Guild Wars 2

Excelsior comrades!

I have just finished the alpha version of my new calculator (the Zalculator). This calculator enables you to plug in your build, your rotation, and see what kind of damage it does. Easy and simple!

So far I have made only an alpha version.This is an ambitious project in its early stage. It is limited to a number of classes and builds, but the framework is in place for future expansions to other classes.

I would like to get feedback on this calculator, any kind of feedback, so I can know what to work on first. If you are interested, could you please download it and try it?

Download
Designed for excel 2007+

Features
Current:
-Build calculator (Elem, warrior, guardian, and more coming soon)
-Combat stats
-Dps calculator (direct damage)
Upcoming:
-Rotation calculator (already present but not yet functional)

Note
Until the rotation calculator is finished you need to manually plug in your rotation coefficients and adjust the other numbers in orange. Take a look at the rotation tab.

If you have any questions or remarks, please let me know! I look forward to making this tool complete, and I can only do so with your help.

Should drop this over in the Mesmer forums. Those guys have been arguing for weeks about Greatsword vs Sword/X DPS. Literally weeks.

I will add the Mesmer data to this calculator and then we can settle this debate

Stuff like this is neat in theory and all but what I’m more interested in is whether or not it actually proves to be correct with field testing. Do you have any evidence that your calculator lines up with empirical data?

Yes all formulas used in the calculator have been tested and approved.

(apart from the behaviour of vulnerability stacks, the common belief is that it is additive)

Zelyhn.8069:

Yes all formulas used in the calculator have been tested and approved.

(apart from the behaviour of vulnerability stacks, the common belief is that it is additive)

Forgive my skepticism, but are you planning on publishing some evidence of that test?

I would surely like to do this, what exactly would you like me to show evidence for?
-The way stats are built and their effects
-The damage calculation formula

Kruciel.5931:

Should drop this over in the Mesmer forums. Those guys have been arguing for weeks about Greatsword vs Sword/X DPS. Literally weeks.

Really? It’s shocking anyone would defend greatsword.

Probably the damage calculation since that’s what it really all boils down to, at least for the dungeon forum folks. You would need to factor in the armor level of whatever mob you’d be using I’d suppose. Good candidates for damage sponges: Abomination in Arah p2, any kind of Giant (the champs in P3, perhaps) The Destroyer in CoE/3. Of course some of those guys are also levels 80+, don’t know if anyone’s figured out the damage reduction for that yet.

Broadicea.8294:

Probably the damage calculation since that’s what it really all boils down to, at least for the dungeon forum folks. You would need to factor in the armor level of whatever mob you’d be using I’d suppose. Good candidates for damage sponges: Abomination in Arah p2, any kind of Giant (the champs in P3, perhaps) The Destroyer in CoE/3. Of course some of those guys are also levels 80+, don’t know if anyone’s figured out the damage reduction for that yet.

Just have a place where you can add in the armor level to adjust the damage numbers if possible.

Honestly, though, I don’t think it’s that important. If the formulae are accurate, you can still calculate the highest DPS build on some set armor level. You could still find out which build is the highest DPS, which is what matters most, since it will also be the highest DPS against a higher armor enemy.

Bright.9160:

Broadicea.8294:

Probably the damage calculation since that’s what it really all boils down to, at least for the dungeon forum folks. You would need to factor in the armor level of whatever mob you’d be using I’d suppose. Good candidates for damage sponges: Abomination in Arah p2, any kind of Giant (the champs in P3, perhaps) The Destroyer in CoE/3. Of course some of those guys are also levels 80+, don’t know if anyone’s figured out the damage reduction for that yet.

Just have a place where you can add in the armor level to adjust the damage numbers if possible.

Honestly, though, I don’t think it’s that important. If the formulae are accurate, you can still calculate the highest DPS build on some set armor level. You could still find out which build is the highest DPS, which is what matters most, since it will also be the highest DPS against a higher armor enemy.

True, armor probably doesn’t matter, but what I’d like to see is a calculator saying I’ll do xxxx DPS, then go into the game, record a fight, and be able to look at my combat log/the floaters and compare.

You can easily adjust the armor of targets in the calculator

As for your test Broadicea, if you put the correct input in the calculator then it will give your correct average dps.

