Guild Wars 2

Any chance this trait will ever see a buff? At 15% chance, I hardly ever notice this trait in action. I’d be happy with a reduction on the bleed duration if the percent chance was significantly buffed, so it would be more reliable and feel a bit less random.

If the chance get buffed, I’m afraid of anet giving it an ICD. This trait is really strong with grenades, because you are throwing 3 at once and so you have a bigger chance to trigger it. That’s actually the problem, this trait is balanced around grenades and so all other explosions, which could trigger it, can’t be efficient with this trait, because they just hit one time. Like mines, bombs, all turret explosions, etc.

Actually, it would be nice if anet would test with the numbers, increase the trigger chance and then slap an icd on it, which would lead to the same amount of bleeds with grenades like before, but still giving significantly more bleeds with other explosions.

Yeah. I mean, overall I love the trait, but at the moment it is so random for me. It either works incredible, or is hardly noticeable.

I think the reason you hardly notice it is because it doesnt stack high, but is quite long in duration.
There are some ways to achieve maximum affect out of it tho. One thing quite some people dont know is that is has a chance on proccing with fire bomb. If you place a fire bomb and then start throwing grenades in the bomb, you will see the following on average:
In a 4s time:
You get 4 ticks from fire bomb, can throw a grenade barrage once, and three times other grenades. This means you hit your enemy 21 times with explosives. This means you will avg proc shrapnel 3,15 times.
Now three stacks of bleeding for all that trouble seems like nothing. However with a base duration of 12s ( 15 with firearms) it does count up. With 1200 condition damage we get:
(1200×0.06+22)×3 = 4230 extra damage.
Now do keep in mind: this extra damage does not immediatly come on top of your dps in that 4s window, but you can continue stacking it.
Conclusion, it is a good dps booster. However when it comes to PvP only reliable with grenades and not bombs.

Shaogin.2679:

Any chance this trait will ever see a buff? At 15% chance, I hardly ever notice this trait in action. I’d be happy with a reduction on the bleed duration if the percent chance was significantly buffed, so it would be more reliable and feel a bit less random.

To be numerically simple 15% is 1/7. If you through out 3 grenades at a time you can simply take say (1/7+1/7+1/7) or (15%15%15%) [rounded for ease] because they are “or” events in probability and it doesn’t matter which grenade triggers the bleed and we are not looking for the odds of more than one bleed or all three triggering bleeding (1/343 by the way) if that is possible. Basically just auto-attacking you have an 45% chance to cause bleeding with a single grenade volley.

This is relatively high considering it is non-conditional. To be blunt about 1/4 grenades will trigger a 12 ~ 24s bleed with condition duration. That’s pretty good.

DGraves.3720:

Shaogin.2679:

Any chance this trait will ever see a buff? At 15% chance, I hardly ever notice this trait in action. I’d be happy with a reduction on the bleed duration if the percent chance was significantly buffed, so it would be more reliable and feel a bit less random.

To be numerically simple 15% is 1/7. If you through out 3 grenades at a time you can simply take say (1/7+1/7+1/7) or (15%15%15%) [rounded for ease] because they are “or” events in probability and it doesn’t matter which grenade triggers the bleed and we are not looking for the odds of more than one bleed or all three triggering bleeding (1/343 by the way) if that is possible. Basically just auto-attacking you have an 45% chance to cause bleeding with a single grenade volley.

This is relatively high considering it is non-conditional. To be blunt about 1/4 grenades will trigger a 12 ~ 24s bleed with condition duration. That’s pretty good.

The effectiveness is limited to grenades though, and results tend to be completely random. Sometimes it works great, but other times I’ve gone through quite a few cycles of grenades with no procs. I’m just one of those people that isn’t a fan of rng affecting combat too much. Personally I’d prefer a higher chance with an icd.

Shaogin.2679:

DGraves.3720:

Shaogin.2679:

Any chance this trait will ever see a buff? At 15% chance, I hardly ever notice this trait in action. I’d be happy with a reduction on the bleed duration if the percent chance was significantly buffed, so it would be more reliable and feel a bit less random.

