help in that slow hard hitting weapons offer more skill damage, except for strength weapons where 2h weapons actually offer less skill damage.


Is there a thread that has the details of the enemies? Like their armor value? For mobs and bosses. That would be another nice project if such a thread does not exist.

Some sort of Pokedex but for enemy monsters lol.

EDIT: That way we can have an idea of what armor we are up against when testing what you have found. :D

Does crit play a factor?

Here you go:
http://www.spacetimestudios.com/showthread.php?18229-Pocket-Legends-Enemy-Boss-and-NPC-Compendium

Crit is just 2x damage and is independent of enemy armor, dodge, and your hit.

Crit damage = 2 x normal damage - 2 x enemy armor

giayuan
04-08-2011, 08:10 PM
This is great! I see the effort and its displayed beatifully! :D welcome to forums btw!!