User talk:Tennessee Ernie Ford/Shortcuts/Guide to Caster Weapons

Health vs AR Mods for casters
You claim that health is often better than armor, and then follow up with noting that +30 health keeps you alive better than +5 armor if the amount of damage it takes to kill you is at most 361. But who dies in at most 361 health worth of damage, other than low levels (who will reach max level before optimizing their equipment) and people with suicide builds (who axiomatically don't care if they die, anyway)? Even assuming no healing at all, it should take substantially over 500 damage to kill you. If you get some healing, armor is that much further superior. Additional health is only better than additional armor if most of the damage you're taking is armor-ignoring, which is pretty rare. Additional armor is better than additional health most of the time, at least for PVE, when there are healers readily available, so that it should typically take thousands of damage to eventually drain healers' energy and finish you off, making +5 armor substantially better than +30 health at countering quick spikes and many times better than +30 health over the course of a battle. Quizzical 20:06, October 3, 2009 (UTC)


 * Whether we end up in agreement, thank you for posting. I'm still trying to make sense of a lot of the game mechanics and separate out the signal:noise from various comments posted. It helps to "talk" through details with people who are carefully analytical (the game seems to fuel a lot of emotional debates, as evidenced by the responses to the ill-fated hench-bar contest).


 * As to the point at hand, the short story is: I accept your conclusion that armor is better if the goal is to maximize the amount of damage reduction over long periods of time. I'm not sure, however, that I agree with the premise: I seem to encounter lots of short battles with more armor-ignoring damage. The extra health gives the team just enough time to remove the most serious threat or two; the remaining foes are too busy surviving on their own to worry too much. My H/H teams appear to be significantly more effective using Health mods.


 * I play with a friend who has also been moving towards high health mods, but based on your ideas here, I suspect he'll reconsider and return to high AR instead. His play style more closely resembles yours than it does mine (I tend to take bigger risks, strong offense is key to my defense, and I assume short battles.) He had lots of trouble in Thirsty River; he could hold out infinitely long against the constantly reinforced teams, but he couldn't take down the Priests. On the other hand, if I wasn't careful about aggro, game over; when I was careful, speedy gonzales.


 * So, based on your comments, I'll update the note on the page here to more clearly present the trade-off between AR and Health. If I word it correctly, it should allow the cautious to see an obvious case for AR and the incautious to choose health. (Presuming, of course, you [or others] don't offer other, convincing arguments against health.)  &mdash; Tennessee Ernie Ford ( TEF ) 23:22, October 3, 2009 (UTC)


