Guild Wars 2

Puandro.3245:

The cast times are correct i can assure you, i used recording software and viewed it frame by frame to get it correct.

I did the same, and we’re only different by .1 seconds, so it’s not exactly a huge difference. Mine are .1s faster for two attacks (3.5 vs 3.6).

I also don’t know what you mean im using abilities before their CD’s are up.

You weren’t for the build you originally had for that rotation. You were counting on a 6 second 100b CD from the 20% reduction via traits which you won’t have with 30/0/0/10/30. I didn’t take the time to go through and reedit the swaps to 100b that were coming in at < 8 seconds (sometimes ~7). When I saw that even before those tweaks, it was only @ 5.4%, I lost energy for the effort.

Puandro.3245:

If you care to add a .2 sec delay for swapping you can edit it in and make your own rotation. If you enable editing you can even edit the buffs/traits ect. Also you cannot swap every 5 secs with Axe/Axe, you can only swap after Chop, Double Chop, or Tripple chop. Swapping mid auto attack will cause dps loss.

I did, and accounted for both, I let full chop chains go for the Axe/Mace + Axe/Mac swaps rather than attempting to optimize mid chain as the whole goal of that was not to find highest possible DPS (knowing GS + Axe/Mace would be that), but to find the highest consistent DPS with lowest rotation management overhead. Essentially, Auto-attack and hit Cyclone/Crushing Blow/Tremor on CD while swapping as close to 5s as you’re able.

Excellent sheet you created for seeing full debuff/buff DPS. The numbers can get pretty crazy. Axe/Mace + Axe/Mace bumping right up against 10k dps is insane.

One time I created a Sylvari character and kept it.

Charr in mostly cultural T3 (get different boots at the minimum) is very hard to beat.

Thief (S/P) / Spirit Ranger are the first two thoughts that come to mind, although it’s hard to find a bad combo in the game at all while leveling.

Nike Porphyrogenita.8137:

Whats hilarious is, the 30/0/0/10/30 trait allocation would be better even in a pure GS swap to a GS with Battle sigil setup than your 30/30/0/10/0 build.

FWIW, you haven’t kept track of the thread very well. As I’d hoped, information came to light (in large part thanks to Paundro and others) that helped establish where you can get gains from 30/0/0/10/30 over 30/30/0/10/0 reliably.

Since that time, the question has circled to which weapons to use with 30/0/0/10/30 for highest raw dps. I also found 2.6% (close enough to three where I imagine even you wouldn’t quibble) improvement from using both GS and Axe/Mace. However, the rotation complexity (not swapping weapons, but maintaining CDs that do not align well on both weapon sets, force cancelling chains or channeleds) to earn that 2.6% increase I do not believe is worth it nor realistic in real world situations.

Beating on a stationary target? Sure. Dodging/watching for red circle spam/moving/switching to an add that spawns… all that added complexity is likely to cause tiny but accumulated delays that eat away at the % gain.

The question of utility is different and one that’s likely to change boss to boss, the goal was to first establish which was the most reliable, consistent highest dps.

You said before you’d said all you had to say on that matter, but am glad to see you back to participating. Now may be a good time, if the thread is worth your time to respond, to read it as well.

Based on your spreadsheet with the very optimized Axe/GS rotation (cutting into chain combos to get swaps dead on the 5s mark every time or to prevent a single double strike of an auto when unwanted) when compared against the simpler axe/mace full (two sets, swapping every 5s within bounds of CDs/Chains), there is a 5.4% difference in DPS when the traits are set to 30/0/0/10/30 with the build mentioned before.

I also noticed some issues where abilities were being used before their CDs would be available, pushing up higher dps moves into the rotation. It’s not a significant change, but enough of one on small timeframes like this to account for some of that % differential between 5.4% and 2.6%.

Hats off for very very good work on the sheet as well as the effort behind it. Another tip of the hat for the demeanor you’ve carried while assisting the community over 20+ pages of posts. None of us keep it 100% of the time, but you’re consistently helpful and respectful.

Looking through it now, found a post where you stated you authored the sheet. I saw the rotation information below, the only thing that stood out immediately were some of the cast times were a little short (there’s a slight delay with every swap, ~.2s), will take a little to unravel the rest as it’s a sheet meant to do -everything- from the ground up.

Is the spreadsheet you’re referencing from here:

https://forum-en.gw2archive.eu/forum/professions/thief/Build-The-REAL-maximum-DPS-thief

?

Puandro.3245:

Unleashed i will provide info when im home. Or if you care enough look through my post history and find my post of my spreadsheet.

I do care, I don’t mind digging for information if I know it’s out there (my willingness to put time into this should be obvious).

I’ll look for your sheet.

aeneq.1760:

Yeah limiting yourself to one weapon set is imo a bad thing anyway…

It can be, but the first thing I wanted to establish was highest possible dps and highest possible reliable dps. I want to compare utility weapons now (most of the DPS can be earned from Axe alone, making it more versatile as you can combo it with another axe, a sword, a mace or a shield, if your secondary is a GS, your rotation, timing and trait advantages change).

