Miqote 4th face nose fudge

[BillyKaplan]

The amount of data transfer needed for each such increase grows exponentially per every category added. Let's say the base data for your character takes X. And when you're in Kugane, that X is sent to all the players who can see you. Now it's X+1. So to begin with, you get (X+1) * N people who see you. That's already N data more than before.
But here's the catch, YOU see THEM, too. So for each of those N people, YOU get that +1 piece of data as well, so that's 2N total extra data already.
But this ratio is true for everyone around you. And soon enough you get a lot of extra data piling up.

What they can and should do, and have done so in the past, is not to create new categories but to bake the possible combinations into the existing categories. For instance, fem-Miqo's have the old Y'shlota hairstyle twice - once with golden beads, another with silver beads. The hairstyles that have a hairbow are duplicated, every time with the bow in different colors.
Unless they made it so that adding even another combination to face options would somehow double the space needed to save that data (actually possible, depending on implementation...) that sounds like a good solution, at least in theory. Them not doing it after all this time makes me think that they did design it in a way that would bloat data if they so much as sneeze at it...

[Clover_Blake]

The amount of data transfer needed for each such increase grows exponentially per every category added. Let's say the base data for your character takes X. And when you're in Kugane, that X is sent to all the players who can see you. Now it's X+1. So to begin with, you get (X+1) * N people who see you. That's already N data more than before.
But here's the catch, YOU see THEM, too. So for each of those N people, YOU get that +1 piece of data as well, so that's 2N total extra data already.
But this ratio is true for everyone around you. And soon enough you get a lot of extra data piling up.

That makes sense, thank you for the explanation.
I can't help but wonder, however, how other games with more hardcore character customization do it and seem to work fine. Even games that were released before FFXIV (like Aion).

[Fredco191]

Lalafell (Plainsfolk) have an option to have their noses be darker than normal. It counts as a facial feature.

Weird how they didn’t do this for that Miqo’te face.

[DeaconMoore]

That makes sense, thank you for the explanation.
I can't help but wonder, however, how other games with more hardcore character customization do it and seem to work fine. Even games that were released before FFXIV (like Aion).

Other games are designed better with beefier server hamsters that can handle the load and not start uncontrollably convulsing 'till their hearts explode.

At it's core ffxiv is flawed. They more than likely instead of a complete ground up rebuild, tore down parts, and put in some new pieces here and there and jury rigged the crap outta it. If you saw the code as a real life manifestation, it'd probably be being held together with an ungodly amount of duct tape. God forbid you bump it.

[Shurrikhan]

That makes sense, thank you for the explanation.
I can't help but wonder, however, how other games with more hardcore character customization do it and seem to work fine. Even games that were released before FFXIV (like Aion).

This. I can see the results of literally 48 separate 1-100 sliders just fine in a game that better handles population density outside of PvP, but... XIV can't do a 5th face for a particular race?

[Genz]

The amount of data transfer needed for each such increase grows exponentially per every category added. Let's say the base data for your character takes X. And when you're in Kugane, that X is sent to all the players who can see you. Now it's X+1. So to begin with, you get (X+1) * N people who see you. That's already N data more than before.
But here's the catch, YOU see THEM, too. So for each of those N people, YOU get that +1 piece of data as well, so that's 2N total extra data already.
But this ratio is true for everyone around you. And soon enough you get a lot of extra data piling up.

N^2 (N players receive the data of the N players)

But it doesn't explain much, since other MMO can give more options with the same limitation. There are a lot of other parameters in the whole picture, and we can only throw guesses at it.
One thing is FFXIV may have a ton more data per character than other games, because we have every classes on one character and big inventory. FFXIV may also have more quests and achievement, and also more sub-systems (chocobo, TT cards, squadrons, etc.). Each of these things aren't big, but put together it makes a lot.
I remember Yoshda mentioned the data transfer between the "worlds" servers and "instances" servers as a limitation, which is why they could add the saddle only by making it inaccessible in instances. Open-world MMO or MMO that relies less on instances may have more room for character data.
Another thing is that FFXIV is more or less bug-free. Maybe other MMO have slightly more bugs because they're slightly more lenient with their data management, but it's not something we can really estimate.
In fine, there is also the possibility that FFXIV is just inefficient with its data, but that's just one guess amongst the others.

