Warhammer Online – Melee Combat Mechanics
INTRODUCTION
Based on my experiences in WoW, I developed a passion for simulating combat, especially melee combat. Finding the gear combinations for attaining my character’s maximum Damage Per Second (DPS) was the foundation of those exercises on both my rogue Discoepfeand and my shaman Disquette. The information in this post is intended to be the foundation for those people who wish to do the same in Warhammer: Age of Reckoning (WAR).
DPS AND THEORYCRAFTING IN WAR
WAR, for better or worse, is primarily a PvP (or RvR, actually) game at heart. As a result, sustained DPS is often not going to be the concern of the players, who will instead be focused on burst damage as much as sustained damage. There are, however, some high-end PvE aspects that the developer (Mythic Entertainment) has alluded to. I don’t particularly care how useful this information is to the “average RvR player”, to be quite frank. My desire in this game isn’t to gain knowledge for the purpose of using it as Mythic intended, but simply to know and explain it for those who might care about the PvE portions of the game. As might be evident from my writing style, I’m getting older and am finding happiness in different things than I used to. As such, this blog is here for my own edification, and if you don’t like it or my motivations, feel free to stop reading – I’ll never know the difference ;-)
So, with that all out of the way…
BASIC STATS TO KNOW
- Strength – this increases your auto-attack melee dps, all other things remaining the same, by 0.1 dps per strength.
- Toughness – this stat, which both players and mobs have as a characteristic, is the inverse of strength (with respect to damage – your strength also effects your target’s chance to parry your attacks. That’s irrelevant for this post). Explicitly, it reduces the dps done to you by 0.1 dps / toughness.
- Weapon DPS – the damage per second inherent in every weapon. It’s interesting to note that WAR is not like WoW – there is no variability from hit to hit. It’s all a mathematical formula with no randomness to the outcome of each hit. Crits may be an exception (I haven’t started work on them yet).
- Weapon Speed – how many seconds it takes for each auto-attack swing of your weapon.
OTHER BASICS TO KNOW
- Your combat log will show not only the damage you did, but also how much was mitigated by armor. As an example, you might see the following: Your attack hits Evil Ogre for 80 damage. (20 mitigated)
- Very Important Disclaimer!!! So far the formulas I’m comfortable sharing in this post have proved to be accurate within 1 point of damage in all the testing I’ve done. I cannot, however, say without a doubt that these are the formulas being used. The best thing we can do is test for ourselves and draw our own conclusions. Having other people confirm these formulas via in-game testing would be nifty, so drop me a line if you do so.
ON TO THE SHOW!
Basic Attack Damage when single-wielding
Your auto-attack damage is calculated by a fairly simple formula. We’ll use some of those stats we covered earlier. I’ll use the following variable names:
- Strength: Str
- Target’s Tougness: Tou
- Strength Bonus: StrBon (I’m going to use this so as to simplify the way the formulas look. So far I haven’t found any damage calculations which use Str but not Tou, or vice versa, and when they’re used together, so far they’ve always been seen in the form Str-Tou).
- Weapon DPS: Wdps
- Weapon Speed: Wspeed
- Armor Mitigation %: Armor
Please note – throughout this whole post I’ll just be using a single term for armor. Weapon Skill affects armor penetration. That’s a very uninteresting stat for me, as it’s essentially a linear effect. A fully accurate formula would use (ArmorMitigation =
TargetArmor%*(1-WeaponSkillArmorPenetration%). If you’re writing a calculator, be sure to include that term, but this thing is complicated enough without that uninteresting term to further expand these large formulas. Thanks to Wodin and Hermit for pointing out that I shouldn’t leave this out all together without and explanation!
- Damage per Attack: Dmg (I’m specifying this so that it’s not confused with Damage per Second).
Dmg = ((StrBon/10) + Wdps)*(Wspeed)*(1-Armor)
Let’s do an example! Let’s say you’re a White Lion, fighting a Vicious Spider (made up name) and the actors have the following stats:
White Lion Str: 350
Spider Tou: 250
White Lion Wdps: 25
White Lion Wspeed: 2.0 seconds / swing
(thanks for catching this labeling error, DPS_squared)
Spider Armor Mitigation: 15% (Incidentally, this is about half of what is typical for even con’d mobs)
Using our formula above, each melee auto-attack will do…
Dmg = (((Str – Tou)/10) + Wdps)*(Wspeed)*(1-Armor)
Dmg = (((350-250)/10) + 25)*2.0*(1-15%)
Dmg = ((100/10) + 25)*2.0*85%
Dmg = (10+25)*2*0.85
Dmg = 59.5
This will appear in your combat log as:
Your attack hits Vicious Spider for 59 damage. (11 mitigated)
Ok, so what’s happening here? Lots of interesting things! To wit:
- Rounding – It should come as no surprise that WAR, like other MMOs, rounds abilities to whole numbers for display purposes. This does not mean that the game doesn’t process fractions – I’ve seen one of my HoTs tick for varying amounts, sometimes 268, sometimes 267, which I can only imagine is the game keeping track of the fractions internally in the server (or perhaps the client, I don’t know).
- There are two forms of damage mitigation, but you’re only seeing one explicitly – Only the armor mitigation shows up in the display window, but remember that the target’s toughness is also factoring in to reduce the damage.
Calculating Mob Damage Reduction
You might ask how to figure out the opponent’s armor and toughness values. I sure needed to know that in order to reverse-engineer the more complicated attack types. Let’s go through that process (Hey, I spent the hours to figure out how to calculate these, so you’re gonna read about it :p )
- Armor - this was the easy one. For every auto attack, on every character, it was pretty straightforward to see that comparing the amount mitigated in the combat log to the damage done always had the same ratio when all else remained the same (target mob, your str, weapon, etc). And that’s that! Ok, except for crits, which I keep coming back to as being special. Other blogs or articles will probably cover how that is working in game at the moment.
- Toughness - this was the harder one. The first thing is to realize that mobs have this! Coming from WoW where the only mitigation stat against physical attacks was armor, I had to expand my horizons a bit to understand that this was not only present, but was *invisible* (ooooh, ahhhhh). What I mean by that is that it doesn’t show in the mitigation portion of the combat log. However, if you do a few tests on mobs, you find that regardless of your weapon dps, your strenght, your weapon speed, etc, you are constantly hitting the mob for a consistent amount less than you would expect if you used just strength in a simple calculation.You might find it helpful, if you want to do dps testing for yourself, to have a set of guidelines to follow for calculating a mob’s toughness.
Here’s what you do – look at your stats, auto attack a mob, and back into the toughness using the following procedure:
- Record your Str, Wdps, and Wspeed
- Auto attack the mob, and add the mitigated and non-mitigated portions together. For instance, for the combat log entry Your attack hits Vicious Spider for 59 damage. (11 mitigated), The number you would want to record is 59+11 = 70. We’ll call this number TDmg (for Total Damage, including armor mitigation)
- Use the formula below, which you can re-create yourself by undoing the process we used above if you like to prove things out to yourself:
tou = 10*((((0.1*Str + Wdps)*Wspeed)-TDmg)/Wspeed)
That might need an example to make sure I got my parentheses right! Let’s use the spider example from above. Remember from before, that:
White Lion Str: 350
White Lion Wdps: 25
White Lion Wspeed: 2.0 seconds / swing
Spider Armor Mitigation: 15% (this is completely unnecessary information for this calculation)
Add to that our two new data points:
Mitigated Damage: 11
Final Actual Damage to Mob: 59
TDmg = 70 (the 2 lines above added together)
Let’s calculate it out using the formulas shown above:
tou = 10*((((0.1*Str + Wdps)*Wspeed)-TDmg)/Wspeed)
tou = 10*((((0.1*350 + 25)*2.0)-(59+11))/Wspeed)
tou = 10*((((35+25)*2.0)-70)/2)
tou = 10*((120-70)/2)
tou = 250
Yay, it works! It matches the toughness we had in the assumptions used for the first example. If you’re theorycrafting/testing dps on mobs, build this formula into a spreadsheet – you’re going to use it a lot ;-)
Calculation Damage of Special Attacks
(Note: I haven’t looked at special attack damage while dual-wielding. This only works for single wielding at the moment)
(Edit for Update! Erdrick figured out the Dmg for special attacks while dual-wielding, yay!)
Unlike DAoC, and like some classes in WoW, WAR intertwines special attacks with your auto attack, as opposed to having a queue’ing system. As a result of not using a queue system, Mythic did something very smart for calculating special attack damage – they treat the weapon speed for the damage calculations as if it were the speed of the ability. Most special abilities are instant. Since each instant attack invokes the global cooldown of 1.5 seconds, 1.5 is the speed value used for most specials, whether you’re using a big 2 hander or a quick dagger. This is in contrast to WoW in which weapon speeds are normalized for instant attacks by their type (2.4 for swords, 1.7 for daggers, 3.something for 2 handers).
Another wrinkle thrown into the special attack calculations is that strength counts for twice as much! Above we noted that, per the character screen, each point of Str added 0.1 dps to your attacks. Also per the character screen, you will see that for special abilities, each point of Str adds 0.2 dps. What the character screen does not tell you is that your opponent’s toughness mitigation is also doubled per point. This makes things very convenient for us, as that doubling for both means that we can continue using our meta-stat: StrBon (Your strength – Your target’s toughness)
Now we have two facts that help us calculate the damage – weapon speed of 1.5 (usually) and StrBonus doubling – but we need more. MORE BLOOD FOR THE BLOOD GOD!
First, we have the ability’s damage itself. An ability may something like “A cruel blow which does 63 damage to your target and causes it to hate you greatly”.
Second, we have the special attack weapon dps modifier. This modifier, from my testing, is set to 150%. This means that if your weapon was 100dps, the damage of your special attack would be increased by 150 due to its weapon dps.
As noted by EasymodeX in the comments section, that’s equal to the multiplier for strength, which is a pleasing number – it again is the swing time of an instant attack.
So, let’s put that all together into a formula that captures these different facts. We’ll use the term Admg to refer to any extra damage you get from the special attack, per its tooltip, and we’ll use the term Sdmg to refer to the total damage of your special attacks after all the calculations are done.
Sdmg = (((StrBon/10)*2 + Wdps)*1.5 + Admg)*(1-armor)
(single wield only)
(note that Wspeed doesn’t factor in here – the special ability damage calculation is completely weapon-speed-free. You’ll be seeing this fact referred to again later in this post.)
As usual, let’s do an example so you can see it working. Assume the following:
White Lion Str: 323
Spider Tou: 264
White Lion Wdps: 51
White Lion Admg: 62
Spider Armor Mitigation: 15%
Sdmg = ((StrBon/10)*2*1.5 + Wdps*150% + Admg)*(1-Armor)
Sdmg = (((323-264)/10)*3 + 51*150% + 62) * (1 – 15%)
Sdmg = (5.9*3 + 76.5 + 62) * 0.85
Sdmg = 156.2 * 0.85
Sdmg = Your special attack hits Vicious Spider for 133 damage. (23 mitigated)
Update as indicated above, big thanks to Erdrick for figuring out special damage while single *or* dual wielding. That dude has more patience than me, that’s for sure :D Here it is:
Sdmg = ((StrBon/10)*2*1.5 + MHdps*150% + OFdps*52.5%+ Admg)*(1-Armor)
(Update on 9/18/08 – OFdps coefficient may be 67.5% now. I’ll have to check when I have a reasonably high level dual-wielder. Right now I only have a WP)
Alrighty, I hope that’s all clear, because the next formula (the last in this article) is the real doozy. Or maybe it’s not. But the fact is, it was horrible trying to figure it the next formula, and I had a victory beer when I finally got it working. However, even though it has so far been within 1 damage of actual tests each time I’ve done this, it looks so odd that I don’t have 100% confidence in its accuracy. I do have enough confidence in it to post it, and hopefully it stands up to further testing. If it doesn’t, so be it, and I’ll be very open to seeing a more accurate formula. So then, without further ado, let’s talk about…
Off-Hand Damage, when it happens, and how it’s calculated
Mythic has taken a novel (to me at least) approach to the mechanics of dual-wielding melee weapons. Instead of the off-hand weapon always swinging as it does in WoW, the chance for the off-hand weapon to swing is a simple proc from the main hand. A certain percentage of the time when the main hand hits, the offhand will also hit as an auto-attack. I don’t know what that percentage is. It’s probably documented somewhere, and if not, that’s a trivial task to empirically determine the proc chance (probably, who knows if they threw in something odd in the determination).
