Attack Speed

I found out that h helmut seems to have same attack frequency as sheryl...
This just doesn't make sense. Can someone help me do the test?

I just tested Sheryl and Halloween Helmut and found that Helmut actually has a slightly higher attack frequency than Sheryl. Hmmm...maybe there's a case for using Helmut after all. I also recently observed that Monica has the same attack frequency as Achillea, a B speed. There just seem to be no fault with her.

I updated my first post a little, I still need to place more people on the list but I added quite a few 5*.
I have pretty much decided worrying about 4*s is a lost cause. None of my allies have 4* leaders anymore so It would be almost impossible to compare them.

As far as Sogno is concerned her attack animation makes her attack ridiculously slow. She is actually slower than Hildegarde which is quite sad. The one benefit that repeated hitters have is that they can change targets mid attack if their first target dies.

Managed to find a few things about attack delay as from zoeb and clock's unit and monster encyclopedia.

It seems that attack delay does not directly affect DPS (e.g. if someone's attack delay is half of the other unit, it does not mean his/her attack would be doubled relative to the other unit).

Also, Melee to melee attack delay matches up. Comparing missile and magical attack speeds are more difficult since some have more varying animation to one another as Whitemidge found out.

Is there a list of attack delays anywhere? I know they are on the Chinese cards but I don't know if there is a convenient place to find them.

That's the only one unfortunately. The good thing is though it's easy to find out melee attack delays since we know several of them already. Take the Werewolf as an example, I bench marked her with Iwan and I found out that they're identical in attack speeds so I knew her attack delay was 0.6.

WW attack delay is not .6 seconds.

I got it wrong? I'll try to get another Iwan. I've tried comparing a Gordon and WW and she's indeed faster than .65. I'll double check again once I get the Iwan.

Edit*

Ok I'm not going crazy. Her attack speed is identical to Iwan's attack speed. Either you have access to her actual attack delay values and you're saying that identical attack speeds does not mean that they have identical attack delays or you're pulling my leg.

I think BKWiz is just trying to cause trouble

Wheatmidge wrote:

I think BKWiz is just trying to cause trouble

I suggest thinking about that accusation for a few minutes before saying it again.

And to prove that WW is NOT 0.6 seconds between attacks, I give the following information from my data gathering/files:


Charles' attack speed delay is 0.8 seconds from end of animation to the start of the next animation (Swing). No Speed Demon modifier included. Note : It's actually a little slower than 0.8 seconds, but I rounded it to the nearest 10th for easier viewing and my own sanity.

Considering Charles is S speed and the only S speed in the game, I highly doubt that the Werewolf at C speed is going to be beating Charles anytime soon. Which means a 0.6 second delay in between swings is impossible.

The values from the Chinese site are a good baseline to start from, but overall they are completely useless other than for comparative purposes.

For calculating actual attack speed values, and thus DPS values as the next step, you can't use the Chinese site to get actual numbers.

As I've stated in a post I edited out, there are more things that go into the 'actual' v. 'listed' attack speed value.

Did some testing on my own:

Rough results:

Kiki (D) 1.38s (11554 per hit at max level)
Remika (D) 1.30s
Meg (C) 1.19s
Angela (S) 0.89s (17732 per hit at max level)

Test setup:
- The target heros are used as allies (no leader bonus)) in Blaze Quest 4 and video recorded.
- Then look for an easy identifiable part of attack animation (1st appearance of projectile, weapon slashes etc) and measure the number of frames between each.

Comments:
- There were some variations in the number of frames (maybe due to phone CPU), but black dragon can give >12 sec worth of results for averaging.
- My numbers seem quite different from posts above (hardly any delay above 1sec, but I doubt few units are faster than Angela), still I think is a these numbers measure cycle more accurately for DPS estimation.
- This is quite some work so not actively continuing for now.

kyaaya wrote:

Did some testing on my own:

Rough results:

Kiki (D) 1.38s (11554 per hit at max level)
Remika (D) 1.30s
Meg (C) 1.19s
Angela (S) 0.89s (17732 per hit at max level)

Test setup:
- The target heros are used as allies (no leader bonus)) in Blaze Quest 4 and video recorded.
- Then look for an easy identifiable part of attack animation (1st appearance of projectile, weapon slashes etc) and measure the number of frames between each.

Comments:
- There were some variations in the number of frames (maybe due to phone CPU), but black dragon can give >12 sec worth of results for averaging.
- My numbers seem quite different from posts above (hardly any delay above 1sec, but I doubt few units are faster than Angela), still I think is a these numbers measure cycle more accurately for DPS estimation.
- This is quite some work so not actively continuing for now.

Yes it does take quite a bit of work. I rounded to the nearest 10th for simplicity sake, but even .1 seconds makes a huge difference in terms of actual DPS of a unit so there will be some variance.

But it's a start at least. Unless someone with lightning fast reflexes or a video camera type program which records to the hundredths and lots of patience wants to have a go at it.

On a related side note to validate your results, I have Charles and Angela at the exact same attack speed.

Both attack every 0.8 seconds. I use the first time a damage number appears for a unit as the stop and start mark. Easier than using animations or when the sword reaches it's max arc or something silly.

The slight variance is usually due to human error moreso than anything else. Unless you are running multiple programs on your phone or are overheating it...

BkWiz wrote:

Wheatmidge wrote:

I think BKWiz is just trying to cause trouble

I suggest thinking about that accusation for a few minutes before saying it again.

And to prove that WW is NOT 0.6 seconds between attacks, I give the following information from my data gathering/files:


Charles' attack speed delay is 0.8 seconds from end of animation to the start of the next animation (Swing). No Speed Demon modifier included. Note : It's actually a little slower than 0.8 seconds, but I rounded it to the nearest 10th for easier viewing and my own sanity.

Considering Charles is S speed and the only S speed in the game, I highly doubt that the Werewolf at C speed is going to be beating Charles anytime soon. Which means a 0.6 second delay in between swings is impossible.

The values from the Chinese site are a good baseline to start from, but overall they are completely useless other than for comparative purposes.

For calculating actual attack speed values, and thus DPS values as the next step, you can't use the Chinese site to get actual numbers.

As I've stated in a post I edited out, there are more things that go into the 'actual' v. 'listed' attack speed value.

A guest post on page 3 already mentioned that these values do not represent for the actual DPS. It is as you say though I think that a numbering system is much more meaningful and far less cumbersome than what we have as a letter system considering it lists almost all of our mages as having attack speeds of D instead having varying attack speeds.

Thank you Blargod for pointing out that BkWiz is ignoring what we had already discussed.

We had already said that the chinese numbers are not the entire attack speed but instead the attack delay between the end of one animation and the start of the next.

Both angela and charles are listed as .2 delay on the chinese site and end up being the same speed. If you say the entire attack takes them .8 seconds then the attack animation takes .6 seconds.

Werewolf seems to have a .6 attack delay. The same as Gerhard. Assuming they have the same timing for their attack animation as charles then their total attack will take 1.2 seconds. Which sounds about right.

Are all of the attack animations for melee units actually the same? Anyone feel like testing that?

So far I've seen they seem to have animations that are identical to one another.

Ignoring units with weird attack animations(Georgine, Sogno, Joachim, Hildegarde, etc) do all other units of the same type have the same animation speed?

I've noticed something really weird. I've assigned leadership to H. Helmut but he didn't increase his speed to D even if I had Speed Demon on. When I made him to level 90, he seemed to change his attack speed.

I don't know if my eyes are playing tricks on me but can someone test this out (if you have a low level H. Helmut)?