All the online resources I can find say alacrity doesn't really affect damage until you reach a breakpoint, and lower the GCD to 1.4 seconds.

Yet adding just one Alacrity Enhancement significantly increases DPS on the character sheet.

Has something changed? How does the math work?

The damage numbers are completely obsolete, they mean nothing.

The way alacrity works is that there are breakpoints at which you will get a lower GCD significantly more often. The general principle is still that alacrity makes you faster, however going a little above the "breakpoint" will "counter lag" which is copium for "it just plays nicer since you almost always have the same GCD". Not every GCD has the same length, ability queues and everything just have nothing to do with it. They just do not always have the same length and you are able to get a 1.4 GCD with less alacrity than the breakpoint, however you will only sometimes have that happen, that is why the recommendation is to go a little bit above.

If you are a casual player, which I assume you are, I would recommend not to worry at all about these breakpoints and just aim for a good distribution. For solo play, alacrity and accuracy values DO NOT MATTER in the slightest. They have no effect, on planets usually you are given the accuracy anyway, nevermind most adds don't even have a defense and alacrity will get scaled down so far you'll not notice it, that doesn't mean to go full crit - don't get me wrong, just aim to balance it out. If you want to do operations I strongly recommend going above 110% accuracy, but usually solo players do not have the resources required to hit the alacrity breakpoint and have a reasonable amount of crit, so just do a 60/40 split crit/alacrity and you'll be set until you can raid nightmare, if you wish to do so.

I don't know how the character sheet calculates the damage number. I have found in most games not to trust those numbers. They never seem to be able to take into account all nuances of the stats as well as the abilities available to the class (passive bonuses, abilities that make other abilities stronger, etc). The numbers you will see in guides have been figured out by doing parses on target dummies and using that to figure out a theoretical max damage on a boss. Each boss encounter will be different. Some bosses will favor one combat style or another depending on the specific boss mechanics.

As for alacrity, the breakpoints are because of a limitation of the game engine. The GCD (global cooldown) which activates every time you use an ability is only calculated down to tenths (1.5, 1.4, 1.3). Alacrity percentages in between the breakpoints don't lower the GCD at all until you reach a breakpoint, where it jumps down a full tenth (e.g. 1.5 to 1.4). Ability activation times are calculated down to hundredths (2.50, 2.49, 2.48), so alacrity percentages in between can reduce cast times for abilities with longer cast times like Force Storm. Abilities that are already 1.5 second activation time will still cast based on hundredths, but you won't be able to cast the next one until the GCD is up, so they are basically limited by the GCD more than their activation time. I'm not even sure the cast bar shows hundredths (don't have the game open at the moment), but it definitely does this behind the scenes. Because so many abilities are 1.5 second activation time, lowering the GCD can have a big impact on DPS. Alacrity also affects regen rate of class resources. This can be important for healers, but not so much for soloing stuff.

The damage stat probably doesn't take into account this limitation. In other words, it assumes if you increase your alacrity by 2% from an enhancement, you're using 2% more abilities, when that might not be the case. Some disciplines are all instant abilities, only limited by GCD. If that 2% more alacrity doesn't reach a breakpoint, you would be getting 0% more abilities from that enhancement, a total waste. To truly calculate damage accurately requires simulating a combat situation with the ideal rotation or priority list for the particular combat style. It's not something that can be done instantly, so they have to use simpler calculations for the character sheet. They probably don't even take into account the discipline you are playing.

Alacrity and accuracy are both uncapped even in level sync. This is a change from 7.1. It means they are the most powerful stats if you're doing low level content because you get the full benefit of them. On the other hand, critical rating, is still capped very low. You won't get much out of it in level sync. For solo content, accuracy only matters for Elite (gold) and Champion (gold/silver) mobs. Elites have 5% chance to resist/dodge your attacks. Champions have a 10% chance to resist/dodge your attacks. So accuracy of 105% is enough to not miss from everything except Champions, and 110% is enough to avoid missing entirely from all mobs. Elites are fairly common in level sync content, especially if you do a lot of heroics it is noticeable. There are a few Champions in heroics as well. Because they have such high health, I find it nice to not miss. It already takes a while to bring them down, missing a lot can be frustrating. Almost all solo content can be done with the base 100% accuracy, but it's definitely nice not missing.