Edit 9/26/08 – The poster Star in the comments sections has been awesome and tested offhand proc rates! Here are his findings (comment copied directly):
Just calculated the proc chance of the offhand hit, and it results 0′46. But should need a more extended review.
The difficulty is that, in the combat log, offhand hits appear just like the mainhand hits (so you have to play with the time of the attack and the damage).
Also, you can get parried or blocked you main hand attack, and the offhand attack may still proc.
The part that I cared about was how to calculate how much damage each off-hand swing would cause when one did happen. So then why go into the proc-chance explanation? It is because that concept is intrinsic to understanding how and why the off-hand damage is what it is. In contrast to the other sections, I’m going to start us off with the damage formula, and then explain it after the fact. First, however, we need some new terminology to add to our stable of stats:
- Our current familiar stats:
- Strength: Str
- Target’s Tougness: Tou
- Strength Bonus: StrBon
- Armor Mitigation %: Armor
- Off-Hand Damage per Attack: OHDmg
- Our new stats:
- Off-Hand Weapon DPS: OHWdps
- Off-Hand Weapon Speed: OHWspeed
- Main-Hand Weapon Speed: MHWspeed
- Off-Hand Weapon DPS penalty: OHWPen
- Off-Hand Strength penalty: OHSPen
Alrighty, so here’s the formula:
OHDmg / (1 – Armor) =
(StrBon*OHSPen/10)*OHWSpeed
+ (MHWSpeed-OHWSpeed)*OHWdps
+ OHWdps*OHWSpeed*OHWPen)
ZOMG! I thought I was going to cry figuring that out. 4 seriously.
At this point, you have to take on faith that through trial and error and a few beers, I figured out the following constant values, and that they’re correct:
OHWPen = 0.9
OHSPen = 0.5
By now, you know what comes next – An Example!
Let’s lay out the relevant stats…
Str: 391
Tou: 291
StrBon: 100 (Remember, this is just Str – Tou)
Armor: 15%
OHWdps: 38.0
OHWspeed: 2.2
MHWspeed: 2.6
OHWPen: 0.9
OHSPen: 0.5
OHDmg / (1 – Armor) =
(StrBon*OHSPen/10)*OHWSpeed
+ (MHWSpeed-OHWSpeed)*OHWdps
+ OHWdps*OHWSpeed*OHWPen
OHDmg / (1 – 0.15) =
(100*0.5/10)*2.2
+ (2.6-2.2)*38.0
+ 38.0*2.2*0.9
OHDmg / (1 – 0.15) =
(100*0.5/10)*2.2
+ (2.6-2.2)*38.0
+ 38.0*2.2*0.9
OHDmg / (0.85) = 5*2.2 + 0.4*38.0 + 38.0*2.2*0.9
OHDmg / (0.85) = 11 + 15.2 + 75.2
OHDmg = 101.4 * 0.85
OHDmg = Your special attack hits Vicious Spider for 86 damage. (15 mitigated)
hey.
HEY.
HEY!!!!! Wake up! You still with me???
Ok, good, we’re almost done. I want to explain what’s happening with the formula above so you can have a sense of how well thought out Mythic’s offhand damage mechanic is.
First, as we discovered before, note that your special attacks aren’t diminished by using a fast weapon since special attacks don’t factor weapon speed into the damage calculations. Since off-hand weapon swings are procs of main-hand attacks, Mythic wanted to make sure that people didn’t abuse this mechanic by having a really fast main-hand to proc a hard hitting slow off-hand. Thus, they prorate the damage of your off-hand attacks by the difference between it and your main-hand weapon speed. The part of the calculation where you see this is in the term: (MHWSpeed-OHWSpeed)*OHWdps. The first term in the formula is also interesting in that you see Mythic halving the Strength bonus compared to a main-hand attack. Both that, and the last term of the formula (in which you see that Mythic has reduced the standard (weapon dps * weapon speed) by 10%) seem to be their attempt to tune the offhand damage to whatever internal goals they have. That’s my guess, at least, because it’s pretty damn complicated if that isn’t why they made this monstrosity of an off-hand-auto-attack-damage-calculation!
Conclusion
Melee DPS calculations in WAR are not nearly as simple as one might expect at first glance. While even these very primal calculations are relatively complex, there are even more complex formulas that will need to be developed to explain things such as off-hand crit damage.
And finally, I must reiterate: These formulas are my attempt at matching up mathematical formulas with observed results in WAR. I cannot guarantee that any of these are correct. However, they match up with my in game experiences so far.
I guess this is then end of my first blog post EVAR! If you have any comments or questions, feel free to post them in the response section (assuming it works, we’ll see if i get this blog thing figured out or not) ;-)
- Disquette
(Shaman, Disciple of Khaine, and ElitistJerks.com forum member)
August 19, 2008 at 11:37 pm |
Actually this is quite in depth and I like that someone has taken the time to figure these formulas out. Without seeing these formulas I would have been totally clueless until open beta and quite some time of testing on how exactly to gear to maximize my damage.
August 19, 2008 at 11:41 pm |
Wow! Great article! very very very helpful!
One question: do you know if two-handed weapons give you a damage bonus? I heard that was the case, but you didn’t mention it at all. Thanks!
August 19, 2008 at 11:57 pm |
Damn dude. That was some hardcore stuff you did there, nice job and thank you.
August 19, 2008 at 11:59 pm |
WAAAGGHHHHH!!! – To each his own – thanks for reading!
Lol@WAAGHH – I’m glad to hear it. Hope you max out your gear and spec.
lasmrah – They don’t give a damage bonus as far as I can tell. I tested with a warrior priest using both a 2handed weapon and a 1handed weapon, and the same formulas held. This doesn’t mean that there’s not benefit. I’ve heard, but not verified for myself, that using a 2H weapon does something like increases your crit chance, or decreases your chance to be parried, or something like that. I’m not sure if that’s in game now or not.
Ironbreaker – Rawr! Hope you’re able to dump some grudge on an unsuspecting greenskin with great vengeance.
August 20, 2008 at 2:22 am |
JAJAJAJAJAJA!!1!
August 20, 2008 at 9:48 am |
Nice!
I love it when people on teh internets do all the work so I can just soak up the benefits!
August 20, 2008 at 10:07 am |
Most DoTs are instant casts. It’ll be worth testing if they gain StrBon*casting_time damage spread evenly over the whole duration or what. casting_time = 1.5 of course.
I’d initially assume that Int works for spells exactly like Str does for melee. We could also speak about IntBon. Interesting that there’s nothing similar for other ranged attacks, supposedly their damage is higher to compensate.
August 20, 2008 at 6:20 pm |
What about Weapon Skill? As I recall from the GUI, this attribute increases your attacks ability to ignore armor. Mine said something like 18%. My assumption is that this would just be a simple modifier to your formula, affecting the armor mitigation. Did this factor in to any of your testing?
August 20, 2008 at 7:09 pm |
Hermit,
You are correct that this is a simple modifier. For my purposes, armor and weaponskill aren’t very interesting. The interesting part is figuring out how the damage before armor mitigation works.
That having been said, you are indeed correct and I should add it, simple though it may be, to the formulas.
August 20, 2008 at 7:22 pm |
Thank you. Having this work as a foundation really helps for understanding how to get the most out of melee.
August 21, 2008 at 4:08 pm |
As one of the first to begin “theory crafting” warhammer expect to see a lot of traffic. Keep up the good work ^-^
August 21, 2008 at 11:17 pm |
“They don’t give a damage bonus as far as I can tell.”
I think the main thing here is that 2h weapons have higher DPS ratings.
Also, one comment on your formula for the Special Attack damage:
The “150%” weapon DPS modifier is simply the 1.5s normalized weapon speed. So you could re-write the special attack formula almost identical to the normal attack formula, except with the 2x STR piece.
Unless you’re deliberately separating the STR and Weap DPS components in order to more smoothly lead into the discussion on offhand damage.
August 22, 2008 at 9:48 am |
Way to much time on your hands mate, please get a girlfriend and get laid.. just play the damn game and have fun, making pointless formulas really isn’t helping anyone other then your own ego, “OOOHH I THOUGHT UP ALL THIS 1337 STUFFZ TO TELL PEOPLE.” When in reality, people are going to up their STR and realize, hey I hit harder.. and then equip a crazy 2h and realize OMFG IM HITTING FOR LIKE 100 HARDER!! Nobody is really going to remember your stupid ass formulas and equations, get laid imo.. and I can’t beleive I just read that idiotic shit.
August 22, 2008 at 10:12 am |
EasymodeX – that’s a really good point!
I didn’t separate them in order to lead into the other discussion, it’s just something that I overlooked. I separated out the components because they *could* have had different coefficients, and then never recombined the terms after finding out that they’re actually the same.
Thanks for catching that – I’ll edit the formulas to reflect this.
August 22, 2008 at 11:06 am |
thanks for the post. this is good info.
August 22, 2008 at 6:26 pm |
Hey never played the game, so all I get to do is theory craft at this point. You are trying to calculate the damage per swing so why so you multiply the reported DPS by the swings per second? Shouldn’t you take the inverse of the weapon speed. Looking at the item db it appears item speed is reported as seconds per swing, and correspond roughly to the number you use, so its more a labeling issue than a calculation issue, but it might confuse some people.
August 22, 2008 at 8:12 pm |
Whoopsies, good catch. I labeled that wrong, it should read “seconds per swing”. Thanks for catching that, and I’ll fix it when I get home.
August 22, 2008 at 11:10 pm |
Awesome work mate, i enjoyed reading that and will be checking back for updates! :)
August 23, 2008 at 4:24 am |
Wow. This is beautiful. Thanks for sharing your hard work!
August 24, 2008 at 8:33 pm |
From tooltip
10 Strength = +2 dps (special), +1 dps(auto)
10 Toughness = -2 dps (special + auto)
are u sure about the autoattack formula
August 24, 2008 at 8:42 pm |
btw if weaponskill works like that its pretty worthless
especially against cloth classes who allready have low armor
say 400 wskill gives 50% armor penetration, and ur fighting against a clothie with 10% armor, 400 wskill would give u -5% armor on the target. am i interpreting this right?
August 24, 2008 at 9:32 pm |
Hi Rhqq2.
The best I can say about any formula in here is… try it out! The reason that this took me so long to figure out is that not everything matches up nice with tooltips. Not that the tool tips are wrong, but that they’re incomplete (toughness is a good example of this).
If any of my formulas are wrong, I’d love to know about it so I can correct the formulas, but as only one person with a limited amount of time, perhaps there’s something that I missed.
I really do welcome any corrections.
As to weapon skill, I agree with your assesment, and that’s why I have stacked str instead of weapon skill – Str works against everyone. WS is pretty poor in dps/point unless it’s a high armor target. Perhaps someone with time can test this to verify.