To be numerically simple 15% is 1/7. If you through out 3 grenades at a time you can simply take say (1/7+1/7+1/7) or (15%15%15%) [rounded for ease] because they are “or” events in probability and it doesn’t matter which grenade triggers the bleed and we are not looking for the odds of more than one bleed or all three triggering bleeding (1/343 by the way) if that is possible. Basically just auto-attacking you have an 45% chance to cause bleeding with a single grenade volley.

This is relatively high considering it is non-conditional. To be blunt about 1/4 grenades will trigger a 12 ~ 24s bleed with condition duration. That’s pretty good.

The effectiveness is limited to grenades though, and results tend to be completely random. Sometimes it works great, but other times I’ve gone through quite a few cycles of grenades with no procs. I’m just one of those people that isn’t a fan of rng affecting combat too much. Personally I’d prefer a higher chance with an icd.

Well that’s actually what makes Shrapnel so great; it is a “bonus bleed” rather than a “dedicated bleed” so you could proc it all the way through and get 3 + 3 + 1 (sigil of earth) + 1 (if crit) if you were super lucky on a shrapnel grenade throw! Woohoo!

But really, no one should use Shrapnel with the intent to apply lasting bleeds. It is specifically designed for grenade damage. Yeah, you can proc it with mortar but with bomb there are better options like the Big Ol’ or you can take Siege Rounds. In reality shrapnel is not for condition damage builds; it adds condition damage pretty regularly if I may say so even with worst case scenario math [(1-X)^Y where X is the probability of the event (.15) and Y is the total number of grenades over that time] but yeah it is inconsistent on purpose and is working as intended.

I would hate for Shrapnel to get an ICD though. That would ruin it. You may as well just take a pistol because it attacks faster if you really want to stack bleed. There’s no meaningful difference between conditions based solely on their longevity.

@DGraves
In my opinion your whole comment makes little sense … :/

You say “it’s great because it’s a bonus bleed and not a dedicated bleed” wich is basicly the same in the long term. So it’s neither good or bad. Just empty words.

You say “no one should use Shrapnel with the intent to apply lasting bleeds, it’s designed for nade dmg”. Uhhh well and how? I guess it increases the nade damage via bleeds, does it not? Even bomb 2 causes multiple procs and it is after all a (if not THE) main source of condi dmg in condi builds.

You say an “ICD would ruin it” wich adapted in the right way would lead to the very same results in the long term. I agree it’d be a nerf for nades, a buff for everything else, but if you take the overall average it doesn’t matter at all. Noone changes their rotation just for shrapnel.

You also say that you should use pistol aa for bleed and not nades with shrapnel with wich I can not agree with, especially with the Sharpshooter trait procing 3 times on nades. Nade aa’s do 50% more bleed per sec than pistol aa’s. Or are you aa-ing with pistol against red Vale Guardian (wich makes sense for the last few % tough, agreed xD)?

Greez!
- Ziggy

Xyonon.3987:

@DGraves
In my opinion your whole comment makes little sense … :/

You say “it’s great because it’s a bonus bleed and not a dedicated bleed” wich is basicly the same in the long term. So it’s neither good or bad. Just empty words.

You say “no one should use Shrapnel with the intent to apply lasting bleeds, it’s designed for nade dmg”. Uhhh well and how? I guess it increases the nade damage via bleeds, does it not? Even bomb 2 causes multiple procs and it is after all a (if not THE) main source of condi dmg in condi builds.

You say an “ICD would ruin it” wich adapted in the right way would lead to the very same results in the long term. I agree it’d be a nerf for nades, a buff for everything else, but if you take the overall average it doesn’t matter at all. Noone changes their rotation just for shrapnel.

You also say that you should use pistol aa for bleed and not nades with shrapnel with wich I can not agree with, especially with the Sharpshooter trait procing 3 times on nades. Nade aa’s do 50% more bleed per sec than pistol aa’s. Or are you aa-ing with pistol against red Vale Guardian (wich makes sense for the last few % tough, agreed xD)?

