Evidence that Macros Do NOT Cause GCD Delay or Lost Casts

[Saraide]

Hold on actually, I looked at the way car manufacturers test stuff and it might be a really good analogy:

https://en.wikipedia.org/wiki/Volksw...ssions_scandal

[LilimoLimomo]

Even a lay person would realize that fights are not just a target dummy

You're right, nobody would actually propose that attacking a target dummy is the same as doing a fight. Which is why nobody in this thread has done that.

But you should also realize that you don't need to be in a full-on fight to figure out the mechanics of how using a single action works. Press a button against a target dummy. Press that same button in a real fight. Try this for a bit and you'll eventually learn that the buttons work the same in both situations.

Correct me if I'm wrong, Saraide, but I get the impression that you are focusing on a different question than the one this thread is:

1. This thread focuses on whether there is anything innate to the macro system that will result in delayed APM regardless of user performance.
2. You seem to be focusing on whether use of the macro system increases the likelihood of user error.

If we conflate these two questions, we end up running around in circles because while these are related, they are fundamentally different and thus have different answers.

For the first question, you can perform some simple tests to figure out the answer.

But based on your very first reply to this thread, that's not what you're interested in talking about. You want to talk about the second question.

There's a reason I haven't focused on that question, and that's because it's trivially easy to answer: it depends on the player.

Because even if macros did cause significant action loss, there are plenty of FF14 players who would still benefit from macros. For example, it's not uncommon for players to have trouble remembering where 30 different actions are on their controller, and when you have trouble finding an action it can end up taking precious seconds to find and use it. Some players will be able to improve and readily access every action in their arsenal, but others will hit a wall. If the use of macros can make actions easier to find for this kind of player, such that they can press the action they want with less delay, then even if the macro loses some frames upon press, the result can be that the player's action per minute improves.

However, there can also be the exact opposite result. A player could find macros confusing, could forget which 2 actions are bound to the same button, and thus macros could actually make their play worse, even if macros don't innately have any action loss.

Which is to say, for some people macros will improve play, for others it will hinder it.

What I've learned over and over again throughout the course of my life is that people have different capabilities. Each of us have different strengths and different weaknesses in a variety of areas, and because of that we have different experiences and different challenges. That's why instead of being prescriptive, I instead encourage everyone to give macros a try and see what the impact is. Then depending on their individual results, they can each decide to what degree — if any — they want to keep using macros.

I hope this helps to bridge the gap in our discussion and help us see eye-to-eye!

[LilimoLimomo]

Some spells such as Blizzard or Holy still have same cast time and recast time.

I identified the source of hotbar copy & set macros. It uses the same method as in the Macrology text.
【FF14】マクロでも先行入力を可能にする方法 (FF14 How to enable queue input using macros)
A macro in the video is like this. It was posted in 2015. It contains a removed action.

/micon "Fast Blade"
/ac "Fast Blade" <t>
/crosshotbar copy PLD 1 GLA 1
/crosshotbar set "Fast Blade" 1 RA3 <wait.1>
/crosshotbar set "Savage Blade" 1 RA3 <wait.3>
/crosshotbar copy GLA 1 PLD 1

This type of macro guarantees no GCD delay in exchange for additional presses. Unfortunately this method cannot be applied to many of my macros for GCD actions. I heavily use mouseover macros. These cannot be used without macros and pseuedo-queue is the only available option.

For some reason the URL for this video is set to reload 10+ times, so to make it easier for anyone else who wants to view it, here's an alternate link that just gives you the video without the reloading: https://www.youtube.com/watch?v=Q-LdICjdnfQ

This video is interesting; using the double-tap to get access to 2 different actions with 0.5 seconds of queuing. Personally, I probably wouldn't use this specific macro myself, because Savage Blade is only accessible for such a short window, meaning that if the enemy moves away after you perform Fast Blade then you might not be able to cast Savage Blade before the button is removed from your crosshotbar. But I wonder if there might be other actions I could use kind of technique with? Perhaps something more time-limited like Ninja's Mudra?

But that's really not the important part; the important part is that this person was creating macros that allowed 0.5 seconds of queuing via a double-tap 9 whole years ago! Credit where credit is due! ^^

I found another video showing GCD delay.
【FF14】GCD遅延検証Ping40代とスキルマクロ(FF14 GCD Delay Inspection: 40ms range ping and macros)

It compares 4 cases side-by-side by repeated casting Jolt on a striking dummy. It was posted in 2023.
(1) average ping 41 normal action
(2) average ping 3 normal action
(3) average ping 3 single line macro
(4) average ping 3 repeated line macro
The video is paused at 0:07. Before the second cast (3) caused a delay. After an explanation, pause is canceled at 1:54. There are no significant delay other than in (3) after 50 casts.