Zelyhn.8069:

You can easily adjust the armor of targets in the calculator

As for your test Broadicea, if you put the correct input in the calculator then it will give your correct average dps.

Independent verification is what I’m talking about. We can trust that the speedometer in my car is correct, but independent verification from another source would prove that conclusively. Your calculator says I’ll do xxxx damage, I’d like to see an independent measurement of that damage to verify the theoretical basis of your model. Not that I believe it’s incorrect, I’m just much more biased towards empirical data and analysis.

Actually, what matters most is internal comparative accuracy. It’s irrelevant if the numbers in game differ, what matters is that is the calculator says Build A is better than Build B that it’s true in game as well.

Exactly

And as for the independent verification, I tested the behaviour of every factor that my calcualtor takes into account. So, using your speedometer analogy, if every component of the device has been approved then the final object must be working correctly. Well, provided I put the pieces in the right place! That’s why I am asking for feedback.

Nike Porphyrogenita.8137:

Actually, what matters most is internal comparative accuracy. It’s irrelevant if the numbers in game differ, what matters is that is the calculator says Build A is better than Build B that it’s true in game as well.

Understandable, but then it would be nice to have as accurate a comparison as possible using actual hard metrics rather than a simple “better than” or “worse than” comparator, since I might be ok trading 5% personal DPS out for the convenience of condi removal on demand, or something of that nature. The calculator should do that, but if the calculator says 5% difference it would be nice to verify that in game as well. 100 seconds to kill something versus 95 seconds to kill it (on average of course) would prove out the theoretical approach.

Zelyhn.8069:

Exactly

And as for the independent verification, I tested the behaviour of every factor that my calcualtor takes into account. So, using your speedometer analogy, if every component of the device has been approved then the final object must be working correctly. Well, provided I put the pieces in the right place! That’s why I am asking for feedback.

You must never have worked professionally with software or electronics before. Or met a QA person.

Guys I’m not trying to rag on anyone doing “theory crafting”, and this isn’t rocket science. If you come up with a hypothesis, it doesn’t matter how sound it is on the blackboard, what matters is the field test. NASA doesn’t go from the blackboard to the launchpad without testing in between. Pharma doesn’t get a product on the shelves until the clinical trials have been reviewed. Toyota doesn’t just calculate G forces caused during a crash, they test that sucker. That’s all I’m asking about.

Hes asking for feedback. He cant be the independant tester for his own project.

Zelyhn.8069:

Yes all formulas used in the calculator have been tested and approved.

(apart from the behaviour of vulnerability stacks, the common belief is that it is additive)

“Approved” by whom?

Broadicea.8294:

Understandable, but then it would be nice to have as accurate a comparison as possible using actual hard metrics rather than a simple “better than” or “worse than” comparator, since I might be ok trading 5% personal DPS out for the convenience of condi removal on demand, or something of that nature. The calculator should do that, but if the calculator says 5% difference it would be nice to verify that in game as well. 100 seconds to kill something versus 95 seconds to kill it (on average of course) would prove out the theoretical approach.

You must never have worked professionally with software or electronics before. Or met a QA person.

Guys I’m not trying to rag on anyone doing “theory crafting”, and this isn’t rocket science. If you come up with a hypothesis, it doesn’t matter how sound it is on the blackboard, what matters is the field test. NASA doesn’t go from the blackboard to the launchpad without testing in between. Pharma doesn’t get a product on the shelves until the clinical trials have been reviewed. Toyota doesn’t just calculate G forces caused during a crash, they test that sucker. That’s all I’m asking about.

Actually that’s exactly what the calculator does

If you put in the right characteristic of this “something” then it will tell you exactly your average dps on it and you can deduct your time to kill it from that.

In fact theorycrafting is even more precise than rocketscience because we “reverse engineer” what the devs have given us. So we are sure that there are a finite number of laws, a design, and we can master it all.

As for the testing, trust me I have done the necessary field work. I tested the formulas in PvP with stable weapons and I am absolutely confident about what I have put in the calculator.

Wow, how useful. Thanks.

Ok thank you all for your interest.

I am now working in putting in the warrior’s data. I am focusing on the most common builds (30/0/0/10/30) so far.
The warrior requires a framework addition because of class mechanics (weapon choices mainly).

