New Healer Job Request

[LilimoLimomo]

Just wanted to say that combat macros like your healing and damage one are generally not recommended, FFXIV doesn't handle them very well it will typically introduce small delays in skill execution, which over time can add up, that you wouldn't have it you executed them manually (or even worse it won't fire if there is a lag spike).

I've heard this kind of thing before, but my assumption has always been that the people who are saying that don't know how to make a macro that simulates the "buffering" that pressing normal buttons has. But perhaps I'm mistaken; I just never see others writing macros that contain the "buffering" that mine do. Is there by chance a source for this claim? I'm guessing if this is the case that someone did some tests, and hopefully they recorded their methodology as part of that? I'd love to review something like that.

And if it doesn't exist, no pressure. I should actually probably just do my own tests anyway. I have a feeling I know what I'll find, but maybe I'll get proof that I'm wrong!

All of that said, all macro use is a compromise. You're losing something (for example, a certain amount of buffering) in order to get something else (like accessibility). And whether the juice is worth the squeeze is going to be relative to each player's needs. If you're a player like me who can only use certain button combinations without discomfort, there's no way to reasonably play most classes without macros. And that's a time when we should be encouraging macro use, because not everyone has the same needs, not everyone has the same ability. At some point for some players, being able to play comfortably is going to be worth the theoretical loss of some DPS. Especially if they wouldn't have been doing that theoretical DPS anyway because without the macros they would perform worse.

All of that said, I do agree that if there are downsides to macros, it's important to note those downsides so that players who are interested in them can make an informed decision on whether such macros are worth it to them.

If I remember, I'll do some tests tonight and report back the results!

[Sebazy]

I've heard this kind of thing before, but my assumption has always been that the people who are saying that don't know how to make a macro that simulates the "buffering" that pressing normal buttons has.

Using macros has a very significant impact on your overall cast throughput, it's been a long time since I mathed it myself, but from memory it was somewhere in the region of 10-20% all told. It's important to stress that this only fully applies if you're keeping your GCD rolling through. If your mobility impairment is consistently preventing you from rolling your GCD then that changes things. At that point the only real issue is that you'll generally be paying a small time penalty any time you try to reactively bop someone with a heal or other ability through macros due to inevitable downtime.

In BCOB I prog raided as SCH using macros to force my fairy to embrace the MT whilst I did just about anything else. As part of my preparations for SCOB, I experimented with other ways to force embraces including using a free spinning mouse wheel, G25 pedals and a stupidly expensive Corsair keyboard that had a ton of macro keys.

To my horror, I found that using in game macros was quite literally costing me somewhere in the reason of 2-3 full GCDs per minute from memory. Switching to pedals and running the macros on the keyboard instead of in game was a mighty throughput improvement.

As such, my recommendation is to go find a target dummy and experiment with what you have at your disposal. If you can find someone to log your throughput, even better.

[LilimoLimomo]

If I remember, I'll do some tests tonight and report back the results!

Alright, I did some tests tonight! Even got some footage of them that I can upload to a YouTube channel as soon as I figure out how to make one for it. Some of what I said was accurate, but I also learned some things I didn't know!

First, so you understand my methodology, I tested using two macros on my level 90 BLM with 1196 Spell Speed (2.36 recast):

/macroicon "Blizzard"
/macroerror off
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"
/ac "Blizzard"

and

/macroicon "Scathe"
/macroerror off
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"
/ac "Scathe"

Now, for the results:

First, there is one transparent benefit of not using macros: increased action queue time.

So for example, if I want to cast two consecutive Scathe attacks, my second Scathe will be queued up if I press the standard Scathe button when my GCD cooldown is just over 3/4 complete; when the GCD cooldown completes at 4/4, the second Scathe will immediately be cast. However, if I press my Scathe macro at 3/4 GCD cooldown, the action will not be queued. For the Scathe macro, the earliest I could queue it was at roughly 7/8 of the GCD cooldown. So in this case, the non-macro button had roughly twice as much of a queue window.

I also discovered that the queue window differed — for both macros and non-macros — depending on what kind of action was previously cast. Specifically, the queue window becomes shorter whenever the preceding action has a cast bar; this is presumably because you aren't just waiting for the GCD to return, but you are waiting for the previous action to be cast. I'm guessing this is just animation lock, although there's a part of me that wonders if it may be something else because this effect also differed depending on whether or not a macro was used.

