We need actual matchmaking in Duty Finder.

[HardcoreRaider]

I'm sick of being matched with players who can barely press buttons and fail to basic mechanics like stack markers that everyone should have seen a thousand times already. Let better players match with better players so we can quickly finish the dungeon and let casuals slow down and pull one mob at a time if they want.

[Burmecia]

...
ERP optional?

[ADVSS]

LMAO Ranked Duty Finder would be hillarious!

[Kaurhz]

Are we back in 2017? Nice... Trolls are running low!

I will indulge though.. Bad idea, simply because people that are more talented and adept at the game would have a much lower pool of players to be matched with, by-product you'll just have longer queue times. Ergo, people as bad will likely end up getting their routines out of the way sooner than those that are adept. You'll start off with 1 metric then find something else to complain about. DPS, iLv, use of skills (e.g. Peleton)

[AphraelAmarantha]

Are we back in 2017? Nice... Trolls are running low!

I will indulge though.. Bad idea, simply because people that are more talented and adept at the game would have a much lower pool of players to be matched with, by-product you'll just have longer queue times. Ergo, people as bad will likely end up getting their routines out of the way sooner than those that are adept. You'll start off with 1 metric then find something else to complain about. DPS, iLv, use of skills (e.g. Peleton)

This is false. Besides FFXIV, I play pso2 on the JP servers and they have an expert matchmaking for their emergency quests (their version of raids). It doesn’t affect how long it takes to get matched with a group. Additionally, expert matchmaking makes raids go by almost two to three times faster than non-expert matching. Believe it or not, ffxiv could actually use something like this. The tales thread alone is proof of that.

Now how to implement an expert match making queue along with requirements to be an expert is the hard part. It can’t be like mentors because obviously there are mentors who aren’t that skilled. There needs to be a really difficult test of one’s knowledge of the game and all three roles in addition on how to handle complex mechs and so on to get the requirement. PSO2 has two pretty difficult (for casuals anyway) quests that are required to get the title needed for expert matchmaking. According to SEGA’s own stats, only around 15-20% of the player base has the title, which is what their target was so it works quite well there. The only negative is if your friends aren’t skilled enough or you aren’t, then you won’t be able to play the game together if one of you don’t want to queue for non expert, not until you actually git gud. But actually affecting queue times? No that won’t happen.

[Kaurhz]

Snip.

Interesting.

My point really was being that you'd start off with 1 metric, then another, and then another until you finally create a metric that is something utterly unrealistic for the content it is aimed at. Pointing this at raids is 1 thing, but also pointing it at your daily roulettes??? Besides, I feel the issue with XIV is not necessarily the player skill, but rather the investment, or lack thereof that people are actually willing to put into the game; in other words, being lazy. Doing this does nothing to ultimately address this glaring issue, this is the equivalent of a poorly woven band-aid. There's a tonne of things the game could do without resorting to this. Do you really want to incorporate a difficult test of the game's intricate mechanics, alongside all 3 roles into regular matchmaking for XIV? This is something you expect in Savage content, maybe your extreme trials. not in a leveling roulette or an expert roulette, or trial roulette, there's simply too much of a variance in the difficulty of a lot of dungeons to warrant this form of testing in any valid way. The only distance they would ever go in this direction is job locking Level 50+ content, which I think is a reasonable level (can't do x content without doing y quest from your job). Let's not forget the possible exploitation this opens up. Let's say you have stack markers as an example for a mechanic, what's to stop me from griefing someone and running away from their stack marker? They died, not me. You can quite comfortably pull this off in Heroes Gauntlet. People already resort to this level of play to prevent others from getting duels in the Bozjan Southern Front. So I hardly see them being able to use something like this in XIV anytime soon.

Creating a test of player skill, ability, and understanding of the game is 1 thing, but actually enforcing that in the long run in any way is not really feasible. There's nothing there to stop someone from being lazy in your roulette or instance after reaching your 'requirement'.