Here's a site with the target numbers for accuracy and alacrity:

https://grophets.com/ask/alacrity-thresholds-for-7-0/

https://grophets.com/ask/how-do-i-get-to-110-accuracy/

I'm sorry but that is just not true.

A GCD can very well have the length of 1.36 or 1.42, its just that the distribution of these will be around 1.4

I will still say that alacrity and accuracy are useless, the benefit from alacrity is ONLY useful when you are able to get more GCDs in than you are doing anyway, which will almost never be the case for story or solo content.

Also I have the feeling that some abilities just get no bonus from alacrity, or that chaining two abilities will rrsult in a longer GCD, but that is really just boomer talk and I have barely anything to back that up. Focused Burst will always have a GCD of around 1.5, provided you do a blade barrage before. It is true, but I got no clue why.

All the online resources I can find say alacrity doesn't really affect damage until you reach a breakpoint, and lower the GCD to 1.4 seconds.

Depends. If your chosen combat style has interesting abilities that take longer than a GCD to launch, you'll see the effect there, because ability cast/channel(1) times obey Alacrity on a "hundredths of a second" basis.

The DPS shown on the sheet might or might not be an accurate reflection of that, but the effect is there.

(1) For those in the audience who don't know the distinction: "cast" abilities are launched, do nothing for a time, then deliver their effect at the end e.g. Grav Round on Commando, while "channeled" abilities deliver packets of effect during the channel time e.g. Sweeping Gunfire on Gunslinger. Alacrity shortens cast times, delivering the effect sooner, and shortens channel times, delivering the same number of packets in a shorter time.

I'm sorry but that is just not true.

 

A GCD can very well have the length of 1.36 or 1.42, its just that the distribution of these will be around 1.4

 

I will still say that alacrity and accuracy are useless, the benefit from alacrity is ONLY useful when you are able to get more GCDs in than you are doing anyway, which will almost never be the case for story or solo content.

 

Also I have the feeling that some abilities just get no bonus from alacrity, or that chaining two abilities will rrsult in a longer GCD, but that is really just boomer talk and I have barely anything to back that up. Focused Burst will always have a GCD of around 1.5, provided you do a blade barrage before. It is true, but I got no clue why.

I'd like to see some combat log parse evidence that demonstrates you can routinely see the effective delay between abilities shorter than the known breakpoints. It's simple enough to do, parse hitting a target dummy for at least 273 seconds and ideally some multiple of that, using only your basic attack, at different alacrity levels. I have never been able to get such a parse to give me results that would indicate I ever have a GCD less than 1.4 seconds when I'm gearing for the 1.4s GCD. Having enough alacrity such that you would have a GCD of 1.36 seconds implies you are exceeding 44 APM on your basic attack, and I've never been able to observe that.

I'd like to see some combat log parse evidence that demonstrates you can routinely see the effective delay between abilities shorter than the known breakpoints. It's simple enough to do, parse hitting a target dummy for at least 273 seconds and ideally some multiple of that, using only your basic attack, at different alacrity levels. I have never been able to get such a parse to give me results that would indicate I ever have a GCD less than 1.4 seconds when I'm gearing for the 1.4s GCD. Having enough alacrity such that you would have a GCD of 1.36 seconds implies you are exceeding 44 APM on your basic attack, and I've never been able to observe that.

No, that's what I was saying though. The assumption that alacrity effects every ability to the same degree HAS to be wrong. Chaining basic attacks will obviously prove that you average out on 1.4 GCDs, which is completely in line with what I said. You will have SOME GCDs lower than 1.4 and some over 1.4. The problem is that always the same ability chains will result in a longer GCD which is not factored in in ANY theorycrafting which makes that whole thing even more worthless.

