Because there is a group of users who are already fanatic about playing the current alpha and which is resisting any changes in the game mechanics they are used to. You could say, they are acting irrational. They also often tend to claim that things must be done the precise way it had be done in Total Annihilation, even if not applicable at all due to the differences in scale and the envisioned focus on high level strategies and massive unit blobs rather than individual unit control. The users in this group don't like to think much. They prefer a static build queue which can be executed in a routinized manner and which ensures the victory in a single effort. They don't like having to fight multiple rounds, in their terms, the game is over once any player has reached the end of the build queue. They don't like choices, they don't like second chances, they don't like complex terrain. They only aim for finding the perfect, universal solution for the game so everything which makes the game more complex or dynamic is not welcome. Skill in their terms is, how precise you are in executing the build order, mix in some micro intensive exploiting of "game mechanics" to rule out these with insufficient APM.
Although perhaps slightly unkind to some folks, I can definitely see some tendencies of that around on occasion. Particularly when some people claim that highly developed micromanagement skills and APM would be mandatory for top-level play.
I'm not entirely sure what you mean by launch timing spread. The use of 'timing' instead of 'time' suggests that you're changing the timing, as in varying a unit's fire rate. Maybe you just mean "inaccuracy"? On a separate note, consider this question: If artillery has longer range, higher damage, and greater accuracy than any other weapon, how do you fight it? You can't dodge, you can't hide, you can't run, and you can't fight back. I think most consider artillery to be naturally weak against movement, both because it's a familiar weakness and because it's the only real weakness artillery has in in an RTS where there is no cave to hide under. I'm also not sure why you've listed 'direct-fire' and 'rockets' separately from artillery and beams. It confuses the issue when the only real differences between them are projectile speed and gravity, which are scalar quantities. (Is a rocket at 0.3 gravity different from an artillery shell at 0.3 gravity? Is a beam with instant speed different from a bullet with nearly instant speed?) And then you have homing missiles, apparently your catch-all for guided projectiles, which could actually be a much more diverse and complex category. Honestly, this thread would make more sense to me if it were titled 'On Making Artillery Better Than Other Weapons'.
I don't know how to express it in proper English, but what I mean is that there is a random factor to the duration of the prelaunch animation, one which can not be accounted for, neither automatically nor by hand. The same jitter is subtracted from the postlaunch animation so the rate of fire remains constant, but the accuracy is reduced against moving targets in general since the linear movement prediction can only account for a median delay. So you are not required to dodge by hand by abusing specific characteristics of the movement prediction, instead the system guarantees that artillery is bad at aiming at single moving targets and you can't cheat around. Rockets and direct fire behave very similar, that true. But there are also distinct differences, such as the detonation on proximity for rockets rather than on direct hit as for direct fire. There are no pure rockets in the current version of PA and there is no proximity trigger either, yet. Low arced artillery shells and direct fire projectiles behave similar, but one is affected by gravity a lot due to the comparable low projectile speed in relation to range, while the other one can ignore it. Artillery also requires a different targeting algorithm, since there are two possible solutions. Also different rules for target selection, for direct fire weapons it would be pointless to aim at empty space on the ground. For artillery, this is a legit option which should be considered by the target selection AI. Beams may look like direct fire, but they behave complete different on a technical level. They are not simulated continuously as a projectile and can therefor never miss, but that results in a more costly calculation since it requires constant LoS tests. Don't mistake them for the starwars style laser shots, which are actually just direct fire projectiles. Guided projectiles are pretty much all the same on a technical layer. While the effect on impact may vary, they all behave the very same way. Turn rate and speed, thats the only thing where they differ. No gravity, but they require a unique steering behavior. The original post isn't about a specific balancing of one weapon type against another, it's meant to show up the various parameters which can be tweaked for each single one of these weapon types to achieve a certain behavior or how the avoid a behavior which could be exploited.
That jitter has to be of the same order of magnitude as the time taken for the target to drive one body-length. I suspect that once the jitter time is less than half the time taken to cover a body-length, it will hit just as well as if there was no jitter. Intuitively, we expect very short jitters to behave similar to no jitter. And very long jitters to have significant effects on accuracy. Of course, very large targets, and very slow targets, are easy to hit, so the jitter has little effect against them. Small, and/or fast targets are where it matters.
You've got it. Thats the idea behind introducing the jitter in that precise point. The actual effects when the jitter becomes bigger depends on the type of distribution used for the jitter (equal or normal distribution), but the characteristic of the hit chance are somewhere along the line of hit chance = min(size / ((jitter * speed) + spread), 100%) Simplified formula, of course, but the actual hit chance should be quite similar.
When it comes to moving targets, you get the same result with angular variation on firing. All your system becomes is a stealthy buff to micromanaging units.
Explain? The dodge mechanic is retained, but it no longer requires micro to execute since any movement is sufficient. Thats not true. Angular variation also decreases hit chance against immobile targets.
Sorry, but please stop and go back. Any unit that is in motion takes less damage (on a statistical level), compared to any unit that is still. Keeping them in motion is hence advantageous.
I can't say that I like that dodge mechanic anyway, but thats something inherited from Total Annihilation. Issue is, the dodge mechanic as it was implemented in TA required even more micro as you had to adapt the movement pattern to the time of flight of the projectile or the turrets turn speed in order to dodge. It would be the easiest to do away with movement based dodge mechanics all together, but I'm not going to risk THAT shitstorm. So the only logical choice is, to simplify the mechanic at least so far, that it becomes usable with reasonable effort at all. Keep in mind, that staying in movement is something the unit AI can do all by itself if required. Active dodging with the TA mechanics is not possible.
You mean the benefit of remaining invulnerable? You can still dodge laser towers with bots indefinitely, never catching even a single projectile. People do bother with this type of micro because it allows to achieve game breaking advantages because the current simulation is not designed to differentiate between fast and slow moving targets and the current workarounds can be abused. So why is it even in the game if not working properly? Because it was in TA. Because the whole bot faction relies on this very mechanic.