[Laphicet]

Creating a test of player skill, ability, and understanding of the game is 1 thing, but actually enforcing that in the long run in any way is not really feasible. There's nothing there to stop someone from being lazy in your roulette or instance after reaching your 'requirement'.

The idea is that if someone would put in the effort to pass said test they would at least have a better understanding of their class, as is needed to pass it, and would be using said knowledge.

A 0-dps cure-spamming healer, zero-mitigation tank, dps using skills improperly, dps that don't know how to AOE, etc, wouldn't be able to pass the test in the first place, that's the idea, to be a way of telling them their play habits are bad and need improvement. And someone who does pass the test would have to know about their class and how to execute their rotation somewhat well and handle role-specific duties properly, and thus would not be a detriment in further content unless they intentionally decide to grief, as you don't go from understanding these things and using them regularly to just, not doing such on a whim, at least not usually. These test instances would also have to be exempt from the echo buff system, of course, so you couldn't fail upward through them.

This of course, assumes that the community perspective of what classes do lines up with the developer perspective, which if one takes one look at the current state of healers and healer gear (piety on savage weps, anyone?), the devs clearly don't see at least one of the roles the same way the community does, and thus would probably make poor tests for such (a healer test with minimal or incredibly lax dps requirements, a SAM test that doesn't take into account the 60s loop and sets the dps requirement lower as a result, or a tank test that doesn't encourage double pulling, for a few examples).

[Kaurhz]

Snip

These are things that work well when written on paper but not so much when actually incorporated as an actual feature.

They would have a better understanding of their class yes, especially as far as passing the test is concerned, but actually maintaining that in the first place and not simply being lazy is an entirely different issue. Again, a player can pass the test, but whether they actually put that into practice is an entirely different issue. Someone doing 0-DPS on a healer, or doing no mitigation is often a serious case of being sloppy, or lazy as opposed to not understanding it, which is the issue a lot of players are having anyway. I've got a friend who outright refuses to even bother doing DPS in Expert roulettes or even normal duty finder content, yet in harder content, he's more inclined to actually do damage. Why does he do this? Because he can, because dungeon content is not really designed with people playing even remotely optimally, whereas savage and extreme at least tries to push this direction. The only real way you're going to address any of the underlying issues is by simply making it to where people are just outright required to perform at such a level at all times or at design unique and creative mechanics wherein it at least incentivizes healers to DPS, or tanks to actually mitigate (e.g. mechanics like Cursekeeper and Infirm Soul being incorporated into more casual content. Features like these just open their way up to potentially a lot of griefing.

Maybe you can keep it as continuous measurement, so each run is measured and you get an average across all of your runs, but that only turns your daily roulettes into too serious of activity for a vast majority of players. Besides, once you're in the rut of being on the 'bad' side, then it'll be very difficult to get out of when going in on a tank, or a healer. The less mitigation your tank does, the less opportunity you have to do damage as a healer since you're busy picking up that slack. Or the less damage your DPS do the longer a given pull goes on for you go closer to the inevitable wipe due to lack of CD availability, and then subsequent pulls just become all the more difficult. The more mistakes your DPS or tanks do the more vulnerability stacks they gain, and the more you're required to actually heal them instead of doing damage. Funny given the topic, but this is a team-oriented game, and the more your team works in harmony the easier it is to get through the content. The less harmoniously they work the harder it becomes, which is why splitting up matchmaking in this fashion just won't work, especially for the intended level/audience.

[HardcoreRaider]

Are we back in 2017? Nice... Trolls are running low!

I will indulge though.. Bad idea, simply because people that are more talented and adept at the game would have a much lower pool of players to be matched with, by-product you'll just have longer queue times. Ergo, people as bad will likely end up getting their routines out of the way sooner than those that are adept. You'll start off with 1 metric then find something else to complain about. DPS, iLv, use of skills (e.g. Peleton)

Separating the playerbase into multiple groups does not slow down matchmaking contrary to what you believe, because the primary bottleneck right now is for a tank and a healer to be matched with 2 DPS. Decreasing the pool of players simultaneously decrease both groups of players and therefore will only increase the variance of the queue time, not the expectation.