*** And Another Update ***

http://www.mrrabbit.net/misc/swtor/swtorglobalcooldown.php

This time around I went through the trouble of pushing my Sorcerer Healer to the 1.3 GCD.

And it was the same painful 2+ hour process of adding, testing, adding, testing just as with the Mercenary Healer.

9.3 / Force Regen Rate was triggered @ 3305 Alacrity Points, but she kept coming in with a quite a few 1.31 /s and 1.32 /s for Raw GCD results with only ONE coming in at 1.3048 /s.

After roughly the same amount of augments, and earpiece and implant changes plus a few key enhancements, I ended up with:

3724 Alacrity Points

9.4 /s Force Regen Rate

1.3030 /s Raw GCD Result

1.3 /s Actual GCD

There were a few tests where she was dropping to 1.3041.

In the past when I've compared my characters to previous SWTOR Theory Crafting charts for diminished returns and the related GCD points and Damage Mitigation stuff I would only have to go about 100-200 pts past the published GCD Tier break points on my OP healer, 200-300 for my Mercenary and Sorcerer healers.

This time around it's 200+ for my OP healer, and 500+ for my Mercenary and Sorcerer healers.

For the lower 1.4 GCD, my characters have always been a lot closer.

Again, this is what the Personal Starship Operations Test Dummy is reporting via Starparse Combat Log running in personal Parse mode.

=8-|

Just thought I'd add that I did a few Polarity Shift tests as well...

....the math on the individual break downs in the log shows:

1.0970 to 1.100 to 1.1700

Those being the most common break downs, suggesting the Polarity Shift on a 1.3 GCD Sorcerer results in a 1.1 GCD.

=8-)

If what the OP is writing is egregiously wrong will you guys please add useful information that actually proves his info is wrong and explain why?

 

How does this help with argumentative posts that don't have any technical information to compare with the OP's?

 

I realize Dipstik is regarded highly here, but that doesn't mean other people cannot provide useful or helpful information on topics Dipstik already wrote about.

 

Thanks OP for trying to be helpful.

Mainly because this is the same attitude behind some of the users of a specific "helping" discord. Egotistical, elitist, bad attitudes if you question them. It's quite pathetic.

They're not there to help, just to prove you wrong, how incorrect you are, belittle you, etc...

In fact a user posted this thread in that said discord mocking it. Pretty pathetic that instead of helping others in a friendly manner they come out to just attack & defend their cocky attitudes & knowledge. Which does show, no matter how much knowledge you have, if you act like a jerk, no one will care.

Most tend to forget it's a game that could shut down tomorrow & life would move on, none the worse for it.

Some need to take a RL break & examine what they're missing outside of a pixel universe. lol

Many thanks Original Mr Rabbit

You're welcome!

Ok, Ive a question that I’m pretty sure I know the answer too already.

My wife plays multiple fury maras. But they have slightly different stats due to different, ear, implants and augments on the left side.

The stats are not majorly different, but she swears one feels more fluid and faster with the CDs when she plays them. They both have the same Armor set as it’s an exact clone as are the weapons (considering it’s the same set )

Anyway, I know the alacrity on one is slightly over the stats for the GCD of 1.4 secs (7.55%) and the other is closer to 8%. The other difference is the one with more alacrity has a little less crit. Both have the same accuracy.

She’s convinced the one with higher alacrity feels smoother and faster, I’ve told her it’s in her head because the 8% won’t make any difference to the GCD she gets at 7.55%. Am I right or am I missing something with how this meta’s alacrity is working?

Going out on a limb here...

Here's my take....with a lot of subjective, off hand and guestimation.

I suspect that if you are spot on with a GCD in accordance with the Theory Crafting Modeling AND the corresponding new level of Force Regen Rate / Heat Dissipation Rate / Energy Regen Rate . . . you can say you have the GCD.

But I suspect that there is a combination of other things that cause a slow down:

1. As mentioned by a previous poster, server <> client update lag.

2. Local client log update lag - your client is always writing the log - if you have it enabled - it's a write to disk taking place..

AND

3. Some unknown RNG that Bioware is tossing into their formula on a class/spec basis.

4. CPU exceptions during calculation forcing a sudden recalculation creating a delay.

You'll see the effects of these "suspected culprits" in the Starparse Combat Log - in the intervals between instant cast abilities:

1.2885 1.3040, 1.2935, 1.3112, 1.3211, 1.2943, 1.3001, and so on...

You'll see those intermittently...over a test over a given duration.

And if you have too many of those in a given test cycle, you've just lost from a DATA ANALYSIS standpoint from that LOCAL TEST and OBSERVATION your 1.3 GCD.

...even though you know that technically you have it.

This is why I'm often adding 1-2 228 Alacrity Augments more than what is needed - to get ahead of this.

In other words, I want what shows up in my testing as a consistent 1.3 GCD.

as and a result, a character that also feels smoother.

It's much easier to achieve on a 1.4 GCD character...much harder on a 1.3 GCD character.