August 24, 2008 at 10:09 pm |
Do offhand weapon attacks proc on special attacks? It is not explicitly stated so I cannot assume.
If not it makes using a Chalice much more appeal as a DoK. If yes are all the effects of the special attack applied with the offhand damage? Like for instance does Consume Essence heal based on both the Main hand and Offhand, or just the Main hand?
August 24, 2008 at 11:53 pm |
Interesting questions, and I’m not really sure. I haven’t tested proc rates at all, and I don’t know for sure how specials damage is calculated when dual wielding. I sorta got sucked into playing the game and now I don’t want to go back to testing >.<
August 25, 2008 at 7:42 pm |
Thanks a lot for this interesting article. Before this I only had fade impressions on what to wear. Now it’s much more simpler. Victory Beer raised for you!
August 25, 2008 at 9:23 pm |
Great job, thanks for the formulas
August 25, 2008 at 10:32 pm |
Great stuff here disquette!
Of note to me are the following:
1) Slower weapons/abilities are less impacted by toughness (log-based).
2) Str to Tou differentials are more important than offhand weapon dps.
Subnote: Dual wielders should stack Strength
Subnote: Dual wielders should go after less tough targets.
3) Armor is handled after Toughness (no multiplicative effects).
I’m going to assume that each class has a base DPS target that is adjusted for utility, range, etc. This leaves a couple of questions still left for the statistical side of analysis:
1) What are the baseline DPS values of skills by class?
2) How much DPS does an attack lose for being ranged?
3) How much DPS does a ranged class get in exchange for autoattack?
4) What % of total melee dps is Autoattack?
5) Does the Block effect scale/compensate for dps lost?
Ok…there are a billion questions left and my inner math junkie can’t wait to make the spreadsheets!
August 25, 2008 at 11:00 pm |
fyi, in the newsletter followup to preview weekend, it was revealed that the WAR global cooldown is in fact 1.4s
August 26, 2008 at 2:19 am |
Great work Disquette!
what I take from this is. Witch elfs stacking str will do far better vs soft targets than hard as is intended I suppose.
To seriosly hurt a tank type a second set of gear will be needed.
pve vs rvr gear sets and skills inc.
August 26, 2008 at 5:00 am |
“fyi, in the newsletter followup to preview weekend, it was revealed that the WAR global cooldown is in fact 1.4s”
Yeah, I saw that! The calculations still seem to be done with a 1.5 coefficient to many things. Perhaps, however, they have changed internally. As with everything, I strongly suggest that people verify, not just trust, these formulas.
Deathdonut – sounds like a pretty good list of deductions and questions. Once you have your spreadsheets up and running, it will be interesting to see how they bear out.
August 27, 2008 at 12:27 am |
Ok I have one question that I dont know can be answered.
Lets say for example I am a tank with 600 toughness (-120 dps).
My opponent has 600 strength (+60 autoswing dps).
My opponent has a (60) dps weapon
Is toughness capped to only negate strength or will i reduce his autoswing damage to 1?
August 27, 2008 at 2:04 am |
That’s a really good question. I’d suggest trying this out by using the toughness formula I listed above after swinging at a mob. Once you’ve calculated its toughness, you could take off all str gear to the point where you have less str than it has toughness and swinging at it again.
My suspicion (or, I should say, how I’d write the game code) is that it would indeed eat into the auto attack damage until it was 1 or 0. But, you never know until you test! If you do decide to test, I’d love to hear how it turns out.
August 27, 2008 at 3:58 am |
Shit! That’s excellent work. Are you by any chance going to make formulas to explain magic damage? =) I totally understand if you didn’t. Either way thanks for the hard work.
August 27, 2008 at 4:13 am |
Since toughness directly negates str/int and is calculated before armor, how useful do you think this is in PvP? It sounds like toughness is not all that great if you already have high armor mitigation/resists, and much better if you have minimal.
In fact, since armor and resist scale linearly, with only a 75% mitigation cutoff, it seems always worthwhile to add more resists than more toughness.
August 27, 2008 at 6:30 am |
thank you very, very much for this!
I got one question:
How will Channeled-Melee-Abilities be calculated?
With “Divine Assault” you hit up to 4 times in 3seconds.
Are this 4 “instant attacks”(dps*1.5*4) or is this a 3sec attack divided into 4 parts? (dps*3/4)
aside from that question your calculation show in order to produce 1 more ability damage(not counting defenses) one need: 0.66 dps or 5 str or 1ability+dmg
this gives a very quick gear evaluation hint!
again, thanks alot for your effort!!
August 27, 2008 at 9:30 am |
Nice work.
I have a few comments based on my experiences. The first being that level difference factors into damage mitigation when fighting mobs above your level. Hitting the same level 35 mob with both a level 33 and 31 character resulted in a noticeable difference in mitigation both before and after armor. Otherwise your toughness formula was spot on for me for equal level or lower.
In regards to special damage I’ve found that some abilities act strangely as far as the contributions they receive from stats. Playing around with an Ironbreaker your special damage formula worked great for some attacks (grudging, inspiring, guarded), but some specials just didn’t jive (kneecapper, heavy blow).
From the casting standpoint I’ve seen similar oddities. For example, rune of striking and rune of fortune both have the same listed damage and cast time, but rune of fortune does considerably more damage.
August 27, 2008 at 2:37 pm |
Well done. I appreciate you taking the time to post the results of your work.
August 27, 2008 at 6:27 pm |
I’m super impressed that you actually managed to reverse engineer these formulas. Woah. You are the first person to ever make me want to learn how to use a spreadsheet. Great stuff, I’m sure I’ll be back.
August 29, 2008 at 7:03 am |
Great work, found a link to this from a post on WHA. Thanks for all the info :)
August 29, 2008 at 12:16 pm |
I just skimmed this and couldn’t find it but did you include the offhand contribution to special attacks for dual wielders?
August 29, 2008 at 12:20 pm |
legend.
August 29, 2008 at 12:35 pm |
Yeah you were close, but you forgot the offhand contribution to special abilities. The correct formula is below, also this formula doesn’t work for bleed attacks.
Sdmg = ((StrBon/10)*2*1.5 + MHdps*150% + OFdps*52.5%+ Admg)/Armor
August 29, 2008 at 6:27 pm |
In the calculation above for off-hand dps, you’ve assumed a slower main hand weapon than off hand weapon. What if this is reversed (2.2 speed main and 2.6 speed off)? Then, wouldn’t the calculation be:
OHDmg / (0.85) = 5*2.2 – 0.4*38.0 + 38.0*2.2*0.9
OHDmg / (0.85) = 11 – 15.2 + 75.2
OHDmg = 71 * 0.85
OHDmg = Your special attack hits Vicious Spider for 60 damage. (11 mitigated)
So if I’m doing that right, it seems that using a slower off-hand than main-hand would mess up your DPS.
But on the other hand, the off-hand DPS number might also change — since the slower weapon will probably have a higher DPS rating. But then I’d assume, though, that the off-hand DPS would be higher — the DPS rating of a 2.6 speed weapon should be higher than the DPS rating of a 2.2 speed weapon, of the same level. So that would make the DPS penalty even greater. . .
Am I missing something or is that accurate?
August 29, 2008 at 8:51 pm |
Great theorycrafting Disquette. First off, just ignore the haters out there like “lol@this” who obviously couldn’t understand rudimentary math to even grasp a quarter of what you deduced and thus are jealous.
Secondly, how would you calculate the DoT damage of an ability like Rend that inflicts 75 instant dam and then 1 dps over a period of 9 seconds with an offhand mutation-claw? By your logic, it would be rounded up to 1 regardless, because you are basically doing 52.5% of the bleeding damage which is still 1 per 9 seconds.
August 29, 2008 at 9:39 pm |
Kharmic – I like testing to confirm theories such as the one you provided. I’m anxious for the servers to come back up so we can do that.
Kestrel_Trebor – I think your answer is exactly right: “OHDmg = Your special attack hits Vicious Spider for 60 damage. (11 mitigated)”
I think I tested this , but don’t remember for sure. Once servers are back up we should be able to do so.
August 29, 2008 at 10:12 pm |
No comment as to the missing offhand contribution to ability damage?
August 29, 2008 at 10:30 pm |
Erdrick, Updated, you rule!
August 30, 2008 at 12:45 am |
No problem, I look forward to possibly working with you in the future.
August 30, 2008 at 10:23 am |
Simply great information, thanks for the work.
I have a question about crit percentages which you may or may not already know, although you said you’d be covering crits in the future.
If a tactic or ability gives +20% crit chance, is this:
Your chance to crit vs. your target’s Initiative + 20
Or:
20% of your chance to crit vs. your target’s Initiative
August 30, 2008 at 1:49 pm |
I’m pretty sure that tactics/abilities get added towards the end. The part that will cap out is crit chance due to level differences.
September 1, 2008 at 7:57 pm |
UM WOW SIR YOU ARE A TOTAL NERD. HOW ABOUT YOU GO SLAY SOME DRAGONS? I TOOK SOME TIME IN BETWEEN BANGING MY SUPER MODEL GIRLFRIENDS ABOARD MY MILLION DOLLAR YACHT WHILE SIPPING ON COGNAC TO PARTAKE OF THIS BLOG AND I MUST SAY JUST READING IT MADE A PAIR OF SPECTACLES AND A POCKET PROTECTOR W/ TI-83+ CALCULATOR APPEAR SPONTANEOUSLY ON MY SCULPTED 6-PACK CHEST!
SO IM SURE YOU ASK, “HEY WHOOSH, IF YOU ARE SUCH A SUPER SUCCESSFUL ALL-STAR WITH SUPERMODEL GIRLFRIENDS, WHY ARE YOU READING A BLOG ABOUT WARHAMMER DPS MECHANICS?” AND I SAY, WELL DISQUETTE, IT’S BECAUSE SOMETIMES WHEN I FEEL DOWN I JUST OPEN THE INTERNETS AND GOOGLE “NERDS WHO ARENT AS INCREDIBLE AS ME”, AND WHAT DO YOU KNOW? “DISQUETTE’S WEB BLOG”.
GOOD DAY SIR AND HAPPY DRAGONSLAYING
September 2, 2008 at 12:21 am |
I have no idea why I’d reply to myself. After all, *I* am Whoosh.
September 2, 2008 at 1:44 am |
How you determine the 150% coef applied to Main hand dps value and the 52.5% coef applied to the Off Hand dps value for Sdps calculation :
Sdmg = ((StrBon/10)*2*1.5 + MHdps*150% + OFdps*52.5%+ Admg)*(1-Armor)
September 2, 2008 at 6:17 am |
Just testing. I can only confirm the 1.5 (150%) parts. Erdrick supplied the 52.5% portion for the offhand contribution. Servers aren’t live, so I’m taking his word for it. He seems to know what he’s doing.
September 2, 2008 at 11:46 am |
The simpliest method for offhand is to just remove the offhand and see the damage difference. The higher the damage you do the easier it is to calculate. The one thing to make sure though is that the offhand doesn’t have any +strength stat to it or else it’ll screw with your numbers.
September 2, 2008 at 11:51 am |
Don’t forget that weapon skill adds to armor penatration. I would say you are still missing the (1 – Armor + SOMETHINGWITHWEAPONSKILL).
September 2, 2008 at 9:15 pm |
While I believe the WAR UI needs polishing its a mistake to use WoW as your measuring stick. The core audience of this game is unlikely to be 3 year vets of WoW, but even if it was if the game is just like WoW then its already failed.
Don’t get me wrong your complaints may be entirely justified (in fact the ones that are clear bugs obviously are) but when you get into the area of it not “feeling right: because its not WoW that seems a lot more shaky.