It's so cool seeing examples of others doing these kinds of tests. It makes it feel more like a community of folks who share similar interests and are all using the scientific method to gain a better understanding of those interests. ^^

Thank you for explaining both of these videos! I know a bit of Japanese but I don't have much vocabulary for a lot of this technical stuff, especially kanji; so I wouldn't have been able to understand these without you!

[Mikoko_Miko]

Thank you LilimoLimomo. I updated the links.

It is fortunate that the scientific method is universal. I have high hopes for AI technology that can overcome language barriers. I tried NoteGPT to summarize the second movie. It gave me a satisfying result. Original Japanese caption was translated well. The word macro was used in three senses here; single line macro, repeating line macro and macros in general. These must be distinguished from the context.

Summarized by NoteGPT

Summary
The video discusses GCD delay testing in FF14, emphasizing the effectiveness of pre-input techniques and the challenges posed by macros in maintaining optimal timing.
Highlights
GCD consistency is maintained despite slight ping variations.
Pre-input can compensate for higher ping, enhancing performance.
Macros can’t utilize pre-input as effectively as manual inputs.
Stacking the same skill in macros helps achieve pseudo-pre-input.
Macros introduce delays, impacting GCD fluidity.
Use macros wisely; they are helpful in specific situations.
Manual button mashing may be necessary for optimal macro function.
Key Insights
Understanding GCD mechanics is crucial for maximizing performance in FF14. Players should be aware of how timing affects their actions.
Pre-input techniques are powerful tools for managing latency, allowing players to mitigate ping-related issues and maintain smooth gameplay.
The limitations of macros highlight the importance of player skill in executing abilities effectively, particularly in high-stakes situations.
Stacking skills in macros can enhance performance but requires careful management to avoid increased delays.
Recognizing the trade-offs of using macros is vital; they can enhance convenience but may lead to performance issues if overused.
The necessity for manual input for certain actions emphasizes the balance between automation and hands-on gameplay.
Ultimately, mastering both macros and manual techniques can lead to a more robust and effective gameplay experience.

[Rin_Sato]

The reason I think a fight would be a good test, is because the whole reason you'd ever macro your actions is so that you could use them in a fight, and not so you can whack a stationary dummy.
As soon as you need to move around and dodge aoes, does the macro still work just as well? What are the issues that may arise?

I believe the reason people in general don't macro their actions is simply that it isn't worth the hassle and is worse.

"Too many variables."

This is because you are, I am sorry to say, not a consistent player.
If you play optimally you will be hitting the same gcd, in the same point in the fight (granted boss rotation), every single time.
There will be no variables aside from time and uptime.

If you can't do this I don't see any reason to even argue against what you're saying, because clearly it doesn't even matter to you if you are losing uptime when playing normally, or doing random rotations.

And I swear to god if somebody mentions some niche cases of ex. multiple misaligned raidwides and dots, please note that the same rotation can be replicated without the buffs (if you even cared about matching it exactly (it may not always matter for uptime), because all I am saying to look at is **Action count** and **Encounter Duration**, which depends on **Skill speed**. (You can simply look at a limited time window and compare the action count within it, provided it is long enough, so before anyone bothers making the extremely bad reply about needing to have exactly the same encounter time, you don't, it is sufficient to manually review a (sufficiently large) portion of the encounter.)

[LilimoLimomo]

The reason I think a fight would be a good test, is because the whole reason you'd ever macro your actions is so that you could use them in a fight, and not so you can whack a stationary dummy.

When scientists perform experiments, they do them in controlled settings that don't perfectly mirror the real world. I genuinely encourage you to think about why they do that, and perhaps to do some research on the topic. Not only is it pertinent to this specific issue, but to the field of science as a whole.

As soon as you need to move around and dodge aoes, does the macro still work just as well? What are the issues that may arise?

Macros function the same regardless of context; they run at 1-line-per-frame unless instructed to do otherwise. So your macro itself won't function differently; a valid target is a valid target. So if we're approaching this question in terms of "how do macros function", then there's no meaningful difference here.

But if you're talking about the possibility of user error, then that's actually something that can be helpful for a person to test, both on dummies and in more nuanced fights. For example, there's more going on in a fight so there's more to pay attention to, and maybe the player is also being affected by stress, etc. Under those circumstances, will the player be able to use the macro as well?

But there's no test that can give us a generalized answer to that, because it depends. It depends on the player. It depends on the macro. Every permutation will result in a distinct answer.

Nonetheless, it's good to do these tests, but just for your own personal use; this is one of the best ways to determine which macros you personally want to use during fights.

If you play optimally you will be hitting the same gcd, in the same point in the fight (granted boss rotation), every single time.
There will be no variables aside from time and uptime.

There's a lot to unpack here.