 * Let's start by distinguishing between health degeneration and direct damage. In most battles, I have quite a bit more health regeneration than degeneration, so if we're considering healing, that, on net, means more health than if we only considered direct damage.  The exceptions are mobs that rely heavily on health degeneration, but this leads to long, drawn out battles.  The simple fact is that it takes a long time to kill anything with health degeneration; even the cap of -10 degeneration takes 25-28 seconds to kill a player with no healing or direct damage in that time.
 * So that leaves direct damage, and most direct damage is armor-respecting. In particular, all vanilla attacks are armor-respecting.  The extra damage of an attack skill for a martial weapon user is typically armor-ignoring, but that doesn't mean that the entire attack is armor-ignoring.  Only the extra damage from the attack skill ignores armor, while the base damage of the attack still respects it.  The base damage is very often more than the added damage of the attack skill, so the bulk of the damage from warriors, rangers, and paragons is armor-respecting.  Dervishes also do a lot of damage with spells, but their spells respect armor more often than not, so the bulk of dervish damage is armor-respecting, too.
 * That leaves assassins alone as a martial weapon class that often does a majority of its damage of the armor-ignoring variety. Even that depends on the particular assassin build, though, and essentially all assassins have a considerable fraction of their damage respect armor.
 * Among casters, virtually all elementalist damage respects armor. Nearly all ritualist damage does, too, with their spirit attacks as the main exception, but there aren't many spirit spammer mobs--and there cannot be packs of them or they would kill each other's spirits.  Among monks, only smiting monks have significant armor-ignoring damage, and a majority of monks are not smiting while even smiting monks do a lot of armor-respecting damage, so the majority of monk damage respects armor.
 * Necromancers are more mixed, with some doing purely armor-respecting damage, and others having armor-ignoring damage account for the substantial majority of what they deal out. Minions do purely armor-respecting damage, as does the vanilla attack from a necromancer.  Casts are a mixed bag, with some respecting armor and some not.  On net, the considerable majority of necromancer damage respects armor.
 * That leaves only mesmers, and nearly all mesmer skills that deal damage make it armor-ignoring damage. A significant fraction of the damage mesmers deal is from their plain attacks, so while the bulk of mesmer damage is armor-ignoring, it's far from all of it.  It's also important to note that mesmers can't spike damage very well.  Quite a few mesmer skills that deal damage only deal damage when the target does something to trigger it.  Most mesmers really can't do much to a target just standing there, so if a bunch of mesmers are going to kill you with armor-ignoring damage, it will probably take them quite a while to do it.  Packs of them can't stack the same hex to spike damage, either.
 * So we have that six classes have most of their damage respect armor, three are mixed, and one is predominantly armor-ignoring. But the classes are not all equal in the damage they deal.  There are vastly more warrior mobs in the game than mesmers, for example.
 * When I have characters die, the most common causes are a bunch of mobs of the same class (most commonly warrior or ranger) all attacking one character with vanilla attacks at once, and elementalists spamming very powerful spells. That's mostly armor-respecting damage.  If one gears to counter the most dangerous mob attacks, that means countering armor-respecting damage.
 * It's also important to consider the effects of hard mode. Anything that works in hard mode will nearly always work in easy mode, but the converse is far from true.  Increasing mob attributes scales up both armor-respecting and armor-ignoring damage by about the same amount.  Increasing mob level scales up armor-respecting damage still further, but does not change armor-ignoring damage.  A strong mob may deal twice as much armor-ignoring damage in hard mode as in easy mode, but three times as much armor-respecting damage.  If the choice between health and armor being better in general in easy mode is even debatable, then it's not even close in hard mode.  Armor will be a lot better.  Armor-ignoring damage from monster skills notably doesn't scale up at all for hard mode.
 * When I said over the course of a long battle, that doesn't only mean a drawn-out three minute battle to wear down some boss. That means long enough for healers to see who is taking damage and get several heals off.  Ten seconds is usually more than enough for this.  It means as opposed to when mobs spike a character dead in two seconds before healers can react.  The relevant difference, after all, is not the duration of time, but whether or not the character gets healed.
 * So why do so many players prefer health to armor? There are several reasons, I think.  One is diminishing returns to scale, which is most easily intuitively grasped by considering extreme cases.  Someone who has only 100 health would benefit greatly from having 30 more, but not so much from +5 armor.  For someone who had that little health, additional health really would be superior to additional armor.  Someone who had 2000 health would not benefit much from having 30 more, but could probably use the extra armor.
 * This matters because it is common for players to say, hey, let's do lots of damage. Most players like to be damage dealers, which is why in so many games, it's easy to find more damage dealers than you need, but hard to find enough decent tanks and healers--at least if you exclude players of tank or healer classes who will actually try to play as damage dealers.  Superior attribute runes for higher attributes can be tempting, but they reduce the player's health.  Players who have 330 health instead of 530 may well benefit more from the extra health.  If this means that the player is dying, then the real solution is not a suffix mod of fortitude, but rather, scrapping the superior attribute runes.
 * Next is ignorance. Players often prefer the benefits they understand to those they don't.  It is not hard to understand what a +30 HP mod does.  The effects of armor are more subtle.  Players used to other games may think that 60 armor on each piece of armor means 300 armor total, and an extra 5 is inconsequential.  Even those who realize that 60 armor on each piece means 60 armor total may be used to other games where the effect of armor tapers off faster and not realize just how powerful extra armor is.  I'd bet that a considerable majority of the players in the game do not realize that the effect of armor is exponential.  In most games, it isn't; in fact, I can't think of another game off hand where it is.
 * Finally, players tend to assume that something expensive must be good, and thus want it. I'd bet that a large majority of the players who have or want a voltaic spear now wouldn't have one if they were given out cheaply by collectors.  Fortitude mods are more expensive than defense mods because they are so much rarer, or rather, because the max version of them is so much rarer.  A fortitude mod that comes on a gold weapon typically won't be max.  A defense mod that comes on a gold or purple weapon is 100% guaranteed to be max.  That means that there are far more of the latter floating around than the former.  With a substantial fraction of the playerbase wanting the former, it drives up prices.  Quizzical 00:35, October 4, 2009 (UTC)


 * Re: most common reasons to prefer health: I agree completely. Prices drive preferences far out of proportion to the value of coolness. Similarly, I agree that +30 health is far easier to make sense of than any type of AR benefit. Exponential functions were hard for people in school; I doubt that people make better sense of them in their pastime. Also, I agree with you that "long periods of time" means something of order 10-15 seconds, not minutes.


