[Sung Entune]

Originally Posted by zenharmony

Alright here we go. I'm using variable names here for clarity:

DPS1 = The bard's dps without haste song.
Damage1 = The bard's damage flat damage output per hit of her dagger.
Time1 = The time it takes the bard to poke a bear with her dagger without haste song.

DPS2 = The bard's dps with haste song.
Time2 = The time it takes the bard to poke a bear with her dagger with haste song.

So in our example, the bard has a Damage1 of say 100, using a weapon that hits every 2 seconds. This gives her a DPS of 100/2 = 50.

Now with haste, the bard is still doing Damage1 per hit (100). But the weapon hits faster. At 7% haste, the weapon hits every 1.86 seconds. This gives her a DPS of 100/1.86 = 53.76.

53.76 is 107.5% of the original 50 dps.


How's that?

That's what I thought you where doing.


First, I have yet to see an adequate dev response stating how haste works and is calculated. Our buff says "increases attack speed by 7%" which doesn't necessarily mean you take WEAPON SPD x 0.93. A variety of methods could be used to calculate haste. This is especially true considering that duel wield is a combined attack and your attack speed is supposedly 66% of your main weapon speed and 33% of your offhand speed. Almost every ability which gives us something to our attack is split. IE...+4 damage from HoW isn't +4 to main +4 to offhand, it's split between the two roughly 2/3s / 1/3s. I wouldn't be surprised if haste is split say 4.5% to main hand and 2.5% to offhand. If a Dev has clarified this a link would be greatly appreciated.

Beyond that, DPS isn't simply DMG / Spd. DPS is a combination of Str, Dexterity, Stances... wide variety of factors, all of which would not be affected the same way as weapon damage and all which change the DPS notation on the char sheet.

[zenharmony]

First, I have yet to see an adequate dev response stating how haste works and is calculated. Our buff says increases attack speed by 7% which doesn't necessarily mean you take WEAPON SPD x 0.93

I really don't think there's any other way to do it... your attack speed IS your weapon speed, plus any speed mods (such as the bard song). Dual wield doesn't change anything. (0.66SPD1 + 0.33SPD2) is still a constant with the same two weapons.

Beyond that, DPS isn't simply DMG / Spd. DPS is a combination of Str, Dexterity, Stances... wide variety of factors, all of which would not be affected the same way as weapon damage.

DPS *IS* simply [damage] divided by [time per attack] (not divided by [speed]...which is [attacks per second], inversely proportional to [time per attack]... or that would really be screwed up...).

The DMG portion is derived from str, weapon damage, and +damage mods. But again, that portion is constant regardless of attack haste buffs.

[Sung Entune]

Once I was able to fill in your numbers I was able to see the error in my math now, thanks for explaining it better. I have a list of bugs vgbards.com is sending in to Talisker tommorow, I'll make sure this get's added to it as a graphics bug and possibly non-functioning haste components.

I've tested haste through parsing and through video timing. In parsed results, I found an increase in melee damage appropriate to the amount of haste added to the point that the melee attack reaches a attack speed of 2.0. Beyond 2.0, using specials, the effects of haste are extremely minor. In video timing (IE...I recorded auto attack using FRAPs and millisecond playback) I could only determine that there was an effect from haste, but not to what extent.

[rabb1t]

As mentioned you can see on the character sheet under the damage ranges.

Test it yourself. Make a new song, with only haste in it and nothing else. Play said song. Poof, about a second or two later, your numbers will go up. Stop the song, or play a different song (such as run speed). Poof, a few seconds the numbers go down. (Be sure you have weapons equiped when you do this.)

You can check for stacking by adding components. As example, put 1 +4 in there, you can see it go up x. Put a second one in and play it again and you see it go higher. Put in a 3rd and again, it still goes higher. Haste, on the other hand, only goes up 1 'jump step' regardless of how many you put in the song. If it were additional hastes, then it should = 7% per bit, which would relate to shorter weapon delays per step if it stacked, thus increasing the DPS rating each time. But again, as I stated before, only the 1st bit increases the DPS rating. The rest seem to not further increase it.

As peeps have mentioned, if you are properly stopping the song between each component change, or if you are sure you are playing the new song, and it doesn't register, then it is bugged. (Or it was changed after my tests. I haven't had access since the 10th.)

[Sung Entune]

Originally Posted by rabb1t

As mentioned you can see on the character sheet under the damage ranges.

