I think you may be on to something as I've been paying closer attention to trying to line dead center on the lead indicator and it does seem to help (firing at point blank is still pretty screwy but I gather that I haven't been unique in experiencing whacky mechanics there).

This just means that you (the player) is aiming better. Your targeting computer misses more often at the edges than in the middle of your firing arc. Think of it this way, the longer the line is from the lead circle to the circle in the center of your firing arc = more miss chance.

That's a valid point. It seems i have to supplement my original statement.

 

"If two craft have the same effective hitpoints, the craft with evasion requires the same number of shots fired to kill it as the craft without evasion" needs to be supplemented with "as long as it takes more than one hit to kill the latter"

 

Luckily, i do not believe oneshotting a craft other than a scout who sacrificed health for evasion is common in GSF, so the original statement holds true for the most part.

Still salty, but I'd like to point out that with shield piercing being so common, any craft that takes some hull damage and flies away to regenerate shields is at risk of being one-shot from their current health (which is what Ramalina said in the quote).

So on the whole, if playing against a ship with evasion, your skill is slightly less valuable than it would be if you were playing against a ship with the same average health achieved with shields. Not by a large amount during normal play, but if you play long enough there will be some memorable unpleasant experiences with strings of misses. The lower the level of evasion, the less area there is under the 'lots of misses' tail of the histogram, so in cases where people aren't stacking the maximum amount of evasion it's not an issue, and even at 41% evasion it's only a teeny tiny issue.

This is what I meant when I said earlier that RNG devalues skill (also your comments about noobs stacking evasion). Can you tell us (or better yet, give a formula telling us) what percentage of the time you need more shots to kill someone with evasion than you would if their evasion were hull and shields (assuming equal EHP and always full hull)? I ask for number of shots because, as you later point out, number of shots to kill can greatly extend time to kill when firing opportunities are short and far between.

Thanks for doing all this math, it's appreciated.

Evasion as far as avoiding debuffs depends on how common and powerful debuffs are.

It also depends on how easily removed those debuffs are. As an example, in Guild Wars, there are two types of debuffs: conditions and hexes. Conditions are very easily applied, but also very easily removed. Hexes are somewhat harder to apply (a skill is generally dedicated to applying a hex, as opposed to applying a condition as an add-on effect), but much harder to remove. In that game, conditions were serious if ignored but generally very quickly removed, and hexes were major problems that could shut down your entire team's strategy because of the difficulty of removal.

Currently debuffs can't be removed at all, but I could see bombers (or someone) getting something to that effect later on.

The other conclusion that came out of my look at this is that front loaded burst damage is really good. At the proper range, burst cannons are the best weapon in game. If you gave Strike fighters burst cannons, I think it wouldn't take long for threads complaining about the stupidly overpowered nature of ion cannons followed by burst cannons at close range to surface.

I'd like to point out that damage now is always better than damage later, except at the extremes (i.e. 10 damage now vs 100 damage later). Considering the shield regen mechanics, I honestly think BioWare would not go wrong to consider balancing ships around how much frontloaded damage they can do, and the ease at which they do it. Then player skill begins to focus more on how to put out your burst damage without getting hit by theirs. Whether that's a good direction for the game is a question I don't want to get involved in, but it's worth pointing out that's how this game (and dozens if not hundreds of others) currently work.

And honestly, if that's what strikes need in order to compete with scouts and gunships, it might not be so bad. They don't have the engine efficiency to reliably get into close range, so it might even be balanced.

Also, while I like extended range heavy lasers on my Starguard, it's possible that damage capacitors are the best possible capacitor for all ships that can use them. At least for any player that actually aims to shoot rather than following a 'spray and pray' policy. I'm definitely changing to damage capacitors on all my ships except the Starguard, for it I'll have to spend some time experimenting to see how the tactical value of the extra range compares against the increased per shot probability of kill that increased damage provides.

I particularly like the design of capacitors as is. The optimal choice for your build (and there's always an optimal choice) depends on a lot of factors: playstyle, build goals, player skill, weapon components, and probably a couple others I haven't thought of. I think the benefits of the range capacitor (larger threat radius, extra shots while approaching the target, advantage in head to head combat, etc) are underrated on most weapons (although in fairness the extra shots will always be relatively low damage), and I especially like that you can "graduate" from damage capacitor to frequency capacitor as you get better at quickly lining up shots.

I'm also much more optimistic about the effect that the minimum charge requirement will have on balancing railguns. Having to wait before you can fire is actually a fairly substantial penalty, though it affects a gunship at range less than any other ship, because their shot opportunity windows tend to be much longer.

Perhaps I'm too pessimistic on this front, but I'm not encouraged by a lack of numbers in the patch notes. I suppose another tool they could tinker with would be the cooldown between shots. This would have a similar effect, but would also potentially give the target's shields more time to regen without the usual paradigm of "more shield regen time = more railgun damage".

In this case the tail of the distribution containing misses is longer than the tail containing best cases.