Greez!
- Ziggy

I’ll explain.

The trait itself exists for those who don’t use the other two skills. It is equivalent to a placeholder, it is the worst of the three traits in actuality, and it really shouldn’t be used. People use it because it’s BiS for them ( a bonus bleed ) but in reality it isn’t the best of any given skill. The fire bomb is not consistent enough (cooldown) to even begin to talk about the value of shrapnel so that’s out, it might proc once or twice consistently during that pulsing. Realistically it is very much so grenade based and basically rewards grenadiers for strong consistent pressure.

Using shrapnel with the intent to create lasting bleeds is just not wise. It gives consistent pressure builds a slight hand up but otherwise it isn’t build to be a truly reliable source of damage. Going off this vein ICD would ruin it for A ) the pulses in the firebomb which negates your own point, B ) explosions in general because you are slowing the rate in which it procs which slows the rate for all explosions and terminates the chance of a greater the normal outcome rendering it consistent but never actually granting meaningful damage.

Shrapnel is a “reward” for pressure not a “strategic maneuver” for condi; you gamble on your condition damage increasing by X over Y time and use pressure and attack rate.

As for this:

“You also say that you should use pistol aa for bleed and not nades with shrapnel with wich I can not agree with, especially with the Sharpshooter trait procing 3 times on nades.”

The amusing part of this is that you attack Shrapnel but are fine with Sharpshooter which has a 33% chance to proc. You are aware that there is a 1/81 (3^3) chance that all three will go off right? It’s incredibly rare and it is conditional (requires a crit) so if you don’t have 100 crit rate it isn’t even 33%. Basically it’s the same problem.

By the way how much damage in one second does a 100 pt. 6s bleed do versus a 100 pt. 12s bleed?

Trick question, it’s 100. It really doesn’t matter. I knew this error would come up too. Whether it’s 2s or 200s a 100dps bleed does 100 damage a second.

when 1111ing with grenades, you can expect around .5 bleeds per throw. since youre taking shrapnel, its safe to assume a condi build, and of course any good pve condi build will obviously be maxing out bleed duration, so you get 24 sec bleeds out of shrapnel. since youre 1111ing in this obviously unrealistic scenario, at 1 throw per sec, you should be getting about 12 bleeds out of shrapnel.

again, thats grossly high. because obviously you need to be dancing in and out of nades in a good pve condi build. so how to account for that hmm… ah i know. since the cd of shrapnel nades is 5 sec and your goal is to use it on cd and each cast takes 1 sec, you should be casting nades a minimum of 20% of the time. but of course, that doesnt account for fire/conc bombs and the occasional mortar 2 and nades 5 and nade barrage. im not gonna be super detailed because im not really interested in being exact here, but overall ill ballpark it at 33% of the max efficiency of this trait (speaking of which, casting firebomb is more efficient for shrapnel by 1 hit, but it has a higher cd).

so you can expect about 4 bleeds from it after wailing on something for a while. 4 bleeds is from 500-700 dps depending on your buffs and such.

compare that to siege rounds. at 100% crit (unrealistic for condi), thats about a 3k crit (buffed) every 40 seconds… 75 extra dps. additionally, your mortar poison goes for 2 more ticks, which is ~1000 damage every 12 secs (but you dont have enough slack in your rotation to cast that on cd)… a max of around 100 dps, being generous.

so in total, siege rounds offers ~175 dps, being generous.

shrapnel offers ~500 dps, being conservative.

evasive powder keg isnt good.

all 3 are free dps skills in that they dont modify cast times and everything that they affect are things you should be casting anyways.

sooooo

feel free to make your own judgment. none of these make or break a build. usually people select with a mind to maximize dps in pve, for reference.

The problem when comparing Shrapnel to anything is it’s random nature. You can’t actually average against it over the course of X minutes because averages require huge amounts of time and iterations to play out. To be frank most battles are over before 300 samples are even taken ( that is, 300 "111"s ) since large enemies are rarely soloed and small enemies just don’t have the HP.

