Curse

DieInBasra · Jul 03, 2006, 07:45 PM // 19:45

Quote:

Assassins are better for single character spiking,

Hammer chain? I think Dev-Fierce-Crushing-Mighty under frenzy blows the standard assassin combos out of the water. I would make Backbreaker-Crushing-Mighty-Irresistable the example instead, but that takes too long to charge, and is pretty energy intensive. It'll get you the kill, though.

I'd agree that assassins are better at spikeing in splits without monks, though.

TheOneMephisto · Jul 03, 2006, 08:15 PM // 20:15

Quote:

Originally Posted by DieInBasra

Hammer chain? I think Dev-Fierce-Crushing-Mighty under frenzy blows the standard assassin combos out of the water. I would make Backbreaker-Crushing-Mighty-Irresistable the example instead, but that takes too long to charge, and is pretty energy intensive. It'll get you the kill, though.

I'd agree that assassins are better at spikeing in splits without monks, though.

They both do approx the same damage, assassin might actually have a small advantage (no numbers, just by what I've seen). However, it is worth noting the assassin's chain also comes in 6 smaller packets, which makes it better vs. prot spirit.

Another advantage about the assassin which I forgot to mention is their mobility, which makes them even better at splitting and NPC ganking. If you want an examply of this, watch WM vs mn game 1 (I love WM). Watch how an assassin with AoD can transport an entire team around the map with recall. Of course, other professions can do this, but AoD is one of those skills that almost every assassin should have.

Wasteland Squidget · Jul 03, 2006, 08:15 PM // 20:15

Okay, I missed this thread before and there's a lot to mention...

Isil`Zha - To be blunt, the posts you're making have very little to do with how non-RA or AB PvP is actually played, and show a serious misunderstanding of the game. While I understand the desire to defend 'your' class, running rigged tests to try to prove the assassin's superiority over the warrior spreads misinformation and hurts the forum. Even ignoring the obvious flaws in your methods (13 axe? Come on.) there are differences between the assassin and warrior in practice that make the test you ran pretty irrelevant anyway.

-Most warrior builds only have a 2 or 3 skill combo, while a standard assassin build has four. I guess you could compare Sever-Gash-Final to Mantis-Jungle-Twisting, but those aren't the best combos the warrior or assassin class offer.

-Your test was on how long it took you to drop a single dummy of varying AL, but this test heavily favors the assassin because assassin damage is frontloaded while warriors have to build adrenaline. In actual play an assassin can spike first, but with their energy demands they're not necessarily using their chainmuch more often than warriors. While the frontloaded damage of an assassin is not to be ignored, especially in small skirmishes like the ones you sometimes see in GvG, in most PvP fights you're not going to see more attack skill chains out of the assassin than the warrior.

-Blocks/Evades/Blinds will screw up an assassin combo a lot more than a warrior. You dismissed this point earlier, but in almost any form of PvP blocks and evades are common.

-Your DPS through degen conditions isn't going to be terribly relevant to the kill in 8v8 because it won't last out its full duration. Any good team will remove these conditions before they become a death wish, and in an environment like Tombs they're actually detrimental because they greatly increase the heal from Restore Condition. There's a couple reasons why assassins don't see a lot of play in Tombs, and that's one of the larger ones.

Also, please do not double or triple post on these forums. I've merged your three consecutive posts on page 4 into one.

Moving on...

Siren -It's useless to say "my build can do X" if you're not willing to post the build. If you don't want to share your builds that's fine, but don't expect anything but being dismissed if you proclaim their virtues but won't post them. Saying that an assassin can do something without proving it is pointless and clogs up threads like these with unproductive debate.

Others - The biggest problem I see in this thread is that people aren't taking into account the gametype with their posts. Assassins, more than most other classes, have their roles and builds vary with gametype. A TA assassin will look and play very differently from a GvG assassin. If you're going to discuss what an assassin can or can't do, try to put it in the context of a certain gametype. Are you referring to the spike damage they can deal out versus warriors in an 8v8? Their effectiveness at movement control in a 2-man GvG skirmish? Their ability to take down scrub monks in RA or alliance battles? Detailing these things would fix a lot of the arguments in this thread, IMO.