But it also means I'm wasting stats in Alacrity that would theoretically and likely realistically be better spend in Critical Rating points.

Do a test like I do, watch that ability bar, and you will actually see a couple periods during your test where it appears the ability cool down sliding effect is slowing down for a few seconds.

What's really bugging me is that the cost is much more extreme on my Mercenary and Sorcerer healers at 1.3 GCD than it is for my Operative healer at 1.3 GCD.

Why is that?

In other words, even I am scratching my head.

=8-P

Going out on a limb here...

 

Here's my take....with a lot of subjective, off hand and guestimation.

 

I suspect that if you are spot on with a GCD in accordance with the Theory Crafting Modeling AND the corresponding new level of Force Regen Rate / Heat Dissipation Rate / Energy Regen Rate . . . you can say you have the GCD.

 

But I suspect that there is a combination of other things that cause a slow down:

 

1. As mentioned by a previous poster, server <> client update lag.

2. Local client log update lag - your client is always writing the log - if you have it enabled - it's a write to disk taking place..

 

AND

 

3. Some unknown RNG that Bioware is tossing into their formula on a class/spec basis.

4. CPU exceptions during calculation forcing a sudden recalculation creating a delay.

 

You'll see the effects of these "suspected culprits" in the Starparse Combat Log - in the intervals between instant cast abilities:

 

 

1.2885 1.3040, 1.2935, 1.3112, 1.3211, 1.2943, 1.3001, and so on...

 

 

You'll see those intermittently...over a test over a given duration.

 

And if you have too many of those in a given test cycle, you've just lost from a DATA ANALYSIS standpoint from that LOCAL TEST and OBSERVATION your 1.3 GCD.

 

...even though you know that technically you have it.

 

This is why I'm often adding 1-2 228 Alacrity Augments more than what is needed - to get ahead of this.

 

In other words, I want what shows up in my testing as a consistent 1.3 GCD.

 

as and a result, a character that also feels smoother.

 

It's much easier to achieve on a 1.4 GCD character...much harder on a 1.3 GCD character.

 

 

But it also means I'm wasting stats in Alacrity that would theoretically and likely realistically be better spend in Critical Rating points.

 

 

Do a test like I do, watch that ability bar, and you will actually see a couple periods during your test where it appears the ability cool down sliding effect is slowing down for a few seconds.

 

 

What's really bugging me is that the cost is much more extreme on my Mercenary and Sorcerer healers at 1.3 GCD than it is for my Operative healer at 1.3 GCD.

 

Why is that?

 

In other words, even I am scratching my head.

 

=8-P

I get what you’re saying. But I don’t think it’s correct.

Let me provide some other information to rule out some of those possibilities.

We don’t have combat logging on for pve stuff.

We have the same PC set up and graphics settings.

This is not like it’s two different PC’s anyway, it’s her own Alts with slightly different alacrity stats.

It’s not playing for 5 mins on each to test, it’s hours on each and when she switches she “thinks” it feels different on the other Mara.

Also, isn’t 7.55% well above the 1,4 second thresh hold or do I have that wrong.,

I’m not trying to start any argument and your insight is appreciated. But is it at all remotely possible the 8% vs 7.55% is giving her slightly faster feel. Which would mean Alacrity is no longer working exactly as we are used to.

Or as I suspect, it’s totally a perception thing and in her head,

I get what you’re saying. But I don’t think it’s correct.

 

Let me provide some other information to rule out some of those possibilities.

 

We don’t have combat logging on for pve stuff.

 

We have the same PC set up and graphics settings.

 

This is not like it’s two different PC’s anyway, it’s her own Alts with slightly different alacrity stats.

 

It’s not playing for 5 mins on each to test, it’s hours on each and when she switches she “thinks” it feels different on the other Mara.

 

Also, isn’t 7.55% well above the 1,4 second thresh hold or do I have that wrong.,

 

I’m not trying to start any argument and your insight is appreciated. But is it at all remotely possible the 8% vs 7.55% is giving her slightly faster feel. Which would mean Alacrity is no longer working exactly as we are used to.

Or as I suspect, it’s totally a perception thing and in her head,

I think what I'm really saying is that my excess alacrity adjustments firm the slow downs up enough that they too stay in the 1.3 GCD range resulting in a smoother feel.

The tiers are still there....you see it in corresponding the dissipation / regen rate change...pretty sure that hasn't changed.

Note that for Maras, Juggs, you don't have that statistic...

And yes, according to MrPhStar, your' 7.55% does indicate some overhead like I do on my characters...

Other than that, I can't really think of anything besides remotes possibilities such as hardware differences - hard drive speeds, i.e, actual differences in machines.

I will say that my Twilek Sorcerer rolled on the Bastion and my Rat Sorcerer rolled on the Harbringer both as healers with same setup have always felt different - the Rat one feeling a tad smoother.

But that's entirely subjective.