How so? Isn't the distribution symmetrical? What reason do you have to assume it isn't? Because if i understand you right, you are basically saying that, for example, if i have 50% accuracy, then the probability 90% of my shots will hit is different than the probability 10% of my shots will hit. That does not sound right to me.

(I wont be able to continue this debate as i cancelled my subscription. Will still play GSF but without forum access. You guys have fun:))

This is what I meant when I said earlier that RNG devalues skill (also your comments about noobs stacking evasion). Can you tell us (or better yet, give a formula telling us) what percentage of the time you need more shots to kill someone with evasion than you would if their evasion were hull and shields (assuming equal EHP and always full hull)? I ask for number of shots because, as you later point out, number of shots to kill can greatly extend time to kill when firing opportunities are short and far between.

The answer to this is, you have to know the exact total health of the opponent, damage done per shot, and level of evasion. Once you have those, you figure out how many shots it takes to kill the ship with shields, this is X shots. Then you use the binomial distribution and calculate, or use a look up table and find the probability that you'll kill the ship with evasion with X or fewer shots ( you have to evaluate each number of shots from 0 to X individually to make the histogram, then you add the values for 0, 1, 2, . . . , X shots and then subtract that from 1, which gives you the probability that you need more shots to kill the ship with evasion, or depending on the reference you may be able to find a cumulative distribution table for the binomial distribution with the appropriate parameters, in which case you just have to look up one number on the table). If any of the factors I mentioned in the first sentence change you'll have to re-compute with the new values.

My take home conclusion was that balanced with the probability of needing fewer shots in some cases, it's not really worth worrying about at the levels you see in GSF right now. If they introduce passive evasion of 50% or more, then I'd start being seriously concerned.

I particularly like the design of capacitors as is. The optimal choice for your build (and there's always an optimal choice) depends on a lot of factors: playstyle, build goals, player skill, weapon components, and probably a couple others I haven't thought of. I think the benefits of the range capacitor (larger threat radius, extra shots while approaching the target, advantage in head to head combat, etc) are underrated on most weapons (although in fairness the extra shots will always be relatively low damage), and I especially like that you can "graduate" from damage capacitor to frequency capacitor as you get better at quickly lining up shots.

Frequency capacitors are mostly good when you have terrible aim. Pair them with a rapid fire weapon to increase your chances in the, 'will I hit' lottery. When you get good at aiming switch to a slow firing weapon with damage capacitors, and you should be much more deadly. (Beginners reading this, this applies to weapons being used at appropriate ranges. If you're in close dogfights for example, heavy lasers + damage capacitors would be a bad choice because of how hard it would be to hit anything due to the narrow cone of fire and tracking penalty on heavy lasers. Burst laser cannons + damage capacitors on the other hand, would be an excellent combo for short range fighting if you have good aim).

For an increase in practical damage, frequency helps when the end of a shot opportunity clips the last shot of a weapon in a case where it wouldn't have clipped the last shot of the weapon if frequency capacitors had been equipped (in most cases this is going to be a rare occurrence). Another case where frequency capacitors can help is where a target doesn't move, then it really can provide a flat dps increase. Otherwise it's mostly good when paired with a rapid firing weapon for someone hoping to replace the ability to aim well with sheer volume of fire.

Damage capacitors: if you're good at flying and good at shooting.

Frequency capacitors: if you're terrible at flying, terrible at shooting, or terrible at both, and remember to pair with a weapon with a high rate of fire.

Range capacitors: Possibly interesting with heavy lasers, otherwise mostly only good if you find that you would have had a shot or a higher damage shot with just a little bit more range. Or special uses like outranging a turret.

Damage capacitors: if you're good at flying and good at shooting.

Frequency capacitors: if you're terrible at flying, terrible at shooting, or terrible at both, and remember to pair with a weapon with a high rate of fire.

Range capacitors: Possibly interesting with heavy lasers, otherwise mostly only good if you find that you would have had a shot or a higher damage shot with just a little bit more range. Or special uses like outranging a turret.

Range is easily the worst of the three choices as it takes next to nothing in time to close the difference.

Damage vs frequency is a little dependent on taste. Though I would argue that a more fair comparison is crit vs power (respectively). With power, people expect better minimum damage output, but with crit they expect better burst possibility. I would say that with frequency increases your chances of hitting, though they will individually be for a lower amount. Where as damage caps add to any hits but you get reduced chances to hit.

Oh and I think that freq cap with a rapid fire laser takes MORE skill than one that stacks all the damage in one shot (aka burst laser with a damage cap). You have to constantly realign your guns to the lead target marker, but if you are really skilled at it, your DPS output is better than the "bursty" combos.

The answer to this is, stuff.

Well, that's butts. I don't remember nearly enough stats to do that right now.

Burst laser cannons + damage capacitors on the other hand, would be an excellent combo for short range fighting if you have good aim.

If I had amazing aim (65%+ consistently), I would seriously consider switching from damage to frequency with burst, because I'd be hitting enough that I could afford to pop off two shots at the guy faster.

Also, range capacitors have a special place with quads and pods, because that puts your "dogfighting range" up to about 6 km with both (assuming you go for dot instead of range on the pods, which is questionable, but hey).

