Alacrity and the GCD

[Magdain]

Somebody here claims that character animations have a higher priority than GCD. The claim is that if your GCD is shorter than the animation of the spell, your next spell will not cast until the animation of the previous one finishes.

I've yet to see any evidence but it seems plausible given my experiences. I have access to the Alacrity adrenal and relic use items, so I will try to test this once the servers are up; If anybody has any advice for methodology I'd appreciate it, as I'm rather new to field tests.

[Spren]

I don't think that's true Magdain. I have ~10% alacrity before using any abilities. If I blow my +20% Alacrity ability (lightning sorcerer) I will cast really fast and it seems like the animations speed up. One weird thing I see is if I'm spamming lightning strike with the above conditions, around 1.0-1.1 second casts, and then I get a proc to use Force lightning at 2x the speed (normally 3.0 seconds, down to 1.5 seconds before all that alacrity), I will begin doing the channel before the bar appears, or I see the effect. So visually it's strange but timing wise I think it's correct.

Overall the huge discussion about character responsiveness is largely subjective. And if there is a legitimate problem that could be objectively shown, at least every other person posting about it is an idiot with a bad computer. The forums are filled with bads, and so while some people may be experiencing something real, and completely separate from FPS/Latency, a lot of the volume over the topic isn't actually from those people.

[Kaedis]

The claim wasn't that animations overrule Alacrity, it was that Alacrity didn't affect the GCD because the animations needed time to run. That said, Whirlwind with a single point in Haunted Dreams makes it clear that animations speed up if Alacrity is present.

Quote:I read this whole thread. It seems to me like we know alacrity lowers the GCD on heals with an activation time but we don't know if it affects an instant cast?

Actually, we know that at this time it does not. Given what we've tested it on, I think it's still up in the air if Alacrity affects the GCD on channeled spells that drop under 1.5. Actually, what I suspect is the case is that the GCD is equal to either 1.5 or the length of the current channel/cast, whichever is lower. So a 3 second channel/cast would have a 1.5 second GCD regardless, but a 1.5 second hasted to 1.3 second would have a 1.3 second GCD, regardless of whether it's a cast or a channel. I have no data as of yet to back this up, as I don't have fraps, but I have sufficient alacrity on my gear that I should be able to test this with the Lightning tree Alacrity cooldown pretty easily once I get back home.

[Daxx]

(12-28-2011 03:43 AM)Magdain Wrote:  Somebody here claims that character animations have a higher priority than GCD. The claim is that if your GCD is shorter than the animation of the spell, your next spell will not cast until the animation of the previous one finishes.

I've yet to see any evidence but it seems plausible given my experiences. I have access to the Alacrity adrenal and relic use items, so I will try to test this once the servers are up; If anybody has any advice for methodology I'd appreciate it, as I'm rather new to field tests.

A perfect example of the animation having a higher priority than gcd is the sith warrior juggernaut force choke with talents that remove the channel on it. You can't do anything till the character finishes the cast animation; might line up with gcd but feels like its much longer.

[theothersteve7]

I thought I was noticing some oddities with using Riposte and Force Kick off the GCD, but I'm pretty sure I can chalk those up to very briefly not having the resources to use them.

I haven't noticed any problems with Force Stasis.

[hatterson]

(12-27-2011 08:44 PM)LagunaD Wrote:  

(12-24-2011 07:17 AM)lostdummy Wrote:  Does alacrity influence DoTs, and how?

Some possibilities:
1) it does not
2) reduce duration
3) duration remains same, but add ticks (similar to wow)

My money is on #1, but #3 seems easy to rule out. Affliction can be talented to have a 21-second duration (7 ticks). If you can get 15% or so from Alacrity Rating, and another 20% from the Inquisitor ability, you should certainly see more than 7 ticks if Alacrity affects the DoT.

And with video capture, you can certainly time the interval between the first and last tick well enough to test #2 (the difference would be several seconds if Alacrity is reducing the duration).

I hadn't seem someone actually test this so I got a buddy and decided to run the numbers. I apologize if someone else already did this, I just haven't seen it on here

I spec'd lightning so I had the increased duration of affliction and access to polarity shift.

I had a stopwatch in hand to time what was happening in real time and ensure that nothing was going on with the game's debuff count timer.

I tested once with no gear on to ensure the baseline looked good. With no gear on an affliction did indeed last 21 seconds and ticked 7 times, once every 3 seconds. The debuff was removed from the opposing player exactly when the last tick happened.

I then put my alacrity gear on which gave me 320 alacrity rating for 9.78% Activation Speed. With this affliction still lasted 21 seconds and ticked 7 times, once every 3 seconds. The debuff was again removed from the opposing player exactly when the last tick happened.