=8-|

Mainly because this is the same attitude behind some of the users of a specific "helping" discord. Egotistical, elitist, bad attitudes if you question them. It's quite pathetic.

They're not there to help, just to prove you wrong, how incorrect you are, belittle you, etc...

 

In fact a user posted this thread in that said discord mocking it. Pretty pathetic that instead of helping others in a friendly manner they come out to just attack & defend their cocky attitudes & knowledge. Which does show, no matter how much knowledge you have, if you act like a jerk, no one will care.

 

Most tend to forget it's a game that could shut down tomorrow & life would move on, none the worse for it.

 

Some need to take a RL break & examine what they're missing outside of a pixel universe. lol

beeing a jerk doesnt render their knowledge wrong, tho.

in fact, i explained multiple times why OPs conclusions are wrong and why he shouldnt refer to the mathematical facts as "magic". all hes doing is defendibg himself, not his point with reasonable counter points.

shouldnt be suprised if he gets mocked at this point......

And again, a strawman.

I never referred to the math itself as "magic".

That you just pulled out of your own posterior in attempt to smear me with a strawman.

Now you are being deliberate in your misrepresentation - you are very dishonest.

I referred very specifically to those who back in the 5.0 day were advertising 1860 as the number to get 1.3 GCD in a broad general manner...

...instead of % break points, and also without any mention of spec differences, gear bonuses, flagship bonuses, etc.

Seriously, you are a very nasty person.

And again, a strawman.

 

I never referred to the math itself as "magic".

 

That you just pulled out of your own posterior in attempt to smear me with a strawman.

 

Now you are being deliberate in your misrepresentation - you are very dishonest.

 

I referred very specifically to those who back in the 5.0 day were advertising 1860 as the number to get 1.3 GCD in a broad general manner...

 

...instead of % break points, and also without any mention of spec differences, gear bonuses, flagship bonuses, etc.

 

Seriously, you are a very nasty person.

 

The problem with you is that you refuse to say what it is. What you are doing seems a flight of fancy and a futile exercise but you present it as if it had some meaning or value. However, what that value would be is entirely unclear.

Your dishonesty lies there. Others have a fair point to question what you're doing because your posting numbers based on a flawed concept. It's not fair to accuse people of using strawman arguments when you do not even explain what the value proposition is of these numbers and what meaning they should have.

I'm convinced after reading this thread that you are in fact trying to do maths based on your anecdotal experience with a dummy. And that experience is based on your pesonal limitations. The issue with that is that people all have their own results because we are all different. And this is the key element why I think what you're doing here is pointless.

Mathematics is about calculations that assume certain circumstances to be equal. That makes it a reference point to which everybody can measure themselves. If you want to take a variable into account, then you have to add it to the equation.

By using your personal experience as the baseline, that means that you take the variables into the equaltion that relate to your performance. And that makes this data rather personal and therefore not useful for others. It takes your mouse, keyboard, playstyle, reaction time etc. into it. But they are likely to not be the same as mine or anyone else.

So when you share these numbers, people assume that you are doing this with a purpose. You want to make a point in some way. I still do not understand what your point is. Your method of gathering data is flawed for general use because it shows only what you achieve. Moreover, by practicing your data might still change because you become faster or better by practicing. So this data is unreliable.

And that's what people tend to dislike about this thread because even though you don't straight out say it, you do present it as if it has some value, even though you never explain what you think this value should be. So if anyone is being dishonest, it started with you as far as I'm concerned. So feel free, to explain what you think people can do with this data of yours. Maybe then people can understand what you're trying to do here.

And again, a strawman.

 

I never referred to the math itself as "magic".

 

That you just pulled out of your own posterior in attempt to smear me with a strawman.

 

Now you are being deliberate in your misrepresentation - you are very dishonest.

 

I referred very specifically to those who back in the 5.0 day were advertising 1860 as the number to get 1.3 GCD in a broad general manner...

 

...instead of % break points, and also without any mention of spec differences, gear bonuses, flagship bonuses, etc.

 

Seriously, you are a very nasty person.

 

Quote from your document linked in your first post:

"Ignore those people, that's the "Magic Number" "Kool Aid" or "Religion" that keeps getting passed around."

might be my reading comprehension since i'm not a native speaker, but it definietly looks like you called the established tresholds "magic numbers" and "religion" and therefor the math they are based on.

if my understanding of that phrase was wrong, i wanna apologize for that.

anyway, if i'm a "nasty" person for pointing out the flaws in your observations and giving you constructive feedback based on actual calculations, hell ill take "nasty" as compliment in this context.

but you should think about your self-perception if the truth triggers you that much.

edit: i figured i should go futher into detail once again:

the main problem is, that you are mixing up two different things here:

people talking about "magical" numbers when asked what the best alacrity value for any given class/spec should be.

