DPS Tables - Massive Update in Progress

dakkon107 wrote:

Yeah, poison depends. There is talk that being able to apply poison from farther away enhances it's value. If you can KB/Slow lock a unit so the melees are hitting no matter what, then maybe, Agrippa and Diana are better. I wouldn't use more than one poison unit anyway.

As far as overall DPS goes, I see Vamp and H.Monica being the best, but when you look just at ground attack, I see Agrippa, Diana, H.Monica, and Cecilia being better. I am just assuming that you just take the the stat and divide by the attacks per second value to get DPS. This method doesn't weigh ground any higher than sea, which seems to be what you are going for correct?

I don't think that it is as much subjective as situational, as in what scenarios can you lock.

I just made a spreadsheet that can compare all of this stuff, but want to integrate Strussdell's list to see how everything compares. My idea thus far is to allow us to quickly compare units for different uses. It will allow one to prepare for battles objectively. For example, to fight Hippolyte in the next event, filter for melee units based on ground DPS and you can see that Vlad, Agrip, and Diana are the best.

While I like Strus' list a lot, I feel it's not as objective as it could be due to skill weightings and land/sea/air weightings. However, it's a better indicator than my total stat comparison of 'what's a good overall unit'.

But for min and maxers, they usually have more than 1 6* unit and thus aren't looking for simply 'what's a good overall unit' but 'what is the best unit that I have in a certain situation'.

That is what this list prioritizes moreso than anything else.

Example : I'm fighting the black dragon who is weak against melee and magic units. Which is the best against him? Refer to my list and find out.

Awesome work, now I can be a lazy bum and figure which units are best to deploy during boss fight without killing my brain cells X3 Thank you!

BkWiz wrote:

While I like Strus' list a lot, I feel it's not as objective as it could be due to skill weightings and land/sea/air weightings. It's a better indicator than my total stat comparison of 'what's a good overall unit'.

But for min and maxers, they usually have more than 1 6* unit and thus aren't looking for simply 'what's a good overall unit' but what is the best unit that I have in a certain situation.

I agree with that, it seems to look at quest and tower events only, and one of those two is useless. I haven't played long enough to know how tower events work to formulate my own opinions on some of these things. So far, I haven't had a good place to try out poison effectiveness and how range actually affects overall damage.

These DPS rankings are really a perfect objective framework for evaluating most other rankings though. If you are looking at boss fights like fires or Halloween, these straight up values are perfect. Range can then be factored in for tower defense events and such, with subjective ability scalars for PvP, tower, etc.

I don't think anyone has really figured out damage in arena events, so these may not work there very well. I think Clock was working on a damage prediction formula for those, but she said that it was still off on a lot of units. Having series bonuses here doesn't really help. I may try working on those when Arena of Condemnation comes out because those events are pretty awesome, but we will see!

Kyte wrote:

Awesome work, now I can be a lazy bum and figure which units are best to deploy during boss fight without killing my brain cells X3 Thank you!

It's for lazy bums like you that I did this list for.

That and I got tired of people questioning me and saying I had nfc what I was talking about...

You know who you are...

Now they can see exactly why I didn't want to explain everything in a short chat...

It's like saying the sun is a bright light in the sky. True, but totally useless.

nice job bro ~ waiting for the result

Very nice work How about a mod sticky this? As this is pure numbers, unbiased and free of opinion.

BTW, those values used for speed, are they just a multiplier, or their actual attack animation in seconds?

BkWiz wrote:

That and I got tired of people questioning me and saying I had nfc what I was talking about...

You know who you are...

Yay, I know who I am!


This is a great list. Thank you.

Just Petra and Cornelia left. Waiting for confirmation on Seth/Abel before posting.

Will start adding my thoughts on misc things regarding dps later. If there is anything missing please let me know so I can review and fix if necessary

BkWiz wrote:

Archer DPS


Land : Vampiress | Anastasia | Saki | Minerva | Nina | Oktavia | Reve | Alex

Air : Oktavia | Vampiress | Nina | Saki | Minerva | Anastasia | Reve | Alex

Sea : Oktavia | Minerva | Nina | Vampiress | Saki | Anastasia | Reve | Alex

Total Stats : Oktavia | Vampiress | Nina | Minerva | Saki | Anastasia | Reve | Alex


To Be Completed : Abel | Seth

Cornellia is missing right sir?

pnoytroi wrote:

BkWiz wrote:

Archer DPS


Land : Vampiress | Anastasia | Saki | Minerva | Nina | Oktavia | Reve | Alex

Air : Oktavia | Vampiress | Nina | Saki | Minerva | Anastasia | Reve | Alex

Sea : Oktavia | Minerva | Nina | Vampiress | Saki | Anastasia | Reve | Alex

Total Stats : Oktavia | Vampiress | Nina | Minerva | Saki | Anastasia | Reve | Alex


To Be Completed : Abel | Seth

Cornellia is missing right sir?

Just Cornelia left and she is a mage.

