zorak ramone, on 04 January 2013 - 07:45 AM, said:
DPS/HST is a measure of weapon damage efficiency. As you have in your revised weapons stats, DPS/HST goes down with increasing range as it should.
What is your model for determining how much DPS/HST goes down with increasing range?
What is your model for determining how much DPS/HST goes up or down depending on firing mechanics (e.g., the SRM6 vs the AC20, which have comparable range)?
DPS/HST is only one of many stats. In a prolonged standoff, DPS/HST becomes increasingly importantly as other stats become less important. Understand first that DPS/HST is only one narrow sliver of the picture.
The other (largely) important parts of the picture are APT (alpha per ton), useful in calculation for burst-DPS, and DPH (damage per heat), an indicator of how efficient the weapon is when firing. As you yourself stated, range is another factor.
Firstly, I approached it with the idea that these values should not have massive variations. Some variation is necessary, or else all the weapons become 'too' similar - but when it does come down to DPS/HST, you do not want massive variation.
The weapon with the lowest DPS/HST is the ERPPC, at 0.1063. The weapon with the highest is the LRM5, at 0.2787. Now, the question in this case is, what advantage does the ERPPC grant you, in exchange for such low DPS/HST, and what shortcomings does the LRM5 have that grant it such an advantage in this one statistic?
In the case of the ERPPC, it applies all of its damage to a single spot on the mech, and is usable from extreme range - such extreme range, some players might not be able to retaliate against it. The ERPPC has very, very easy application of firepower, and from a massive range.
The LRM5 spreads its damage everywhere on the target mech, and even with my suggested alterations to hardpoints, mounting a bunch of LRM-5's will eat up all of your criticals. In fact, all LRM's suffer from the issue of poor damage application in this manner.
I balanced the individual weapons on a case-by-case basis, analyzing their intended role (and if any modifications needed to be made to their role, and in most cases, none was necessary). Then I analyzed how best to go about making that particular weapon perform its role without being either narrowly constrained or overpowered. This is really a matter of trial and error; I typed different numbers into the spreadsheet until I was satisfied with the derived values. For weapons among one group, I made sure to give their derived stats a predictable 'curve' along which their performance falls. Case and point: Look at how smoothly the standard autocannons transition in terms of DPS per ton, alpha per ton, damage per heat, etc. While the statistical curves of all weapon groups cannot be made absolutely perfectly smooth (at least, not without some very fancy mathematics), there are no statistically significant deviations.
Sadly, there is one value which I had to abstract, since no number can be readily placed on it: Application of firepower. How the f*** does a laser's instant-hit and damage-over-time stack up against Autocannon and PPC damage application to a single body part, but with a non-instantaneous projectile flightspeed? And more importantly, with hit detection and projectiles in their present state, it makes applying damage with autocannons and PPC's harder by some unquantifiable amount, so it's not as if it can simply be playtested for valid data.
In the end, lasers take less skill if you need to do only a slight bit of damage to core a target, and don't particularly care about spread. However, they take more skill than a PPC if you need to apply all the damage to one body part. Historically, I find that one needs to do the latter of the two more often. I designed the weapons so that this fact is reflected in their alpha-per-ton. The PPC does less, but always to a single part. The large laser does more - but unless your aim is absolutely crazy-awesome, not all of that damage is going to make it onto one body part. Chances are, the entire damage of the beam might not even make it to your target mech.
Best of all, since the values don't fluctuate that much, it would be extremely difficult (hopefully impossible) to find an overpowered combination of weaponry to use on a mech - and the non-radical deviations ensure that even if some team does come up with a metagame loadout that is definitively superior, it will only be by a percentage point or so - which is well within the standard deviation of player skill.
Unfortunately, I cannot stamp these spreadsheets as 'balanced' until there is experimental proof through extensive playtesting that the mathematical analysis of these weapons translates flawlessly into actual gameplay as intended. These spreadsheets are only a starting point. Things would inevitably have to be tweaked up or down a few percentage points; I am almost certain of it.
However, I can definitively say that these spreadsheets are
damn close, and are way closer to an ideal balanced state than the game currently is. Not only that, but with the sole exception of ultra-autocannon firing mechanics, everything on the spreadsheets can be typed in to the servers and (theoretically) immediately hotfixed, with very little effort on the part of the devs.
Also note that ECM did not exist whatsoever when these were composed, and I have my fair share of problems with the current ECM system (I think it's total ****). Missile boats do not deserve free reign over open areas - but nor do ECM users deserve immunity from bombardment.
Note additionally that these balance changes assume that non-artemis SRM spread at maximum effective range (270 meters) is not so wide as to be completely useless. It was a little screwy when SRM spread was so tiny as to be identical to autocannons, but nailing an Atlas torso (the whole thing) with most rockets at 150 meters should be the norm.
I actually advocate perfect SRM accuracy (no spread), but sequential firing - to match the lasers' damage application model. Dealing all damage to one body part should either require a stationary target or a massive display of skill.
Edited by Xandralkus, 05 January 2013 - 04:51 AM.