while you are right, that there shouldnt be THAT one number regardless of spec, set bonus, guild perk for each GCD treshold, you are missing the fact that there ARE indeed many numbers for alacrity stats each taking exactly these variables into account. informed people will ask for these variables before giving you the exact number you need. some of them can be found here:

https://docs.google.com/spreadsheets/d/1R20FOfta3jXS-zQqs40jcKU8zEou5fujMW_-71-micI/edit#gid=0

you notice the graph for alacrity are in fact 3 different graphs taking the 3 different possible spec alacrity bonus into account, 0%, 3% and 5%, and provide accurate numbers for 1.3GCD and 1.4GCD for all three of them.

all other stats numbers for alacrity guild perk, set bonus etc. can be calculated with the formula on the left by transfering the formula for the "alacrityRating" and puting the desired percentage value (less any alacrity bonus from any source) into the new formula.

point is: just because some uninformed people mostly tell you the 0% treshold numbers without asking/considering what spec you play, doesnt mean there are no fixed numbers for these other cases.

the second thing, which has no relation to the previous point above, is, that you try to show that even these fixed numbers arent relatable because you observed larger GCD intervalls than the usual 1.3/1.4 sec with minor diffrences at the third or forth decimal.

i tried to explain that there are many many many varialbes you gotta take into account when drawing conclusions. the least (im-) possible conclusion is, that the alacrity formula is wrong and there are ACTUAL longer GCDs while using the mentioned treshold values. which could be countered by adding more alacrity, to take the average GCD way below 1.3/1.4, so you dont get above them even with the messured amplitudes.

the more reasonable conclusion would be, that the math is actualy right and the amplitudes/inaccuracies you are observing are casued by technical and human boundries such as input lag, proccessing time at the server, read/write timings from and into the combat log, etc.

that does mean that any additional alacrity over the calculated tresholds wouldnt change anything, since the game iteself would still calculate your GCDs according to the mentioned tresholds and the inaccuracies would still be added after that.

in short: math states, that the game calculates you GCD to 1.4sec baseline, no mether if you are using 1213 stat points or 2000. (for 0% alacrity bonus of any source ofc.)

and since you cant simply deny accurate math, your observations are at least meaningless, if not wrong.

Quote from your document linked in your first post:

"Ignore those people, that's the "Magic Number" "Kool Aid" or "Religion" that keeps getting passed around."

 

might be my reading comprehension since i'm not a native speaker, but it definietly looks like you called the established tresholds "magic numbers" and "religion" and therefor the math they are based on.

 

if my understanding of that phrase was wrong, i wanna apologize for that.

 

anyway, if i'm a "nasty" person for pointing out the flaws in your observations and giving you constructive feedback based on actual calculations, hell ill take "nasty" as compliment in this context.

but you should think about your self-perception if the truth triggers you that much.

 

edit: i figured i should go futher into detail once again:

 

the main problem is, that you are mixing up two different things here:

people talking about "magical" numbers when asked what the best alacrity value for any given class/spec should be.

while you are right, that there shouldnt be THAT one number regardless of spec, set bonus, guild perk for each GCD treshold, you are missing the fact that there ARE indeed many numbers for alacrity stats each taking exactly these variables into account. informed people will ask for these variables before giving you the exact number you need. some of them can be found here:

https://docs.google.com/spreadsheets/d/1R20FOfta3jXS-zQqs40jcKU8zEou5fujMW_-71-micI/edit#gid=0

 

you notice the graph for alacrity are in fact 3 different graphs taking the 3 different possible spec alacrity bonus into account, 0%, 3% and 5%, and provide accurate numbers for 1.3GCD and 1.4GCD for all three of them.

 

Never said otherwise, never questioned it, it's what I'm using as a starting reference. That's why I keep telling you to knock it off with the Strawmans.

 

all other stats numbers for alacrity guild perk, set bonus etc. can be calculated with the formula on the left by transfering the formula for the "alacrityRating" and puting the desired percentage value (less any alacrity bonus from any source) into the new formula.

 

point is: just because some uninformed people mostly tell you the 0% treshold numbers without asking/considering what spec you play, doesnt mean there are no fixed numbers for these other cases.

 

the second thing, which has no relation to the previous point above, is, that you try to show that even these fixed numbers arent relatable because you observed larger GCD intervalls than the usual 1.3/1.4 sec with minor diffrences at the third or forth decimal.

i tried to explain that there are many many many varialbes you gotta take into account when drawing conclusions. the least (im-) possible conclusion is, that the alacrity formula is wrong and there are ACTUAL longer GCDs while using the mentioned treshold values. which could be countered by adding more alacrity, to take the average GCD way below 1.3/1.4, so you dont get above them even with the messured amplitudes.

 

the more reasonable conclusion would be, that the math is actualy right and the amplitudes/inaccuracies you are observing are casued by technical and human boundries such as input lag, proccessing time at the server, read/write timings from and into the combat log, etc.

