Indoorsman, on 07 November 2012 - 01:11 PM, said:
The way the game is played is not equal or even similar to the way TT was played. The stats are "proven" to work for a board game, not a computer game. Are you ignoring the variables of how this game is played? It's 100% different than a board game as far as HOW you play it. You can't just change the weapon stats so that you still do the exact SAME damage over a 10s window. That's just a dirivative/ratio of TT stats. The only way it would make sense is if this game wasn't first person, you DID control multiple mechs, it still had a tile layout etc.. In otherwords if this was TT on the computer then you could use TT values. It's not so it shouldn't.
It is true that aiming with the mouse is completely different than 2d6 to determine chance of hit based on some environmental factors, and 2d6 to determine which locations successful attacks do damage to. This does not necessarily mean that there is something inherent about moving to a mouse aiming system that throws using TT values for weapons out of a desired hierarchical balance.
Let us have a look at what effect the differences between TT and MWO's aiming systems have on chance to hit.
TT's environmental factors that affect the chance to hit are: Cover (Woods, water, hills, and building), Attacker speed and stance, target speed and stance, Damaged actuators, Damaged sensors, and Heat.
MWO's environmental factors are the same, but lack stances (no crouch or prone currently), critical hits and Heat. All pretty self explanatory, we all know that trying to hit a Jenner running at 128 kph is harder than one standing still, and similarly but to a lesser degree trying to hit as that Jenner running at 128 kph is harder. Cover also quite clearly has an effect. The trick comes in trying to determine if the real time mouse aiming nature of MWO makes some or all of these factors more pronounced, or less pronounced. We can agree that they do have an effect in both gameplay methods.
Do these differences in how MWO's environmental factors affect hit chance change the hierarchical balance of weapons? No, because all of these factors affect all weapons equally (I hear an argument brewing, wait for a few points), thus do not move any individual weapon up to down in hierarchy.
Beyond environmental factors, let us look at weapon factors.
TT's weapon values that affect the chance to hit are: Short, Medium and Long range and that is it.
Does the distance from the attacker to the target affect hit chance in mouse aiming? Yes, it is harder to consistently hit a target that is smaller in pixel/screen-percentage size than one that is larger in size, but it is not the weapons range value in MWO that affects hit chance, but the raw distance between the attacker and target, making this more of an environmental effect in MWO. MWO's weapon range value does not affect hit chance, so hit chance by the weapon's range value is the same for all weapons. This is the first factor that throws a monkey wrench in the hierarchical balance of weapons, where some weapon have greater or lesser ranges (and thus hit chance at those ranged) than other weapons to balance them in respect to their damage, weight, heat production, or ammo consumption. In an attempt to keep TT's weapon balance PGI makes weapons do less damage over range past the weapon's range value. The specific effects this had on the weapon balance hierarchy, I really could not say, and honestly have little desire to attempt to fully analyze if it caused any repositioning of weapons in the hierarchy. I am going to guess damage drop-off has little appreciable effect on the hierarchy due to any notable complaints about it in the CB or OB forums.
MWO does throw in an additional weapon factor in chance to hit that does not exist in TT by adding in additional elements such as the various ballistic properties of each weapon's shot, making some weapons easier to hit with than other weapons as some projectiles take longer to travel the same distance as another weapon. This is the second factor that throws a monkey wrench into the desired weapon hierarchy, and again I am not going to really analyze it, but I know it has a notable effect on the overall weapon balance hierarchy, such as AC/20s being harder to brawl with at 270m than a Gauss rifles.
Moving on from chance to hit, to location of a hit:
In TT there are no weapon values that affect the location of a hit. No value of a weapon affects the 2d6 roll. It could be argued that the scatter tables for multiple submunitions of LRMs, SRMs, SSRMs, LBX ACs, and UAC have an effect here.
In MWO, mouse aiming affects all direct fire non-guided weapons equally. So by itself, mouse aiming does not change the hierarchical balance of individual weapons compared to each other in where they hit when they hit. Just like in chance to hit, the MWO factor that does have an effect on the balance of where weapons hit compared to each other is the ballistic qualities of the shot as a slower projectile might strike a moving mech farther in the direction it is moving away from. In this regard, PGI added to laser weapons damage over the duration of beam time (making them more likely to multiple locations rather than one single location) to try and bring them back in-line with the other weapons that do not instantly hit.
We should be able to agree that there is nothing inherent in mouse-aiming that causes individual (non-guided) weapons in the balance hierarchy to shift up or down in relation to the other weapons, as they all benefit equally. Where they do not benefit equally is when other elements not specifically related to aiming with a mouse are added, such as differing ballistic properties of a weapon's shot.
The heat generated by a weapon and the heat dissipation rate of heat sinks have no effect on aiming, hitting, or where a hit deals damage. Trying to use these two factors to try and balance issues perceived to be caused by shifting to a mouse driven aiming system is fundamentally flawed. Should we get rid of mouse aiming? No. Should we get rid of varying ballistic qualities of weapons? Maybe...
PGI introduced different ballistic qualities to shots being a factor in hitting, so any effects these have on a weapon's place in the weapon balance hierarchy is due completely to PGI's design decisions related to the speed of the projectile, but are not an inherent issue with mouse aiming, and certainly not related to using TT weapon values with mouse aiming.