DieInBasra · Jul 03, 2006, 09:17 PM // 21:17

Quote:

They both do approx the same damage, assassin might actually have a small advantage (no numbers, just by what I've seen). However, it is worth noting the assassin's chain also comes in 6 smaller packets, which makes it better vs. prot spirit.

Yeah, both are also quite chainlike and get hurt by blocks, but the assassins even more. Either way, as long as the hammers around, you can't say that assassins are definitly better then warriors when it comes to single target spikeing. I'd settle for equal, maybe. Both have pros and cons.

TheOneMephisto · Jul 03, 2006, 09:23 PM // 21:23

Quote:

Originally Posted by DieInBasra

Yeah, both are also quite chainlike and get hurt by blocks, but the assassins even more. Either way, as long as the hammers around, you can't say that assassins are definitly better then warriors when it comes to single target spikeing. I'd settle for equal, maybe. Both have pros and cons.

Yes. Of course, this thread was mostly talking in comparison with axes, so that's what I based my post off of. I would have to say that hammers and daggers have much different usages/pros/cons, and that they both have their place.

I think that's the main problem with this thread... people aren't realizing that the assassin IS NOT DESIGNED TO REPLACE THE WARRIOR, he is designed to do a very different thing that the warrior. Their roles may overlap, but they both have their individual strengths and weaknesses.

kryshnysh · Jul 03, 2006, 09:39 PM // 21:39

Quote:

Originally Posted by TheOneMephisto

You might want to put the DPS while under frenzyfor the wars, since frenzy should be on every serious warrior's skillbar. Otherwise, nice.

I posted a link to Why Nuking Sucks which contains those numbers. Assassins can similarly buff values using skills and I think a general comparison is more useful than resorting back to skill -> counterskill arguments which I think would do better in a separate thread.

The point of those numbers can be seen in the first paragraph in Why Nuking Sucks. Assassins do comparable damage to warriors as a baseline, so the question comes down to slightly less tangible reasons as to why assassins are not generally considered frontline 7v7/8v8 warrior replacements.

The thread that started the whole "Why Nuking Sucks" will have some insight on answering this question (a lot has also been said here): http://www.guildwarsguru.com/forum/s...=110881&page=5 in particular on page 5 that I have linked to. If you pick just one post, I would say #86 (by Ensign in case others are removed), although I think proper use of Dark Escape should allow an assassin to get out of trouble and overextend pretty heavily, similar to how a frenzying warrior uses sprint/rush to improve defense and get out of trouble.

Isil`Zha · Jul 03, 2006, 10:38 PM // 22:38

Quote:

Originally Posted by Wasteland Squidget

Okay, I missed this thread before and there's a lot to mention...

Isil`Zha - To be blunt, the posts you're making have very little to do with how non-RA or AB PvP is actually played, and show a serious misunderstanding of the game.

And yet, we do very well in there. Most of what I've said comes from how it actually performs in the game. The fact is you are going to have to take on the other team to vie for control of points, assuming both teams know how to play. I did note that we definatly know how to play AB PvP, and it's noted in one of my first posts in the thread, here.

Quote:

While I understand the desire to defend 'your' class, running rigged tests to try to prove the assassin's superiority over the warrior spreads misinformation and hurts the forum. Even ignoring the obvious flaws in your methods (13 axe? Come on.) there are differences between the assassin and warrior in practice that make the test you ran pretty irrelevant anyway.

My defense is against the "Assassin's Suck" mentality. Again, I'm not saying that an Assassin can replace a Warrior, and I'm not trying to downplay Warriors.

Quote:

-Most warrior builds only have a 2 or 3 skill combo, while a standard assassin build has four. I guess you could compare Sever-Gash-Final to Mantis-Jungle-Twisting, but those aren't the best combos the warrior or assassin class offer.

-Your test was on how long it took you to drop a single dummy of varying AL, but this test heavily favors the assassin because assassin damage is frontloaded while warriors have to build adrenaline.

Exactly. Though I don't see how it's biased towards the Assassin. At the start of a battle, the Assassin will have full energy, and the Warrior will have no adrenaline. There's one key difference between Assassin's and Warriors, and why Assassin's aren't "useless."