After playing a game and becoming very familiar with its UI over a course of 3 years, playing a same genre game that isn’tt the same is going to feel awkward and annoying at first, but its likely that once you’ve played the new game some fraction of three years it will become just as natural.
The thing is though, I don’t think your all wrong either. I’m sure there ARE elements of the UI that are not good enough and need to be better. whether or not becoming a WoW interface is the answer depends on each individual case.
One thing is certain. UI itterations happen at the end of the project all the time. Often a simple one or two line code change can have a large effect on the players game-play experience. Good job bring up these issues as it can only result in a better game.
September 2, 2008 at 10:21 pm |
“whoosh” 6 pack chest…. lol
September 2, 2008 at 10:54 pm |
Great information exactly what I was after..
Also nobody else has mentioned, but …. Thanks for disco and disqo mods :P
I’m guessing your the guy who created them.
If you are, thanks~~ I still use disqo on my shaman in sunwell :P
September 3, 2008 at 4:32 am |
Nice work.
Have you considered rating the different mDPS classes output at a basic level?
Like Ironbreaker versus Witch Hunter versus White Lion, etc
September 3, 2008 at 2:21 pm |
This sort of test will be a bit difficult because it requires sames stats, same Weapon Skill, same Wspeed etc etc on same mobs with the same type of melee damage (slash, crush …) bonus, debuff and so on, but it should be nice to determine the coefficient table of each career for 1 handed and 2 handed hits.
Concerning the type of damage, do you checked if there is a special damage reduction due to the combinaison between the type of damage (slash, crush, thrust) and the target’s armor (Cloth, Leather, Mail, Plate).
September 3, 2008 at 11:38 pm |
This has been a very interesting read especially since I didn’t get into the closed beta and I’m waiting anxious for the open beta to start.
I’m looking forward to doing some tests with the OH proc rates. I’Ve always found the it fun to deal with “random” events. It keeps the game from getting to routine ;)
September 4, 2008 at 5:26 pm |
would the ballastic stat be exchanged for the str stat for ranged dps attacks (both auto and special)? been a while since i was in the beta (looking forward to OB on the 7th) but I am pretty sure the tooltip for ballastic said something about increasing damage done similar to the str stat but for ranged attacks.
is toughness still applied for ranged physical damage or is there another stat that covers those attacks?
September 5, 2008 at 2:15 am |
I think that ballistic and str are indeed completely analogous, and that toughness affects all damage, melee, ranged physical, and ranged magical. But, if anyone has done exhaustive testing to confirm or reject this, I’d like to hear about it.
September 5, 2008 at 4:54 am |
[...] foward in comparison I think. I was very impressed with this guys work. I hope you find it useful. Warhammer Online – Melee Combat Mechanics « Disquette’s Weblog Cya in game. __________________ Charisma is my dump [...]
September 5, 2008 at 4:50 pm |
ranged magic damage is reduced by the three magic resists that you can get on your gear. from what i could tell from wardb looking at some rank 40 toons 16 pts of resist was around ~1% of damage reduction….
where as i think willpower increases your chance to resist/disrupt a spell completely.
September 9, 2008 at 4:08 pm |
One quick question, because you obviously have a much better interpretation of the math involved than I do.
I play a disciple and I’m wondering if weapon speed is really a factor, or if DPS is the only thing of importance. Here’s an example.
Weapon1:
14dps 2.4speed
Weapon2:
12.4dps 1.9speed
I assumed the slower higher dps weapon should be my main hand, and the fast weapon should be my off-hand. Is this correct?
Should I just dump the 12.4 dps fast weapon and buy another slower 14dps weapon for my off-hand?
Great work, and thanks for doing the math so I don’t have to!
September 11, 2008 at 11:00 am |
Hey, thanks for answering my “Does weapon speed actually matter and wtf should i be stat stacking” questions. I’m still a bit lost about Toughness = Str negating statistic. Does this mean that tanks who stack it totally cancel out ALL incoming damage or is there a cap to the mitigation?…
Or am i reading that wrong… perhaps a player with 400 str gets 400dmg taken off the incoming special damage landing on the tank.
-thanks, still working out if its worth being a tank and just stacking toughness and str for an unstoppable shielded powerhouse :P
September 11, 2008 at 8:36 pm |
I took the liberty of linking your blog as a news item on our new fansite for WAR. I hope this was OK.
Feel free to drop by and chat with us, we launched just the other day and are trying to get things rolling as a go-to source for all things theorycraftish.
September 11, 2008 at 10:00 pm |
Yep, sounds fine. Good luck with the site!
September 14, 2008 at 8:54 am |
good work there mate altho im pretty sure you can simplify the equation a bit when dueal weild.
From some tests with WP / DoK head to head using weapons the same levle the damage was almost identical.
This means that main hand DPS + % of off hand dps = 2h DPS
which led me to believe formulae was calculated using a straight addition of both weapons then modified to bring in line with 2h and caclulated as a single weapon not as 2 individual parts.
my initiial formulae worked out to be
(Str – Tou + Wdps) x normalised speed + ability damage / (armor / penetration)
on all tests it came out pretty close but i know they have altered to coefficients couple patches ago.
bottom line non mathematical ppl need to know is simplt
STR adds to damage TOU subtracts from damage
Armor is a divider and can be reduced thru weapon skill.
Toughnesss is more effective against low sustained dps – armor / resists perfom better agains high burst dps.
(ie toughness doesnt scale as damage goes up which is why tanks can tank heros and everyone else gets splattered.)
Roar
September 14, 2008 at 5:05 pm |
Happen to know whether abilties with cast times shorter than 1.5s which aren’t instant are considered 1.5 in the damage algo?
For instance would spiral fletched arrow (1.0 speed) actually hit for less than it would if it were an instant cast spell?
September 14, 2008 at 5:19 pm |
Roaring – thanks – I’ll look into that kind of simplification when I get a little time. I’m not convinced that you can flat out add weapon dps, but I could be wrong (I’m always amenable to fixing formulas if people find bugs in them).
Crim – that’s a question that I’ve thought about but never answered for myself. If you happen to find out, I’d love to hear the answer.
September 15, 2008 at 2:17 am |
[...] This is a great post for those who like to understand the math behind the game. original post: Warhammer Online – Melee Combat Mechanics Disquette’s Weblog edited for space: DPS AND THEORYCRAFTING IN WAR BASIC STATS TO KNOW 1. Strength – this increases [...]
September 15, 2008 at 5:50 am |
So has anyone actually plugged this into a nifty spreadsheet so my brain hurts less while trying to figure out what weapons to choose between?
If not maybe I’ll have to take on that project when I have a little time
September 15, 2008 at 7:52 am |
Someone built a webpage for special attacks, pretty sure I saw it linked somewhere in the warhammeralliance forums.
It’s not straightforward in practice, however, since you have to know so many variables, some of which you can’t control or have a hard time figuring out on the fly (opponent’s armor mitigation and toughness).
September 15, 2008 at 6:58 pm |
re weapon skill, it’s important to pay attention to the fact this is a sliding percentage, with the slide based upon the target’s level. For those who want to include it in Disquette’s formulas above, substitute the following for the section that reads (1-armor):
(1 – AM + ((WpSkl / (OL * 5 + 50)) *.25) )
Where WpSk = player Weapon Skill, OL = Opponent Level, and AM = Armor Mitigation value in decimal format.
At level 10, our White Lion has a WpSk ~105. Assuming a level 10 spider with 15% armor mitigation and the same auto-attack for 70 Disquette generated above the final result goes from 59 (11 mitigated) to 70. ( 1 – .15 + (( 105 / (10*5 + 50)) * .25)) = (1 – .15 + ( 105 / 100 * .25 )) = 1 – .15 + .2625 :: cap of 1.
In comparison, assume all above EXCEPT our spider is level 40. NOW:
(1 – .15 + (( 105 / (40 * 5 + 50)) * .25))
(1 – .15 + ( 105 / 250 * .25 ))
(1 – .15 + .105)
(1 – .045)
.945
We get 66 (4 mitigated) as a result this time.
Final note : I do not think the formula for armor de-mitigation is correct, though it’s the current best known value. I also suspect there may be a cap. The reason is as follows:
Beta demonstrated that the base WpSk for a WL is 221 at level 40. Facing other level 40 players, this gives us a de-mitigation value of:
221 / 250 * .25 = .221, or 22.1%.
Assuming un-enhanced armor mitigations for the four basic levels (robe/light/medium/heavy) are 0/20/40/60%, this makes the armor penetration for WL (and WH and SM) 100%/100%/85%/65%.
For those counting from Destruction, the high WS is 197 for BOrc, DoK, and WE. They get 19.7% de-mitigation from level 40 WS vs level 40 players.
September 15, 2008 at 7:00 pm |
Sorry, I should have pointed out one obvious element of the above mess.
Facing level 40 opponents, the de-mitigation provided by your Weapon Skill = WpSk / 1000. 105 WpSk == .105 (10.5%) de-mitigation, as example.
September 16, 2008 at 12:21 pm |
Great article!. Quick question since im at work and dont have tiome to analyze formulas. How big impact on damage does armor have?.
In other words – Marauder has -75% debuff. How does it translate into damage gain?
September 17, 2008 at 10:50 pm |
[...] http://disquette.wordpress.com/warhammer-online-melee-combat-mechanics/ [...]
September 19, 2008 at 7:29 pm |
[...] Melee Combat Mechanics [...]
September 20, 2008 at 3:25 pm |
Are you certain that the de-mitigation is flat?
ie: WL vs Chosen, WL 25% de-mitigation from weapon skill, Chosen 60% armour mitigation.
So is the final armour mitigation for the Chosen now 60-25 = 35%?
Or is it 60*(1-25) = 45%?
September 21, 2008 at 5:57 pm |
One additional item open for investigation still, relevant to weapon choice for dual wieldders:
Does weapon speed impact the chance for the off-hand to proc?
In DAOC, Mythic normalized things so that you could not simply choose a fast weapon in order to have a powerful proc go off all the time, as I recall. In order to figure out what speed and DPS weapon to wield where, it would be interesting to know whether this is the case in WAR as well.
Hope that made sense ;)
September 22, 2008 at 4:40 am |
This is awesome. I can’t imagine how long this took but thank you for taking the time. Math for the win.
September 22, 2008 at 10:54 pm |
Kestrel, I agree. I only have a single wielder at the moment, but if I ever roll an alt, perhaps I’ll try it out.
The three open issues that I haven’t seen definitive answers to are:
- How are DoT damages calculated?
- How are Crit damages calculated?
- What is the offhand proc chance, and do either MH and/or OH weapon speeds affect it.
September 23, 2008 at 11:56 pm |
I was curious about picking out weapons, for example: trying to decide between a weapon with higher dps but less str.
For special attacks: how much str would it require to replace 1 dps less on a weapon? Is it the same as the tooltip addition to dps that str shows? It doesn’t seem like it for abilities as your formula show a much different calculation, and the tooltip itself states for auto-attacks I believe.
Do auto-attacks go off during special abilities? If so, how would this affect weapon choice once again?
September 24, 2008 at 7:10 am |
the funny thing is, this formula is almost exactly how the table top game works —> Str vs. tough – armor = damage done. It’s just more complex
September 25, 2008 at 11:33 am |
The global cooldown is 1.5 seconds on the abilities.
How does that make it the weapon speed or is used in conjunction with the weapon speed ?
If my weapon has a speed of say 3.4 seconds, how could I possibly hit with it more than once within 3.4 seconds ?
Technically any ability which uses my weapon should thus have a cooldown of 3.4 seconds and all other abilities which don;t use my weapon stay at 1.5 seconds.
I’m not saying the calculations are wrong.