You have two axe/mace equipped, both with one weapon having a SoB.

If the confusion was on relative complexity (Nike consistently assumes I think it’s hard just to weapon swap), the doc should help highlight the difference.

https://docs.google.com/spreadsheet/ccc?key=0AqCHNWXHft5SdFR6WkR4R2JxSS1xcHlZc1FEa195R0E#gid=7

Look at the AxeGS tabs to see what you need to do to get the most optimal timing, it’s a PITA thanks to imprecise ability times/CD synchs.

aeneq.1760:

Should never use the animation to stopgap the chain, suggest disabling auto attack and use the skill availability instead since these should be locally coordinated and thus lag independent (ofcourse you would need to calculate each chain step separately since the input lag is relevant).

Yes, you could work towards shortcutting an animation if you have an ability that breaks in. That could be why you’re stating 3.2 rather than 3.5. Will look later, but I doubt you can safely go for 3.2 rather than 3.5, possibly could get it down to 3.4 without ever running the risk of clipping that final strike.

I’ve done the work for all in terms of “I’m standing at a dummy and have the following debuffs on boss”. Complexity is easy to measure. How long does it take to find a pattern in what abilities to activate?

Pure Axe/Mace has orders of magnitude less complexity (you will -still- swap on every CD for Sigil of Battle). It’s going to have a higher % of situations where it’s usable.

aeneq.1760:

Also with regards to rotations;

• The axe chain is backloaded which is 3.6s for the full dmg effect
• The 100B is 3.2s long and evenly distributed

Interesting that these are your numbers. I used a stop watch in ‘real time’ and then took recorded video and compared frame times. I used samples of 50 executions or more to come up to an average.

100b’s time from starting to being able to execute the next attack was 3.5, the only attack to be perfectly on the money with its stated time. Axe chain also was 3.5, which seemed longer than they intended.

I wonder how much latency is playing a role there, as it doesn’t see intuitive that you’d have less latency for one ability and more for another than I do. What race are you using (sneaking suspicion this could be a root cause of timing differences). I’m using Charr.

And just as a note, there seems to be a misconception that the goal should always be to shoot for the highest theoretical number possible when looking to clear content, and anyone who isn’t doing that isn’t looking to “min/max” or be leet.

That’s not accurate.

Take the following example: A 150 key sequence results in 100,000 damage. Failing the sequence results in 0 damage. The same class has a 5 key sequence that does 3,000 damage. 10% less damage for the same time period, yet anyone serious about clearing content is going to stick with 5 key sequences because there are normally many other factors to consider than just the attack. You need to move, dodge, buff teammates, possibly revive a teammate if something goes bad. Maybe you need to pick up an add that spawns or be kiting.

The 150key sequence could do 1,000,000 damage and the choice would be the same. Unless you’re standing still punching a stuffed dummy, complexity counts.

This is why a video showing that it is possible to perfectly execute moves exactly as necessary to reach that potential is important. How a rotation performs in the real world is more important than simulation, and ease of execution plays a significant factor in that.

Anecdotes are anecdotes and evidence is evidence. They’re not the same. There are people that have been abducted by aliens, saw Elvis at Walmart and can speak to your dead relatives if you ask them.

I could do a “simple warrior rotation” if that’s what it was, but that’s not what it is. It’s not a rotation. The times just don’t add up. You say it’s about 10% more DPS than pure axe. Excellent, point to the math showing that to be true and I’ll gladly agree.

I set out to get evidence to reach a conclusion. I’ve posted the evidence. I’ve put up spreadsheets that show it’s 2.6%, not 10%, and it also shows it’s not a “rotation” so much as it is slapping abilities on CD which will be out of order for several cycles before they resynch into anything you could call a rotation.

I took the time to do the work and to share the work. You were helpful in highlighting a non-capped vuln rotation that has very very good dps. That was information that was testable, usable and informative. Saying “is about 10% more dps than pure axe”… how did you measure that? Time to death for mob once? 10 times? Felt like 10%? Logged extensively? Laid out the math in a simulation?

As I said before, if someone has a video of the following, I’ll gladly eat crow:

- Full boss encounter where they never strike more than 1 time with Sword Auto
- Full boss encounter where they never accidentally slide through a boss trying to land 4 hits via WW
- Full boss encounter where there is never a weapon swap CD, Vuln Debuff CD like say, crushing blow or Cyclone, unused for more than .1 ms and no more than a total of 10 times are there any delays at all.

The video won’t appear. It won’t appear because that isn’t how the game works and it’s not how people work.

Nike Porphyrogenita.8137:

… where the cool downs mostly line up and do the work for you.

They line up poorly with actual cast times taken into account. The waiting for them to actually line up eats up the % gains from making the attempt.