Actually that’s the major problem with any comparison regarding random events. Averaging them, because of the nature of averaging, creates unrealistic conditions. Furthermore when talking about Siege Rounds it hits twice, so I’m just going to presume that your number is for two hits in your favor but otherwise you halved your dps calculation.

Then you have the question of what type of damage fits the situation. Generally speaking regardless of DPS in a pragmatic setting you actually don’t care how long a bleed is. To explain if you have a regular orrian enemy are you going to go headstrong in there with your 20s bleeds at 50 dmg per tick or just blow the thing up for 6k damage right away. Another major problem is that generally speaking you can’t equate the two tactics anyway; fighting a highly evasive enemy or an enemy that keeps you on the move may be easier with long-term conditions while fighting multiple grunts is more effective with burst. Even if one has a higher calculated DPS (which happens all the time) it doesn’t necessary have the most pragmatic solution.

I think that the biggest problem in most games is people play to maximize numbers which is not necessarily effectiveness. They ignore so many conditions and realities and tack on many other elements that are just outright unrealistic. To be blunt sometimes dealing 3k ASAP is worth more than dealing 6k over 24s. It requires more than just staring at a bunch of end values.

But the good news is that at least everyone understands that the grandmaster traits for the explosives line is horrible. There is no best. Partially because they are all sort of terrible.

no, averaging something literally tells you what to expect. it is not unrealistic. stop trying to debunk a claim by saying averages are unrealistic. it makes you look really silly.

furthermore, what i am talking about with regard to siege rounds is how much dps the trait adds to whats already there. it adds 1 hit on orbital, and it adds 2 ticks to the poison field. it does not add anything else. and since were discussing condi builds here (because the dps increase from shrapnel is pretty laughable in a power build, especially in the unbuffed state you espouse), i went and smacked a pocket raptor with orbital strike on my full ascended viper/sinister engi and saw a 2k crit, which is laughable, so i multiplied that number by 1.5 to simulate full buffs without spending more than 10 secs thinking about it.

it is not my problem if you cant understand how estimating works. it is not my problem if you cant understand that in pve, 6k damage is far better to deal than 3k in a 5-10 minute fight such as any high fractal or any raid, and i will kick you for being an obstinate pug when you object.

and actually, while these grandmasters arent playstyle defining in pve the way anet stated they intended for them to generally be, they are pretty good and definitely not horrible. perhaps you remember how people complained how dominating grenadier used to be when it was a trait that gave more than 15% total dps to any build that took it, since you remember how much power a full stack of might used to give.

insanemaniac.2456:

no, averaging something literally tells you what to expect. it is not unrealistic. stop trying to debunk a claim by saying averages are unrealistic. it makes you look really silly.

The problem is that people don’t understand what averages actually do in models. Averages tell you what to expect under specific conditions presuming no changes but this doesn’t account for any of changes that do actually occur. There is no average “my character is dead again” time in the mix because no one accounts for their character going down. There is no average “my boons are depleted because key source alpha died or was interrupted” because no one mixes that in there. There is no average “things are just going balls to the walls wrong why is this happening!?” or “I am new to this content and don’t know what I am doing and it turns out neither does that guy.” either.

Averages are not the magic and people make them out to be. You can model many things but the less specific your model the more abstract the values. Averages actually are really not the best way to go for short-term engagements and this game is filled with generally short-term engagements. Math is only as good as those who know how to use it; like any tool it can be made to state the obscene because people can simply misuse it to prove a point. It is no different than using any other form of statistics to create havoc in the world.

furthermore, what i am talking about with regard to siege rounds is how much dps the trait adds to whats already there. it adds 1 hit on orbital, and it adds 2 ticks to the poison field. it does not add anything else. and since were discussing condi builds here (because the dps increase from shrapnel is pretty laughable in a power build, especially in the unbuffed state you espouse), i went and smacked a pocket raptor with orbital strike on my full ascended viper/sinister engi and saw a 2k crit, which is laughable, so i multiplied that number by 1.5 to simulate full buffs without spending more than 10 secs thinking about it.