 * So, let me rephrase my argument: I am finding that +30 health has been more important to my mission survival than has +5 AR. Any stated theories have been attempts to explain my experience. I'm persuaded by the force of your argument (if not the volume ;-) that I should revisit this more systematically. It could be that my tactics create the relatively unique conditions in which health beats AR. It might be that I'm overpowered (in skillbar and tactics) relative to the mission; arguably, if I changed kit, I would do better still. Any thoughts on a mish/area which should favor health and yet will demonstrate AR's superior benefits?


 * For what it's worth, these are the arguments above I find compelling, although I cannot reconcile them with my experience:
 * Scale: H+30 doesn't scale; AR benefits do, independent of damage source (except for armor ignoring).
 * Diminishing returns: H+30 provides less benefit with increasing health; AR+5 always provides same % protection.


 * A couple of other disjointed thoughts:
 * I suspect that we form different team types; could it be that, e.g. Rt healing favors +30 Health compared to +AR?
 * I probably spend 20-30% of my time in HM, so it's possible I am not seeing enough situations in which the difference matters.
 * Doesn't monster-skill-based damage (armor respecting or ignoring) scale up from leveling differences?
 * You are almost certainly a more skilled player than I, so it's entirely possible that I'm encountering conditions that don't apply to you. As a very poor analogy, the fastest gun in the west worries about the top 20 gunfighters; the 142nd fastest mostly worries about 143 and not shooting self in the foot.
 * My approach to the plain-vanilla, single-prof mob has been to avoid the need to survive it for very long (pulling, cornering, disabling, distracting, overwhelming, etc.) That approach works in GW because the game is static; I get be more clever than designers b/c they never change the design. It won't work in PvP and it is less useful in some other games I've played.  &mdash; Tennessee Ernie Ford ( TEF ) 03:11, October 4, 2009 (UTC)


 * In PvP, more often than not you're going to deal with: Visions of Regret, Backfire, Lingering Curse, Soul Bind, Barbs, Insidious Parasite, Faintheartedness, Empathy, Defile Defenses, Suffering, Corrupt Enchantment, bleeding, burning, holy damage, and skills that can make anyone squishy. +30 health won't do much against all that, but +5 armor will do nothing. --Macros 03:45, October 4, 2009 (UTC)


 * As Macros brings up, there may be a need to give different recommendations for PVE and PVP.
 * Both armor and health offer diminishing returns to scale. A theoretical character with 1000 armor and 500 health would be immune to armor-respecting damage anyway (as everything would round to 0), so adding 5 armor wouldn't make a bit of difference.  It's not that +5 armor is superior to +30 health in general, but only that that tends to be the case in the ranges where players operate, at least after reaching max level.  For a level 1, +30 health really would be better most of the time.
 * It's hard to come up with a place that will clearly demonstrate the difference because the difference is so small. In most situations, you could take neither armor nor health, but just leave the mod slot empty and still beat the mission.  The theoretical reasons why having either +30 health or +5 armor are superior to having neither are obvious, but it would be hard to measure the difference intuitively in the game.  This is especially so when it's just one party member out of eight, and the other seven are identical in both configurations.  Quizzical 04:42, October 4, 2009 (UTC)


 * And don't forget about shields. Even if you don't meet the req, a max shield will still give +8 armor. Throw in an inscription with +10 vs a certain damage type, and suddenly you have +18 armor. In order to be considered "pro" these days, you need to have a set of shields, one for each damage type, and swap them out when you start getting hit by that damage type. It requires a lot of micromanagement, true, but I remember back in the day when people would swap their entire armor set (or sometimes just a few pieces, which made it really hard to keep track of everything) in the same way, before Anet nerfed that. --Macros 05:35, October 4, 2009 (UTC)

Stuff

 * HSR speeds up repeat castings of the same spells, e.g. Flare <- doesn't this particular spell have no recharge time making HSR irrelevant <_<
 * Spears allow the caster to remain at ranged distances. <- could be clearer here; melee doesn't prevent you from remaining ranged to cast  Jennalee 15:41, October 6, 2009 (UTC)


 * Argh. Shows to go I should think twice about starting something new when I'm sick. Ok, yeah, Flare has activation time, not recharge (so, I will find a more appropriate HSR spammable). And, yes, I will have to find a better way of showing that minor advantage of spear over Sword/Axe (I prefer not having to think about c-space). Thank you for taking the time to edit and point out these deficiencies.  &mdash; Tennessee Ernie Ford ( TEF ) 17:37, October 6, 2009 (UTC)