How so? Isn't the distribution symmetrical? What reason do you have to assume it isn't? Because if i understand you right, you are basically saying that, for example, if i have 50% accuracy, then the probability 90% of my shots will hit is different than the probability 10% of my shots will hit. That does not sound right to me.

 

(I wont be able to continue this debate as i cancelled my subscription. Will still play GSF but without forum access. You guys have fun:))

Ah, no that's not exactly it. The probability of missing 90% or hitting 10% should be the same for any given number of shots. What I'm talking about though is, "how many shots is it likely to take me to kill the target." With shields as soon as damage per shot times number of shots is greater than or equal to total health you have a kill. If you have fewer shots the probability of a kill is zero, if you have more the probability of a kill is 100%. So the number of shots to kill is static. Dealing with evasion, while there is a minimum number of shots needed to kill, there is no maximum number of shots needed to kill. Each number of shots, from the minimum all the way up to the number of shots that can be fired in a GSF match before a time limit or server reset ends the match, has it's own probability of a kill.

In this case the distribution is not symmetrical. The minimum number of shots is one, the average depending on the weapon and ship in question might range from one point something to maybe 12 or 15 or something like that. With evasion it's possible, if unlikely, to have it take a thousand shots more than the average to kill a ship with evasion. Therefore there is a very small probability that exists at that point in the distribution. There is no chance that it will take a thousand shots less than the average to kill the target, because the minimum number of shots is one, and the average number is less than one thousand. So the distribution is forced to be asymmetrical.

Oh and I think that freq cap with a rapid fire laser takes MORE skill than one that stacks all the damage in one shot (aka burst laser with a damage cap). You have to constantly realign your guns to the lead target marker, but if you are really skilled at it, your DPS output is better than the "bursty" combos.

The thing is that in GSF burst damage is a lot better than sustained dps due to the typically short duration of opportunities for shooting at an opponent with any chance of hitting. Railguns and burst laser cannons are wildly popular for a reason.

If you're hitting with most of your shots, a slower firing high damage per shot weapon gives the maximum burstiness in a dogfighting environment.

Frequency upgrades are a buff to sustained dps (as long as you don't run out of power). They can out dps a bursty weapon over time, but that's hard to do with a fast moving erratic target. So say you're taking a Novadive out to hunt turrets and gunships that don't run when they should. Then frequency can be pretty good, and partly that's because the baseline scouts don't have much in the way of bursty options for blasters. Frequency upgrades just are not good for burst in a dogfight if the other pilot is maneuvering at all, and to match a bursty weapon with sustained dps you'd have to be insanely good at keeping your cursor over a moving point even accounting for server lag. If you have trouble hitting at all though, the additional shots from frequency upgrades can be very useful. Another case where frequency might shine is if you have distortion field and do a lot of head to head passes with people who don't have distortion field.

It's worth noting that the size of your computer screen, quality of internet connection, and how you have mouse sensitivity adjusted can have a large impact on your baseline ability to land shots, which in turn affects if an increase in chances to hit or increase in damage of hits that you land is more valuable.

With evasion it's possible, if unlikely, to have it take a thousand shots more than the average to kill a ship with evasion.

I believe the graph looks something like this, with our mode being somewhere around 3 and our mean (labeled "average" on that chart) closer to 4.5.

This is, of course, assuming we're shooting at tofu (i.e. the target does not weave or use any cooldowns that increase survivability). In a realistic fight, the curve skews to right (I believe that is the correct term), meaning that evasive maneuvers, component abilities, and shield regeneration all increase shots to kill. Skill (that is to say, accuracy) on the part of the attacker skews the curve to the left (for example, if you have 65% accuracy at the end of the match, you take fewer shots to kill someone than if you had 35% accuracy).

My problem with evasion is that this tail (I believe that is also the correct term) is immune to player influence but still increases the average (mean, median, and mode) number of shots to kill. In the extreme circumstances, the number of shots to kill can be very large.

I particularly like the design of capacitors as is. The optimal choice for your build (and there's always an optimal choice) depends on a lot of factors: playstyle, build goals, player skill, weapon components, and probably a couple others I haven't thought of. I think the benefits of the range capacitor (larger threat radius, extra shots while approaching the target, advantage in head to head combat, etc) are underrated on most weapons (although in fairness the extra shots will always be relatively low damage), and I especially like that you can "graduate" from damage capacitor to frequency capacitor as you get better at quickly lining up shots.

imho Range Capacitors are next to useless. Damage Capacitor is a great place to start when learning GSF, Frequency Capacitor is the top choice once you get some skills on all ships.

imho Range Capacitors are next to useless. Damage Capacitor is a great place to start when learning GSF, Frequency Capacitor is the top choice once you get some skills on all ships.

I went with range capacitors on both of my strike fighter pilots, I've found that the extra range matches my heavy lasers up to my concussion missile range fairly well (heavies still end up having slightly shorter range, but it's nice having the reticule turn red when I'm able to start locking on with missiles) and it gives me an extra second of fire while closing on gunships. Also makes hitting those fleeing scouts easier.