Test it yourself. Make a new song, with only haste in it and nothing else. Play said song. Poof, about a second or two later, your numbers will go up. Stop the song, or play a different song (such as run speed). Poof, a few seconds the numbers go down. (Be sure you have weapons equiped when you do this.)

You can check for stacking by adding components. As example, put 1 +4 in there, you can see it go up x. Put a second one in and play it again and you see it go higher. Put in a 3rd and again, it still goes higher. Haste, on the other hand, only goes up 1 'jump step' regardless of how many you put in the song. If it were additional hastes, then it should = 7% per bit, which would relate to shorter weapon delays per step if it stacked, thus increasing the DPS rating each time. But again, as I stated before, only the 1st bit increases the DPS rating. The rest seem to not further increase it.

As peeps have mentioned, if you are properly stopping the song between each component change, or if you are sure you are playing the new song, and it doesn't register, then it is bugged. (Or it was changed after my tests. I haven't had access since the 10th.)

It's definately not changing the character sheet even after 10 seconds. i've listed it in our bug list and sent it off. It will probably be another week or so though before they start working on bards again. Look for an update then.

[rabb1t]

Odd, that most certainly sounds bugged then cause I'm sure that it changed with the first bit, but no other bits. (I think I had a Longsword and Shortsword at the time, so I don't think I was hitting a possible 2.0 delay cap.)

[Spatzimaus`]

Well, it's also possible the "only 1 haste" thing is/was a display bug. Considering that there's been mention of a 25% haste cap, I'm pretty sure we're not intended to only use a single haste component. And it's a matter of Occam's Razor; considering the number of other bugs we've run into, I'm inclined to think that a simple UI display bug in a value most people don't pay much moment-to-moment attention to could easily slip under the radar.

When using a 3-haste song in a group, I definitely notice my teammates swinging more often. The RIGHT way to do this would be to time how long it took a person with a given weapon to swing X times and divide. (i.e., have them swing 11 times, take the time between swing 1 and swing 11, and divide by 10). As long as you don't use any special attacks, it should be pretty consistent. I'll try a bit of this tonight.
Same goes for proc percentages; it's easy enough to count, although the randomness involved requires a large sample size. (I used to play CoH, and the "Hero Logger" program someone whipped up was essential for these sorts of discussions. I let a mob beat on me for two hours straight on the test server just to measure the debuff percentage of Cloak of Fear... then they changed it.)

If it's not a simple linear increase across all weapons, I expect it's something like the separation of animation time from recharge time that most games suffer from. That is, if your weapon has a 2.5s speed, maybe 1.0s of that is the animation, and 1.5s is a "wait" time. The haste might only reduce the "wait", since it can't do anything to the animation. So, your 7% haste might only boost your DPS by 4%.
This would help explain the diminishing returns some people have been claiming when using light, fast weapons. (I use dual axes.)

[Spatzimaus`]

Okay, I just did some testing.

First of all, the Emboldening Embellishment (the one that adds to DEX and VIT) doesn't have any stacking problems, at least with a single person. At level 20 it adds 13 to each, and when I put three of them in a song it added 39. It's possible that the 3-component stacking was fixed when they changed it from a flat +10 to a 10-18 scaling, since I KNOW it worked fine when I put two in a single song before the fix.

Next, haste stacking. Here's what I found:
A = Hazoc's Harmony of Haste (embellishment, 7% haste for 2 energy/sec)
B = Hazoc's Hasty Lyric (lyric, 7% haste for 3 energy/sec)
In all cases, I just used fists, dealing a consistent 41-42 damage per hit and a listed DPS of 20.75. The enemy of choice was a Grass Beetle (non-aggressive, level 9, 2 dot) in Three Rivers.
The seconds per attack is taken by counting from the time I hit the attack key, counting time until the death blow landed, and dividing by (N-1). The -1 is because of how the game front-loads the animation.

In EVERY CASE, the UI listed a DPS of 20.75, regardless of the amount of haste applied. Since the fists dealt 41-42 damage, this'd imply an attack rate of exactly 2.0 seconds.

CASE 1: No song running
47 sec for 21 swings = 2.35 sec
50 sec for 22 swings = 2.38 sec
Okay, this is a decent baseline. We'll use the average, 2.365.
CONCLUSION: The UI DPS is worthless.