 

Never said otherwise, again why I keep telling you to knock it off with the Strawmans.

that does mean that any additional alacrity over the calculated tresholds wouldnt change anything, since the game iteself would still calculate your GCDs according to the mentioned tresholds and the inaccuracies would still be added after that.

in short: math states, that the game calculates you GCD to 1.4sec baseline, no mether if you are using 1213 stat points or 2000. (for 0% alacrity bonus of any source ofc.)

 

Never said otherwise, that's why I keep telling you to knock it off with the Strawmans.

and since you cant simply deny accurate math, your observations are at least meaningless, if not wrong.

All I'm doing with my wasted Alacrity Points is I'm using it to make certain the GCD intervats that are occurring below my targeted TIER get firmed upward into the TIER I am targeting from MY CLIENT PERSPECTIVE.

In other words if I consistently get for a 1.4 GCD test:

1.3819 1.4010 1.3987 1.3812 1.3930 1.4011 1.4012 1.3998 1.4202 1.4390 1.4100 1.3923 1.3844 1.3769 1.4041

Adding 1-2 Extra 228 augments has consistently done the following for me:

1.3844 1.3905 1.3834 1.3911 1.4033 1.4001 1.4039 1.4001 1.3888 1.3965 1.3802

In other words, whether it be some UNDISCLOSED RNG, network delay, log writing delay, Server <> Ciient delay - you name it, what I do firms up the results during slowdown periods that would have otherwise resulted in lower GCD intervals from MY CLIENT PERSPECTIVE.

At 1.3 GCD, this effort is much more costly - 200+ points on my Operative Healers BUT 500+ points on my Mercenary Healer and Sorcerer - beyond that suggest by the 6.0 Theory Crafting Charts.

As I state very clearly in my main page and show very clearly in my chart document opening paragraphs - THERE IS NO IN BETWEEN!!! Either you are committed to a TIER or you are not.

So knock it off with your Strawman attempts on that as well....

Now having said the above...

The 300 point difference in what it take to get a CLEAN 1.3 GCD from my client perspective on my Mercenary Healer and Sorcerer Healer versus my Operative Healer suggest that there is some hard coded differential treatment of those three healing specs when working in 1.3 GCD territory.

Chew on that for awhile. Go do your own test...watch how the Mercenary Healer and Sorcerer Healer respond to additive augments VERSUS an Operative Healer after they have been set at the threshold noted in the Theory Crafting Charts.

Note rather interestingly that after the first bug patch, they took away the +2% Alacrity Rating boost in the Sorc Healer set - never mind it wasn't working anyway.

Note also that the +2 Alacrity Rating boost on the two sets available to Merc DPS or Healer don't work as stated either - nor do they stack.

Working as intended?

=8-|

It's not just your client perspective, it's also you as a player perspective. Your skill or performance with the 3 healer classes may not be consistent across all three. So it may not be hard coded in the game but it may be an issue on your side as a player. Maybe it's because the operative suits your style of play more.

That's impossible to say unless we could be sure that you play like a machine effectively. But we cannot use the human factor as a non-variable. That's the difficulty with what you're doing here. You see yourself as consistent and of equal performance on all 3 healers even in playstyle. That's a rather big assumption to make for anyone.

It's not just your client perspective, it's also you as a player perspective. Your skill or performance with the 3 healer classes may not be consistent across all three. So it may not be hard coded in the game but it may be an issue on your side as a player. Maybe it's because the operative suits your style of play more.

 

That's impossible to say unless we could be sure that you play like a machine effectively. But we cannot use the human factor as a non-variable. That's the difficulty with what you're doing here. You see yourself as consistent and of equal performance on all 3 healers even in playstyle. That's a rather big assumption to make for anyone.

Did you even read the test conditions?

Have you even read the thread?

=8-|

I’ve been watching this thread and the other since it’s been posted. I’m certainly no maths expert and I’ve not done break point testing before. But I have done extensive bolster testing and I can say that doesn’t always work the way it should (yes I know it’s a different system and has been fubar most of its life).

What I’m getting at is Bioware aren’t perfect in designing their systems. Real world testing vs theory crafting could have some variations that only show up when you can practically demonstrate them as an experiment.

Instead of you guys just relying on the maths and practical experiments separately and arguing and trying disprove each other , why not join forces and make a fool proof practical experiment that can definitively show what is happening.

Now, I want to state this so no one takes offence, I’m not saying either of you are wrong, but I can’t tell from this discussion who is right and who isn’t because both of you are making good points. I also can’t test because I’ve over 200ms lag and I’m sure that would skew my results.

I’m also sure that I’m not the only one sitting on the fence trying to figure out what is what. And you guys trying to explain it again why you are right and the other person is wrong is frankly getting tiresome.

So as I said earlier, Bioware’s systems are often fubar from how we expect they should work on paper. That opens up the possibility that MrRabbit might have a point or it could mean the design of the experiment is flawed. But we don’t know either unless you guys can work together and design a test that practically shows the results more than just theory,

I’ve been watching this thread and the other since it’s been posted. I’m certainly no maths expert and I’ve not done break point testing before. But I have done extensive bolster testing and I can say that doesn’t always work the way it should (yes I know it’s a different system and has been fubar most of its life).