You misread. It’s not as simple as just swapping weapons on CD. I can do that, easily. It’s getting read of .2 to -.2 ms gaps in cooldowns that are necessary to get the extra % points you gain. There is no “rotation” until you reach several cycles of swaps because the cooldowns (I did the best I could with the actual ability times in those spreadsheets to land on a pattern, full Axe + GS ‘rotation’ would take many cycles to repeat), when counting actual animation time and latency, simply don’t line up well enough. This means rather than becoming very good at having mastered a rotation, like say you would a build order in an RTS, to get those extra % points you are focusing on “which ability is ready and most beneficial?” fairly constantly.

If you’re not doing that, if you’ve settled into a rotation over a small # or even a single cycle, you’re wasting milliseconds to allow that consistency and much of those gains (possibly all, feel free to throw up a repeatable rotation that you use) are lost.

If someone has a video of the following, I’ll gladly eat crow:

- Full boss encounter where they never strike more than 1 time with Sword Auto
- Full boss encounter where they never accidentally slide through a boss trying to land 4 hits via WW
- Full boss encounter where there is never a weapon swap CD, Vuln Debuff CD like say, crushing blow or Cyclone, unused for more than .1 ms and no more than a total of 10 times are there any delays at all.

Done and spreadsheet is updated.. The abilities being used to gain vuln, if you already have 25 vuln, are a DPS loss over a simpler rotation. You can see this with the NV and NVn tabs in the spreadsheet.

https://docs.google.com/spreadsheet/ccc?key=0AqCHNWXHft5SdFR6WkR4R2JxSS1xcHlZc1FEa195R0E#gid=7

GS + Axe/Mace is always the absolute best, whether vuln capped and simplifying rotation or not vuln capped and working to help keep it capped. It’s 2.8% to 3.4% better if your execution and timing are perfect.

Really, with a # that small in difference, I’d rather have the simplified rotation and have more situational awareness as a result, but that’s just me.

Updated. Next I want to look at each without the benefit of vulnerability, should you be in a group that already hits 25 stacks on its own.

I don’t know why, but I’m unable to see this post when logged in anymore. Will start a new post.

Never mind, I’m able to see it after I posted this. Interesting forum bug.

So, final spreadsheet update. GS + Axe is 2.6% better than straight up Axe. 2.6% comes at a tremendous cost in managing the rotation as due to animation times and latency, it takes many cycles to reach a repeated pattern unless you’re moving through it slowly enough where you’re giving up those gains anyway. Axe rotation on the other hand is quite simple. Will post that update in another hour or two, need to get back to dinner.

jul.7602:

I looked at some of these Spread sheets and I’m curious to where you’re getting this information. Do you have any sources?

A while back it was common knowledge that % damage modifiers stack did not multiplicative but additive. Whats the current Effective power formula?

You can’t compare Axe/axe to GS with effective power because they have different coefficients.

I’m using the method of damage multiplier stacking that the game does. This can be confirmed by multiple sources, including crit chart on the wiki. It is additive for each stage as you say.

As for power scaling, that’s already taken into account at the base damage listed. That way, I don’t have to complicate the after effects.

The math is all there, if you see something in error, I’m happy to have it pointed out so I can get it corrected.

I stop watched every ability over 20+ times and have the average time if you’d like it per ability. Could differ slightly depending on your latency (used the display of the bar for the next ability whenever available, did the best I could for attacks without it).

This is taking a few minutes to put together the rotations in the same type of spreadsheet, but -so long- as you do not already have full vulnerability stacks, there is a dps increase for Axe weaving in cyclone even. You shift between 4 to 12 stacks of vuln using that and mace 4.

I’ve done GS rotation and Axe rotation solo. Unless there is some interplay between the vuln stacks of the GS rotation and Axe rotation that will emerge from their being combined, Axe rotation by itself is still going to be your best raw dps.

I’ve uploaded the updated docs spreadsheet to reflect it. I showed 6 stacks where midway through the chain 4 would fall off. It’s not perfect, but until we get actual combat logs some approximation is necessary. If Cyclone strikes more than twice if you’re right up on something, I’ll update it to reflect that (would make Axe even more attractive).

By the way, major kudos to you guys for going through this after a year. I may be retreading old ground here, but my end goal is to have a definitive, detailed set of facts to keep up to date patch to patch.

I need to fix the GS rotation, still using the full auto chain, will do so after I can get back to this, couple of hours from now.

I realize this makes me something of an annoyance, but I’d rather be slightly obstinate than lazy. With the assumption coming down that teams easily fill vulnerability, boons and conditions with their normal abilities, why would you ever switch to a lower dps strike from the highest dps strike?

It seems like the goal with that rotation is: Get a stack of invuln up, hit with your hardest burst, WW (4 strikes is great @ a 2800+ base in just over a second of anim time, but moving like that isn’t always possible nor the right call, like lupi, still, it’s fine to count it, I’ll include it) and adjust.

