Star Trek Online

Star Trek Online (http://sto-forum.perfectworld.com/index.php)
-   Builds, Powers, and Game Mechanics (http://sto-forum.perfectworld.com/forumdisplay.php?f=219)
-   -   How are the following shield resistances handled? (http://sto-forum.perfectworld.com/showthread.php?t=594311)

ussultimatum 03-25-2013 09:02 AM

How are the following shield resistances handled?
 
As I understand it, shield resistances are multiplicative.

So my question is the following (assuming 0 shields power, and no EPTS 1):

With a Res A / B Resilient shield we have:

> 15% to 3 energy types
> 5% absorb (which as I understand it functions as resistance)
> 2% x 10 stacks of the ADAPT property

Players typically look at that and think "40% resistance" - which is not the case.

However (Edited to include stirling191's and frtoaster's fixes, thanks!):


1-(((1-0.15)*(1-0.05))*((1-0.02)^10)))

= 34% resistance

Did I even do this correctly?

Feel free to let me know how wrong I am if I did!!


My main question is with regards to the ADAPT property and if it does add 10 individually handled, multiplicative 2% resistance bonuses.

stirling191 03-26-2013 06:05 AM

That *looks* correct, but it's been awhile since I've dug up the formulae for shield resists.

That being said, based on your operating assumptions (IE: multiplicative instead of additive stacking, and zero shield power based resists) the equation as intended will get you to a 34.1 % shield resist number.

Though upon further review I believe you are missing a set of parenthesees around the entire resistance portion.

I believe it should be:

1-(((1-0.15)*(1-0.05))*((1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)))

To contain all the resistances into a single multiplicative operation. Elsewise you get a .7752 number that doesn't really jive with anything (too high to be a resist, and too low to make sense given the additive stacking value of 40%).

Hope that was in some way helpful.

EDIT: Just for kicks I ran the equation as though the ADAPT was a single resist number. At 10 stacks it works out to a 35.4% resist (assuming a 20% resist "chunk" as opposed to single modifiers). Only a 1.3% difference.

ussultimatum 03-26-2013 07:52 AM

Quote:

Originally Posted by stirling191 (Post 8818321)
That *looks* correct, but it's been awhile since I've dug up the formulae for shield resists.

That being said, based on your operating assumptions (IE: multiplicative instead of additive stacking, and zero shield power based resists) the equation as intended will get you to a 34.1 % shield resist number.

Though upon further review I believe you are missing a set of parenthesees around the entire resistance portion.

I believe it should be:

1-(((1-0.15)*(1-0.05))*((1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)))

It worked when I plugged it into Excel, but Excel probably fixed it for me. :o


Quote:

Originally Posted by stirling191 (Post 8818321)
To contain all the resistances into a single multiplicative operation. Elsewise you get a .7752 number that doesn't really jive with anything (too high to be a resist, and too low to make sense given the additive stacking value of 40%).

Hope that was in some way helpful.

Yes, thank you.


Quote:

Originally Posted by stirling191 (Post 8818321)
EDIT: Just for kicks I ran the equation as though the ADAPT was a single resist number. At 10 stacks it works out to a 35.4% resist (assuming a 20% resist "chunk" as opposed to single modifiers). Only a 1.3% difference.


Good call! I didn't even think to try that as a comparison. Thanks again!

stirling191 03-26-2013 08:02 AM

Quote:

Originally Posted by ussultimatum (Post 8819671)
Good call! I didn't even think to try that as a comparison. Thanks again!

My pleasure. A little early morning math to get the synapses firing :)

I was originally planning on doing a side by side comparison of "stacks" versus "chunks" at each stack level (IE: 4, 6, 8, 10, 12, 14, 16, 18, 20%), but give how relatively small the difference at 10 stacks was it felt a mite unnecessary.

ussultimatum 03-26-2013 09:30 AM

Quote:

Originally Posted by stirling191 (Post 8819811)
My pleasure. A little early morning math to get the synapses firing :)

I was originally planning on doing a side by side comparison of "stacks" versus "chunks" at each stack level (IE: 4, 6, 8, 10, 12, 14, 16, 18, 20%), but give how relatively small the difference at 10 stacks was it felt a mite unnecessary.


Agreed.

I think the benefit for a lump 20% bonus vs. stacks of 2% handled individually would maybe show a bit stronger if we assumed non-resilient (losing 5%) as well as opponent using a non-protected energy type (i.e. Disruptors vs. your Res A shield, which loses another 15%).

But as my OP already discounts EPTS 1 and takes 0 shield power into consideration, this is already unrealistic enough.

I'll probably continue to calculate it as 2% stacks vs. a single lump of 20% (or 10% for 5 stacks), but that's just my gut feeling on how it's probably handled.

stirling191 03-26-2013 09:43 AM

Quote:

Originally Posted by ussultimatum (Post 8821271)
I think the benefit for a lump 20% bonus vs. stacks of 2% handled individually would maybe show a bit stronger if we assumed non-resilient (losing 5%) as well as opponent using a non-protected energy type (i.e. Disruptors vs. your Res A shield, which loses another 15%).