Seth, Abel, and Petra completed. Thanks xSmall!

petra 1.3 anas 1.4 but anas always hit 1st( tested it w/o setting petra ldr )...mage atk animations r slowest compared to others @@

nocturned wrote:

petra 1.3 anas 1.4 but anas always hit 1st( tested it w/o setting petra ldr )...mage atk animations r slowest compared to others @@

Yep, but here is where you are looking at it wrong, I'm not looking at who hits first to get attack speed. I looked at the first 'hit' to the next 'hit' to the get attack speed between hits.

This way, my figures already incorporate missile speed, character animation and attack speed delay into the figures.

And I don't have to do a whole new set of calculations/data gathering such as animation speed per character, attack delay between characters, etc.

It's a lot of work for nothing when all we are looking for as our end goal is unit DPS.

In DPS, we only need to know the delay in between attacks and have missile speed and animation speed factored into the attack delay to get the actual attack speed value.

To put it into a formula : ATTACK DELAY + MISSILE AIR TIME + ANIMATION FRAMES/ANIMATION SPEED = ATTACK SPEED VALUE.

This is a really simplistic formula and doesn't incorporate the fact that Seth and Abel, and possibly Alex have a sliding scale on their missile speed depending how close or far the unit is from them unlike other archers or mages.

Quote :

The points used in the attack speed calculation are the points of impact. As soon as the number pops up during an attack, the first point is noted. When it comes up again, the second point is noted, etc. The attack speed is then estimated based off a stop watch and the average of the times between the two points.

Already pointed that out in my misc post portion as you can see. I'll edit it and add more of an explanation at that bullet point so there is less confusion.

snappycow wrote:

Very nice work How about a mod sticky this? As this is pure numbers, unbiased and free of opinion.

BTW, those values used for speed, are they just a multiplier, or their actual attack animation in seconds?

Forgot to reply to this, I apologize, the values for attack speed are the actual times in between attacks. So the first time a damage number pops up, that is one 'point', the second time a number pops up that is the second 'point'.

As I said earlier, once I have the time between those two points, I can use the non modified stats from the japanese wiki and simply divide them by the attack speed in order to get the damage per second the unit puts out.

This is awsome BK, thanks for putting in the time to do this

hip hip hooray- its BK!

BkWiz wrote:

Forgot to reply to this, I apologize, the values for attack speed are the actual times in between attacks. So the first time a damage number pops up, that is one 'point', the second time a number pops up that is the second 'point'.

As I said earlier, once I have the time between those two points, I can use the non modified stats from the japanese wiki and simply divide them by the attack speed in order to get the damage per second the unit puts out.

I thought so too, just asking as you didn't indicate them on the table earlier, and also since other people's rankings put an arbitrary multiplier/points for speed.

I'm totally dumbfounded by the work you done there... Amazing work! Appreciated the work you have compiled there. *Bookmark page* =D

Finally, a dedicated findings about DPS where some people just do not have an understanding how it works (no offense)

Props to you BK to get this done. Appreciated it much.

Very informative, it's amazing to see how much the game's developers deviated from actual time data for some of the units. Blethi is a big one, Revoke his C status immediately. I like seeing that my own work isn't too far off base. My S to D ratio of 2:1 remains to be a fair approximation for example (except for those extreme cases referenced above). Wonder why the developers can't get the speed data right? The sprite animation frames can be counted and the letter system could just correspond with the number of frames. Not really too hard since they are the ones doing the animation.

Mad props to you BKWiz! Tremendous amount of time and effort must have been used for those helpful data!

snappycow wrote:

I thought so too, just asking as you didn't indicate them on the table earlier, and also since other people's rankings put an arbitrary multiplier/points for speed.

While BK's testing is more accurate I think calling other's work arbitrary is a bit of a stretch. in My spreadsheet for example although I used the game's flawed letter system to derive my speed bonus from I ran many tests to develop that that bonus and didn't just apply values at my whim. Though knowing the game's system is flawed was a subject of some personal frustration I did not have the resources or time to do individual tests on each of the game's 212 units.

dnonamus wrote:

snappycow wrote:

I thought so too, just asking as you didn't indicate them on the table earlier, and also since other people's rankings put an arbitrary multiplier/points for speed.

While BK's testing is more accurate I think calling other's work arbitrary is a bit of a stretch. in My spreadsheet for example although I used the game's flawed letter system to derive my speed bonus from I ran many tests to develop that that bonus and didn't just apply values at my whim. Though knowing the game's system is flawed was a subject of some personal frustration I did not have the resources or time to do individual tests on each of the game's 212 units.

Personally I like your spreadsheet best since it closely matches my own data without having the actual data.

As I said on your thread, there was only one major discrepancy that I noted.

Bumping this because it need not be lost

Looking for Dana to run a quick test if anyone has her.

Using this as a notepad for Arena DPS.


Under Chip Effect

Charles - 107735
Reinhard - 129674
Antony - 112525
Ellen - 190748
Frank - 76717
Belethi - 171907
Oktavia - 194532
Reve (AE) - 320563
Reve - 183179
Reinhard - 129674
Nina - 168284
Agrippina - 213212
Vampiress - 183260
Faith - 163509
Faith (AE) - 286141
Georgios - 240186
Monica - 222520


Player Unit Attack Order

Charles
Reinhard
Ellen
Agrippina
Belethi
Nina
Oktavia
Georgios
Vampiress
H.Monica
Monica
Reve
Gloria
Faith