I am also working on a preliminary rotation calculator for the elementalist. I will wait until the 15th of October to see if the devs actually release accurate tooltip information before attempting to make a complete rotation calculator.

Don’t hesitate to hit me with any kind of feedback on anything!

Can someone please give me the usual dps rotation for warriors?

If you could give it to me with details (like for my fresh air rotation in my calculator) that would be lovely.

Edit: also what are the most used utility skills for warriors?

All depends of the situation but I mainly use GS / Axe-Mace
I start the fight with mace #5 while moving to the ennemy, then axe #2, mace #4, switch to GS, HB, WW if the ennemy is pushed against a wall, else a few AA (till I can switch again to Axe-Mace), switch to Axe-Mace, Axe #2, 1 or 2 Axe AA chain, Axe #2, switch to GS, etc… and I use Mace #4 when it’s off cooldown.

For Utility, it also depends of the situation, but for organised group, I use:
- For Great Justice
- Signet of Fury / Might / Frenzy (depending of the situation)
- Banner (Strength / Discipline) or On My Mark
- Signet of Rage / Battle Standard

100b, ww, axe auto and repeat assuming max vuln stacks and against a wall. Otherwise you cast axe 2, mace 4 at the start or end of axe auto chain. If not against a wall you just 100b and use bladetrail or rush until you can swap back to axe.

OK thanks, I am going to try and put this on paper, to find the proper c/s (four figures: against a wall and in open ground, with and without applying vuln). If you already have a the c/s figures (modified for relative weapon strength on GS) then feel free to fill me in on that so I can cross check with what I find myself

There is a lot of information I cannot find:

-100b’s coefficient (the wiki state 1.78, but it also says 1642 damage which is a 4.48 coeff)
Edit: I didn’t include the last blow in what I just said. Anyway I really have no reliable figure for the coefficient.
-100b’s exact animation duration

I can assume WW’s animation is 1 second (0.25 cast + 0.75 evade, if no aftercast)

-Animation duration of GS’s AA (each of the three stages)

-Axe 2 animation duration

-Mace 4 animation duration

If anyone can give me this data, or estimates, I would appreciate it thanks

Nike Porphyrogenita.8137:

Kruciel.5931:

Should drop this over in the Mesmer forums. Those guys have been arguing for weeks about Greatsword vs Sword/X DPS. Literally weeks.

Really? It’s shocking anyone would defend greatsword.

There are a lot of bad players over there.

Zelyhn.8069:

There is a lot of information I cannot find:

-100b’s coefficient (the wiki state 1.78, but it also says 1642 damage which is a 4.48 coeff)
Edit: I didn’t include the last blow in what I just said. Anyway I really have no reliable figure for the coefficient.
-100b’s exact animation duration

I can assume WW’s animation is 1 second (0.25 cast + 0.75 evade, if no aftercast)

-Animation duration of GS’s AA (each of the three stages)

-Axe 2 animation duration

-Mace 4 animation duration

If anyone can give me this data, or estimates, I would appreciate it thanks

How would coefficients be found/calculated? Just curios :s

I think you can work them out from the tooltips. Assuming the tooltip is accurate.

Since one day i had every single skill on my GS with perfectly round numbers (down to nearest 100 or 10 i forget) and took kitten and stuff… so idk if that would be useful :/

How to find a coefficient from the tooltips:

coefficient = damage displayed * default armor / lvl 80 power / weapon damage

-default armor: 2600
-lvl 80 power: 916
-weapon damage: 952.5 for most 1h / 1047.5 for most 2h (exotics, haven’t had time to 1111 enough WB yet to ascend)

When i calculated it before i used the power i had displayed on my warrior and the tooltip it showed on her bar. So im guessing you can use base power if you can find the base tooltip. But if you have a lvl 80 warriors its easier to just use the numbers it gives you ingame.

axe auto coef is 0.7 for all hits except the last which is 1.502

spoj.9672:

axe auto coef is 0.7 for all hits except the last which is 1.502

The wiki states the following damages: 252, 538, and 858.
This gives the following coefficients: 0.75, 1.6, and 2.56
The total coefficient over the 3.6s animation is therefore: 4.91
This makes the warrior’s axe AA have a coefficient per second (c/s) value of: 1.36
(equal to elem’s LH AA)

Id check it yourself if you can. I wouldnt trust the wiki values to be entirely accurate.