In fact the math has nothing to do with it…as such and they could even be 10000% correct :)
It just seems very unrealistic by any means to to have a weapon with a speed slower than 1.5 seconds but still can swing the weapon every 1.5 seconds.
just weired,… i think.
September 25, 2008 at 1:12 pm |
This isn’t DAoC with a queing system where abilities happen along with your next swing. Your autoattack, if you have no haste tactics/buffs/etc would indeed happen at
0 seconds
3.4 seconds
6.8 seconds
10.2 seconds
etc etc.
Your instant attacks get interspersed with the auto attacks. Let’s say you’re just spamming as fast as the global cooldown would let you an instant attack called “clobber”.
you would clobber at
0 seconds
1.4 seconds
2.8 seconds
4.2 seconds
5.6 seconds
7 seconds.
(remember mythic has stated that the GCD in WAR is 1.4 seconds, unlike the 1.5 second GCD in WoW).
So, if you have autoattack on, and are spamming the clobber, your stream of attacks would look like this:
(first number is what time the event happened, auto = auto attack swing)
0.0 auto
0.0 clobber
1.4 clobber
2.8 clobber
3.4 auto
4.2 clobber
5.6 clobber
6.8 auto
7.0 clobber
8.4 clobber
9.8 clobber
10.2 auto
etc.
September 25, 2008 at 5:26 pm |
This may seem like a silly question, but is autoattack started automatically once you strike someone with an ability?
September 25, 2008 at 7:23 pm |
Very interesting results. It’ll be interesting to see if this applies to all abilities and all stats.
Now the special damage here, is in relation to what’s listed on the tool tip, correct? And this is only for instant abilities. Have you had time to test this against abilities that have a ‘casting time’ such as heals and regular direct damage attacks by changing the weapon speed to equal the casting time? Or does that not work?
September 26, 2008 at 4:26 am |
***Dual wield. Nearly same DPS. Different Speed.***
I ignore armor, toughness, parry, dodge…
My STR is 205 (witch elf 16).
Weapon_1: 20DPS, 2 WeaponSpeed.
Weapon_2: 21DPS, 1′6 WeaponSpeed.
***First Case***:
Mainhand=weapon_1. OffHand=Weapon2.
Mainhand:
DMG_MH = (205/10 +20) * 2 = 81 dmg
DPS_MH = DMG_MH / WSpeed = 40′5 dps
Offhand:
DMG_OH = (205*0′5/10) * 1′6
+ (2 – 1′6) * 21
+ 21 * 1′6 * 0′9
DMG_OH = 16′4 + 8′4 + 30′24 = 55′04 dmg
DPS_OH = DMG_OH *(proc_rate/WMH_Speed) = 27′5*procrate dps.
DPS_total_1 = 40′5 + 27′5*procrate
**Second Case**:
Mainhand=weapon_2. OffHand=Weapon1.
Mainhand:
DMG_MH = (205/10 +21) * 1′6 = 66′4 dmg
DPS_MH = DMG_MH / WSpeed = 41′5 dps
Offhand:
DMG_OH = (205*0′5/10) * 2
+ (1′6 – 2) * 20
+ 20 * 2 * 0′9
DMG_OH = 20′5 + (-8) + 36 = 48′5 dmg
DPS_OH = DMG_OH *(proc_rate/WMH_Speed) = 30′3*procrate dps.
DPS_total_2 = 41′5 + 30′3*procrate
Here you have a chart:
http://i35.tinypic.com/zt6res.jpg
Looks like, even with the penalization for wielding a fast weapon in the main hand, it’s still worth it at same DPS weapons.
September 26, 2008 at 9:42 am |
Just calculated the proc chance of the offhand hit, and it results 0′46. But should need a more extended review.
The difficulty is that, in the combat log, offhand hits appear just like the mainhand hits (so you have to play with the time of the attack and the damage).
Also, you can get parried or blocked you main hand attack, and the offhand attack may still proc.
September 26, 2008 at 10:29 am |
QUESTION: how toughness and resists work when fighting multiple opponents? Is it applied to all attacks or it only reduces that which comes as first? So in short if 3 low damage players would hit high toughness tank would they smack him down easily or would they need some burst damage pal to do the job for them?
September 26, 2008 at 12:46 pm |
star – that rocks. I copy/pasted what you wrote into the article (with attribution, of course).
Others, I’ll do a closer read through soon but for now I’m in a bit of a rush, I’ll check back in soon.
September 26, 2008 at 2:54 pm |
I have a question disquette. Have you found the order that defenses are calculated in? Example: When you attack someone, does their parry, dodge, block, etc.. come into play first?
September 26, 2008 at 3:56 pm |
No Toof, and that’s a pretty good question. I also don’t know if defenses are additive or multiplicative. That will take a lot of testing which will be a lot easier with better gear (where tanks can stack enough defensive stats to be unhittable if the stats are additive).
Additionally, there’s the outstanding item of how using a Great weapon affects defensive stats. Using a 2 hander supposedly gives you +10% chance to avoid parry, or +10% chance to crit – I forget which. But, that should be taken into account when people have enough time and gear to do the testing as well.
September 26, 2008 at 3:58 pm |
Georgio – Interestingly enough, when I was walking to my car last night, I started thinking about how Mythic implemented the hit code in DAoC, in which more people attack gave higher and higher bonuses to getting through a mob (or player’s?) defenses.
I’m pretty sure, based on “feel”, that toughness isn’t negated by having more than one attacker. But, that’s all it is – “feel”. We’ve all seen the ironbreaker or black orc that holds of 6 or 8 attackers at once. If toughness didn’t work on all attackers simultaneously, I don’t know that we would see that result.
But, I haven’t methodically tested that. Obviously Mythic is capable of making a system like that since they did so in DAoC, but I’d be surprised if they did it here, for RvR at least.
September 26, 2008 at 3:59 pm |
@ Zeal:
“This may seem like a silly question, but is autoattack started automatically once you strike someone with an ability?”
Yes.
September 26, 2008 at 4:00 pm |
Grr, I should have consolidated these, and will do so when I get home, but another thought is the cone of attack.
As you may know, in WoW specials had a larger cone of attack than auto attack. So, Zeal, you may be able to do specials without autoattacking (if that’s important to you) by turning a certain number of degrees away from a target so that your specials will work but your auto attack doesn’t – I don’t know.
September 26, 2008 at 4:41 pm |
disquette, You have numbers for the auto-attack for a single weapon but what is it for Dual Weapons? Does the offhand DPS effect the damage like it does for special attacks?
And you mentioned that your not sure about Two-Handed weapons. Or are they like Single handed weapons, only with a much higher DPS for the level range you get one at?
Also, Star suggests that you still gain a benefit for having a faster weapon in the off hand compared to the main hand. Do you agree with that?
September 26, 2008 at 4:46 pm |
Oh, and one last question. Does the DPS of a weapon for say, a Sorcerer, also change the damage done from special attacks? Or are we sticking to just melee classes right now, in terms of calculations.
September 26, 2008 at 5:29 pm |
Hi Sentack. There’s a whole section on how your second weapon autoattack damage is calculated, and you’ll see that Erdrick figured out how to incorporate offhand weapon dps into special’s calculation.
Yes, I do believe that other than some wierdness in either parry or crit %, 2H behaves like a 1H.
So far it seems that a faster weapon in OH is better, but I wouldn’t say it’s conclusively proved since I haven’t seen the proc rate proved over various weapon speeds (maybe they prorate proc chance per weapon speed, who knows).
For the last part, about weapon dps on spells, I’d guess not, but since I don’t play a caster, it’s probably better for you to test this and tell us! Use a sorc/BW/whatever, equip a weapon, cast. Unequip, cast on the same type/level of mob, and let us know if the dmg differs.
September 26, 2008 at 6:33 pm |
The question is not the auto-attack damage of the off-hand. The question is, do off-hand weapons effect the damage of main-hand auto-attacks like it does for Special attacks. Or is it, given that special attacks never proc an off-hand attack, so the off-hand DPS is added to the special attack, giving dual wielding weapon a minor benefit over single weapons for doing special attacks.
And indeed, initially, your calculations seemed to suggest that the best thing to do is have the fastest off-hand weapon for the biggest offhand bonus. So you want the slowest main hand, the fastest off hand and speed doesn’t really matter at all when it comes to Special attacks.
As for staffs and special attacks, what I’ll have to do is find 2 staffs of identical speed and different DPS and that have the same or Intelligence modifier as well. Since it’s typical for staffs to have an Int bonus. Weapon speed shouldn’t play a factor, since the casting speed of the spell is what you factor in for the damage. Or so we hope.
September 26, 2008 at 7:28 pm |
Oops! I see what you mean. No, I haven’t seen offhand weapon characteristics affect the main hand in the tests that I did, but I wasn’t thinking about that as a possibility, so perhaps such a thing exists and I didn’t think to look for it. I think I would have seen this in the spreadsheets when matching game performance to predicted performance, but I can’t say for sure.
And it sounds like you have a pretty good starting setup for the weapon dps testing for staves and spell damage.
September 27, 2008 at 7:43 pm |
Hi, just dropping by to say thanks for this analysis. I’m actually trying to code up a UI addon that can reverse out information from combat logs on the fly to firstly validate the formulas and subsequently let us do more interesting calculations in-game.
However, in my own (brief so far) testing in-game, I seemed to get good matching numbers if I took the minimum seconds-per-attack to be 1.5.
eg. 20dps 2.5s swing melee weapon, 76 strength, autoattacks on a level 11 venom bat produce 35 + 10 mitigated. That autoattack shows 18 dps, I expect to be putting out 27.6dps, so I assume that toughness is blocking 9.6 dps.
(As an aside, this appears to confirm that toughness can block more than strength-supplied DPS. I’m ranged so my strength is tiny)
Same mob, spanner swipe which is 87 damage no cast time or cool down, does 86 + 25 mitigated. If I then call that a 1.5s attack, that’s 74dps exactly. Add the 19.2dps from toughness (assuming toughness is twice as effective against abilities as against autoattacks) I’m putting out 93.2dps. Spanner swipe’s 87 damage over 1.5s is 58dps, 20dps weapon plus 15.2dps from strength produces 93.2dps as well.
I get similar results for firebomb, which is an ability with a 1s casting cost and no cooldown, again using 1.5 as the seconds to calculate the dps over. Sadly, I didn’t get whole the same exactly number results with firebomb (130 damage including mitigation, so over 1.5 seconds that’s 86 and two-thirds dps) but then again WAR hides fractional values from us as well in the combat logs.
I was going to do some testing with multiple-second casting time abilities, but Darklands crashed or something.
As a side note, it looks to me like toughness VS all, block VS all, parry VS melee, dodge VS ranged and disrupt VS spells, armour VS untyped damage and resists VS their own type of magic damage.
There’s some interesting combinations out there. Engineer grenades do corporeal damage but use ballistic skill for DPS and no weapon, so I expect you’d dodge not disrupt those.
I guess that explains why the renown gear appears to trade in other things for more toughness.
Also, it looked to me like criticals are applied to the part of the damage that survived the rest of the trip. Criticalling against the above-mentioned venom bat produced ~48% higher damage but the same mitigation value reported (ie autoattack went from 35 + 10 mitigated to 51 + 10 mitigated, ability went from 86+25 to 128 + 25)
Unless it _is_ 50% extra and there’s very late rounding going to to get the displayed numbers. I can’t see how anything that rounds to 35 can be multiplied by 1.5 to get something that rounds to 51 though.
Hmm. That turned out longer than I intended.
September 27, 2008 at 9:54 pm |
Awesome work dude! I’d love to see a mod that could be used to validate or contradict any formulas. I’m on the go for now, but I’ll look at it again later. Great stuff here!