Quote:

In actual play an assassin can spike first, but with their energy demands they're not necessarily using their chainmuch more often than warriors. While the frontloaded damage of an assassin is not to be ignored, especially in small skirmishes like the ones you sometimes see in GvG, in most PvP fights you're not going to see more attack skill chains out of the assassin than the warrior.

Critical Strikes helps rather well with regaining energy. Basically the opposite of a Warrior in that reguard. The Assassin has to rebuild energy after unleashing a combo, and the Warrior has to build adrenaline before the combo.

Quote:

-Blocks/Evades/Blinds will screw up an assassin combo a lot more than a warrior. You dismissed this point earlier, but in almost any form of PvP blocks and evades are common.

Dismissed only based on the introduction of utility skills/team dynamic adding a virtual unlimited number of variables. I can't have an A/Mo and remove blindness? It could go on forever, so for the test it wasn't considered into it.

Quote:

-Your DPS through degen conditions isn't going to be terribly relevant to the kill in 8v8 because it won't last out its full duration. Any good team will remove these conditions before they become a death wish, and in an environment like Tombs they're actually detrimental because they greatly increase the heal from Restore Condition. There's a couple reasons why assassins don't see a lot of play in Tombs, and that's one of the larger ones.

I never really count on the degen anyway, and it was thrown out in the tests too.

Quote:

Also, please do not double or triple post on these forums. I've merged your three consecutive posts on page 4 into one.

Sorry

Quote:

Others - The biggest problem I see in this thread is that people aren't taking into account the gametype with their posts. Assassins, more than most other classes, have their roles and builds vary with gametype. A TA assassin will look and play very differently from a GvG assassin.

This is also true, and I did state the context of my Assassin build(s) at the beginning. Then others came in saying "that'll never work in GvG like it does in RA" when I've never once stated or even used it in RA. I even specifically stated that I have never used it in RA, and some people continue with the straw man.

Quote:

If you're going to discuss what an assassin can or can't do, try to put it in the context of a certain gametype. Are you referring to the spike damage they can deal out versus warriors in an 8v8? Their effectiveness at movement control in a 2-man GvG skirmish? Their ability to take down scrub monks in RA or alliance battles? Detailing these things would fix a lot of the arguments in this thread, IMO.

I detailed long ago that it was in AB/TA. I also detailed how it's effective in the only place that really matters in the end, in actual practice. My Assassin works very well assassinating things. Nothing like out-flanking the enemy for a surprise AoD, a quick kill of a nuker/monk/mesmer in the back lines, then teleporting back to saftey.

And yeah, there are quite a few people that aren't worth bragging over killing. You can definatly tell if the Monk you're going after is a scrub or not. Same goes for finding scrub Warriors that don't know what they're doing. But I've definatly run into plenty of experienced players and I can still take them on too.

Totally different from a Warrior, but still a useful profession.

I think my point as been made.

Quote:

Originally Posted by DieInBasra

Yeah, both are also quite chainlike and get hurt by blocks, but the assassins even more. Either way, as long as the hammers around, you can't say that assassins are definitly better then warriors when it comes to single target spikeing. I'd settle for equal, maybe. Both have pros and cons.

That's basically the jist of what I was going for - I was never trying to say "OMG ASNs PWN WARRS," just pointing out that they're not a weak profession. Going one on one with a Warrior in melee and showing that an Assassin can win was probably the best way to show their usefullness in straight up combat performance - and this is based on my personal success with it, not that numbers test.

Slayers Xian Long · Jul 05, 2006, 05:17 PM // 17:17

Siren, if Anet had your logic, the Assassin class would be in a big heap of trouble.

Simply put, that has to be one of THE WORST SKILL bars I've ever seen for assassins. No self healing means your split potential is weakened.

No teleportation means you have no mobility advantage over your targets.

Even assuming you're smart enough to take a snare (which I doubt), your hex gives away instantly who you're spiking which ... defeats the purpose of a spike.

Not only that, without a 13 crit strike, your energy management relies entirely upon your hexes and given the recharge time on the skills, you're to be hell of a lot useless against any real team with any sense of how to play.

Assassins are so frustrating because they're pigeonholed to a single elite because AoD by far the most viable option. (I'd say their anti caster is decent but too easy to bypass) Mobius Strike has some potential because it lets them be more versatile with their chains but really means they need to clog up even more of their skillbar with attack skills.