Unless they made it so that adding even another combination to face options would somehow double the space needed to save that data (actually possible, depending on implementation...)

I don't see how, even trying to do it very badly. Adding one on/off choice is 1 bit.

[MsQi]

The amount of data transfer needed for each such increase grows exponentially per every category added. Let's say the base data for your character takes X. And when you're in Kugane, that X is sent to all the players who can see you. Now it's X+1. So to begin with, you get (X+1) * N people who see you. That's already N data more than before.
But here's the catch, YOU see THEM, too. So for each of those N people, YOU get that +1 piece of data as well, so that's 2N total extra data already.
But this ratio is true for everyone around you. And soon enough you get a lot of extra data piling up.

What they can and should do, and have done so in the past, is not to create new categories but to bake the possible combinations into the existing categories. For instance, fem-Miqo's have the old Y'shlota hairstyle twice - once with golden beads, another with silver beads. The hairstyles that have a hairbow are duplicated, every time with the bow in different colors.
Unless they made it so that adding even another combination to face options would somehow double the space needed to save that data (actually possible, depending on implementation...) that sounds like a good solution, at least in theory. Them not doing it after all this time makes me think that they did design it in a way that would bloat data if they so much as sneeze at it...

The data is minuscule, a few more bits in a byte that is probably already in use. It's just letting your computer know what face to render, the data for which is on your hard drive.

Enough of this "Won't someone please think of the Servers!" nonsense. This is 2018 and we aren't using dial-up!

[Skivvy]

Enough of this "Won't someone please think of the Servers!" nonsense. This 2018 and we aren't using dial-up!

Sadly, SE hasn't received that memo yet.

I could be wrong, but judging by your join date, it looks like you may be fairly new to the game. Over the last several years, it's become almost laughable how many times they have come back saying they cannot implement something due to server stress/load/data/limitations/etc.

[Hyperia]

Sadly, SE hasn't received that memo yet.

I could be wrong, but judging by your join date, it looks like you may be fairly new to the game. Over the last several years, it's become almost laughable how many times they have come back saying they cannot implement something due to server stress/load/data/limitations/etc.

From what I understand, in order to get FFXIV 2.0 out in time, they had to repurpose a lot of the FFXIV 1.0 code and as a result, its a mess of spaghetti code behind the scenes.

Yes, if they were creating modern up to date code from the start, they could probably have over 1,000 customizations without worrying about overloading the data transfer systems from instance server to zone servers but they do not have the luxury. They are stuck with what they have and are trying to clean it up as best they can. its similar to trying to perform an oil change on a car driving down the highway without locking the engine.

In order to replace all of the code with proper cleaned up code, it would be a monumental effort requiring a massive amount of development time. My guess is that they would essentially have to recreate the entire game to clean it up properly. This isn't a missed expansion but probably two to three, were talking 5 to 6 years. You don't just replace the infrastructure on the back end, you have to redo everything from server coding to graphic assets and more to allow for what everybody wants.

Trust me, I want loads of options, vastly improved graphics but to get there, were talking essentially a new game like FFXIV-2. Can they do it? Sure, but they need the financial and developmental resources that are being spent on KH3, XIV DLC and VII teams to do it. They will make incremental optimizations when they can but that's the best pace they can manage with such a difficult code base.

Yoshi saved the game back in 2.0 but at the cost of a difficult future, he knew that going into this, I'm sure he has a way out.

[Genz]

From what I understand, in order to get FFXIV 2.0 out in time, they had to repurpose a lot of the FFXIV 1.0 code and as a result, its a mess of spaghetti code behind the scenes.

No

...or at least that's what I'd like to say, but their communication is so shitty that they let this kind of "fake news" slides without a care in the world ; maybe it's right in the end.

But there is no reason to think they re-used the architecture of 1.0. 2.x is obviously vastly superior on a ton of technical aspects, even if there are still some, maybe inevitable, limitations.

Also, there seems to be a confusion between code and assets, and some may think code was re-used since assets were re-used, but of course it's separated (especially on the server-side, which doesn't care about assets)