Axe is where this could be a matter of needing to dig into cyclone. If you have 1 creature and strike it 1 time during the cyclone, you’re doing 688 base damage, just -barely- more than auto attack chop 1, and you’ve delayed getting to chop 2 and chop 3. Mace yeah, I get it, if you need invuln, the 1k+ base and invuln stacks are worth it. But why would you use 5 if your’e not fighting trash? Stripping defiant is normally timed for a purpose on boss fights, better to not risk messing with that.

Going to go do some reading on Axe Cyclone to see what’s up, but will change GS rotation immediately to reflect it. (WW will still not make up for DPS gained from sticking to Axe exclusively unless those modifiers surprise me).

By the way, this isn’t me challenging you any more than this thread was a ‘challenge’ to anyone else. This is my attempt to get this information out into the open. I’ve seen that reddit thread on all the calculations. They involve different stats, but they’re no different from the theorycrafting done for other MMOs and games for many, many years. Everyone understands effective health and effective strikes etc.

It is useful for ballparks on damage, it’s even better at gathering information on defense, but it isn’t the same as breaking down every traits interaction (not the stat increases from the trait points, but the trait abilities/passives).

Puandro.3245:

I see no math on how you arrived to the base damage calculation, all i see is a made up number for base damage and then you added damage multipliers to it. Your rotation is also wrong for GS and Axe.

http://gw2skills.net/editor/?fIAQNAR3ZjcOtvZPKPMxBEzjhIIKC1D7YDbjUByA-jQyAYLBZiQgoCwJvioxW4KiGraBTVuoalLg3gRrGA-e

Base damage calculation is the base damage listed for each ability on gw2skills.net with the gear builds linked. I’m not near a game client to see what they are in game. Either way, they changed (when I compared to the two in game and on the site before) relative to one another. Whether the base is 100 or 641, the increases will be similar in terms of percentage. As Disc does not work off of bleeds, nor does it add any other flat damage amounts but always percentages, the base is meaningless. 100 going to 150 is no different from 641 going to 961.5.

What do you feel is the optimal rotation for either? It’s quite easy to edit. (Calling Axe a rotation is hilarious too, I know, it’s just auto-attack endlessly…).

Puandro.3245:

Pure axe does not do more dps than Axe/GS and has never done so.

I welcome the math on that as well. Spreadsheets are up, tell me where I’ve blown it.

Well, this is where I stumble on the obvious thing that had eluded me before. Looking at the spreadsheet as I put it up there I noticed that much of the issue with the hard logging came from my trying to use the 30/0/0/10/30 at an ability level above my own, timing swaps to 100b and back out perfectly, wasting 100b strikes.

What if I played 30/0/0/10/30 the same way I’d recommend playing 30/30/0/10/0?

Full Axe + 30/0/0/10/30 as it turns out is the absolute highest dps. If you’re a pro at timing swaps, it’s still better DPS than switching to 100b.

Even better, its worse case DPS #‘s are in line with Arms best case while it’s best case numbers are higher than anything else I’ve calculated. I’ve updated the google doc with a tab called “30 Disc Training Wheels”.

https://docs.google.com/spreadsheet/ccc?key=0AqCHNWXHft5SdFR6WkR4R2JxSS1xcHlZc1FEa195R0E#gid=3

Conclusion: Get good at swapping as you can, carry axe/mace and axe/axe or axe/sword, leave GS for fights were the mobility is required.

Nike Porphyrogenita.8137:

I really, really don’t want to play this game, but I’m bored at work so here goes…

No problem, I get it, it’s a lot of work to get down into each attack/trait and lay them all out against one another. If that’s all you have to say, it’s all you have to say.

8 stacks of might is great, which you get so long as you perfectly time every swap (6 from SoB and 2 from trait) for the sigil and trait CDs, which don’t line up nicely with abilities. I’m old and find myself alternating between 4-8 for roughly an average of 6. 6 is still better than force, weaker than night.

(you can get all but 2 of those using double axe/mace with arms anyway if your timing is that awesome).

Anyway, thanks for what time you did put in, spreadsheet is up for others with more time to pour through if they wish.

https://docs.google.com/spreadsheet/ccc?key=0AqCHNWXHft5SdFR6WkR4R2JxSS1xcHlZc1FEa195R0E#gid=0

Fixed a small problem with one of the fields on Disc.

Gathering up some of the data and pulling it together into a simple rotation, Disc shows more gains than Arms against pure trait math when looking at ‘real world’ scenarios for perfect execution on a simple highest output rotation.

A worse case scenario (perfect execution) Disc is better than worse case scenario Arms by ~4%. A best case scenario Disc (perfect execution) is better than best case Arms by 7.3%.

To reach those numbers, you will need to never miss a swap to 100b CD on cooldown, hitting it within the millisecond it needs to be hit, and swapping back to Axe the moment you can back to Axe. You’ll never be able to drop the end of a 100b either to dodge etc (this is also true of Axe chop chain, but as those are less bursty, missing one hurts slightly less).

