Archer Attack Speed Basics

kalden wrote: »

Round to the nearest valid Attack Duration (0.05 increment). Demon sparked 25%;

2.00----1.43----0.70----default
2.00----1.67----0.60----weapon
2.50----2.00----0.50----bracer
3.33----2.22----0.45----set
3.33----2.50----0.40----accessory
5.00----3.33----0.30----armor
5.00----4.00----0.25----cape
5.00----5.00----0.20----tome

Round to the nearest valid Attack Duration (0.05 increment). Red Bubble 20%;

1.67----1.43----0.70----default
2.00----1.67----0.60----weapon
2.50----2.00----0.50----bracer
2.50----2.22----0.45----set
3.33----2.50----0.40----accessory
5.00----3.33----0.30----armor
5.00----4.00----0.25----cape
5.00----5.00----0.20----tome

Round to the nearest valid Attack Duration (0.05 increment). Demon Quickshot 30%;

2.000----1.43----0.70----default
2.500----1.67----0.60----weapon
2.500----2.00----0.50----bracer
3.330----2.22----0.45----set
3.330----2.50----0.40----accessory
5.000----3.33----0.30----armor
5.000----4.00----0.25----cape
10.00----5.00----0.20----tome b:shocked

Based on these numbers, I think it is fair to say that the person who claimed the attack rate was capped at 5.00 attacks a second was fooled by the illusion created with such truncations and roundings.

Elenacostel - Heavens Tear wrote: »

Were those theoretical numbers or real ones?

The person PM'd me saying he had a lv150 EA fist build and had told me it was impossible to pass 5.00 attacks a second.

If your method of calculation is correct Elena, and Kalden's numbers are correct, I can draw a logical preliminary conclusion, unless he provides visual evidence, that he was fooled by this illusion and that the max attack rate is not capped at 5.00.

Elenacostel - Heavens Tear wrote: »

-- Appendix: Examples of Attack Rate % Increase Computations --

This is the set of steps you need to compute your new Attack Speed. Remember that all math is done using Attack Durations, not Attack Speeds.

1) Convert Attack Speed to Attack Duration (use The Chart above).
2) Use this formula to compute your new Attack Duration:
____New Attack Duration = Attack Duration * (1.00 - Speed Increase %)
3) Truncate to 2 decimal places.
4) Round to the nearest valid Attack Duration (0.05 increment).
5) Convert rounded Attack Duration to Attack Speed (use The Chart above).

There is an flaw in this math I believe, as it was cited on my DPS thread (dead now, made under fourm name) I used this example:

Using the fist decide w/ -.1 intervel

.7 -.1 = .6

New attack duration = .6

Now, Increase attack speed by 100% with a (fictional) new uber skill:

.6 * (100 - 100) = 0

So now attack duration is:

# of attacks/0, it's impossible to divide by zero.

To finish this off, your attacking infinate times per second because you increased attack rate by 100%, however, logically increasing anything by 100% doubles it, not makes it infinity b:shocked.

As a counter example I propose a new formula:

1: attack speed of fists is 1.43, that converts to .7

2: .7 - .1 from decide = .6

This is where we differ

3: change back to attack speed .6 = 1.666... thats 1.65

4:1.65 * (1.00 + 1.00) = 3.30 attack rate

3.30 is double, thats logical.

I just got really lucky and crafted hypothetical bracers with -600 seconds of interval.

I kill stuff 10 minutes before I start attacking b:victory

Elenacostel - Heavens Tear wrote: »

Do you have an example that supports your math? I basically reverse-engineered the algorithm based on examples people have posted on this thread, as well as from my own experience. I am quite interested in any examples that prove my formula false.

Also, I too can create a fictional Bracers that give -1.5 seconds to attack interval. Given that a crossbow has an attack interval of 1.5 seconds, that would be an attack interval of zero seconds.

How would your formula process handle this division-by-zero case, given a fictional piece of equipment with such statistics?

