Realistic Solution to Inventory Space (Programming idea that would lower server data)

Realistic Solution to Inventory Space (Programming idea that would lower server data)

First and foremost, I want people to read the full idea before jumping to any conclusions. This is an idea meant to solve current glamour issues in game while at the same time reducing the amount of data that is stored on the server for every character out there. To explain this I want everyone to understand a few things about how Square Enix is looking at things and how they make their decisions:

1) Money spent on the game must be kept within current budget
2) The company is charged for their data usage
3) The higher the data usage is the higher the bill and the more money spent on maintenance versus money spent on design and game-play
4) To redesign a whole system of glamour would cost an incredible amount of time and resources taking away from new dungeons and content

That being said, there let's look at WHAT data is used in the inventory system currently and why simply adding more inventory slots creates problems (the metadata attached to every item):

1) Dye Color
2) Durability
3) Spiritbond
4) Materia
5) Who Crafted the item

Simply speaking there are at least 4 tags with literally hundreds of variables per tag, and one tag with THOUSANDS of variables (although not used on all items). Now multiply that by the hundreds and hundreds of bag slots and you start to see why adding more bag space takes up more data. You simply have more and more metadata that has to be collected for every item. It's not as simple as a stack of crafting mats which only have the tag of “quantity owned.”

Now that we know the problem with just making more bag spaces, and the fact that it costs a LOT of programming resources to make a whole new system to solve this, lets look to what IS programmed in the game already:

The Armoir.

So the Armoir is different from most items in the game in that it doesn't actually collect metadata. Basically anything put into the armoir is put on a list of “owned” or “not owned.” Any information such as dyes, durability, spiritbond, materia, or who crafted it is NOT collected. The item is actually DELETED from your inventory and put on the list of “owned” in the armoir. When you remove it, it simply duplicates the item for you from a non-modified state. In essence it's a cloning device, which uses the game's default settings for an item as it's reference. All data for all items is standardized and no person gets something different or unique. Adding things to the armoir is as easy on their end as adding the item to the list of things can and cannot be included. The programming exists already. So lets look at the Pros and Cons of the Armoir:

Pros:
1) The system is in place.
2) No new programming is needed, only a modification to something already existing.
3) The system saves an INCREDIBLE amount of server data as no Metadata is collected for items stored.
4) Amount of items that can be stored per person can be as large as the item pool currently in game, without adding more data per character.
5) Items can be stored and removed AS NEEDED when people want to use the item for glamour
6) The system already exists (yes I said it again, because this is REALLY important)
7) Huge savings on a programming perspective since minimal modifications are needed to make the system work.
8) Huge savings on Server Costs, as less information is needed per character.

Cons:
1) Items stored lose all customization such as dyes and materia.

So basically there are a lot of Pros to this system but with one MAJOR con. So lets discuss that con, and a solution to solve it. Dying items in this game is EXTREMELY important, and is as the core of customization in Final Fantasy. The above solution to inventory would basically remove any personalized touches you've put on your hard earned gear. So how do we solve it without creating a mess for programmers? The key to making this whole system work is making it EASIER for gamers to re-apply their dyes after having put them in storage. Right now as the system stands, gamers would be punished for storing items with rare dyes because they would simply be ERASED. So lets look at why we need to make it easier for gamers to apply dye:

1) The whole point of allowing glamour to go into the armoir is because it eliminates metadata tags, thereby saving data usage and programming time.
2) You want to encourage AS MANY PEOPLE AS POSSIBLE to use this system because the more that do, the lower the monthly server fees are, thereby saving money and allowing more time and resources to be spent on content creation.
3) The easier the new dye system is, the more motivated people will be to use it.
4) Recreating the dye system from scratch would cost time and resources, so using the current system in some way would be optimal.

Understanding that using current systems in a different way is key to fixing the Dye system. Now I have several proposed solutions, and all would be viable:

1) Use the current dye system, except move rare dyes to be purchased through a vendor via tokens, a resource, or gil. (in game sinks are always good for the economy as a whole)
2) Use the current dye system, except make ALL (including rare) dyes obtainable via crafting, and not just some. (would encourage more crafters)
3) Finally my favorite: Remove all dyes as a reagent to actually dye an item. Basically KEEP the dye palette as is, but when you go to actually dye an item, it uses Shards, Crystals, and Clusters as the reagent for your dye. This would solve several problems in game, as it would give a use for those lower level shards which people now have thousands of, and you could make “rarer” dyes cost Crystals or Clusters. It would also encourage more people to gather, which is also positive.