So for example, if I want to cast two consecutive Blizzard attacks, my second Blizzard will be queued up if I press the standard Blizzard button when my cast bar has roughly 0.3 seconds remaining; then when the cast finishes, the second Blizzard will start charging. However, if I press my Blizzard macro at 0.3 cooldown, the action will not be queued. For the Blizzard macro, the earliest I could queue it was around 0.1 seconds of the cast remaining.

So to summarize this section, I'd recommend trying macros as they still allow actions to be queued, but if the amount of queue window you're getting from them is below what you need to feel comfortable and/or effective, know that using a non-macro will result in at least double the queue window!

Next, I did a test to determine whether macros actually resulted in lost GCD's. Earlier in the day I had seen the video that the Balance Discord links to in order to demonstrate that your GCD can slip when using macros: https://youtu.be/VgpUCgExvbA

When I initially watched that video, I was skeptical. I noticed how the player seemed to be mashing the macros, which you should never do, as pressing the macro again resets rather than letting it finished. So that was one red flag. And while the video didn't show the macro being used — which is vital to the methodology of the experiment — my best guess was that the macro being used probably looked like this:

/macroicon "Glare"
/macroerror off
/ac "Glare"

...because if it only had a single line of Glare, that would explain why the player felt the need to mash the button. If that was true, it would mean the macro wasn't doing much, if any, queuing, which might explain why the GCD was slipping.

However, even with those suspicions, I'll be honest: after my unexpected discovery that macros had shorter queue windows than non-macros, I was worried that my own results might mirror those in the video. So I was a bit surprised when I found out that no, my macros were not resulting in any GCD loss!

My methodology was as follows: I went to the level 90 Stone Sky Sea and started The Final Day (normal). Then I cast Blizzard repeatedly until 1 minute had passed on the timer. Afterwards, I saved a video of the attempts and manually counted the number of Blizzard spells that hit the target before the timer had been reduced from 3:00 to 2:00.

The only difference between my two tests was:

• For the non-macro Blizzard test, I took full advantage of the larger queue window in an effort to ensure no slippage between casts. I aimed to press the button again at roughly 0.25 seconds remaining of each cast.

• For the macro Blizzard test, I needed to press the button later because of the smaller queue window, so I pressed it at the time I am used to pressing the button when playing a caster: a split-second before the cast bar fills (kind of like a rhythm game).

To my surprise, both tests resulted in 25 casts.

However, what was even more surprising was that the non-macro test's final Blizzard hit with 0 seconds to spare, while the macro test's final Blizzard hit with 1 second to spare. Somehow, the slippage actually occurred on the non-macro side. I have absolutely no explanation for this, because in theory, the best case scenario seems like it should have been equal. Perhaps this suggests that some other mechanic is happening that none of us are aware of?

If I hadn't done it myself, I'd be very skeptical of what I was reading here. Which is another reason why I'll be uploading my videos as soon as I'm able to, so that anyone who's interested can check my work. If there were flaws in the test, I hope someone can point them out! And if there aren't, then I hope to find a way to spread this knowledge to the rest of the community. But before any of that can happen, we'll need some peer review!

Anyway, that's my update for today. If this topic interests you, I hope you'll take a look at my videos when I find a way to upload them to verify what I've written here!

[FuzzyJCats]

Lilimo, interesting! It makes sense then to make all GCDs macros like the above, but I think the problem may be when you try to combo 2 actions in 1 macro? Again, all of this can be done away with if SqEnix implements my proposal ofc (I intentionally made it OP). Joking aside, I hope this post will open the developer's eyes to challenges that people with disabilities, like me, have.

That being said, the community is really kind, I only need to say I have slow reflexes as I die the 100th time in a Nier raid, and they STILL rez me, even though their MP/dps/healing spells can otherwise be used more effectively. Even so, it feels awful to feel like you're holding back progress.

[FuzzyJCats]

Alright, I did some tests tonight! Even got some footage of them that I can upload to a YouTube channel as soon as I figure out how to make one for it.

Anyway, that's my update for today. If this topic interests you, I hope you'll take a look at my videos when I find a way to upload them to verify what I've written here!

I would LOVE to watch your video, let me know when you post it on YT!