I've tried so many different assassin builds in the past and so far none has outperformed the standard AoD, GPS ...Twisting Chain in 4 man, and AoD, Shock ... Twisting in GVG, AB.

Aug · Jul 06, 2006, 09:23 PM // 21:23

What is the speed of a dagger? I've seen it listed at 1.33 (same as Sword/Axe) and 1.25.

I'd also be curious to see the math done to get at the values for your average sustained damage by a Warrior and Assassin. When I plug the numbers into my Excel spreadsheet, I'm close, but not exactly at the same #s that Krysh and Ensign are at.

kryshnysh · Jul 06, 2006, 10:12 PM // 22:12

Quote:

Originally Posted by Aug

What is the speed of a dagger? I've seen it listed at 1.33 (same as Sword/Axe) and 1.25.

I'd also be curious to see the math done to get at the values for your average sustained damage by a Warrior and Assassin. When I plug the numbers into my Excel spreadsheet, I'm close, but not exactly at the same #s that Krysh and Ensign are at.

1.33. See guildwiki among other places.

I'd be curious to see how you did your numbers then. Mine match Ensign's on the Axe, Sword, and Hammer. I know I was slightly off on the dagger numbers I reported earlier, and I've gone back and fixed that. Given that my numbers match very closesly to those reported elsewhere, I feel they're close enough to give the right general feel.

Here is the strict formula:
((7+17)/2*(1-0.37)+17*0.37*SQRT(2))*2^(8/40)*1.2*1.15*1.34

Non-critical hits:
(7+17)/2 is the average damage
1-0.37 is the chance of a non-critical (this is with 16 DM, and 13 Critical Strikes)

Critical Hits:
17 is maximum damage
0.37 is the chance of a critical hit
sqrt(2) is the multiplier for a critical hit (beyond just being maximum damage)

2^(8/40) is the multiplier for being 16 DM (against armor level 60)
1.2 is the multiplier for 20% customized
1.15 is the multipler for 15% (15>50, 15/-5, etc.)
1.34 is the multiplier for double strikes (2 + 32% chance you will do this again for each hit you do + 100% chance to do this on the primary hit obviously)

I then multipled by 60/1.33 for the DPM and 1/1.33 for the DPS. This is all obviously averaged data as double strikes can cause a rather large increase in damage on any specific hit, but averaged out it is 0.32 of a strike/regular strike that lands (at 16 DM).

I initially had the double strike multipler entered in wrong, it should be 1.34, I had in 1.32, so the DPM for 16 DM, 13 Critical Strikes is:
1572.9 (26.2 DPS) (changed in the original post as well).

References:
http://gw.gamewikis.org/wiki/Dagger
http://gw.gamewikis.org/wiki/Dagger_Mastery
http://gw.gamewikis.org/wiki/Critical_Strikes
Double Strikes Stack at team-iq.net
Ensign Explains a few points - he used 24% vs. 22.9% criticals at 16 weapon mastery

Please note I also used the formulas rather than the estimates for damage for whenever I could find a formula.

Aug · Jul 07, 2006, 02:20 PM // 14:20

I was confused about dagger speed because guildwiki has 1.25 on http://gw.gamewikis.org/wiki/Double_strike.

Quote:

((7+17)/2*(1-0.37)+ (17*0.37*

Mine's identical to this point.

Quote:

sqrt(2) is the multiplier for a critical hit (beyond just being maximum damage)

I don't believe that is what guildwarswiki states on http://gw.gamewikis.org/wiki/Critical_hit:

Quote:

Whenever a critical hit occurs, the damage done is equal to the maximum damage of the weapon multiplied by approximately √2, the equivalent of striking for maximum damage with a weapon attribute 4 higher than actually possessed

I read that as: SQRT(2) == striking with weapon attribute 4 higher

SQRT(2) is 1.41. 1.41 is also the same thing as a crit with a weapon skill of 12 using the MAX_DMG * 2^((5 * 12 - 40)/40) equation here: http://www.guildwarsguru.com/content...nics-id674.php. I believe we should be using MAX_DMG * 2 ^ ((2 * 20 - 16)/40), which yields 25.77 damage per crit.

I don't think they meant you do a max hit multiplied SQRT(2), and then calculate damage as if you had a 16 weapon skill yet.