My rough math (.98 ^ 10 = .817 vs .80) says that you'd only see a 0.4% difference (the "chunk" vs. "stacks" difference in the earlier equation was 1.3%, straight stacks versus straight chunk is a 1.7% difference) in resist loss versus factoring in Res + Resilient shields.

Without factoring the additional resists from EPtS, TSS, Shield Power and (potentially) RSF into the equation I obviously can't be 100% certain, but the trend certainly seems to point towards a diminishing difference between the two calculations (chunk vs. stacks) as more resists come into play. So while I can absolutely understand the desire to understand the correct mechanism of action, functionally (based on everything that's been presented here) there shouldn't be a noticable difference.

ussultimatum 03-26-2013 09:55 AM

Quote:

Originally Posted by stirling191 (Post 8821551)
My rough math (.98 ^ 10 = .817 vs .80) says that you'd only see a 0.4% difference (the "chunk" vs. "stacks" difference in the earlier equation was 1.3%, straight stacks versus straight chunk is a 1.7% difference) in resist loss versus factoring in Res + Resilient shields.

Without factoring the additional resists from EPtS, TSS, Shield Power and (potentially) RSF into the equation I obviously can't be 100% certain, but the trend certainly seems to point towards a diminishing difference between the two calculations (chunk vs. stacks) as more resists come into play. So while I can absolutely understand the desire to understand the correct mechanism of action, functionally (based on everything that's been presented here) there shouldn't be a noticable difference.


Agreed. It's small enough that I'm happy to not worry about it. :)

bareel 03-26-2013 09:52 PM

Didn't double check the precise math but yes you seem to have done it properly.

The thing is the game doesn't look at damage resisted, it looks at damage that gets past the resists so 20% resist to the game is you take .8 damage. And yes they multiply which makes it look like they don't stack as effectively but that is merely a perception thing at work.

For example lets say you have 20% shield resist from option A and 10% from option B. The math would be .8 * .9 giving you .72 damage hits you, or 28% resistance. And that works if you get hit with 100 damage you could apply them separately as .8 of 100 is 80 and .9 of 80 is 72, or directly as .72 of 100 is 72.

Hope that made sense but the TLDR is you did it right. Oh and their was a rumor by a dev that their are diminishing returns applied similar to resistances with armor and that maco 10% effect is not subject to that. But no real proof nor any true desire to test it.

frtoaster 03-26-2013 11:07 PM

Quote:

Originally Posted by ussultimatum (Post 8800661)
However:

1-((1-0.15)*(1-0.05))*((1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02)*(1-0.02))

= 34% resistance

FYI, there is a mathematical operation known as exponentiation:

1 - (1-0.15) * (1-0.05) * (1-0.02)^10

I wish they would show our shield resistances in the UI. It would make it easier to figure these things out. However, I don't know that anyone has actually tested whether the fleet elite shield bonus stacks multiplicatively or additively.

ussultimatum 03-27-2013 10:16 AM

Quote:

Originally Posted by bareel (Post 8835261)
Hope that made sense but the TLDR is you did it right.

It did, thanks. :)


Quote:

Originally Posted by bareel (Post 8835261)
Oh and their was a rumor by a dev that their are diminishing returns applied similar to resistances with armor and that maco 10% effect is not subject to that. But no real proof nor any true desire to test it.

:eek:

This is a monkey wrench the size of an asteroid.

Ok, let's add that in to the discussion and also go back a bit for some other assumptions I made so at least all of us are on the same page (or can correct me if I am wrong):

1) MACO 10% may or may not be subject to the above formula. If it is not subject to it, does it simply get added in at the end? The beginning?

2) Resilient shields have a 5% absorb property, it's my understanding that this gets factored in (STOked episode mentioned it).

Does it in fact get factored in to total shield resistances (my math above includes it)?

If it does, is this also not subject to the diminishing returns formula?

3) Elite Fleet Shields have 15% vs. 3 Energy types.

Does this function like the rumored MACO 10%, in that it is also not subject to diminishing returns?






Quote:

Originally Posted by frtoaster (Post 8835871)
FYI, there is a mathematical operation known as exponentiation:

1 - (1-0.15) * (1-0.05) * (1-0.02)^10

Well yes, that is a bit cleaner thanks. :P


Quote:

Originally Posted by frtoaster (Post 8835871)
I wish they would show our shield resistances in the UI. It would make it easier to figure these things out. However, I don't know that anyone has actually tested whether the fleet elite shield bonus stacks multiplicatively or additively.

I wish it was in the UI as well, unfortunately the UI team is like some secret team that is locked in an underground fortress bunker.


We did get shield regen in the UI though! Even if only some shields show up (Borg ) and some shields do not show up (MACO :().




Obviously few if any players can actually answer my questions above, maybe we will get a lucky visit from a dev.


All times are GMT -7. The time now is 01:50 PM.