Simple, I would use logic. I understand you have tested this formula several times, but it simply doesn't hold up in the extreams where it logically should. What is flawed in theory is flawed in practice, this diffrence may well be important at higher attack speeds, let me explain:

With your example, increasing something by 100% makes it attack undefined times, I'm simply stating that that makes 0 sense to me. If you increase something by 100% it doubles. If I was paying you $20 an hour, and increased your paycheck by 100%, you would have $40 an hour earnings now, no?

As for your example, thats also simple. First of all I think you are using a bow, as a crossbow's attack intervel is closer to 1.6. Regardless let me use an example. If I can pick an apple at 1 second per apple, and I get new gloves that let me pick apples 1 second faster per apple, then as soon as I begin to pick apples all the apples have been picked. The diffrence is that this flaw is grounded in reality. X - X = 0. This is true in all cases. Try it out on a x,y cordinate plane, with Y being number of actions and X being time. Find the slope and subtact the slope from it, you have a vertical line, the slope in undefined and infinate actions happen in 0 seconds. Its worth mentioning your formula would have the same result as mine.

Don't think this is so far-fetched, it is already possible to decrease attack intervel to .05 with perfect equipment, 0 attack intervel in just one item away, barring an arbitrary limit on attack speed.

I'll level with you, my theory is based on logic and real world examples, I lack access to numerous examples, but let me give you an experiment to test this with:

items/skills: -.1 intervel bow, -.1 intervel wrists, demon hell spark

Your formula:

1.5 -.1 -.1 = 1.3

1.3 * (1.00-.25) = .975, aprox. attack speed 1/1.03

My formula

1.5 -.1 -.1 = 1.3

1.3 = .77

.77 * (1.00 + .25) = .9625, aprox. attack speed 1.04/1.05

So try that out if you can find the time, post screenys too. For all I know you may be right and PWI simply works outside the realm of reason. b:surrender

Elenacostel - Heavens Tear wrote: »

I believe this is more of a translation issue than a math issue. Instead of "Increase Attack Rate by X%", it should be phrased "Decrease Attack Interval by X%". At least, that is what I have been able to figure out.

That... makes sense. b:surrender That statement pretty much destroyed my argument, thanks for answering my questions.

By the way, I thought I should mention it is now possible to attack 20 times per second b:shocked observe:

Fists have .7 attack intervel, so to reach .05 intervel unbuffed you have to achieve -.65 intervel

-.15 fists (custom made fists, 3 -.5 mods, 1/26,200,000 chance of creating)

-.15 wrist (same story ^^)

-.1 R8 top

-.05 Life everlasting

-.05 Lunar glade cape

-.05 TT legs recast at palace of nirvana

-.05 TT99 Ashura legs + boots

-.05 TT99 Lionheart necklace + belt

Thats -.65, tremble in fear, TREMBLE!!!b:angryb:shocked

Legerity - Sanctuary wrote: »

Actually... I don't think you can get Ashura's bonus if you recast the legs to nirvana legs. 2 slots are taken by lionheart, one is taken by rank8, legs are taken which leaves wrists and boots.

Going for either is only -0.15 (Ashura wrist + Ashura bonus or 3x-0.05s wrist).

Meaning total is -0.6s interval.

Maybe, maybe not. I've seen to proof for or against recast leggings retaining their set skills. If they don't, thats all well and good, If they do, theres a small chance of some serious OP archerism going on there...

The attack rate posted on the first page are correct but for one exception. The game does not allow you to exceed attack rate of 5.00 no matter how much -interval you have.

I have a base attack rate of 5.00 using lunar claws and it does not change at all if i use skills or apocs that increase my attack rate.

I have no idea wtf none of that means, its like greek to me. but good job.

if you wear as much -0.1 reducing stuff as you can then you can take almost 40% off your attack time. thats a huge amount more if you use spark burst+ in pvp

Olba - Sanctuary wrote: »

You're missing a 0 from a lot of stuff there, you're totaling -2.9, not -0.65.

Yes... Sorry about that, fixed it.