September 28, 2008 at 4:05 am |
Just did a quick test with Gunblast, which is a 2s cast time, 186 damage attack on the same mob. Total damage including armour was 280, (for reference, armour is applied last in case that wasn’t obvious) giving 159.2dps over 2s or 205.867 dps over 1.5s.
I have 66.8dps from Ballistic + weapon dps, and the gunblast damage is worth 93 dps over 2s or 124 dps over 1.5 s.
66.8 + 93 is pretty close to 159.2dps
66.8 + 125 is way too low to approximate 205.867dps.
So the 1.5s is not magic to abilities, it’s just the minimum seconds over which dps is calculated.
Also, looking closely at my action bar, the GCD countdown starts at 1.5s, not 1.4s. ^_^ I guess they turned it back up as part of the lag-reduction and responsiveness-improvement stuff they did in late open-beta, which with what appears to be their current combat system, doesn’t actually drop your dps because it’s all calculated in dps anyway.
September 28, 2008 at 6:28 am |
Keep in mind that my character is an Archmage, and testing has been done just with that character. As such, my focus is mostly on spells and healing in particular.
A couple of observations -
- Crit seems to be a 33% multiplier to the base damage after toughness, but before reduction from armor. This is different from WoW where the crit multiplier is applied at the end (or more accurately, everything is multiplicative so everything is associative). This also means that crits in a sense ‘go through’ armor and resists to some degree (e.g. if regular hit is 100 post-toughness, armor mitigates 20%, and 80 net damage, a crit would be 133 post-toughness, armor mitigates 20% of BASE DAMAGE with a net damage result of 113). Effectively, crit is -33% armor/resist.
- HoT and DoT damage bonus seems to be dependent on cast time rather than duration. Since almost all HoTs and DoTs are instant, it means that they receive only 1.5s worth of modifier from their respective stats. For example, my instant-cast 15s HoT that ticks 5 times only receives 0.3 per tick per healing DPS, while my 3s direct heal receives 3 per healing DPS.
I have a theory that the healing debuff operates like ‘armor’ against healing. If this is true, it would mean that a crit heal would still go through a healing debuff (essentially it would operate like -33% healing debuff). Unfortunately, I have no easy way to test this.
Another topic that might be worth exploring has to do with AP regeneration. My anecdotal experience is that cast spells seem to burn AP much faster than instant spells; the one sure observation I have made is that you do NOT regenerate AP while you are in the middle of a cast, regardless of how long the cast is (this is easily observed by simply chain casting any spell with a cast time longer than the GCD).
Furthermore, there is some period after the cast where AP regen is suppressed (this is observed by chain spamming an instant spell). My current theory is that instant spells seem cheaper in practice since you are forced to wait 1.5s after casting the spell, and if (as an example) AP regen ticks occurred 2s after the last cast, you are much more likely to hit that tick than if you’d gone over the GCD with a spell cast.
This is potentially important as it means that the true cost of a spell is actually very different from the list cost. In particular direct heals in practice seem to burn AP extremely fast during RvR or any time you are getting set back.
September 28, 2008 at 1:35 pm |
Well, I’ve uploaded what I’ve got on my addon so far to Curseforge at http://war.curseforge.com/projects/libcombatcalcs/. Only after I did that do I realise that there’s no repository viewer there, and it’s not good enough yet to make a release out of.
If you are willing to use git, you’re welcome to check the addon out and try it. It doesn’t work with DoTs yet (I can’t seem to make the math work on any of the DoTs I have access to) and it also fails on Doombolt for some reason.
I only have an Engineer and a Sorceress so far.
It outputs to /debug, the interesting part of the output is where it tries to calculate the toughness of the target depending on what kind of attack the ability is. It has to guess, since there’s nothing in the client-visible information that tells you what type of attack it is. >_<
Anyway, the autoattack numbers look pretty solid to me, and most abilities I’ve tried come out fairly close to matching for toughness calculations, so there’s probably some fudge for rounding going on.
I’d post it to the WHA Mechanics forum, but I don’t qualify to post there yet.
Hmm. Just looked more closely at my numbers, after reading someone on the WHA Mechanics assign a 66% coefficient to Doombolt…
My doombolt hit for 237 before toughness, toughness from melee was about 17dps. If I assume doombolt’s being calculated over two seconds, then that’s 135.5 dps. 100dps comes from the 200 damage assigned in the spell description, and my intelligence DPS bonus is 35.6.
So maybe the issue that that the time used to calculate DPS isn’t actually casting time?
I might see if I can enhance my addon to record known toughness from melee/ranged autoattack and then try and back-calculate spell times if we don’t get good numbers to start with. Or even better, add a window that lets you tweak things until the numbers add up.
September 28, 2008 at 11:05 pm |
I have a question about how % armor is factored into this equation. I have a 14 squig herder that can have up to 556 armor, which gives him 31.6% mitigation. If tanks at level 40 have over 3000 armor, that equates to well over 100% mitigation. Since tanks are obviously able to be damaged at that level, how does this 100+% armor mitigation factor in? In your formula, anything over 100% mitigation becomes a negative number.
I believe weapon skill would be factored into your formula as:
Sdmg = (((StrBon/10)*2 + Wdps)*1.5 + Admg)*(1-armor+(weapon skill/100))
So, is it simply that every class has enough weapon skill at rank 40 to make up for a tank’s incredibly high mitigation? Or is there more to the formula?
-Travail.
September 29, 2008 at 3:40 am |
Travail: The armour mitigation percentage formula has the player level as a divisor. So higher level players need more actual armour value for the same mitigation.
Weapon Skill then reduces that percentage by a percentage of its own, again with level as a divisor.
From the look of things, all the percentage-based combat values use level as a divisor, while DPS-based numbers are all pretty much raw values irrespective of level. (Except damage done by abilities, which scale with ability level)
I’m pretty sure there’s a cap on effective armour mitigation as well, but haven’t looked experimentally for it. (Others have, I’m sure)
However, armour’s not that interesting to me since it’s a easily visible mechanic compared to toughness and stat-based DPS contribution.
September 29, 2008 at 12:29 pm |
Hola, the question of armor mitigation came up in one of the warhammeralliance.com threads, and a guy there tested and verified that 75% was the max mitigation via armor. See the post here:
http://www.warhammeralliance.com/forums/showthread.php?p=1169667#post1169667
September 29, 2008 at 6:31 pm |
Hey m8,
Ive had a shammy forever… enhance for most… always been a fan of your work in that field, so when i saw your post i was well chuffed… and more confident now in what ill need to do as I’ve just started a WH …
cheers
and gw
September 30, 2008 at 8:41 am |
Is there a difference in handling of one handed and two handed weapons ? 2H gets more “built-in-dps”, so far so good. But do 2H weapons get more bonus out of strength or other things ?
September 30, 2008 at 2:12 pm |
Disquette:
I think whether proc chance is “normalized” has implications not just for dual wielders, but for any weapon swing that has a chance to proc an ability.
Star:
If I read correctly above, your conclusion was that, given two weapons of the same DPS but different speeds, fast main-hand and slow off-hand dealt better damage. BUT in your testing setup, your weapons are different dps, correct?
Weapon_1: 20DPS, 2 WeaponSpeed.
Weapon_2: 21DPS, 1′6 WeaponSpeed.
So I’m confused about your conclusion, I think testing has to be with the same dps weapons of different speeds.
September 30, 2008 at 4:50 pm |
I notice when I level up the Parry, Dodge, Disrupt and Armour Penetration stats reduce while the core stats such as weapon skill etc increase.
Anyone know why?
September 30, 2008 at 6:03 pm |
Bay City Roller:
Because those core stats are actually gaining base points as you level up (look at the log messages when you gain a level) while the derived stats such as parry, dodge, disrupt and armour (all percentages) include a term that is divided by your level, so as your level goes up, their values go down.
October 1, 2008 at 9:31 am |
ITs interesting, that you suggest that the STRbon is applied before multiplied with WeaponSpeed!
Weapon 1: 20 DPS and 4seconds per swing >> Pure weapon damage is 80
Weapon 2: 10 DPS and 2seconds per swing >> Pure weapon damage is 20
Its a major difference if you add Str-Bonus to the DPS or to the Pure weapon damage. More logic for me is to add it to the weapon damage.
October 2, 2008 at 12:10 am |
Sirak: Even the in-game tool tip says strength provides DPS, and my own testing supports this.
This way, the speed of the weapon is actually uninteresting to your damage output, your total DPS is the sum of your DPS contributions. Whether your weapon is fast or slow just determines how often that DPS gets turned into actual damage. For your two weapons, the first is twice as damaging as the second overall. Weapon two hits twice for 20 in the time it takes weapon one to hit for 80, and both get 4 seconds worth of StrBon added over that four seconds.
October 2, 2008 at 1:20 am |
So until we know what the division factors of the different levels are, it’s pretty hard to convert a straight armor rating into % mitigation. Interesting, well, that makes my own testing harder. Ah well..
Thanks for the reply.
-Travail.
October 2, 2008 at 9:21 am |
@Travail
Taken from WarhammerAlliance forum
Armor: = Armor / (Level * 1.1)
It appears that Armor is capped at 75% mitigation
For example, at Level 40:
3300 Armor = 75% Mitigation (This is the Armor Cap)
2200 Armor = 50% Mitigation
1100 Armor = 25% Mitigatio
October 2, 2008 at 9:23 am |
above should read:
Armor (Percent Damage Reduction) = Armor (Rating) / (Level * 1.1)
October 2, 2008 at 10:23 am |
Great article, I feel a lot happier looking at weapon stats now.
WRT off hand attacks.
Off hand proc chance 46%
Main hand DPS modifier for special attacks: 150%
Off hand DPS modifier for special attacks: 67.5%
It is interesting to see that 67.5% = 45%*150%. It seems like having an off hand weapon adds 45% to your DPS (I’d accept 46% as a slight statistical glitch).
I’d also consider changing the off hand damage calculation to…
OHDmg / (1 – Armour) = OHWdps * MHWSpeed
…and seeing what comes out. It seems like it’d produce very similar numbers, is consistent with wanting to add 45% off hand DPS to your own DPS and is certainly much simpler. On your example it’s short by 2.6 DPS but without your actual raw data I couldn’t say anything for sure.
You don’t have any raw data to share by any chance?
October 2, 2008 at 11:14 am |
Very good article.
Anyone have some writeup on specifics of parry/dodge/shield mechanics?
October 3, 2008 at 3:00 pm |
Great article. Are the numbers still current? The warhammeralliance forums are filled with info saying 5-6 str = 1 dps, while over here from how i understand it, 10 str = 1dps.
October 3, 2008 at 5:13 pm |
Marauder, from this article 10 STR = 1 DPS for autoattacks, 5 STR = 1 DPS for special attacks
October 4, 2008 at 10:59 am |
[...] is that WAR’s combat calculations are not fully understood yet. An excellent primer is available at Disquette’s Weblog and Warhammer Alliance has a Mechanic Analysis forum as well. I’ve posted some comments at the [...]
October 5, 2008 at 8:05 pm |
I’m not sure if this applies to the general, but I think that a weapon’s DPS does not factor into the formula for ability attacks. At least not for ballistical weapons such as the Witch Hunter’s pistol:
I have two guns with these relevant stats:
Gun 1: 28.0 DPS, +27 Ballistics
Gun 2: 30.9 DPS, +18 Ballistics
Translating these into DPS improvements according to your formula I would get:
Gun 1: Auto-Hit: +30.7 DPS, Abilities: +33.4 DPS
Gun 2: Auto-Hit: +32.7 DPS, Abilities: +34.5 DPS
So Gun 2 would be in any situation superior.