First, fight mechanics have variation, and these variations often impact how the player should respond and thus what actions they should cast. For example, I main BLM, so getting targeted by an AoE means I might need to interrupt a cast and move out of the way, but if that AoE targets someone else I can just keep casting. And you might think that means I just cast the exact same spells after I stop moving, but that's often not the case since Astral Fire needs to be fed periodically. Needing to move instead of standing still will generally result in an adjustment to the cast schedule. Add up these types of moments over the course of an entire fight and you don't just have a meaningful variation, but a prominent variation. And that's what I love about BLM: needing to actually react to what the boss and my teammates are doing keeps the game feeling fresh.

But let's imagine what it would be like if fights were static, such that a player is either never or always going to be targeted by that AoE. Well, there will still be variation, because players aren't computers. We're humans, and just by virtue of being human we introduce variation into these systems. Often the clearest way to observe that is the fact that players are making mistakes all the time. They're getting hit by stuff they didn't mean to get hit by, they're messing up their rotations, they're dying, they're wiping. And thank goodness players are capable of making mistakes and playing sub-optimally, otherwise there would be no challenge to any game, no way to fail, and thus no satisfaction to achieving success. The game industry would not exist were it not for this fundamental fact of life.

In short, there's a lot of variation in combat encounters, and that variation comes from numerous places, which makes them a poor setting for answering the question of whether macros have some manner of innate delay.

[LilimoLimomo]

Thank you LilimoLimomo. I updated the links.

It is fortunate that the scientific method is universal. I have high hopes for AI technology that can overcome language barriers. I tried NoteGPT to summarize the second movie. It gave me a satisfying result. Original Japanese caption was translated well. The word macro was used in three senses here; single line macro, repeating line macro and macros in general. These must be distinguished from the context.

Summarized by NoteGPT

Thank you so much for the translation! ^^

[LilimoLimomo]

A long time ago, circa 2015 or 2016, there were several threads on macros here. I've completely forgotten them, but according to my post history I've posted pseudo-queueing macros and hotbar changing macros. They were not my own invention, so such tips were already widely known in JP community. At that time there were more macro users and less aversion in English forum than today.

WHM Macros

I took the opportunity to read this thread today, and it was nice. I liked the mix of viewpoints, and how even the folks who didn't like macros tended to speak with nuance and weren't prescriptive about their use; most of them were like, "yeah, these don't queue, so they feel clunky so I stopped using them". It's a shame that none of them seemed to try to manual-queuing technique you provided at the tail end of the thread; it would have been interesting to see how the discussion of that went.

I did find it surprising that someone was advocating against macros that call out rez targets! ^^

[Venks]

So I'm still very new to macros and was trying one out in M1N.
https://youtu.be/ebS7LZc7cs4

So the macro I'm using is:

/micon "Fast Blade"
/ac "Fast Blade"
/hotbar copy PLD 1 GLA 1
/hotbar set "Fast Blade" 1 1 <wait.1>
/hotbar set "Riot Blade" 1 1 <wait.3>
/hotbar set "Royal Authority" 1 1 <wait.3>
/hotbar copy GLA 1 PLD 1

Essentially this cycles between all of my 1-2-3 combo actions and ends back on the 1st. I personally find it nice to be able to have all of that on one button, it's less to think about mid-fight and potentially could reduce the actions on my hotbar. Oh and it's worth noting the actions cycling around on my bar are technically not macros themselves, so I get a larger queue window compared to other macros.

There are two things to look out for when using a macro like this:
1) If the boss becomes untargetable it can potentially mess up the combo.

2) Simply with how PLD's combo structure works there will be times where higher priority actions should be pressed first, which will lead to the macro resetting to Fast Blade even though I haven't finished the 1-2-3 combo.

So how useful this type of macro is definitely depends on the type of content you're doing and what your job's rotation is like.

It was fun giving this a go. I made a few rotation mistakes and had some issues with clipping with my oGCDs, but those were all on normal actions, had nothing to do with the macro. And even with those mistakes I was still the highest performing person in the party by a large margin.

There is a lot of talk about optimization in this thread, but this fight is another reminder to me that the most important aspects to dealing high damage are:
1) Not dying (aka doing mechanics correctly)
2) Having good gear

Doesn't really matter how tight your rotation is if your gear is worse or you're dying throughout the fight cus you got snapshotted going for some uptime positional.

Edit: Oh! That also reminds me. Another issue is building the correct muscle memory. I'm so accustomed to my rotation being on particular keybinds that using macros to do something new does take time to adapt to and improve at. I'm likely gonna try out different macros in the future so it'll be a lot of learning as I go.

[Collin_Sky]

If you're afraid to show gameplay of actual content, it stands to reason that it might not work well in real situations.
If you're only demonstrating on a training dummy, it suggests it may only be effective in that controlled environment. But players don’t just fight training dummies all day, they engage in real content, where mechanics create situations that are far from perfect.
You need to prove that it performs as well as not using macros in actual gameplay before you can confidently claim it works in those situations.