Issues that can affect things either way:

Bosses with low boons. Weak bleeds overwriting strong bleeds or bleeds already at 25 stacks. 25 vuln stacks without your help. 25 might stacks but not bleed capped on boss (giving you additional dps from bleeds).

As I’m old, I was impacted on tests by swap time to meet CDs for 100b during tests and found myself wasting the tail end of some 100bs. This can have a substantial impact on Disc for the raw logged numbers and accounts for why hands on testing -for me- put pure axe/mace and 30 in Arms ahead.

However, if you’re fast, can hit every one of those swaps and never waste a 100b in the process, 30/0/0/10/30 is the higher damage build with 1 boon. A very tiny amount of slop and you’ll do better if the boss has 2 boons. If takes 4 boons to get over being old like me.

Wethospu.6437:

I’m not sure if you can call yourself Warrior without Fast Hands. :P
Build looks solid to me but Fast Hands is just huge utility.

Yeah, I’d really like to have it, especially for fights were you need mobility or some utility on a second weapon set for an encounter. However, it comes at the cost of at least getting Versatile Rage (meh) and Signet Mastery (nice for when you can run SoR, meh otherwise) or if mobility is really a problem Warrior’s Sprint.

If Fast Hands was at 5, it’d be a no brainer. Empowered Allies just feels like too much to sacrifice to get Fast Hands.

Lobo, I apologized, twice, in the same post for that. My reaction was to his first post being that the efforts I put in were “lazy”, that my desire was to auto-attack with GS and 100b on CD, etc. etc. Let me know what I need to do more than those two apologies to make it seem like I’m not blowing it off and I’m happy to do so.

I didn’t creep through his message history to find anything more than a post showing the math and build description he said was out there. It’s not. He does have a nice spreadsheet to help look at the default boosts you get with each trait point spent and how to weigh them. He also has several posts out there that show his “you’re lazy because you disagree with me” approach is not unique to this thread.

I’m happy to continue the discussion. I’ve apologized where I wronged him and provided details necessary to show exactly how I’ve erred in calc’ing builds. I’ve put down the details of what I’m seeing in two posts now, one for Disc and one for Arms.

Btw, if your weak bleeds can somehow overwrite stronger bleeds, it appears you could knock 2.% off of worse case scenario and 4% off of best case scenario damage IF the person you’re knocking off bleeds from is wearing full rabid, full runes of the undead, malign infusions, has bowls of truffle soup and master tuning crystal active.

30 in Arms Provides:

+15% Critical Chance
+300 Condition Damage
+33% Chance on Bleed (100% Uptime, its duration + # of hits per second, conservatively put 1.3 stacks from this for dps, 2 for raw)
+33% Chance of Vuln (Same as with bleed on uptime, 2 for raw #’s)
+10% Damage on anything bleeding, 100% uptime.

We’ll ignore the last two trait slots. They’re “meh” no matter what weapon you’re using for reliable calcs and get worse the better you get at dodging and timing bursts/adren boosts.

300 Condition damage is 15 dps per bleed stack, bleeds have a base of 42.5. Counting 1.3 for DPS, 75dps pickup. Let’s be conservative and call this 2% dps increase.

Vulnerability is 1% per stack, raw count at 2. 2% dps for you. 2% DPS for 4 party members. Let’s say you’re that good, and each only do 75% of your DPS, that’s 6%. (We’ll exclude this for totals later and add it in conditionally as we do for boons on 30 in disc).

+10% from the obvious trait.

We know the chart is accurate, moving 50% to 65% is worth 10.2% damage increase @ 70% critical damage from zerker gear.

We have in a worse case scenario (barring bleeds being broken), 24.2% damage increase from 30 in arms no matter what weapon you’re carrying. We have a best case scenario of 32.2% if those 2 stacks of vulnerability are useful to you and your party.

In worse case scenarios, you have (Disc) 20.8% vs. (Arms) 22.2% even if bleed is broken. In best case scenarios you have (Disc) 32.8% vs. (Arms) 32.2%.

If it turns out that Heightened Focus is also bugged, Arms would edge ahead in both worse and best case.

Again, please show where the math or assumptions are wrong!

Nike Porphyrogenita.8137:

“I’m more than somebody.”

I apologize, I was reading through very quickly and misread a post. You said you trust yourself more than “somebody”. I let bias bleed through on that read.

I was wrong there, so back to just recommending higher civility in general.

As for the calculations, EP is a good estimator, but I’m looking for in depth breakdowns trait by trait. I just provided it for discipline and look forward to your showing where I erred. (That’s not sarcasm).

Concise, trait by trait breakdown, before focusing on rotation.

Starting stats for the character based on full zerker, eagle runes, accuracy and force sigils for axe/mace and accuracy for GS: +70% Critical Damage, 220% Critical Damage. A hit that would otherwise, without a critical, land for 100, would hit for 220, 50% critical chance, +5% damage w/ axe.