But while this is true for Auto-Hit, Gun 2 does actually do less damage than Gun 1 during the ability Absolution.
Not sure whether this only applies to guns. Will have to do further research on melee weapons too.
October 5, 2008 at 8:11 pm |
Thanks for the teting Frederic – I assume you made sure this was on the same mob, same level, and you other stats didn’t change?
If so, this would be odd, but good that someone tested. Can anyone else with a ballistics-based character test to confirm Frederic’s findings?
October 6, 2008 at 2:22 pm |
I suspect in Frederic’s example the wrinkle comes from the fact that the WH gun is an offhand weapon. I’d guess that absolution damage bonus is based 150% off (mainhand) rapier damage and 67.5% off (offhand) gun damage, just like for any other dualwielder, and applies ballistics as a modifier. The 9 points of ballistics is worth 1.8 dps, whereas the 1.9 dps from gun1 is reduced to 1.3 dps by the special ability damage bonus offhand formula. Absolution should then do 0.5 dps extra.
This is a total guess – but it should be easy enough for someone with a WH to test it.
October 7, 2008 at 7:11 pm |
[...] these formulae are right Warhammer Online – Melee Combat Mechanics Disquette’s Weblog then strength can be thought of as the opposite of toughness. Weapon skill reduces the effect of [...]
October 7, 2008 at 10:25 pm |
exact gwyntar,
except the main hand weapon doesnt count in SA OffHand, it s basically :
SAGun = ((Bal-Thou)/5 + 0.45 * Gundps) * 1.5 + SAdmg
and
SASword = ((Str-Thou)/5 + Sworddps + 0.45 * Gundps) * 1.5 + SAdmg
disquette,
your OHdmg formula doesn’t work for WitchHunter :(
Data :
SwordAutoAtk : M
GunAutoAtk : O
RazorStrike is a Sword SA (221 base dmg) : SAM
Absolution 1 is a Gun SA (145 base dmg) : SAO
dmg noted : dmg (mit) [total]
All atk versus spider level 13 (i am lvl 31), some hits are differents vs the same mob (ex: M : 156 143 143 143), often the first hit, so i exclude this data
Set 1 :
Str 526, Bal 416
Sword 36 dps, 2 speed
Gun 36.8 dps, 3.1 speed
M : 143 (11) [154]
O : 88 (7) [95]
SAM : 391 (31) [422]
SAO : 240 (19) [259]
Set 2 : Changing sword from set 1
Str 500, Bal 416
Sword 9 dps, 2 speed
Gun 36.8 dps, 3.1 speed
M : 88 (7) [95]
O : 88 (7) [95]
SAM : 344 (30) [374]
SAO : 239 (20) [259]
Set 3 : Changing gun from set 1
Str 511, Bal 395
Sword 36 dps, 2 speed
Gun 9 dps, 2.5 speed
M : 140 (11) [151]
O : 41 (3) [44]
SAM : 370 (30) [400]
SAO : 218 (17) [235]
Set 4 : Changing sword & gun from set 1
Str 485, Bal 395
Sword 9 dps, 2 speed
Gun 9 dps, 2.5 speed
M : 85 (7) [92]
O : 41 (3) [44]
SAM : 323 (28) [351]
SAO : 217 (18) [235]
Set 5 : same speed for sword & gun
Str 507, Bal 395
Sword 31 dps, 2.5 speed
Gun 9 dps, 2.5 speed
M : 162 (13) [175]
O : 50 (4) [54]
SAM : 362 (29) [391]
SAO : 218 (17) [235]
So,
- Mdmg depends only from sword with variance by str in my data : (StrT + Sword) * Swordspeed
- Odmg depends only from sword & gun but i dont know how, may be by inspiring from your idea with two distinct penalties : (0.5*BalT + 0.9*Gun) * Swordspeed. It s the closest formula I found but it s strange for me.
- SAM dmg depends from sword and gun : (2*StrT + Sword + 0.45*Gun) * cdspeed
- SAO dmg depends only from gun : (2*BalT + 0.45*Gun) * cdspeed
Btw sorry for my english, it s not my native language
Ragh
October 8, 2008 at 1:59 am |
Quick question.
Would it be possible for someones toughness to reduce your DPS to 0 ? or does it just effect abilitys ?
October 8, 2008 at 12:38 pm |
I don’t know Vesh. There may always be a 1 dmg minimum. So much stuff to test.
October 8, 2008 at 7:05 pm |
The dual wield formula (sans the left side, that’s simple enough as is), if you play with it enough, becomes
((StrBon * OHSPen / 10) – (1 – OHWPen) * OHWdps) * OHWspeed + MHWspeed * OHWdps
IMO this formula looks more complicated but I prefer it because it isolates OHWspeed and we can see the effect it has on damage.
The biggest chunk of your dps comes from (MHWspeed * OHWdps).
When OHWdps > (StrBon * OHSPen) having a faster offhand weapon is going to result in (very slightly) more damage (because in this case the speed of the offhand penalizes damage), otherwise a slowwer offhand is going to produce (very slightly) more damage.
Using the stats in the WL example (I know it’s not a DW class, but it’s not really important, just the stats) the difference in damage per hit between a 1.0 spd OH and a 5.0 spd OH is 4.8 (before armor) – 100 vs 104.8. Essentially not enough to really matter, but enough so that if you’re a min/max type of person there’s something there to hang on to.
It feels like “(MHWSpeed-OHWSpeed)*OHWdps” should be “(MHWSpeed-OHWSpeed)*OHWdps*OHWPen”, which means the entire formula would simplify to “(StrBon*OHSPen/10)*OHWSpeed + MHWSpeed*OHWdps*OHWPen”. In this case a slower weapon would be better when StrBon > 0.
October 8, 2008 at 10:09 pm |
Allright thanks Disco errr Disq :P Hafgrim says hey btw.
And another question toughness is worked out before armor is for damage reduction right ? just making sure i understand not so great at reading the forumla’s
October 9, 2008 at 4:03 am |
Yah Vesh – toughness comes before armor, so tanks actually get less out of toughness proportionally than a cloth wearer.
Man, that’s awesome about Hafgrim. Tell that dude hello from me too :-)
Oh, and I found a good thread about your other question, regarding minimum damage:
http://www.warhammeralliance.com/forums/showthread.php?t=139187
Looks like if you hit, you’re going to do at least 1 damage, even if you should be doing 0 according to the math.
October 9, 2008 at 8:28 pm |
This is truly a terrific piece of work disquette. I’ve been trying to do something similar for DoT calculations (specifically Magus DoT calculations) and while I have some very approximate figures, I would be eternally grateful for some guidance in working out forumlae such as yours.
I’d really like to find out, in accurate terms, how well DoT spells scale with level, what affects Resistance has, what effects Toughness has, etc.
I’ll try and get a link up to it when I can but you can find it in the WHA Magus forums – “Magus DoTs – some numbers at last”.
October 9, 2008 at 8:32 pm |
Oh, and for magic at least, crits do indeed have a random element to them – the DoTs at any rate.
October 10, 2008 at 9:02 am |
Ahhh thanks disco.
October 11, 2008 at 2:34 pm |
Question about Weapon Skill Armor Pen % and how to use it correctly in your formula.
–My stats–
STR= 474
Wdps= 34
Wspeed= 2.5
After hitting a mob with autoattack i get the results with a single weapon of:
Mitigated Damage = 31
Actual Damage = 145
Total Damage = 176
Using the formula to calculate Targets ArmorPen% and Toughness i get:
Toughness 110
ArmorPen% 17.6% (0.176)
Using **Dmg = ((StrBon/10) + Wdps)*(Wspeed)*(1-Armor)**
The DMG = 145 which is what the games shows in combat log. Good.
Now when i add my Weaponskill ArmorPen% of 17.8%. The formula now shows me doing 150.5 damage per hit. But in game i’m only doing 145. So this means that the mob has higher Toughness, or Armor pen% or both. Which lowers my 150 dmg to 145 dmg.
How do i find out the mobs toughness or armor pen% when also using my weapon skill armor pen% in the formula?
Thank you.
October 12, 2008 at 4:42 am |
Theres like a 100 people posting usable discussion for every retard telling people to get laid and putting things “ON MY SCULPTED 6-PACK CHEST”.
Tnx for the blog, had immediate effect on my play.
October 15, 2008 at 1:39 am |
Good stuff. Especially liked how you explained to process. Smart but not cocky.
For the offhand/mainhand equation, is the (MHWSpeed-OHWSpeed)*OHWdps part of the equation supposed to be the absolute value of (MHWSpeed-OHWSpeed)? Otherwise I’m getting negetive damage output in my spreadsheet if I use a greater weapon speed weapon in the offhand.
October 15, 2008 at 4:24 pm |
Hi Folks,
I’m wondering how many WoW Theorycraftler are posting in this board… jesus :D
I mean… wake up guys – THIS is not WoW and you don’t need sustained DMG to be sucessfull in this game. But wait… if you’ve rolled on the blue side just ignore my post and fell free and stack STR like hell. I’m sure this is the best thing you can do for the red side^^
Instead of stacking your offensive stats (most of them are working linear and won’t affect the base DMG from your spells at all) take a lock at your defensive stats. HP (healps your healer to react), Toughness (decrased every DMG taken by X amount), Resistance (decrased DMG taken by X% amount), and initiative (trust me, Initiative is one of your bread an butter stats to survive in RvR![1])
[1] I’m wondering why nobody has already figured out the impact of having a low amount of initiative, when it comes to initiative debuffs. I mean read the forums, you’ll find a lot about HP and TH and even Resist. But information what happens when you have low initiative in combination with initiative debuffs? Nope, not a single information. Srly, people are blaming in the forum how OP Boiling Blood etc. is. But did you know what a initiative debuff does if you have not enougth initiative stacked?
Try the formula below with your own values… an see what happens when someone debuffs 120 initiative.
= [(Level * 7.5 + 50) / Initiative] * 10
To make it clear, if you have only 200 initiative @ lvl 40 you’ll be an instant gib in RvR (~45% critchance from all sources). *hint* the formula behind getting critted is exponential *hint*
-Punch Helmgard [DE]
October 15, 2008 at 4:36 pm |
Seems this website has some sort of CSS protection. The formula i’ve posted is should be…
%Critchance = [(Level * 7.5 + 50) / Initiative] * 10
-Punch Helmgard [DE]
October 22, 2008 at 9:52 am |
Please check my post on Warhammer Alliance.
http://www.warhammeralliance.com/forums/showthread.php?p=2292823#post2292823
I have screenshots which show that 2h ability damage is NOT adhering to the formula but instead is identical to 1h damage.
October 22, 2008 at 2:01 pm |
Yep, that’s a neat one. I posted my respone in that thread. If you post here again with updates, I’ll get emailed that you did so. Right now we know, based on your testing, that either the formula is incorrect, or that different specials act in different ways. Once you figure out which it is, I’ll edit this blog article to add/edit according to your findings, linking to your work, of course!
October 25, 2008 at 11:16 pm |
My new test is going to blow you away. :)
http://www.warhammeralliance.com/forums/showpost.php?p=2292831&postcount=2
October 29, 2008 at 1:28 pm |
Very, very interesting post. Playing a sorc, spell dmg calculation would be even more useful to me, but still …
One question: Your examples calculate StrBon as StrDPSeffect -TouDPSreduction
Have you tested if this difference can also become negative in case of the opponent’s toughness being larger than my strength? (Maybe not menaingful in PvM, but if I itemized adequately in PvP, this might well become more than a theoretical question …).