 

What I’m getting at is Bioware aren’t perfect in designing their systems. Real world testing vs theory crafting could have some variations that only show up when you can practically demonstrate them as an experiment.

 

Instead of you guys just relying on the maths and practical experiments separately and arguing and trying disprove each other , why not join forces and make a fool proof practical experiment that can definitively show what is happening.

 

Now, I want to state this so no one takes offence, I’m not saying either of you are wrong, but I can’t tell from this discussion who is right and who isn’t because both of you are making good points. I also can’t test because I’ve over 200ms lag and I’m sure that would skew my results.

 

I’m also sure that I’m not the only one sitting on the fence trying to figure out what is what. And you guys trying to explain it again why you are right and the other person is wrong is frankly getting tiresome.

 

So as I said earlier, Bioware’s systems are often fubar from how we expect they should work on paper. That opens up the possibility that MrRabbit might have a point or it could mean the design of the experiment is flawed. But we don’t know either unless you guys can work together and design a test that practically shows the results more than just theory,

The problem is Trixie, is I'm not arguing against the Theory Crafting Charts or the Math.

I actually use 'em each release, and make local adjustments and inform where my characters are at.

It is OTHER people perceiving an argument - and ragging on me.

In all honestly, the only issue I have with the latest Theory Crafting paper is in regards to:

Shield

Absorb

...and Powertech Shield Tech Tanks

Previous iterations of Theory Crafting documents going all the way back to 3.0 made distinctions for Powertech Tanks apart from Immortal Juggernaut Tanks and Darkness Sin Assassin Tanks.

That's about it....

I suspect this may be why I'm seeing quite a few new Powertech Tanks set up with:

1000 Shield Points

2500 Absorb Points

instead of let's say:

2100 Shield Points

1400 Absorb Points

It's not a big deal in SM OPs, but I've run into quite a few in HM OPs where I AND the other healer get tired of focusing 80-100% heals on him, an little or no heals for rest of the raid.

So I end up asking them, "How are you using your Absorb if your Shield is not triggering for a lack of Shield points?"

And when they argue, I then point out: "You can boost Shield rating by a few % in rotation, but it's Absorb rating you can boost by 30% via rotation. Please reconsider..."

=8-|

 

In other words if I consistently get for a 1.4 GCD test:

 

1.3819 1.4010 1.3987 1.3812 1.3930 1.4011 1.4012 1.3998 1.4202 1.4390 1.4100 1.3923 1.3844 1.3769 1.4041

 

Adding 1-2 Extra 228 augments has consistently done the following for me:

 

1.3844 1.3905 1.3834 1.3911 1.4033 1.4001 1.4039 1.4001 1.3888 1.3965 1.3802

 

thats a ballsy claim. what you are saying right now is, that the game actualy changed its mechanics from hard roundig up to the first decimal, to actualy calculating the GCD precisly to the 3rd or 4th decimal for instant abilities.

otherwise you wouldnt see any significant changes in intervals at 3rd or 4th decimal at your end by adding some alacrity.

this wouldnt just affect your client side anymore...

that the devs reworked a fundamental system which has been in place for almost a decade seems unlikly to me.

but hey, provide some screenshots or better upload your parses to parsely.io where you are constantly affecting the intervals of GCDs (of INSTANTS) by adding alacrity without hiting the next known treshold and i will believe you.

just noticed: what makes this even more unlikely to work is the fact, that the low amplitudes apparently didnt got affected by more alacrity at all. statistically the shortest intervalls in the second sample should be significant lower than the shortest intervalls with less alacrity in the first parse. in fact only the high amplitudes got brought down.

seems odd from a mathematical view, but could be the small sample size.

anyway, even if you could affect only the client side by adding stat points, what would be the point of adjusting your side if all that matters for dps/gameplay are calculation on the server side?? whats the point of the post then? why provide this information to other people? i dont get it anymore...

just wondering: IF you knew the tresholds from the beginning, why are ALL of the shown specs in your document waaaaay above them with their alacrity stat? we are not talking about 200-300 points to smooth out (to be proven) amplitudes in GCDs anymore. some of them are 500, 800, or even 1k alacrity points above the last know treshold (or 200-500 below the next). in the same table, you state that these exact characters are about....20-30 alacrity points above the sweet spot. how does that work out? they arent even close to any treshold at all!?

and now the interresting part:

IF what you are saying is true:

why even bother with GCD tresholds at all?

if you can actualy change the GCD to the point where you are constantly getting an average calculated to the 4th decimal, why are more points wasted?

it would just mean you could lower the average GCD intervall even more by adding more points and we would need to talk about even more exact tresholds and calculate at which point the soft diminnishing return of alacrity kicks in to get new sweet spots like it works for crit and mastery already. but i dont see these calculations anywhere, what drives you to the conclusion that your characters wasted alacrity points then?

again: i´m not trying to mock or prove you wrong whatever it takes. first your statements and claims just didnt make sense to me (and honestly they still dont yet), but i´m really interrested if you found an actual game changer here right now. but to prove that point, we gotta go through the hard facts we (believe to) know and prove them wrong. unfortunately that didnt happen at all to this point.

thats a ballsy claim. what you are saying right now is, that the game actualy changed its mechanics from hard roundig up to the first decimal, to actualy calculating the GCD precisly to the 3rd or 4th decimal for instant abilities.

otherwise you wouldnt see any significant changes in intervals at 3rd or 4th decimal at your end by adding some alacrity.

this wouldnt just affect your client side anymore...