Put 30 points into discipline, grab heightened focus and destruction of the empowered.

You gain:

15% critical chance (until a bug is confirmed to make it 12%, sample sizes for hits just aren’t enough nor is logging accurate enough to confirm, going to assume tooltip is correct as it benefits the 30 in disc anyway).

30% Critical Damage

3% damage increase per boon.

There are two ways to look at the math from resources available. You can go with the wiki critical chart guide, move both sliders along for average damage, and see 1600 vs 1975 and determine it’s a 19% damage increase for the 30% CD and 15% critical chance.

That’s the cleanest. You can arrive at the same numbers doing the math the hard way, as the average damage increase would be:

100 * ( 1 – (.50 + .15)) + 100 * ((.5 + .15) * (.3 + 1.5)) = 197.5

over

100 * ( 1 – (0.5)) + 100 * ((0.5) * (0.7 + 1.5)) = 160

Again, 19% increase if looking at it from a base damage of 100. The chart is right.

To get beyond 20%, you add in the damage peer boon. With 1 boon, 22%, with 2, 25%, 3, 28%, 4, 31%.

Now, take rotation into account. Let’s say you’re ok with micro-ing the CD for 100b on weapon swap and don’t mind the loss in either crit rate or damage from missing sigil on that swap (great if you have 100% 100% crit update anyway, just use force). You’re swapping every 5 seconds gaining a 10 second stack of might you don’t need, but let’s count it anyway. 2 stacks of might is about 1.3% damage increase based on 70 pwr/70 condi.

You are now able to also switch to 100b 20% faster (you only benefit from 2 seconds of the 5 second swap reduction as the CD is @ 8 seconds). Before counting in other group party buffs (they’d affect all other strikes anyway, less complex to keep them all out initially), you gain approximately 1250 dps (counting a perfect 100b before team bufs to reach 15000 damage over 3.5, which it actuall times out to via stop watch) for 3.5 seconds out of every 11.5 seconds. Spread over those 11.5, it’s a 108dps increase to swap to 100b. Given you only get 20% increased 100b uptime, you get 108dps vs. 92.5 dps without the weapons swap speed increase. That’s 17.5 dps, less than 1%.

In an absolute best case scenario (4 constant boons on boss), 30 in discipline is providing an extra 32.8%.

Subtract 3% for every boon below 4.

Arms to follow, need a minute to get some other things done.

Nike Porphyrogenita.8137:

You can insult me personally all you like.

I commented on a lack of civility. If describing your behavior to you is insulting, change the behavior. My intent wasn’t to insult, it was just some advice. In fact, the posts contain compliments. You’re a very good player, of that I’m certain. My recommendation is that you change how you deal with people as you’d go from a great player to a great contributor if that’s what you wish to be. You post often enough it seems to be so. Though quotes like “I’m more than somebody” can lead an old man like myself to wonder how you view yourself.

You provided nothing but a spreadsheet and blanket statements. I’ve put the math behind every trait up in this thread, how it interacts with zerker, how I arrived at average dps gains for each. I’m looking for someone to do the same or point to where it’s already been done.

You may well be right, you and others if they have the math, and that is the point of this thread. I expressed it in the first and nearly every subsequent post. I am looking for someone to explain why I’m wrong. That’s not holding on stubbornly to a conclusion reached out of laziness, that’s saying “I spent all this time and didn’t get the same results… someone mind throwing some maths my way so I can see where I’m wrong before I spend a ton of time taking my mess of information and making a consolidated post with it?”

http://gw2skills.net/editor/?fIAQNAR3ZjcOtvZPKPMxBEzjhIIKiLDqAATFIDA-jQyAYLBZiQgoCwJvioxW4KiGraBTVuoalLg3gRrGA-e

Is the build I’d settled on. The 5 points can go anywhere, but I usually put them into arms for the precision as I wasn’t thrilled about any other 5 points spent elsewhere anyway.

I’d normally go with

http://gw2skills.net/editor/?fIAQNAR3ZjcOtvZPKPMxBEzjhIIKizDqAAJvUByA-jQyAYLBZiQgoCwJvioxW4KiGraBTVuoalLg3gRrGA-e

As a result.

The gear is the gear I’m actually wearing. I’m not 100% certain on the runes, I did switch around between Scholar’s/Eagle etc., the usual suspects, but Eagle’s played in nicely with zerker and high critical chance so kept it for now.

http://gw2skills.net/editor/?fIAQNAR3ZjcOtvZPKPMxBEzjhIIKC1D7YDbjUByA-jQyAYLBZiQgoCwJvioxW4KiGraBTVuoalLg3gRrGA-e

Was the comparison build.

By the way, the bleed overwrites is a worry. It’s one of the more difficult things I had to try to account for when calculating. If you’re with someone that can stack bleeds to 25, and they’re riding on much more condi damage than you, you are obviously actually costing your group damage rather than helping it. Same is true for vuln, if the boss is sitting at 25 vuln at all times, you’re not granting any benefit with that trait and the selected traits in arms are pretty lackluster. That would shave 2-8% off of a 30/25/x/1x/x build’s effectiveness. I didn’t run with engineers or others bringing those bleeds to 25 though anywhere outside of QP, same with vuln. I’d seen 20+, but not 25.

