EebstertheGreat

exactly 1500 character LOL

Hello Eebster,
what do you think of the logic below? is there any reason why it should not replace the existing PRD logic? i.e. does it have some logical flow? the reason why I thought of this is that PRD to begin with makes it fair but look random, but when % gets very high, such as 80% or anything above 50%, it seems it went back to true random.

The theoritcal C value, for 80% (C=0.75) is bad. It makes not that many possibilites. It just says (no two in a row can miss block).

so the current is:
Block
Block
Block
hit then Block (because starting C is 0.75 will become 1.5 = 100% as soon as one block fails to proc)

thats all 4 possibilities, which is very close to true random (i.e. PRD purpose failed)

So what if it was coded like the 20% chance to bash but in reverse way, as follows:


For 20% chance to bash, C = 0.05570
the chance will increase everytime a bash fails (as you already know

Now for Vanguard 80% chance to block, why shouldnt it be something like:
20 chance to "Not Block", C = 0.05570
whenever "Block" succeeds, the chance to not block increases by C=0.05570

so at most it will be 18 hits not blocked in a row

This can be rephrased to equivalent statement of
Co = 1 - C is the starting probability of attack modifier (in our example VanguardBlock)
Cn = 1 - nXC is the Nth probability of attack modifier (chance to block after N attacks when it all fails)

difference with original model is that starting C value is different from the increment

what do you think?

alright

it makes sense

thanks! XD

Hello Eebster

Is there good mathimatical model to analyze MoC equivalent IAS in dota2? (It creates an attack like ember everytime it procces)

I had some simple thought as follows: if MoC procced every 1.7 seconds, it provides 100 IAS: if it procces every .85 seconds it increases 200 IAS. and generally it adds 1.7/proc time IAS

At best proc time is CD and rough estimate can be half that time.

Now this model feels too simple and has too many assumptions

Do you have some better math model on your mind? I know it can vary a lot from depending on units attacking. Iy can also vary of all units attacked at same time or not (lets say 100 rangers striking once every 2 seconds all at same time, then it can inly proc once every 2 seconds)

So, how do we tackle this problem? I am thinking one of the best ways to know is to make a prgram register all duels and how many times MoC procced

But is there some ok theoretical model for this?

May I ask, Did you study maths or something similar? (since you use qac at the end, and they way you formulate your statements)

What "Reborn"? Mind paging me?

One week without any activity in AM, a month in Programming... I'm scared to remove the permatabs in my FF, they've been there for so many years :(

Well, looks like this place is slowly going to sleep. Didn't you have 10k posts years ago already? :P
Edit: Looks to be a year ago.

hello EebstertheGreat

I have a question about PRD

it was mentioned that other orders can interrupt the stacking of the chance: i.e. the counter reset if you receive another order other than an attack order

does that mean if skeleton king is jungling it is better for him to auto attaack instead of attack move and kiting the creeps (because he want more critical strike)

also when a camp is cleared and you move to another camp, does the counter reset again? or when does it exactly reset?


also: does it reset in warcrat III in the same manner it resets in dota 2?

that we want attack speed and damage to be equal? ( I am actually surprised becasue i see lot of people say phase boot is better farming etc etc whereas it provide less dps in most heroes

what do you think?

also, not sure if it should be in adv mechanics xD

Hello

Apparently Buriza critical strike chance stack the same way evasion does. but when I did the math it seemed to me that buying 2 butterflies is much better than buying 2 burizas (according to my match total bonus was about 58-59% average increase but buttefly is constat 53% for every butterfly, unliss I did something wrong? xD)

what do you think?

Do you know if Robzor and Shorttail still visit Playdota forum or not?
Did they change their names?

Does the bonus damage from Storm's Windwalk require me to attack the enemy unit from the back?

Do orb effects mechanics change in DotA 2?

Example
Phantom Assassin with
Mjollnir, Desolator, Skadi, S&Y, Diffusal Blade, Helm of the Dominator

According to you, what items should Brewmaster get?
Imo Rod of Atos is a must since it's such a great item, huge casting range for an item that slows movement speed by 60%

Dispel Magic is usually used for what?
I know it dispels Storm's cyclone and deals a lot of damage to illusions or summonned units
Tell me about the not so well known uses of Dispel Magic

Do you think Storm's Windwalk is useless? It makes you deal less DPS because of the fade time for a bonus damage that you can actually deal in the same amount of time without having to use Windwalk and I can only see it useful for catching up to escaping targets and cyclone them

How does DotA 2 Alchemist' aghanim work?
1.For Alchemist himself
2.For his allies

----------
When Moment procs, it just adds a +1000% IAS Slow Aura and a 20/40/60/80% lifesteal Vampiric Aura for 0.5 seconds.
----------
but the attack speed is still limited to 500 in total yes? does that mean in dota 1 Tresdin does not benifit from the attack speed buff of moment of courage if he already has max attack speed, but only benifit from life steal?

I didnt know u could edit posts in other forums :shock:

http://www.playdota.com/forums/showthread.php?p=2649978#post2649978
hows that? :D

Make a movie: http://www.hiveworkshop.com/forums/arena-226/cinematic-contest-6-a-271641/ :D:

The lengendary Anitarf (and Infrane) is a guest judge!