Vininator, on 15 January 2013 - 04:20 AM, said:
*sigh* I guess you can't please everyone.
Paul's idea seemed fine to me. I plan to dust off my 4xMG/LL cicada one this rolls out. How about we see what this does before we bash it? You know, the whole "try it and you might like it" mindset?
It isn't simply about pleasing. It's about a question of how the game rules work and how probabilities work.
If Paul told you that to buff PPCs, he decided to increase their damage to 20 per shot, would you say this is a good idea and just give it a try? Or would you be able to predict that this is probably a bad idea and give feedback accordingly?
Your intuitive grasp of the game is probably good enough to give you a bad expectation about this.
And a similar grasp of the game - not necessarily just intuitive, but actually also based on doing a few math exercises, can tell others that buffing MG crits won't make them worth it either.
These factors influence the importance and effect of crits:
How much armour does a mech carry? As long as you have armour on a hit location, you don't take criticals.
How much internal armour does a mech, especially compared to actual armour?
How much hit points does a component have?
How likely is it, that for any given critical hit, a component takes damage?
How much damage does a component take from a critical hit?
A Catapult for example has 10 points of internal armour in one arm. It can carry 4 times this amount in regular armour.
That means you may need to deal up to 40 points of damage to its ear before you can damage any internal component.
Now, it has 10 points of internal armour there. That already tells you - unless you have a significant crit modifier, you will take out the entire hit location before you take out just one component.
But that's not all - each component has 10 hit points right now. You have 12 crit slots in each arm. So there 's only an 8.33 % chance per crit slot that a single component is hit. One could say that on average, each item takes only 8.33 % per crit slot from critical hits. So an Artemis LRM 20 installed in a Catapult would take about 50 % of all critical damage.
It has 10 hit points, so you would need an effective total crit modifier of 2 to destroy the LRM20 before you destroy the entire hit location.
Of course, if the Catapult really has only one Artemis LRM20 in its ear, the chances are better- only crit slots that actually are filled can be target of a critical, so you only have to deal with 1 or 2 arm actuators where you could waste your damage. So it's more a 75 % of crit damage that will got to the LRM, so an effective total crit multiplier of 1.25 would be sufficient.
Now, this is the caclulation that describes the MG reasonable well. Other Ballistics deal a lot more damage per shot, so just relying on the average will probably give us off results - a single AC/20 that lands a critical hit has a good chance to taking out any component it hits, as long as it does hit.
Note that in the above calculations the effective total crit modifier is dependent on the chance to have a 1-hit, 2-hit or 3-hit critical, and the damage multiplier Paul talks about. So if you have a 50 % chance to score a crit a 1-hit critlca (and no chance for a 2-hit or 3-hit crit for simplicity's sake), a damage multipler of 2 for the MG would basically give it a total crit modifier of 100 % (dealing its full damage on average as crit damage).
Now, that was a Catapult. For lighter mechs, the internal armour is even lower, so it's even less likely you experience a crit before your entire location is destroyed. For heavier mechs, the chances get a bit better.
And the Artemis LRM20 is a pretty large weapon. Some ballistics get larger, but energy weapons are all much smaller. And the same is true for other types of components. Most are simply smaller than this.
So the chances of losing equipment to crits is mostly limited to heavy and assault mechs that utilize large weapons like AC/20s, AC/10s, LB10-X-ACs, LRM20s and Gauss Rifles (which have the added "benefit" of less hit points).
And unfortunately, on the mechs most likely to wield such "fragile" weaponry, you will have 4 or more times armour on a hit location than a single component has hit points. So at best you will have 1/5th of your time in destroying a mech where you can take out a component, and it happens to be the last fifth, when the enemy mech is closed to dying. The other 4/5ths he has his full firepower.
It seems wiser to equip weapons that deliver most of thei firepower also in the first 4/5ths of an engagement.
Edited by MustrumRidcully, 15 January 2013 - 05:05 AM.