If it turns out that’s the key difference, that the groups people are running with when they recommend this are able to 25 cap vuln, bleed and might without their assistance, then everything I’d done would show it also that 30/0/0/10/30 is the best way to go under most circumstances (talking 2 boons then required, that’s reliable).

My original purpose was not to test builds at all by the way, but to decide which legendary to acquire first, Frostfang or Twilight. The build issues came to light from that.

Testing was done via a mixture of recording/logging boss fight strikes (looking at the frame to determine active boons and reversing them out), striking different dummies/NPCs, and then comparing all the notes against information gathered from available online spreadsheets for things like critical hit (nice chart on the wiki), bleeds, as well as looking at tooltip info for a final spot check to see how different presumed base #’s were against the tooltips, knowing that tooltips could be in error.

Once I had what felt like a reasonable average hit, before taking group boons/debuffs into account, and then looking at what the likely boons/debuffs would be in typical 5 mans, I placed each hit in the rotation on a spreadsheet separately. I applied the changes from traits to each strike. I accounted for animation time (the stated cast times are not accurate, I stop watched each ability).

This gave me a pretty decent idea of DPS available with weapons and trait setups. I did -not- do this with different gear outside of sigils/runes. I could, but my goal was not to run out questioning conventional wisdom of every aspect of the game.

I put the dps at 20 and 60 rather than 15 and 42.5 to account for 1.3 stacks as a rough average.

As for %’s to damage increase, each item was laid out, but I could put down the formula for each as well.

It’s clear that in a particular environment where bosses are commonly wearing more than 3 boons that 30/0/0/10/30 will outpace others. 12% damage is fantastic and it’s hard to overcome that with Heightened Focus sitting there as a reward for those last 10 points. In all other environments, it seems to be very situational. Whenever something becomes situational, maybe I’m just old, but ease of use starts to factor in. I don’t believe there are enough people out there able to weapon swap on CD without missing a beat to catch every 100b CD, never mistiming a 100b so that you lose that final end to the chain because you need to dodge, etc. etc.

So, whenever a different in damage is < 5%-10% different, reliability feels pretty important. How easy is it to get the peak?

Nike, I looked through all the posts that appeared in your post history. One, I see patience and civility have been a difficulty for quite some time. Two, I saw no post highlighting any in depth math or testing. I’m not saying you haven’t done this math or testing, I’m certain you have and even if not, you’ve spent significant time playing very high end content. I’m only saying, I saw no post or information sharing the math.

I played for years in top 50 worldwide guilds in ‘that other game’ that had some of the best theorycrafting, content clearing players I’ve ever gamed with. I’m certain I didn’t deserve to be in their company, but they kept me around likely because I’m quiet, do what I’m told and didn’t get in their way. I had the pleasure though of being with people playing at that caliber that realizing being a prick wasn’t required. Reminds me of a quote from the movie Patch..

“You know, sometimes I forget how young you are Mitch… that you think you have to be a prick to get things done… and that you actually think that that’s a new idea.”

I didn’t belong in that quality group of players because they were just better than I am at the game. I’m old, my reflexes have faded and my time to play has been neutered by life. You wouldn’t have belonged with them either. You’re probably good enough, but they played with the best, and to that end, the best meant skill and realizing that it was just a game and that the internet is not a mask through which you can discard the same politeness and consideration you would have if you were sitting next to the person in the actual world.

Easy to forget, I do (especially when wanting to poke fun at something in the game… like the story writing), but you know…

Back to builds, someone please highlight the math! Curious if someone has a breakdown of the # of boons active per boss in all dungeons?

To be clear, I did look at the spreadsheet. It does not address traits.

Belligerence is a poor substitute for conversation. It was neither a lazy effort (to avoid yanking out measuring sticks that can be as invented as we wish, let’s just say the effort is best measured in hours, not minutes) nor one intended to attack anyone for their previous efforts. In fact, this post is a call out to those who may have done -more- extensive testing as I said early in the thread. Perhaps I missed something non-obvious. Adrenaline you won’t need on faster weapon swaps that provide a single might stack… Three automatic traits appear to be lackluster.

My assumption is that the 30/0/0/10/30 was deemed highest dps by using GS to 100b or WW for free evade and then switch back to Axe anyway, making use of the fast weapon swapping to avoid getting stuck in GS longer than needed. Thus the reduced swap time benefit, a free adren bump if you wanted to evisc just before 100b, and a free mostly useless might stack.

I also am not proposing greatsword. Axe appears to have slightly (~3%) higher raw dps potential when traited.

No, this was not the longer thread, this was my request for more information to make sure I’d tested as thoroughly as I needed to before putting up the post.

