Gaan Cathal, on 21 January 2014 - 12:53 PM, said:
[/size]
Where are you deriving your 1/12 chance from? Assuming a side torso with all critslots fully occupied? If so, I did specify that I'm talking in that specific case about isolated ammo caches - this is exactly why I used the "two tons in a leg" example in my first demonstrative calculation - because the leg has no empty slots if it has two tons ammo.
In cases where ammo is non-exclusive, the formula I used holds for single attacks with a minimum 10 damage:
For six tons ammo with two tons heatsinks:
0.42*(6/12)*0.1 = 0.021 = 2.1%
[/size]
I was agreeing with you that it's not worth our time. And, frankly, I find it extremely unlikely PGI have done so, which was rather my point. These are the developers of the infamous 3s Jenner logic, remember.
I was taking the worst possible chances, which is one ammo crit out of twelve possible crit locations. These are the worse numbers you can find.
Once more, you are short cutting the math, as you also have chances of doing two crits and three crits in a single hit.
You have a 25% chance of getting a single crit.
You have a 16 % chance of getting two crits with the same hit.
You have a 3% chance of getting three crits with a single hit.
You need to do the math seperately for each chance, and add the chances together.
Another possible formula for this could be:
((Chance for a single crit x chance of hitting the ammo crit) + (chance of a double crit x (chance of hitting the ammo x 2)) + (Chance of dealing a triple crit x (chance of hitting the ammo x3))) x (the chance of causing an ammo explosion) = the total chance that a single shot has for hitting an ammo crit and causing that ammo crit to explode per shot.
(I suspect that the number end result will be the same as my other version, or very close.)
Even if you go by crits against ammo in the legs, there is still a better chance than the "worst case" I've been presenting.
The leg has 6 crits. 2 are open. If those two crits are filled with ammo, that's a 2/6 (1/3) chance that it will hit the ammo (if actuators do count for the crit "rolls" in MWO that is). So, the odds are still better than 0.32%, and worse than 6.6% (or possibly is the max 6.4% from two ammo crits available math, depending upon if Actuators can get hit or not).
You summing the whole crit system into 42% (should be 44% if my numbers are right anyway) as a SINGLE CRIT POSSIBLE. That isn't correct, as you only have a 25% chance of causing a single crit with a single shot. Then you have a farther 16% chance for a double crit, and a 3% chance for a triple crit. All these other chances INCREASES the likely hood of hitting a 1 in 12 ammo crit, as well as (if you have more than one ammo) increases the likely hood that you will hit more than one ammo crit and make one of them explode. Often times, you only need one to blow to kill a mech.
Lets give this some practical examples, shall we? My Battlemaster:
http://mwo.smurfy-ne...d83edc6e97ad3eb
Lets say, best chance for you, you hit my side torso which is striped of armor. (My Battlemaster has what is probably considered average ammo for the weapons I carry.) I have 5 tons of ammo in there, with 4 crit slots of "buffer". a heat sink and two med lasers. So, that's a 5/9 (0.56 or 56%) chance of hitting ammo with a crit.
I'm just going to say you hit me with a single AC10-20/PPC/Gauss/etc that is large enough to destroy the crit slot it hits, if it hits.
Lets do the math now, as per individual chances of all possible crit types:
(Crit chance) x (chance of hitting the ammo crit) x (number of crits) x (ammo explosion chance) = (total chance of ammo explosion per shot)
0.25 x 1.00 x 1.00 x 0.10 = 0.025 = 2.5%
0.16 x 1.00 x 2.00 x 0.10 = 0.032 = 3.2%
0.03 x 1.00 x 3.00 x 0.10 = 0.009 = 0.9%
Now, add in the "chance it's going to hit the ammo" into it, instead of setting that to 100%:
0.25 x 0.56 x 1.00 x 0.10 = 0.014 = 1.4%
0.16 x 0.56 x 2.00 x 0.10 = 0.01792 = 1.792%
0.03 x 0.56 x 3.00 x 0.10 = 0.00504 = 0.504%
Total actual chance of hitting, critting, and exploding an ammo crit in my right torso is actually:
1.4% + 1.792% + 0.504% = 3.696% chance. (Not your simplified math of (0.42 x 0.56 x .10 = 0.02352) 2.352%)
Take the left side now. I have 3 crits ammo out of 12 (9 other crits are taken by 2 med lasers, 2 SSRM2s, BAP, and one DHS). This is a 3/12 (0.25 or 25%) chance of hitting ammo for every crit applied.
0.25 x 0.25 x 1.00 x 0.10 = 0.00625 = 0.625%
0.16 x 0.25 x 2.00 x 0.10 = 0.008 = 0.8%
0.03 x 0.25 x 3.00 x 0.10 = 0.00225 = 0.225%
The chances of a single shot causing a crit, hitting an ammo crit, and making it explode is:
0.625% + 0.8% + 0.225% = 1.65% chance. (Not 1.05%.)
So, the odds are still better than 0.32%, and worse than 6.6% for each side torso, as I stated before.
Now, lets do this for my Dragon build:
http://mwo.smurfy-ne...97a9f01e57ba7ef
Same thing here, but slightly different loadout. Same math formula.
(If anyone can confirm if Actuators can get crit in this game, it would be good to know. Otherwise, this set of math will presume that one can crit them as "waste" crit chances).
The right arm of my Dragon has 3 tons of ammo and an UAC5. That's 3 ammo crits and 8 other crits. So, a 3/8 (0.375 or 37.5%) chance of hitting ammo.
0.25 x 0.375 x 1.00 x 0.10 = 0.009375 = 0.9375%
0.16 x 0.375 x 2.00 x 0.10 = 0.012 = 1.2%
0.03 x 0.375 x 3.00 x 0.10 = 0.003375 = 0.3375%
Total: 0.9375% + 1.2% + 0.3375% = 2.475% chance of hitting, critting, and exploding my ammo. (Not 1.6875%)
Next spot with ammo, right torso. It's got 1 ton of ammo, an XL crit slot, and BAP. So thats 1/6 (0.167 or 16.7%) crit slots. This is also presuming that engines can take up crit hits, even though they don't do anything right now.
0.25 x 0.167 x 1.00 x 0.10 = 0.004175 = 0.4175%
0.16 x 0.167 x 2.00 x 0.10 = 0.005344 = 0.5344%
0.03 x 0.167 x 3.00 x 0.10 = 0.001503 = 0.1503%
Total: 0.4175% + 0.5344% + 0.1503% = 1.1022% total chance of a single hit shoring a crit, hitting ammo, and making it explode. (Not 0.7014%)
Still sits between 0.32-6.6%.
All I have to ask is, don't make me start adding in the concept of "several weapons hitting at once" and what "they will threaten" for crits/explosions... (AKA: Take the total number, and multiply by number of weapons hitting. We do have instant convergence, don't forget.)
Also, none of these calculations include the risk of when a section with ammo is destroyed and what the chances of an ammo crit exploding under those situations... (we still have to include it in these numbers to be fair to the current system, before you arbitrarily decide to chance some numbers.) Personally, I think bumping it up to 20-30% chance to explode would be far more reasonable of a risk, as one should consider the amount of ammo we much place into our mechs to keep them working in the same manner as TT.
(Edit: Added in totals using your math.)
Edited by Tesunie, 21 January 2014 - 02:32 PM.