 

that the devs reworked a fundamental system which has been in place for almost a decade seems unlikly to me.

but hey, provide some screenshots or better upload your parses to parsely.io where you are constantly affecting the intervals of GCDs (of INSTANTS) by adding alacrity without hiting the next known treshold and i will believe you.

 

just noticed: what makes this even more unlikely to work is the fact, that the low amplitudes apparently didnt got affected by more alacrity at all. statistically the shortest intervalls in the second sample should be significant lower than the shortest intervalls with less alacrity in the first parse. in fact only the high amplitudes got brought down.

seems odd from a mathematical view, but could be the small sample size.

 

anyway, even if you could affect only the client side by adding stat points, what would be the point of adjusting your side if all that matters for dps/gameplay are calculation on the server side?? whats the point of the post then? why provide this information to other people? i dont get it anymore...

 

just wondering: IF you knew the tresholds from the beginning, why are ALL of the shown specs in your document waaaaay above them with their alacrity stat? we are not talking about 200-300 points to smooth out (to be proven) amplitudes in GCDs anymore. some of them are 500, 800, or even 1k alacrity points above the last know treshold (or 200-500 below the next). in the same table, you state that these exact characters are about....20-30 alacrity points above the sweet spot. how does that work out? they arent even close to any treshold at all!?

 

and now the interresting part:

IF what you are saying is true:

why even bother with GCD tresholds at all?

if you can actualy change the GCD to the point where you are constantly getting an average calculated to the 4th decimal, why are more points wasted?

it would just mean you could lower the average GCD intervall even more by adding more points and we would need to talk about even more exact tresholds and calculate at which point the soft diminnishing return of alacrity kicks in to get new sweet spots like it works for crit and mastery already. but i dont see these calculations anywhere, what drives you to the conclusion that your characters wasted alacrity points then?

 

again: i´m not trying to mock or prove you wrong whatever it takes. first your statements and claims just didnt make sense to me (and honestly they still dont yet), but i´m really interrested if you found an actual game changer here right now. but to prove that point, we gotta go through the hard facts we (believe to) know and prove them wrong. unfortunately that didnt happen at all to this point.

Again I am merely reporting what the Test Dummy shows via Starparse Combat Log.

Again I am merely reporting what my characters look like after I put in excess alacrity points to make sure from a Client perspective that I get a "clean" GCD.

I am NOT questioning the math...the theory charts...the GCD tiers, the rounding practice...

I am not making any "ballsy" claims...

if YOU don't like what I'm posting, as I keep telling everyone around me: "Do your own testing on the Test Dummy."

It's on your own personal starship across all characters as a legacy perk.

Starparse is free.

How hard can it be to simply do your own testing?

=8-|

Anyhow, I'm probably going to strip 300-400 alacrity points out of both my Sorcerer and Mercenary healers.

Both triggered a new Force Regen Rate and Heat Dissipation Rate at 3700 points...in order to get what I view as a "clean GCD".

BUT, in just a few ops, both felt nice and smooth BUT it also felt like they were throwing Tissue Paper heals instead of Kolto Bombs and Blessings of Life.

So I'm going to drop them both right down to where my Operative healer is - much closer to the chart - and the associated Force Regen Rate / Heat Dissipation Rate so they can have 300-400 points back in Critical Rating.

For both at 1.3 GCD in accordance with the charts, I will just have to accept the occasional 1.4 GCD cool downs that occur during rotation on the client side of things.

=8-(

 

Did you even read the test conditions?

 

Have you even read the thread?

 

=8-|

Oh good, I got your attention finally. So here's what I did. I followed your set up. So got on starparse, got the test dummy and whacked it with my basic lightsaber attack as you described and did a few runs of it. My sorc healer with 3235 points in alacrity for a stat of 15.48% for reference. This is 27 points over the "magic" 3208 number.

What I did is I took the timestamps for each impact of my skill and measured the GCD between each hit. Of course if I averaged them then I would end up where you would end up, but what's actually much more interesting is to see how the GCD operates from hit to hit.

What I noticed is that if I'm focused and hit the no. 1 key with my keyboard as much as I can then I am constantly around 1.3 GCD. But when my focus lapses then I instantly get significant differences. So I can get a series of mostly 1.294-1.304 seconds and then suddenly get one or two that are slower instantly going to 1.38x or even 1.4xx occasionally. I also got a 1.109 in there which I thought was interesting.