I’ll look through your post history to see if I can locate a thread that already goes into this in great detail as it appears even the slightest question you feel you may have answered gives you license to drop whatever civility you may have available.

Are there more active forums elsewhere on the internet for GW2?

Also, saving Eviscerate for Healing Surge on CD offers a DPS increase. Cast time if right on boss for Evisc is ~.8 seconds and another 1.2 for healing surge animation. This hits for 12k-20k depending on setting (5 man vs. QP). This puts it as a very conservative 6k dps up to 10k dps if you use that combination which allows for 100% adren full uptime.

Arms offers:

+18% crit chance
+300 condition damage (adds approximately 20 dps to precise strikes proc)
+33% chance on critical to cause bleeding. At the crit rates you’ll have with 30 in arms and gear, it’s close enough to 100% uptime to call it 100%, you’ll rarely, rarely see no bleed on the target after the first few strikes. This is ~60 dps however, nothing stellar, but as a % of dps the combination is around 2%.
+33% chance for vulnerability, and like bleed, this has 100% uptime. In QP, mostly worthless, but in 5 mans this can be another 2% dps for you and your entire party.
+10% damage thanks to bleed uptime, critical or no
+Fury and Vigor when taking any typical boss hit should you miss a dodge

So Discipline has: +20.8% to +32.8% (if boss has 4 steady boons without much effort) and 12.5% more rage uptime. Arms has +23% damage and free fury on 10% or greater hit, making up for SoR CD allowing for Banner and thanks to invuln also gives the team a 2% damage increase barring your team being already able to maintain 25 stacks rather than 23.

32.8% is certainly better, but if a boss does not have 4 boons, if it has 0, Arms is well ahead. If it has 1 boon, they’re just about even. At 2 boons it’s still even with a slight nod to Arms accounting for RNG. At 3 boons it will depend on group DPS (if you add 2% to 4 party members doing 75% of yours, ‘cause you’re that awesome, Arms is still slightly ahead). At 4 boons Discipline is clearly ahead, as your personal DPS increase outpaces the group DPS impact.

I’m not counting any GS traits as Axe outperforms GS slightly either way.

Videos hitting upleveled/cloth wearing PCs aren’t going to help in this case. It’s not that it isn’t useful information, but the intent behind this question is more for PvE than WvW.

I’ll get into details with the future post, but it involved several hours of beating on the same mob solo with various builds/runes and manually tracking the results. Given damage has a range rather than simple fixed number, some inference from talents was required. In the end, I took overall averages for each strike with the builds and found their full rotation dps.

I worry about posting what I found to be highest here as it will distract from the responses I’m looking for.

Initially, I only set out to make final confirmation on Greatsword vs. Axe/Mace, but builds had to be taken into account also so I tried a less conventional (but possibly more on the nose setup) build and found better numbers.

I’m certain there is a Warrior export that can fill me in on what I missed about the 30 in discipline being best. In benefits, I see:

-15% crit chance
+30% crit damage
+0 to 12% damage depending on boss boon count
+12.5% uptime signet of rage (assuming your group is so pro they don’t want warbanner) thanks to signet recharge 20% faster
+1 reliable might stack if you’re fine weapon swapping on cooldown

Crit damage has a 50% damage increase over base, meaning you gain +30% to 150%, not +30% of 100% or even above 0%.

Example, if a base damage skill hits for 100, a critical hits for 150, and a critical with 30 in disc hits for 180. You gain a flat 30 damage, not 45 damage as 30% of 150 would be. The higher your crit damage, the less as a percent of total damage another % is worth in this way. Considering zerker sets can easily get you to 220% crit damage before traiting, the 30% is really about 13.5% extra total damage.

“With them you get the highest possible DPS out of the Axe and GS”. is a conclusion, I’m curious how people reached it given that some extensive testing didn’t show that to be true. I’m thinking I either missed something non-obvious or previous testing wasn’t exhaustive enough.

Update – Conclusion: 30/0/0/10/30 is your highest dps. A full GS + Axe/Mace rotation is 2.6% to 3.4% better than straight up Axe/Mace, but comes at a tremendous management cost. I will be consolidating all the information into a single post in the near future, but the spreadsheet is located here for those that want to peek at the process used to come to this conclusion.

https://docs.google.com/spreadsheet/ccc?key=0AqCHNWXHft5SdFR6WkR4R2JxSS1xcHlZc1FEa195R0E#gid=7

I’m putting together a longer post after some extensive testing. However, in the meantime, I’d like to get opinions as to why 30/0/0/10/30 and 30/25/0/0/15 are desired over other builds.

Specifically, the automatic traits from discipline appear lackluster and require micromanaging to get the most of. Looking forward to reading the reasons!

Allow us to save our dye setups and reload them as easily as we change titles. I actually though that was what the favorites system did at first and was disappointed when I realized each time I wanted to change looks, even back to a look I’d already found and settled on, I had to redo 15 different little boxes.