Guild Wars 2

I was doing some steady weapon testing and I noticed something with the phantasms.

1. They don’t do steady damage when summoned from a steady weapon (so they get new non-steady weapons at summon i guess)

2. The Tool Tip damage only adjusts with power on the Swordsman, Duelist and Zerker, not the Warden, Warlock and Mage

3. Only the Tool Tip damage on the Swordsman is remotely accurate

4. Power of the summoned phantasms is Fluid. If you gain a stack of might after summon, the power increase transfers to the Phantasms as you stack them, and do not need to be stacked before the summon.

5. Power is strictly based off of yourself. Any and ALL might stacked on a phantasm does absolutely nothing for it, the power will just be what yours is. Though other boons still do affect it.

Just thought that was interesting. But since I was trying to calculate Actual DPS of phantasms because of all this forum hype about weapons ( I like real numbers, not poorly tested ones with too many extraneous variables). I was wondering if anyone knew of a source code or Anything that provided a base coefficient of the their Damage. Otherwise I have to do naked testing and extrapolate the coefficient as an average based on Damage done. Which isn’t all that fun as to get a decent sample size, is 30+ tests per phantasm.

yeap all as u said.. i tried some experimental cuz couldnt find anywhere updated.. but i cant reach exact time numbers as activation, recharge sec,conjectures
somebody should do it.. exactly the co-efficienties and time(action per sec, etc..)
probably it should be done naked.. its all cuz of uncorrect tool-tips..

Phantasm recharge rates – http://wiki.guildwars2.com/wiki/Phantasmal_Haste

but anybody approved that it is ok? updated of that page is still 22 May?

Xavi.6591:

Phantasm recharge rates – http://wiki.guildwars2.com/wiki/Phantasmal_Haste

Yes, and they do appear to be quite accurate. But there is no numbers for all the attacking times. Zerker, Swordsman don’t channel, but my prelim tests show 1-1.5 seconds for attacking before recharge starts. These need to be added to calculate true DPS. So far my tests show:
4.5 seconds between attacks for swordsman
7 seconds between attacks for Zerker

The others have wiki channeling times that seem accurate so just need to add them to recharge rates.

The phantasmal haste page is 100% accurate…and 100% useless.

For purposes of dps, you need to know the time from attack beginning to attack beginning, and that chart shows you the time from attack end to attack beginning.

Pyroatheist.9031:

The phantasmal haste page is 100% accurate…and 100% useless.

For purposes of dps, you need to know the time from attack beginning to attack beginning, and that chart shows you the time from attack end to attack beginning.

Correct me if I’m wrong here, but is it as useless for one shot attacks like iSwordsman/iWarlock and to some degree the iZerker (since their attack is so quick)? I can see what you’re saying concerning the iDuelist and iWarden.

skcamow.3527:

Pyroatheist.9031:

The phantasmal haste page is 100% accurate…and 100% useless.

For purposes of dps, you need to know the time from attack beginning to attack beginning, and that chart shows you the time from attack end to attack beginning.

Correct me if I’m wrong here, but is it as useless for one shot attacks like iSwordsman/iWarlock and to some degree the iZerker (since their attack is so quick)? I can see what you’re saying concerning the iDuelist and iWarden.

It’s definitely better for those attacks, but still not as precise as it could be.

Yes, for one shot and channeling. The swordsman for instance, when it begins the attack, takes 1-1.5 seconds to jump in, hit, jump out. Once out it begins its recharge timer. So the chart does not account for this.

Haste is ok, but not at that trait and at 25 it’s not really acessable. If you want phantasm damage you won’t do shatter, so you don’t need that path, also phantasm damage is at 10/X/X/25/X. +30% damage is sure better, especially if they die after the first attack.

btt: phantasms copycat your own stats, thats why the tooltip sux i guess. it could be because its “your damage + base damage” whereas the base damage is the tooltip and doesn’t change. Or its just buggy I guess last one xD They definitly get stronger with more power, don’t worry.

Getting stronger with power and traiting for them isn’t the issue. The issue is it is impossible to calculate True Base DPS without knowing all of the factors, mainly the Base Damage Coefficient. Without it, it is impossible to create a graph-able function or chart for DPS comparisons between the phantasms. Just doing random tests on Golems in Mists can not scale the DPS comparison to anything other than those specific stats, and even then then, your sample size has to be quite large to compensate for the evenly distributed random distribution of Damage between fixed Min and Max possibilities.

Ahh, now I see your problem. But why do you need this? You got your stats, now go to Lion’s Arch and cast phantasms on the dummies. You can see e.g. the iDuellist going for 2500 damage there, whereas the iSwordsman does 2200. If you got x% more power, they both will deal x% more damage.

I once tested all sword and GS skills with 0 power and 200 power traited. I noticed that every skill gets the same amount of additional % damage. E.g.:
Sword 1 and 2 (random numbers, just to get you the point)
S1 1: 500
S1 2: 500
S1 3: 750
S2: 2000
now with lets say 200 power they “do”
S1 1: 550
S1 2: 550
S1 3: 825
S2: 2200
they all increase similiar by x% (10% in this case for an easy example).

I’ll continue my research about this and all phantasms. I’ll let you know.

It’s not that I cant do this. As I said above, it is quite possible to test the Damage, and work backwards to determine the Coefficient. The issue is that the phantasms don’t use steady weapons when you do. Say for instance your weapon Damage is between 900 and 1000, you have a chance to change your Damage by 100. The difference becomes more vast when its multiplied with Power.

To compensate you have to run multiple tests to get a sample size large enough. Say you run 1 test, and get the 1000 Damage, it is not representative of the population. Basic stats says a sample of 30 is comparable to a population if it’s randomly distributed . But to be accurate, 50+ would be required per phantasm. I just don’t want to sit there testing… I was wondering if anyone knew of a source to find the numbers. Looks like I have no choice because it’s just going to bug me now otherwise.