I’m just going to say “okay”.

it is not my problem if you cant understand how estimating works.

You’d be surprised … But I have no interest in trying to claim what you do and do not know.

it is not my problem if you cant understand that in pve, 6k damage is far better to deal than 3k in a 5-10 minute fight such as any high fractal or any raid, and i will kick you for being an obstinate pug when you object.

The reality however is you tend to be dealing with very small factor differences, where the question is whether 5.5k vs. 6k is really a big deal. It takes a lot to halve an appropriately built class even with just a small or moderate rotation. It’s rarely ever that bad if a player is just trying.

and actually, while these grandmasters arent playstyle defining in pve the way anet stated they intended for them to generally be, they are pretty good and definitely not horrible. perhaps you remember how people complained how dominating grenadier used to be when it was a trait that gave more than 15% total dps to any build that took it, since you remember how much power a full stack of might used to give.

Well what makes them pretty good? I can think of plenty of things that make them pretty bad and definitely not worthy of being Grandmaster at all.

DGraves.3720:

I’m just going to say “okay”.

this is why your walls of text need to be ignored. :/

insanemaniac.2456:

DGraves.3720:

I’m just going to say “okay”.

this is why your walls of text need to be ignored. :/

I agree. We shouldn’t speak.

DGraves.3720:

insanemaniac.2456:

DGraves.3720:

I’m just going to say “okay”.

this is why your walls of text need to be ignored. :/

I agree. We shouldn’t speak.

im not the one unconditionally refuting numbers, math, and logic just because i can. you are. in about 4 threads.

insanemaniac.2456:

DGraves.3720:

insanemaniac.2456:

DGraves.3720:

I’m just going to say “okay”.

this is why your walls of text need to be ignored. :/

I agree. We shouldn’t speak.

im not the one unconditionally refuting numbers, math, and logic just because i can. you are. in about 4 threads.

Actually I am not refuting the data I am questioning the method of modeling. I don’t think it is accurate. My reasoning is that as a player I watch plenty of people play these cookie cutter builds and not do anywhere near what is professed.

It is not a matter of whether the collected data is valid but of whether the projections are accurate. A good way to explain this is to the loot chance of an item from a specific chest; after 10 tries you get data but would you average off of it? No. After 100? No. After 1,000? It’s not terribly advisable. After 10,000? You may have an idea. After 100,000? It’s getting closer. And you can add more and more zeros for more and more accurate results but the reality is that the data is taken into consideration from a very specific set of scenarios which may have implicit elements; for instance you have questions like “Do teams do better if they are using teamspeak to communicate?” and “Are they friends or is it a PUG?”

Blah blah economics.

DGraves.3720:

insanemaniac.2456:

DGraves.3720:

insanemaniac.2456:

DGraves.3720:

I’m just going to say “okay”.

this is why your walls of text need to be ignored. :/

I agree. We shouldn’t speak.

im not the one unconditionally refuting numbers, math, and logic just because i can. you are. in about 4 threads.

Actually I am not refuting the data I am questioning the method of modeling. I don’t think it is accurate. My reasoning is that as a player I watch plenty of people play these cookie cutter builds and not do anywhere near what is professed.

It is not a matter of whether the collected data is valid but of whether the projections are accurate. A good way to explain this is to the loot chance of an item from a specific chest; after 10 tries you get data but would you average off of it? No. After 100? No. After 1,000? It’s not terribly advisable. After 10,000? You may have an idea. After 100,000? It’s getting closer. And you can add more and more zeros for more and more accurate results but the reality is that the data is taken into consideration from a very specific set of scenarios which may have implicit elements; for instance you have questions like “Do teams do better if they are using teamspeak to communicate?” and “Are they friends or is it a PUG?”

Blah blah economics.

its ok to be skeptical, but you take it to a whole new level of ridiculous.