CASE 2: 1A
59 sec for 28 swings = 2.19 sec
62 sec for 29 swings = 2.21 sec
For an average of 2.20, the DPS increased to 107.5%, which means the delay was reduced to 93.02% of its original value.
CONCLUSION: Hazoc's Harmony of Haste works correctly when put into the song once.

CASE 3: 3A
48 sec for 27 swings = 1.846 sec
(second trial gave exactly the same)
This is a delay of 78% of the baseline.
CONCLUSION: Hazoc's Harmony of Haste also works correctly when you put three copies of it into the song.

CASE 4: 1B
49 sec for 24 swings = 2.13 sec
(second trial gave exactly the same)
This is a delay of 90% of the baseline (a DPS increase of 11.0%)

CASE 5: 2B
48 sec for 25 swings = 2.0 sec
42 sec for 22 swings = 2.0 sec
This is a delay of 84.5% of the baseline (a DPS increase of 18.3%)

CASE 6: 1A+1B
71 sec for 34 swings = 2.15 sec
52 sec for 25 swimgs = 2.17 sec
50 sec for 24 swings = 2.08 sec

CASE 7: 3A+1B
50 sec for 28 swings = 1.85 sec
46 sec for 26 swings = 1.84 sec

-----------------

CONCLUSIONS:

1> The UI's DPS value is just plain wrong, at least in the case of unarmed attacks, as it seems to be rounding the attack rate to 2.0s regardless of the actual value.
2> The UI's DPS does not change as haste effects are applied. (The only times it changed during my testing were the times I accidentally turned on the +damage form or Humming Blade)
3> Hazoc's Harmony of Haste is a linear 7% reduction in attack time, and stacks with itself (at least within a single song).
4> Hazoc's Hasty Lyric seems to be larger than the 7% listed reduction. It COULD be a ~10% increase in attack RATE, but that's just a guess. It stacks with itself, but it doesn't seem to stack linearly.
5> Hazoc's Hasty Lyric and Hazoc's Harmony of Haste do not seem to stack with each other. It seems to be that you find the total effect of one (including any self-stacking) and compare to the total effect of the other, and the larger effect applies. (3 A's are better than 1 B, while 1 B is better than 1 A.)

[Tem]

If haste in VG works like it does in EQ1 (same crew, so it would make sense), then a 7% haste mod equates to 7% more swings, not a 7% lower weapon delay. They aren't the same thing, and the reduction to delay is much more powerful.

For example's sake, lets say you found a way to get 100% haste. A 100% decrease in delay would mean that you win, instantly, all the time. You have no delay, and get unlimited attacks. Attacking 100% more often means that you attack twice as much, equating to a 50% reduction in your delay.

That being said, haste is nowhere near as powerful in this game than in most. The global cooldown is pretty high and cuts drastically into your autoattack DPS unless you can get your delay below 2.0 (go daggers!). I think a rebalancing of autoattack and skill damage would go a long way to making the classes easier to balance.

[Leishiu]

Originally Posted by Tem

If haste in VG works like it does in EQ1 (same crew, so it would make sense), then a 7% haste mod equates to 7% more swings, not a 7% lower weapon delay. They aren't the same thing, and the reduction to delay is much more powerful.

For example's sake, lets say you found a way to get 100% haste. A 100% decrease in delay would mean that you win, instantly, all the time. You have no delay, and get unlimited attacks. Attacking 100% more often means that you attack twice as much, equating to a 50% reduction in your delay.

Uh, considering EQ stacked up a possible three sources of haste (worn, spell and bard), where the two first were capped at a sum of 100% and the last could add up the sum to 130%, the system you're refering to wasen't one that was in EQ. ~

[Tem]

Originally Posted by Leishiu

Uh, considering EQ stacked up a possible three sources of haste (worn, spell and bard), where the two first were capped at a sum of 100% and the last could add up the sum to 130%, the system you're refering to wasen't one that was in EQ. ~

I think you misread me. It certainly was the system in EQ. The fact that you could get above 100% (when I said pretend that you can I was referring to Vanguard, where you cannot) means that it has to be programmed as in increase in swing speed rather than a straight reduction in a delay. A 130% (possibly more for monks, if I recall) reduction in delay means nothing; swinging 130% faster does.

[Spatzimaus`]

Originally Posted by Tem

If haste in VG works like it does in EQ1 (same crew, so it would make sense), then a 7% haste mod equates to 7% more swings, not a 7% lower weapon delay. They aren't the same thing, and the reduction to delay is much more powerful.

