Evidence that Macros Do NOT Cause GCD Delay or Lost Casts

[LilimoLimomo]

You need to actually do way more than 25 casts to test the real difference

The video used as evidence showed a difference in only 23 casts. I did more casts in the same window, which should have shown a more pronounced effect due to the increased number of casts. However, my tests showed the opposite.

send this to the Balance theorycrafters

I had never heard there were Balance theorycrafters who would be important to send this to. Tell me a way to respectfully get in touch with them and I'll be happy to.

or are you worried that your test doesn't actually hold up under scrutiny ?

From two different parts of my original post:

But I also don't just expect you to take my word for it. I encourage you to be a critical thinker, to look over my data and my methods. If you see problems, please, let me know!

If I hadn't done the tests myself, I'd be very skeptical of what I was reading here. Which is another reason why I want to encourage anyone who's interested to check my work. If there were flaws in the test, I hope someone can point them out!

[Rein_eon_Osborne]

[...]I had never heard there were Balance theorycrafters who would be important to send this to. Tell me a way to respectfully get in touch with them and I'll be happy to.

Their #rules section have a minutiae that specifically encourages theorycrafting & "challenging the meta" from its members.

Go try your luck.

[AvoSturmfaust]

[QUOTE=LilimoLimomo;6333965]The video used as evidence showed a difference in only 23 casts. I did more casts in the same window, which should have shown a more pronounced effect due to the increased number of casts. However, my tests showed the opposite.


"I had never heard there were Balance theorycrafters who would be important to send this to. Tell me a way to respectfully get in touch with them and I'll be happy to."


Balance Discord, Top Channel Selection, there you can selet the Theorycraft Battle Channels, its your best place for it

[Darkobra]

You need to actually do way more than 25 casts to test the real difference, send this to the Balance theorycrafters, or are you worried that your test doesn't actually hold up under scrutiny ?

Feel like taking up the mantle and doing more than 25 casts or would that interfere with your forum time?

[Espon]

It's more noticeable when performing oGCD abilities as macros do not queue up if you're locked in frames of animation.

For example as Paladin, if I want to put Intervention on the other tank with a macro, I usually have to mash the button a few times for it to go off, even though the macro has the skill listed, and I'm be in the middle of using another ability, where as hitting Holy Sheltron will queue it up to use at the next available moment, making it always go off. Both of these are off the global cooldown, yet the macro is more cumbersome to use.

There is also the fact that you cannot use the /wait command with fractions of a second, and the GCD is 2.5 so trying to use multiple skills in a row with a macro will always be a DPS loss as you can only have skills set to go off every 3s seconds.

[LilimoLimomo]

It's more noticeable when performing oGCD abilities as macros do not queue up if you're locked in frames of animation.

For example as Paladin, if I want to put Intervention on the other tank with a macro, I usually have to mash the button a few times for it to go off, even though the macro has the skill listed, and I'm be in the middle of using another ability, where as hitting Holy Sheltron will queue it up to use at the next available moment, making it always go off. Both of these are off the global cooldown, yet the macro is more cumbersome to use.

Do you have /ac "Sheltron" listed multiple times in your macro? The more rows of that you have, the greater your queue window will be. With a macro like that, you shouldn't need to mash the button (in fact, mashing the button will make a macro like that perform worse) provided that you are accustomed to the rhythm of when your animation lock lifts. But if you're less in-tune with the proper timing, then non-macros can be really helpful because of their larger queue window.

There is also the fact that you cannot use the /wait command with fractions of a second, and the GCD is 2.5 so trying to use multiple skills in a row with a macro will always be a DPS loss as you can only have skills set to go off every 3s seconds.

Yeah /wait macros are a whole can of worms. There are certainly use cases for them where they can be helpful, but most of the ways we as players would like to use them don't work great due to the limitation you mentioned!

[Iscah]

And as I've demonstrated, the above is untrue: macroing GCD's does not add up to lost casts over time.

You've only proved something that isn't the actual point of the warnings against macros. Why would people be making a macro to do nothing but a single GCD spell? Unless you really want your Fire/Fira/Firaga spell naming back or something, in which case I salute your dedication.

The concern with macros has always been that people think they could use macros to set up an entire sequence of attacks with a single button press, with "wait 2.5 seconds" commands in between them – which all actually turn to 3 seconds, thus the warnings that it will cause you to drift.

[Mr_Gyactus]

You've only proved something that isn't the actual point of the warnings against macros. Why would people be making a macro to do nothing but a single GCD spell? Unless you really want your Fire/Fira/Firaga spell naming back or something, in which case I salute your dedication.

The concern with macros has always been that people think they could use macros to set up an entire sequence of attacks with a single button press, with "wait 2.5 seconds" commands in between them – which all actually turn to 3 seconds, thus the warnings that it will cause you to drift.

with this message you have killed the hopes of OP to win a noble prize. Unfortunately it is something that had to be done.


Cloak of shame for everyone else.

[Jeeqbit]

Yeah /wait macros are a whole can of worms. There are certainly use cases for them where they can be helpful, but most of the ways we as players would like to use them don't work great due to the limitation you mentioned!

/wait only working on whole numbers does mean a DPS loss but of course it may be the only option for certain disabilities.

There is a way to get around the issue of wait not working on fractions of a second, which is to do (assuming your GCD speed is above 2 seconds) -

/action "Combo Action 1" <wait.2>

Then repeat the second combo action lots of times

/action "Combo Action 2"
/action "Combo Action 2"
/action "Combo Action 2"
/action "Combo Action 2"
/action "Combo Action 2"
/action "Combo Action 2"

And eventually one of them works because it lands on whatever your GCD speed is at instead of losing almost a second like with a 3 second wait.

However, there are issues with this. If you press another macro, it cancels the macro. If you weave an ability or do mechanics, it can cause it to be unreliable or delay the GCD. And this is why even with this method, macroing combos is unreliable.

But I think the issue with macroing is more about combos or chains of attacks/abilities than single GCDs or abilities, because as you point out with your tests, that doesn't make much practical difference.

Why would you want to macro a single GCD? There is a reason which is that when you have /autofacetarget on, you target enemies for attacks (required), but also players for heals (not required). So you can make it disable /autofacetarget when healing but not when attacking by macroing single GCDs or abilities.

[Hasrat]

This appears to be a major flaw in testing, or in interpretation of the original problem.

You're testing by stacking the same spell over and over into the macro. Nobody is trying to macro in this way. When players are attempting to create combat macros, they're trying to put together a whole chain of skills.

The queuing is absolutely a problem here, and you already did demonstrate that there are differences in queuing. But what you briefly noticed but failed to recognize the implications for, is the "wait" command cannot increment in fractions of a second. Any player creating any combat macro that uses a "wait" command is naturally introducing further drift into their rotation, and given how the premise of combat macros is to chain several skills, several "wait"s will naturally need implemented.

If you want to actually test this, you need to create an actual viable macro, one that unaware players would likely attempt to create to save space or make combos easier. And then there's the additional consideration of whether or not players intend to macro their entire kit, or if they're using specialized niche macros, and how those interact with GCD of non-macro'd skills.

Idk. I think you were fully aware and understand what you were looking at, but the misstep was in designing a macro for one singular spell. That's never the use case that the macro argument is up against.