You are right: they are not accurate. Also if I remember correctly the coefficient for the axe AA were split between PvP and PvE, weren’t they?

Erm i know they split the rework of axe offhand skills. Not sure about AA. The damage is backloaded in pvp aswell so I would assume its the same as pve.

Ok I figured it out:
- 100b has a coefficient of 5.5 over about 3.7s (3.5s + aftercast estimated at 0.2s)
= 1.48 c/s
- Axe AA has a c/s of 1.36
- GS AA anmation times are unknown to me but it does not matter because when you swap to GS you are always going to swap out after 5s, so the total GS use time is 5s, during which 3.7s are for 100b, which leaves only enough time for 2 GS AA single hits (maybe only one, please confirm for me) so the total coefficients = 5.5 + 0.7 * 2 = 6.9 over 5s so 1.38 raw c/s (1.51 modified for relative weapon strength: 1047.5 versus 952).
- two AAs in axe (full dps rotation) = 9.8 coefficient over 7.2s (c/s=1.36)

Therefore I get:
- GS: 6.9 (7.6 modified) coeff over 5s
- Axe: 9.8 coeff over 7.2s

Total coefficients: 17.4 (modified for relative weapon strength)
Total rotation time: 12.2s

Rotation c/s: 1.42 (modified)

On a side note this is particularly interesting because it means that an All-damage Fresh air elementalist (10/30/20/10/0) would deal 95% of the damage of a warrior, under speedkill conditions.

But i’ll get back to you later on that since I need to polish my calculator before its results are to be trusted

Edit: and that’s not even a good speedkill spec I chose for elems …
Edit2: and I did not include the 70% burn uptime the fresh air elem can maintain …

Zelyhn.8069:

Yes all formulas used in the calculator have been tested and approved.

(apart from the behaviour of vulnerability stacks, the common belief is that it is additive)

Uh… wouldn’t that be really easy to test by hitting a golem with a steady weapon and say… 2 warriors either putting 0, 1, or 2 sets of OMM on the target?

Yes it would, but I not been playing much lately, so I have not been able to find the time to set up this test.

It would require at least two warriors, because we need 25 stacks to be able to notice a difference between additive and multiplicative behaviours (25% VS 28%). I’ll be back in game on Sunday evening so if anyone is willing to help me test this feel free to let me know!

With a steady weapon you would still be able to notice 20% vs 21%, there’s no variability.

Get an engie.

Player Character.9467:

With a steady weapon you would still be able to notice 20% vs 21%, there’s no variability.

Tooltips are rounded so there’s usually 1-2% variance between the tooltip number and what shows up when you actually hit something. It’s not a big enough difference to make the tooltips unreliable but it will make it hard to notice a 1-2% damage difference. in vuln stacks.

Guanglai Kangyi.4318:

Get an engie.

Player Character.9467:

With a steady weapon you would still be able to notice 20% vs 21%, there’s no variability.

Tooltips are rounded so there’s usually 1-2% variance between the tooltip number and what shows up when you actually hit something. It’s not a big enough difference to make the tooltips unreliable but it will make it hard to notice a 1-2% damage difference. in vuln stacks.

Why would you even be looking at tooltip numbers? Steady weapon, no variability in resulting damage except for crits. If your hits go 200, 200, 200, 240, 240, 240, then it’s 20%, if they went 200, 200, 200, 244, 244, 244 then it’s (1%)^20. Done. No tooltips, no spreadsheets, 30 seconds of effort.

I’ll do the test this evening.

Apart from the obvious 30/0/0/10/30 for warriors, what spec should I look into including in my calculator?

I am also adding a build comparator in the calculator. In the future, when I have added all the classes I intend to make a “Group overview” tab showing the dps of each member and the buffs the group is getting etc.

Maybe 30/25/0/0/15 or 30/10(Rending Strikes)/0/0/30 ?

Ok, i’ll make sure they work.

Any relevant weapon set apart from GS-A/M ?

Sword in off-hand if your group can stack vulnerability well enough that you don’t need to.

Yeah but that would be relevant only to the rotation, not the build, since S and M both benefit from the same trait, am I right?

Well maybe include Rifle / LB for those who prefer to range stuff ^^. But I don’t know any rotation with those weapons ^^’.