Saying that... it's just unlucky, or the person is clicking and they might just be slow cannot really be true for the topparse of DPS leaderboard can we agree? So then how come he has this long GCDs on these abilities most frequently when they are chained?

And then how come he has some GCDs lower than 1.4s? That would make even less sense?

Hope that proves it. Truth be told, if someone tells you they know how alacrity works you can be sure they are wrong. Can't just be a logging error, but when parsing sent you can feel the slow GCDs I'm not gonna lie.

Edit: To be fair, it's a bad example for the lower GCDs yes I know why leap has a 1.365 GCD. Here are better values.

No, that's what I was saying though. The assumption that alacrity effects every ability to the same degree HAS to be wrong. Chaining basic attacks will obviously prove that you average out on 1.4 GCDs, which is completely in line with what I said. You will have SOME GCDs lower than 1.4 and some over 1.4. The problem is that always the same ability chains will result in a longer GCD which is not factored in in ANY theorycrafting which makes that whole thing even more worthless.

 

Saying that... it's just unlucky, or the person is clicking and they might just be slow cannot really be true for the topparse of DPS leaderboard can we agree? So then how come he has this long GCDs on these abilities most frequently when they are chained?

And then how come he has some GCDs lower than 1.4s? That would make even less sense?

 

Hope that proves it. Truth be told, if someone tells you they know how alacrity works you can be sure they are wrong. Can't just be a logging error, but when parsing sent you can feel the slow GCDs I'm not gonna lie.

 

Edit: To be fair, it's a bad example for the lower GCDs yes I know why leap has a 1.365 GCD. Here are better values.

You can't use competitive parses because many abilities are off the GCD, and certain abilities (Zen/Berserk; Mental Alacrity/Polarity Shift) may get you into shorter GCD temporarily if you carry sufficient underlying alacrity percentage. The only way to test what your effective GCD actually is, is to do a dummy parse with your basic attack.

You can't use competitive parses because many abilities are off the GCD, and certain abilities (Zen/Berserk; Mental Alacrity/Polarity Shift) may get you into shorter GCD temporarily if you carry sufficient underlying alacrity percentage. The only way to test what your effective GCD actually is, is to do a dummy parse with your basic attack.

thanks, very insightful - I picked classes that do not have, to my knowledge, any alacrity buffs.

If of GCD abilities would be taken into account you'd see zen in the screenshot. I am obviously not talking about the jump, or the strides.

I'm sorry but that is just not true.

 

A GCD can very well have the length of 1.36 or 1.42, its just that the distribution of these will be around 1.4

 

I will still say that alacrity and accuracy are useless, the benefit from alacrity is ONLY useful when you are able to get more GCDs in than you are doing anyway, which will almost never be the case for story or solo content.

 

Also I have the feeling that some abilities just get no bonus from alacrity, or that chaining two abilities will rrsult in a longer GCD, but that is really just boomer talk and I have barely anything to back that up. Focused Burst will always have a GCD of around 1.5, provided you do a blade barrage before. It is true, but I got no clue why.

agreed with the underlying premises but you forgot one of the most important thing about high alacrity.

if some has 2200 points in alacrity = they will have faster CD (many seconds) on a lot of main hard hitting abilities e.x hardstun, CC and vanish

VS someone who has 0 points in alacrity.

So you forgot to mention these points when you say "alacrity is useless". Its not only about how many attacks you can get in under a gcd. Faster CD VS default CD can be critical under some circumstances.

agreed with the underlying premises but you forgot one of the most important thing about high alacrity.

 

if some has 2200 points in alacrity = they will have faster CD (many seconds) on a lot of main hard hitting abilities e.x hardstun, CC and vanish

VS someone who has 0 points in alacrity.