I noticed that there is a difference between using the mouse and the keyboard. I noticed a difference between how focused I was on hitting that button or slightly let go of that focus.

And if I can hit something like 1.109, who knows, there may be a latency or delay issue between SWTOR and Starparse and your inputs somewhere.

Bottom line, after having tested your method, I can see the flaws in the method due to my consistency and even hitting impossible numbers so there must be an issue somewhere technically as well.

That's the sort of thing you notice when you look beyond the averages.

Again, human factor and technical variables that exist out there that will be different for each individual. I just noticed that when I focus and stay on top of the key then I hit the 1.3 GCD consistently and one tiny lapse and it's off. It's perfectly normal to have such human deviation because as far as I know our timing isn't down to the micro second. The devil is always in the details and just taking an overall average doesn't tell you the details.

So really, by using your method I do not need 500 more points to get the 1.3 GCD, I just need consistency on my part. And that's been the point since the beginning.

Okay, so let me get this straight.

U r trying to figure out for each class/spec what alac rating they need in order to hit 1.x GCD w/in the log itself, right?

Here is the issue with that: as u said, even with zero alac rating, u get activations that r over 1.5 secs. That comes from the myriad type of lags out there: input lag, server lag, u name it.

As a result of that, u will always be a little slower than u r supposed to be, it is exactly for this reason that when u r doing a static rotation, and there r abilities with original CD like 15 or 12 secs, u will see that the ability activation b4 u use them again will have less than 1.X secs left on the CD (1.X being whatever GCD u r running).

It is an undeniable fact that u get a major boost in APM once u jump from just shy of 1.x to over the breakpoint for 1.x GCD. That alone is proof enough that what u r doing here is fundamentally flawed.

Oh good, I got your attention finally. So here's what I did. I followed your set up. So got on starparse, got the test dummy and whacked it with my basic lightsaber attack as you described and did a few runs of it. My sorc healer with 3235 points in alacrity for a stat of 15.48% for reference. This is 27 points over the "magic" 3208 number.

 

What I did is I took the timestamps for each impact of my skill and measured the GCD between each hit. Of course if I averaged them then I would end up where you would end up, but what's actually much more interesting is to see how the GCD operates from hit to hit.

 

What I noticed is that if I'm focused and hit the no. 1 key with my keyboard as much as I can then I am constantly around 1.3 GCD. But when my focus lapses then I instantly get significant differences. So I can get a series of mostly 1.294-1.304 seconds and then suddenly get one or two that are slower instantly going to 1.38x or even 1.4xx occasionally. I also got a 1.109 in there which I thought was interesting.

 

I noticed that there is a difference between using the mouse and the keyboard. I noticed a difference between how focused I was on hitting that button or slightly let go of that focus.

 

And if I can hit something like 1.109, who knows, there may be a latency or delay issue between SWTOR and Starparse and your inputs somewhere.

 

Bottom line, after having tested your method, I can see the flaws in the method due to my consistency and even hitting impossible numbers so there must be an issue somewhere technically as well.

 

That's the sort of thing you notice when you look beyond the averages.

 

Again, human factor and technical variables that exist out there that will be different for each individual. I just noticed that when I focus and stay on top of the key then I hit the 1.3 GCD consistently and one tiny lapse and it's off. It's perfectly normal to have such human deviation because as far as I know our timing isn't down to the micro second. The devil is always in the details and just taking an overall average doesn't tell you the details.

 

So really, by using your method I do not need 500 more points to get the 1.3 GCD, I just need consistency on my part. And that's been the point since the beginning.

I know this is a grey area, but a macro to spam the number one key would eliminate the possibility of human error or lapse in concentration. Just saying

Again I am merely reporting what the Test Dummy shows via Starparse Combat Log.

 

Again I am merely reporting what my characters look like after I put in excess alacrity points to make sure from a Client perspective that I get a "clean" GCD.

 

I am NOT questioning the math...the theory charts...the GCD tiers, the rounding practice...

 

I am not making any "ballsy" claims...

 

if YOU don't like what I'm posting, as I keep telling everyone around me: "Do your own testing on the Test Dummy."

 

It's on your own personal starship across all characters as a legacy perk.

 

Starparse is free.

 

How hard can it be to simply do your own testing?

 

=8-|

ok so i gave you the chance multiple times now to just link the uplpaded parses you"ve done or provide some specific evidience that your claims are true. you denie that by telling me to test it myself. thats npt how any of this works. if you claim s.th. gamechanging like that you are in charge to prove it, as you already tested and observed it as you claim.

i mean, in your words: how hard can it be to upload the parses instead of copy/pasting the numbers manually to the thread?

sorry but i give up at this point and just dont believe you got any hard evidience for your claims anymore.

i can just hope that everybody else doesnt believe you as well and this thread is dead soon. i will start by not bumping it up with seemingly useless replies anymore.