That's true, the two aren't the same. But, look at the numbers I posted. Granted, it's a tiny data set, but since there's no randomness to the timing of the autoattack, there's not really any room for discrepancy.

I saw a 7% reduction in attack time, which translated to a 7.5% increase in DPS, when a single haste component was applied. You can argue that the half-percent was a rounding thing and that it was really a 7% increase in DPS (meaning a 6.5% reduction in attack time).

BUT, that doesn't fit with what happened when I stacked three haste components into the song. I saw a 22% reduction in attack time (a 28.2% increase in DPS). To me, this clearly shows that VG haste reduces attack speed linearly, NOT increasing the number of attacks linearly.

So, it's not the same as EQ's haste.

[Tem]

Originally Posted by Spatzimaus`

That's true, the two aren't the same. But, look at the numbers I posted. Granted, it's a tiny data set, but since there's no randomness to the timing of the autoattack, there's not really any room for discrepancy.

I saw a 7% reduction in attack time, which translated to a 7.5% increase in DPS, when a single haste component was applied. You can argue that the half-percent was a rounding thing and that it was really a 7% increase in DPS (meaning a 6.5% reduction in attack time).

BUT, that doesn't fit with what happened when I stacked three haste components into the song. I saw a 22% reduction in attack time (a 28.2% increase in DPS). To me, this clearly shows that VG haste reduces attack speed linearly, NOT increasing the number of attacks linearly.

So, it's not the same as EQ's haste.

Indeed. Unfortunately I only read your post after I had created mine. I find it odd, though, that they went this route, as it, I think creates scaling issues. I wouldn't put it past them to revamp haste in it's entirety when the inevitable bard revamp comes down the line. Hopefully they'll let autoattack persist through skill usage as well.

[Miff]

Originally Posted by Tem

Indeed. Unfortunately I only read your post after I had created mine. I find it odd, though, that they went this route, as it, I think creates scaling issues. I wouldn't put it past them to revamp haste in it's entirety when the inevitable bard revamp comes down the line. Hopefully they'll let autoattack persist through skill usage as well.

There is a total haste cap of 25%.

[bandon]

Originally Posted by Actus

I'm a level 19 bard, and I tried experimenting with my newly acquired song pieces to try to create a sort of defensive-type song that gives up some haste/damage elements for regen and shielding and I'm getting some numbers that just cannot be right. I must be doing something wrong.

For example, the song that absorbs damage is only absorbing 8 points of damage PER PULSE? Per pulse?!?!? When the things trying to kill me are hitting for 200? This song would be barely worth playing at level 1, and I got it at level 18. I'm not sure I would include it in any lineup if it absorbed 80 points...

And the damage shield song burns attackers for 4 points of damage? FOUR?!? When I'm regularly hitting for 800? Am I doing something wrong?

No one notices the regen elements of a song, even when I stack them.

And the songs which look like they should be AE are only single target? And they hit for 20 damage PER PULSE?!?!? What is going on? Am I bugged?

I don't come here to complain, or to start drama (check my posting history), and I'm generally happy with the game. But these things seem way out of whack. Thanks in advance.

Ya im a 46 bard, and the only song i ever run is max haste, stack the str/AC debuff, and 3 of the agi/vit buffs.
everything else is a complete and utter joke. I mean i just got a lvl 45 coda or something that increase all group resists by TWO. lol.
The one that protects against cold/heat increase resists by FOUR.
Like i said, waste o time.
Best defense is a good offense, just ignore all the defensive stuff.

[Muggi]

Originally Posted by bandon

Ya im a 46 bard, and the only song i ever run is max haste, stack the str/AC debuff, and 3 of the agi/vit buffs.
everything else is a complete and utter joke. I mean i just got a lvl 45 coda or something that increase all group resists by TWO. lol.
The one that protects against cold/heat increase resists by FOUR.
Like i said, waste o time.
Best defense is a good offense, just ignore all the defensive stuff.

Yeah ah...might want to look at your attributes when you use that Coda..well, it's really a Bridge, Suann's Transformation is close to the name. There's a reason your resistances are going up by two.

Don't trust tooltips, most are outdated. Also, if you're running haste and the agi/vit buffs instead of Mel's, you might want to reconsider..you're gimping your group's DPS.