So you forgot to mention these points when you say "alacrity is useless". Its not only about how many attacks you can get in under a gcd. Faster CD VS default CD can be critical under some circumstances.

I did not forget it I didn't mention it because it doesn't matter.

Not only is this a false friend in many raiding situations, brontes burn is more of an exception to this and there someone wouldn't get e.g. another fuel without a merc raidbuff no matter how much alacrity is used (well maybe if a DPS dies), because dps should very rarely use offensive cooldown on cooldown. Nevermind in a situation in solo content, most of the time using offensives for anything but boss fights just kinda wasted.

A casual player, someone without even purple or blue augs, will not have a statpool large enough to actually play 2700 accuracy and 2200 alacrity and end up with more than 1k crit. Losing this much crit is just not compensated, especially if the DPS is new or doesn't know the fights, by alacrity. Alacrity is good if the dps already knows what they are doing, how the cooldowns align with the DPS on the boss and when to DELAY them, not how to spam them.

Alacrity has always been one of those stats we don't know 100% how it works. I play swtor at different locations and alacrity has an impact on dps due to connection that is to be sure. At my home that has one gig internet, I can get away with lower Alacrity then some place that only has a 100 meg connection for the same dps parse.

It's almost impossible to prove exactly how it works. We would need thousands of combat logs with different players playing at multiple locations. Then some one to sort though all that data etc.

Long story short. I would recommend parsing where your going to play swtor and using the alacrity break points as a general guide, not a set in stone number.

Alacrity has always been one of those stats we don't know 100% how it works. I play swtor at different locations and alacrity has an impact on dps due to connection that is to be sure. At my home that has one gig internet, I can get away with lower Alacrity then some place that only has a 100 meg connection for the same dps parse.

 

It's almost impossible to prove exactly how it works. We would need thousands of combat logs with different players playing at multiple locations. Then some one to sort though all that data etc.

 

Long story short. I would recommend parsing where your going to play swtor and using the alacrity break points as a general guide, not a set in stone number.

Latency could have an impact if you have the Ability Activation Queue turned off. When off, each ability used triggers the GCD and the next ability can't be used until GCD + latency (when the server gets the request). I'm not sure if this is reflected in the combat log. It depends on how the client writes that out. If it puts data based on the server's responses, latency would be reflected in the numbers when Ability Activation Queue is off. If it put data based on when the player pressed the button client side, latency wouldn't be a factor.

On the other hand, if Ability Activation Queue is turned on (the default), it should remove latency from ability times assuming your latency is less than the setting. For example, if you have <100 ms latency and set it to 0.1s, all the abilities will be chained without latency between them (as long you're pressing abilities within 100ms of the GCD ending). This means latency wouldn't appear in the combat log.

Latency could have an impact if you have the Ability Activation Queue turned off. When off, each ability used triggers the GCD and the next ability can't be used until GCD + latency (when the server gets the request). I'm not sure if this is reflected in the combat log. It depends on how the client writes that out. If it puts data based on the server's responses, latency would be reflected in the numbers when Ability Activation Queue is off. If it put data based on when the player pressed the button client side, latency wouldn't be a factor.

 

On the other hand, if Ability Activation Queue is turned on (the default), it should remove latency from ability times assuming your latency is less than the setting. For example, if you have <100 ms latency and set it to 0.1s, all the abilities will be chained without latency between them (as long you're pressing abilities within 100ms of the GCD ending). This means latency wouldn't appear in the combat log.

Logging is for everybody so it has to be server side, when someone had the game in e.g. german their combat logs for you, provided they join the parse group, would be in german too. They changed that now so everything is the same now, so it loggs when it receives and then sends it back to you.

Factoring such a thing in is a bit... weird. As I said the alacrity isn't easy, the number that is the breakpoint will still have a lot of higher GCDs in it since its close to the breakpoint, adding one or two augs will shift it to having a more consistant 1.4, and sometimes lower, GCD.