Thanks & please keep up posting & finding formulas
Sequana
October 29, 2008 at 2:12 pm |
Check the tests by absalon here: http://www.warhammeralliance.com/forums/showthread.php?t=162317&page=6
Looks like once StrBon goes negative, it gets hard floored at 0.
November 5, 2008 at 12:04 am |
Hey disquette!
I hope you don’t mind, but your formula has inspired me to create a flash Calculator.
http://www.taylorbixler.com/ZomgDmgCalc.html
If you would like to work on a more advanced one with me, just lemme kow ^_^
November 5, 2008 at 12:06 am |
… er *know*
November 5, 2008 at 1:34 am |
I have a few questions about hidden ability-dmg modifiers for equipping a non weapon in the off-hand.
I’ve also seen some skewed results when testing against red+ mobs, maybe I could send you my equations + variables for QA? XD
November 5, 2008 at 4:23 pm |
So um… how do I calculate “WeaponSkillArmorPenetration%” in
:ArmorMitigation = TargetArmor%*(1-WeaponSkillArmorPenetration%)
without knowing the targetArmor%?
November 7, 2008 at 4:17 pm |
Explain in your formula’s why increased weaponspeed, lowers the dps.
the longer the speed the higher the dps.
this is NOT correct obviosly
November 7, 2008 at 11:41 pm |
Explain in your comment what formula you’re talking about, and I can either correct the formula, or explain why you’re wrong.
this is NOT a good comment without any specifics, obviously.
Also, learn the difference between damage per swing and dps.
November 8, 2008 at 8:12 am |
Dmg = ((StrBon/10) + Wdps)*(Wspeed)*(1-Armor)
You use Wdps (weapon DPS) and than use (Wspeed) aswell
you can’t use Wspeed AND dps, its either Wspeed + AVERAGE damage, or just DPS.
I’m srry i wan’t clear enough, not trying to dis you or something.
I hope you can adjust ut formula.
Keep up teh good work
November 8, 2008 at 8:22 am |
In WoW we had Min and Max damage on a weapon, we used teh average damage and multiplied it with the actual speed of teh weapon (which makes sence)
In WAR only the DPS is stated on a weapon therefor it’s not needed to use teh weaponspeed in any calculation.
I’m not certain how to fix your formula but i’m sure u need to remove the weaponspeed bit.
Just add your formula intro Excel and leave all numbers teh same, just change teh speed.
You’ll see that increasing teh weapon speed (lowering the number) lowers the dps output in your formula.
Again gl and keep it up
November 12, 2008 at 7:12 am |
teh teh teh
November 12, 2008 at 11:03 am |
Ok i remeber from old Eq1 times, if your offhand weapon had more delay on speed then the main hand it would only swing every second time.
So it was important to have the slow weapon in main hand or same speed.
Any know if its like that here as well ?=
November 14, 2008 at 5:12 pm |
Your calculations are very impressive, and I am impressed by the amount of effort and time everyone contributed to making these calculations.
I do have one question regarding initiative and its lack of presence in the above equations. The tooltip in the character screen implies that its used as a bypass mechanism for armor mitigation, however I don’t see that present in your formula’s.
I would assume both the player and NPC have this statistic. Is this statistic useless to players, or only used in PvP combat?
November 18, 2008 at 10:54 am |
Thanks for the great post disquette, very useful.
November 18, 2008 at 11:00 pm |
This formula for simple, straightforward melee damage is fantastic. But have you (or anyone) stumbled upon a useful formula for DoTs yet? Or a healing formula for that matter?
November 24, 2008 at 1:36 am |
I believe have been able to figure out / answer both of my questions. In the process, I believe I have found one small optional addendum to the ability formula:
Sdmg = ((((StrBon/10)*2 + Wdps)*1.5 + Admg)/ticks)*(1-armor)
The very simple divisor of ticks (before armor mitigation) allows us to account for both instant damage as well as DoTs. For instant damage, the ticks is obviously an irrelevant 1, which is why the current formula works. This modified one will produce the DoT numbers represented in the combat log, as the log shows DoTs on a per tick basis. So far under my testing this has been accurate to within 2 points. If an ability has both instant damage and DoT, they are calculated independantly. Alternatively, the formula can be left as is, which results in the total damage that a DoT will do if all of its ticks successfully run.
Healing is easier to test due to the fact that you can injure and heal yourself:
Heal = ((willpower/5)*time + Ability)/ticks
Time, as Disquette has noted, is usually equal to 1.5, more if there is a casting time. Pretty simple. Maybe everyone else here already assumed these formulas.
December 22, 2008 at 9:48 am |
Hi Disquette and thank you for sharing all the informations.
I am wondering if the original formula : tou = 10*((((0.1*Str + Wdps)*Wspeed)-TDmg)/Wspeed) is still viable as this : TDdmg = – Wspeed*(0.1*(tou-str)+Wdps+Wspeed). I find it difficult to isolate Mitigation, but if TDdmg = Effective Dmg + Mitigation Dmg and that this formula is right : Dmg = (((Str – Tou)/10) + Wdps)*(Wspeed)*(1-Armor), perhaps it might be possible to isolate Effective Dmg in a formula.
Still, I am new to this, and I like what you did for the War community Disquette. Thank you.
December 22, 2008 at 1:38 pm |
err, the right formula is :
TDdmg=-Wspeed*(0.1tou-0.1str+Wdps)
December 22, 2008 at 8:33 pm |
Hi Disquette,
I am working on a program using your formula to determine under what conditions wounds or toughness is better, also simulating multiple attackers and healing. Most of the discussions on this subject do not take into account your formulas, and are consequently based on false premises. I have a few questions:
- Are your formulas confirmed post-1.1?
- Do you know anything more about crit and spell damage formulas?
Thanks.
December 26, 2008 at 7:49 pm |
TheDoktor,
If the formula above is correct there is no contest between Wounds and Toughness.
Lets take 100 Wounds vs. 100 Toughness
100 Wounds provides 1000 hps
100 Toughness mitigates 30 damage per ability use (((100/10)*2)*1.5). So you’d have to sustain 34 hits to equalize. This isn’t taking your armor into account either since the the subtraction happens BEFORE the armor reduction. So with 20% armor reduction you’d only really get 24 damage saved for a total of 42 hits to equalize. The higher your armor and resistances, the less valuable your toughness.
As long as 1 Wound = 10 HPs, it is the winner hands down.
December 29, 2008 at 8:56 pm |
Using the above calculator, I actually ran some numbers thru the system, trying to get an idea of the difference between using a 1h weapon and a 2h weapon with my Black Guard.
Some information. My BG is level 29 with 11 Ranks in Malice spec. So all Malice based abilities are level 25 (I think, I’m not logged in). My tank has 279 Strength and I’m assuming the target mob has 200 Toughnes. (General Rough guess)
I’m using an Special ability that has a value of 200 (Monstrous Rend) with a 1h sword with 34 DPS and 2.2 speed. Vers a 2h Spear with 49.3 DPS and 4.0 Speed.
On a per hit basis, the 1h came up with 78 dmg (after mitigation) and 194 dmg with the 2H. That’s almost a 2.5x jump in damage. But, if I factor in the weapon speed and adjust for that between the 1H and the 2H, I get a 1.36x jump in damage. Basically, a 33% increase in damage over time.
But, the interesting part is the special damage. Using the above numbers, the 1H special damage is 233, and the 2H damage is 253. That’s only a 8.5% increase. Very very minimal.
This is starting the match what I’ve heard about 2h weapon damage in general. When using special attack, you rarely see a huge jump in DPS by going to 2H. So that’s why Sword and Shield tends to be preferable over Two Handed for most tanks, if they have an option to do so. In some cases, the good abilities require 2H, so you don’t have a choice. Still, very interesting numbers. Anyone else think so? Am I reading anything wrong?
January 1, 2009 at 5:24 am |
A question about what to use MH and what OH. I have had a discussion with a person in game that having Bloodfist Makhaira MH and The Matron’s Fang OH is better than the other way around. I kept trying to explain the whole “OH atk is a proc of the MH” but the person pointed me to Elitist Jerks forums which when I loaded in my webbrowser turned out to be WoW-related. :/ So I’m pretty much sure the person has a wrong impression of how dual wielding works in WAR, but just in case which one of these is better as MH and which one as OH weapon?
http://www.wardb.com/item.aspx?id=640236
http://www.wardb.com/item.aspx?id=475142
The Bloodfist Makhaira also has a 5% chance to proc extra 267 unmitigatable damage which is not showing on WARDB tooltip.
January 5, 2009 at 1:27 am |
Amazing work disquette! Just curious if anyone has figure out a DoT formula.
Also, I used this formula and confirmed it accuracy with Razor Strike and Torment. But, with Fervor (on the instant damage), it did not match. I am not sure if that has to do with the DoT issue. The formula predicted a much higher value than was actual. When I took out for the off hand contribution, it got to with in 7 pts (240 actual vs 247 predicted), just FYI.
February 13, 2009 at 10:39 pm |
[...] in this post that hasn’t already been said. In fact in might be in your best interest to read Disquette’s Weblog. It I’m simply explaining some things I think to be important so that by the end you will [...]
February 23, 2009 at 4:52 pm |
That is great work Disq. Really helpful to all of us who want to know how it works but don’t have the time to figure it out.
It is obsessiveness like this that got men to the moon, and worked out that our forefathers hung around eating bananas a lot.
Love your work :)
February 25, 2009 at 10:16 pm |
A very nice writeup, I would like to say you’re a gamer after my own heart or vice versa. I don’t play a single game without spreadsheets lol. I am very impressed at your write up, good job and keep it up!
March 26, 2009 at 8:59 pm |
Thank you for the work you have done and posted, I am curious however about the value of weapon skill. At what point dose X weapon skill = Y strength. Seems as if high weapon skill will do more against a highly amoured enemy since it’s percentage based. But as far as a damage formula i think it would be of value to see weapon skill’s contribution.
March 27, 2009 at 12:42 am |
Disco you silly theorycrafting ho. I just bought Warhammer today after enjoying the shit out of it on the trial. I’m planning on playing DoK, and it looks like you’re already playing, so I just might see ya around.. oh also thanks for answering all my questions =D
March 27, 2009 at 11:18 am |
Nara! I actually stopped playing about 2 weeks after release, but I wrote this up in Beta and posted it when the NDA was lifted. The next thing I’m looking at playing is the Old Republic, probably with Neera, but I still read the Warhammer Alliance forums because I like what this game could have been. My WoW susbscription is long gone, so it’s just work, reading, and rockband for me these days.
Thanks so much for posting something, it’s awesome to see you here :D
March 29, 2009 at 12:38 pm |
Great work disquette! Its awesome to finally have everything sorted and on paper. :)
March 29, 2009 at 6:09 pm |
Got a question: Anytime the main hand hits, there’s a chance the offhand will do damage as well, as explained. But what if a special ability is used, does the offhand have a chance to auto hit or does the offhand factor into the special ability damage?
March 31, 2009 at 4:51 pm |
Thanks for posting this, awesome work. We had begun trying to feel this out via non standard testing, arguing and beer, but your method seems much more precise. Cheer!
April 8, 2009 at 8:22 pm |
[...] Zu allererst 2 Links mit weiterführenden Informationen zu dem Thema: http://www.warhammeralliance.com/forums/showthread.php?t=54306 http://disquette.wordpress.com/warhammer-online-melee-combat-mechanics [...]
April 28, 2009 at 6:27 pm |
[...] de blogpost opgezocht die ik laatst gelezen had over weapon dmg: Warhammer Online – Melee Combat Mechanics Disquette’s Weblog dps van een wapen wordt dus idd gebruikt __________________ What a difference indifference can [...]