Quote:

1.34 is the multiplier for double strikes (2 + 32% chance you will do this again for each hit you do + 100% chance to do this on the primary hit obviously)

Why are you adding 2 to 32? Guildwarswiki states you have 0% chance to double strike at 0 Dagger Mastery. Edit - I see... a poster at team-iq.net indicates you have a 2% base rate.

Using my formula:

Code:

=(((7 + 17)/2 * 2^((2*16-24)/40) * (1-0.37)) + (17 * 0.37 * (2^((2*20-16)/40)))) * 1.2 * 1.15* 1.32 / 1.33

I get Dagger avg DPS of 24.95

I get an average axe of 29.06

However, guildwarswiki says 23.3% is the correct crit rate, using the .0144x + .0027 formula, which gives me a dagger dps of 24.84 and an axe dps of 28.89.

I've made my Excel spreadsheet accessible here: http://people.msoe.edu/~ormonda/GW/wpn%20calcs.xls

kryshnysh · Jul 07, 2006, 05:51 PM // 17:51

Quote:

I was confused about dagger speed because guildwiki has 1.25 on http://gw.gamewikis.org/wiki/Double_strike.

Thanks. I've corrected that one. Some errors slip through, but thats the first reliable source I've seen that used 1.25 and it disagrees with every other reliable source.

My numbers were twisted to conform to Ensign's. I stated specifically that I used 0.24 as the critical hit chance in defiance of guildwiki to make that so. Considering that guildwiki cites Ensign for that table, and Ensign used 0.24 more recently than he posted the numbers that went into that table, I'm going to just use 0.24 for now. Why not try to be more precise? Because we really don't have a good measure of the statistics to really know how good it should be. In fact its so off that every confidence interval I've tested so far (or even seen) has at least 25.5% as the upper limit based on the data used for the table at guildwiki (90%, 95%, etc. for binomials, the one Ensign used, etc.).

If you read the Dagger Mastery page on guildwiki, it states daggers have a 2% flat chance of a double strike that stacks with the dagger mastery chance (even if you don't have the dagger mastery attribute you can still get a double strike). Since this agrees with the post I referenced at team-iq.net (which I hadn't read before my initial calculations), I recently changed to add that in.

Quote:

I read that as: SQRT(2) == striking with weapon attribute 4 higher

SQRT(2) is 1.41. 1.41 is also the same thing as a crit with a weapon skill of 12 using the MAX_DMG * 2^((5 * 12 - 40)/40) equation here: http://www.guildwarsguru.com/content...nics-id674.php. I believe we should be using MAX_DMG * 2 ^ ((2 * 20 - 16)/40), which yields 25.77 damage per crit.

I don't think they meant you do a max hit multiplied SQRT(2), and then calculate damage as if you had a 16 weapon skill yet.

I think what you are saying is to use the diminishing returns portion above 12 of weapon mastery with a critical hit. The part in bold above is a simplification that only works for attributes below 8, above that it is wrong. 2^((5*12-40)/40) * sqrt(2) = 2^(8/40) * 2^(1/2) = 2^(28/40) which is not equal to 2^(24/40) = 2^((2*20-16)/40).

I believe the sqrt(2) is the more accurate use here. It conforms to numbers Ensign has published, and almost all of the research and articles we are citing here was performed and written by Ensign.

In addition, guildwiki's article on Damage states that the effective bonus is sqrt(2) (DLBonus of 20 is +20 in the numerator of the exponent for the multiplicative factor).

I'm not going to make my excel sheet available here. I trust my own excel sheet, but I don't feel like promoting poor computer security by having as man people as guru offers the ability to download it. If specific people want it, I'll send the pertinent parts to them specifically (its a rather large set of worksheets dealing with a variety of game mechanics).

Ensign · Jul 08, 2006, 07:01 AM // 07:01

No one really knows what the correct critical hit formula is. I did some tests about a year ago that gave me a ballpark figure of about 1.5% per attribute level, at least over the range that I'm interested in. Since then Kuntz, using a tweaked version of GSTATS, ran a whole lot of empirical tests that are the origin of that 23.3% number (I believe). I've been using 24% (and 18% at 12) out of habit, the 23.3% figure could be used without changing the results significantly.