I also had access to polarity shift and I had a relic that gave 225 alacrity rating. With both of these popped I had a total alacrity rating of 545 which resulted in an overall activation speed of 4.68%. Polarity Shift added a flat 20% onto this resulting in the character sheet stating 34.68% activation speed. Under these conditions affliction again lasted only 21 seconds and ticked only 7 times, and the debuff expired exactly as the last tick happened.

All times above lined up with both the game clock (counting on the debuff) and the stopwatch I had.


This same behavior held true without the 6 second duration increase talent. All 3 copies of affliction lasted 15 seconds and ticked 5 times, expiring at the same time as the last tick.

If alacrity affected the length of the debuff (increase):
Even if the alacrity only affected the base 15 seconds, a 34.68% increase should have resulted in a 5.2 second increase to the debuff which would have netted at least 1 extra tick. If it did the entire 21 it should have resulted in a 7.28 second increase to the debuff and added at least 2 extra ticks.

If alacrity affected the length of the debuff (decrease):
The 15 second buff should have lasted only 9.86 seconds and the 21 second debuff should have lasted only 13.8 seconds.

If alacrity affected the frequency of the debuff ticks:
It should have resulted in a tick every 1.96 seconds rather than every 3 seconds which would result in 7 ticks on the 15 second debuff or 10 ticks on the 21 second debuff.

In all cases neither the relic nor polarity shift affected the amount affliction ticks were hitting for. Each hit for 221 when geared and 62 when out of gear


Given that none of these effects were observed I believe it is safe to conclude that alacrity does not affect DoTs.

[Freehugs]

That's good work Hatterson.

[Lord_Indomitus]

(01-05-2012 06:30 AM)hatterson Wrote:  Given that none of these affects were observed I believe it is safe to conclude that alacrity does not affect DoTs.[/size][/b]

So it works exactly as the tooltip suggests. Good to know, thanks for your time to test.

[HunterCelric]

Good news everyone! Alacrity effects the GCD! A poster earlier that he thought that if you have a cast time that's 1.5 seconds and you have Alacrity to bring it down to 1.3 seconds, that the GCD would go fast enough for -that- ability to begin casting again at 1.3 seconds and not have a hang-time of .2 seconds. This is correct. Here's my findings.

My BH, 50, currently has an 8.76% activation speed time with 120 rating + 4% stand alone from talents and an additional 5% from talent procks on crits (rapid shots -always- will crit so it's theoretically always up). This comes out to being 13.76% haste. It brings down my Tracer Missile from it's base 1.5 second cast to 1.3 seconds. Now I know it's hard to imagine that .2 seconds is even noticeable, so I'll sweeten the deal. My on-use-relic grants an additional 270 Alacrity rating. It almost doubles my rating, but not actually my speed (I suspect DR here, but not what I'm here to talk about). All said and done, my activation speed is 20.9% with it all up, bringing my Tracer Missile spam to 1.2 second casts.

So, am I standing there after every cast for .3 seconds? Nope. The instant my missile goes off, another is already casting. I can see that the GCD is moving faster to keep up with the Tracer Missile, which is on a lower 'cooldown' time than the standard GCD. So for all instinctive purposes, the GCD is affected by Alacrity.

However, there's one issue I've come across that has me grinding my teeth. I -think- the GCD is only brought down when you use the same ability that has the lower cooldown than the GCD itself. When I'm letting loose my Tracer spam, and I get a Barrage prock for Unload, on my next GCD I'll hit unload. Often enough I'll get a hang-fire. It'll show me shooting, Unload's channel bar is going down, but I'm not actually firing. I believe, perhaps, that I'm clipping the GCD. I could be wrong, it could also be that the animation of Tracer and Unload are wrestling it out on screen. If I follow up a 1.2 Tracer with Rapid Shots or another instant cast, it goes off with out a hitch.

I hope this sheds some light on the little debate here. Would love to see some other confirmations when you guys get a chance to do some more testing. Thanks!

Celric, Master Zhar Lestin server. GM of Nemesis

[Kaedis]

Quote:Good news everyone! Alacrity effects the GCD! A poster earlier that he thought that if you have a cast time that's 1.5 seconds and you have Alacrity to bring it down to 1.3 seconds, that the GCD would go fast enough for -that- ability to begin casting again at 1.3 seconds and not have a hang-time of .2 seconds. This is correct. Here's my findings.

Old news, actually. We figured this out just before launch. However, this only applies to cast time spells, not instants. The bit about the delay is new, but I think is just a graphical glitch, in all reality.