Talking with Izzy about it, the 'correct' critical hit formula is 1% + 1% per attribute level. That the 'correct' formula has failed to describe critical hit behavior from every empirical test I know of is the big mystery of Guild Wars mechanics.

Critical hits are always +20 in the exponent. I write something differently in the text about +4 levels, but that wasn't strictly true and made the whole bit ambiguous. If I ever revise the guide I'll clean that up. I'll probably update if critical hits ever make sense.

The advantage of assassins is that their attacks do not need to be charged which makes them better in a quick duel - get in, combo, get out. Once a warrior has his adrenaline charged he's going to unleash more dangerous combos, no question.

Peace,
-CxE

Elaine Donnerbalken · Jul 08, 2006, 06:14 PM // 18:14

Maybe Izzy was refering to the Assassins attribute Critical Strikes?

And now an (only theoretical) Assassin Skill Combo for spiking:

Critical Strikes 16
Dagger 13

Palm Strike : 90 damage
Twisting Fangs: (2* 23 *3/4 [Flurry] ) +42 + Deep Wound
Flurry (skilldamage is unaffected from the damagereduction)

= ~166 (132 armor ignoring) damage +Deep Wound, better than the Evis/Exe Combo.

But of course, if you'd take this Assassin only for spiking, you would sacrifice the other benefits of warriors, and I don't think that 16 points more damage are worth that.

(Well, you could use it for spiking heavy armored targets, but since im not in GvG I have no idea if that would be practical.)

Aug · Jul 10, 2006, 02:18 PM // 14:18

I ran a simple test using an Assassin with 16 DM with a customized weapon, but not a +15% damage modifier, and got consistent crits of 33 vs the 60 AL dummy. This would suggest the formula for calculating crit dmg is:

17 * SQRT(2) * 2^(((16+4)-24)/40)) * 1.2 = 33.14

I've updated my spreadsheet to reflect that for crits, and I now get an average sustained DPS of 28.22 DPS for 16 Dagger + 13 Crit Strikes, and 28.17 for 16 Axe.

Strength would put the Axe Warrior again in the lead, but the difference is barely noticeable.

kryshnysh · Jul 10, 2006, 03:43 PM // 15:43

Quote:

Originally Posted by Aug

I ran a simple test using an Assassin with 16 DM with a customized weapon, but not a +15% damage modifier, and got consistent crits of 33 vs the 60 AL dummy. This would suggest the formula for calculating crit dmg is:

17 * SQRT(2) * 2^(((16+4)-24)/40)) * 1.2 = 33.14

I've updated my spreadsheet to reflect that for crits, and I now get an average sustained DPS of 28.22 DPS for 16 Dagger + 13 Crit Strikes, and 28.17 for 16 Axe.

Strength would put the Axe Warrior again in the lead, but the difference is barely noticeable.

That formula is off, plugging it in I get 26.92 as the damage from a critical. 17 * sqrt(2) * 2^((2*16-24)/40) * 1.2 = 33.14.

Please note that strength only triggers upon skill usage.

The game does not round, for attack speeds you should be using 4/3 (see a post of Ensign's among my references earlier).

Your numbers seem a bit high. I haven't changed my numbers (except for a typo

) and don't match the same numbers as you (yet I do match with Ensign). A decrease in critical hit chance should decrease the dpm/dps and thats the only thing I can see you should be doing differently at this point. How are yours so high?

Aug · Jul 10, 2006, 06:00 PM // 18:00

I shouldn't be adding the +4. It should just be:

=17 * SQRT(2) * 2^((2*16-24)/40) * 1.2

Now I get 26.22 now if I use .24 base crit rate.

I'm pretty sure I have the wrong equation for crits of weapon skills <=12. No way is it

MAX_DMG * SQRT(2) * 2^((5*WPN_SKILL-40)/40)

kryshnysh · Jul 10, 2006, 06:10 PM // 18:10

Quote:

Originally Posted by Aug

I shouldn't be adding the +4. It should just be:

=17 * SQRT(2) * 2^((2*16-24)/40) * 1.2

Now I get 26.22 now if I use .24 base crit rate.

Sounds like we are now in agreement in how we achieve our numbers, with varying critical hit values giving varying numbers obviously, but all negligibly so.