I encourage people to come up with other ideas as a solution to dying the gear which is removed from the armoir, as the easier it is to do, the more motivated people will be to use the armoir in the first place.


As a final note, please understand that I am not trying to tell the devs how to do their job. I am simply trying my best to come up with a CONSTRUCTIVE solution to a problem that exists, and do so with enough thought to their situation as well as ours. Sure we want LOTS of things, but we also need to be conscientious of what they can logistically do within a reasonable time and budget. Sometimes we don't need to reinvent the wheel when they've already created a great system that would function GREAT with just a little modification to it's purpose! I love this game dearly, as I know everyone else posting on this forum does, and I encourage people to consider this idea and let the devs know what your thoughts on it are!

Thank you all!

-Seraphim Sokar
-Exodus

Ultimately I wish they would give us a glamour log to help clear out all the inventory eaten up by items saved for glamour but expanding the armoire would at least help til we could get a true glamour log.

You have my full support for this. This would have zero cons for me, while also having some additional pros that are probably only relevant to me personally.

I've been playing a lot of GW2 and I wish the dye system in XIV worked like that. Buying rare dyes off the mogstation would "unlock" the dye in the pallete, and you would simply need to craft the recipe (with shards as you said) in order to dye the gear.

I like this idea and I wish the dev team would listen.

I don't think they're necessarily charged for data usage (as they have dedicated servers and aren't cloud-based to our knowledge), but reducing the amount of data in memory would reduce some load on the systems. At the same time, I'm not sure if what you're suggesting would help with that (especially since most of the more expensive pieces should be living in memory as references rather than independent objects--or else what are the devs thinking?).

I'd be very curious to see how they have it coded and organized. The idea that characters move around between zones seems strange to me, when the more practical solution would be to put coordinates on the character object, and change memory references on the character and zone for one another.

Idk. I get the sense that Yoshida isn't a programmer, and that, to explain things to him, the devs had to use language that isn't technically precise (which is why some things he says about server load and such don't make a lot of sense from a programming standpoint).

Quote:

---

Originally Posted by Lambdafish

I've been playing a lot of GW2 and I wish the dye system in XIV worked like that. Buying rare dyes off the mogstation would "unlock" the dye in the pallete, and you would simply need to craft the recipe (with shards as you said) in order to dye the gear.

I like this idea and I wish the dev team would listen.

---

Many items have more than one dye location as well, up to four.

I believe just adding a glamour log would significantly cull inventory issues. Then like many MMOs just larger decor storage areas (Wildstar has up to 2.5k exterior and 2.5k interior and a 1k decor storage box lol) and crafting spaces (although with our larger inventory and a glamour log it might be okay anyways). 140+ slots and 2x 175 for crafting / tokens / leveling gear (no need to keep for fashion).

As for the system, unless I misunderstood, I would massively prefer a new glamour log rather than piggyback the armoire- the interface would be horrible for glamour / item managing vs all other competitive MMOs. Would likely be one of if not the worst glamour systems lol. Although I respect that you're (OP) trying to be feasible and friendly to the devs. Imo, It's not good enough a system to use for a glamour log and if its not meant to be a glamour log then a glamour log could do it all better anyways (imo lol).

As much as this sujestion to add to your idea OP can get... exploitable... u could have it so any "Base" gear (IE: gear that doesn't have a glamour cast apon it) that gets put in the Armoir gives the player 1 copy of the dye back

I say "Exploitable" cause then gear would get sold way higher than others on MB just cause it comes with a free rare, possible Mog-store, dye that you then get to use... granted I personally see it as a small CON for the high.many PRO's it may give.

You do there is a thread that has been made for new idea's to improve the game right?

Not trying to be rude but a lot of you new guys in the fourms need to learn how to use the search bar.

I think you're misunderstanding what a variable is. A variable is a single space that can hold one of many values. Each of the five qualities you listed could be a single variable, except for the materia on an item which could be six. So that's about 10 variables per item.

You're making an assumption about the armoire that may or may not be correct: that items in it are stored by ID. If they are, then adding things to the armoire shouldn't really be any problem. You'd basically just flag it as armoire-able and presumably the system sorts itself out. However, for efficiency's sake, I strongly suspect the armoire contents are stored as boolean bits, one per item. If that's the case, then to put more items in the armoire, they'd have to increase all user data for every character by a number of bits equal to the number of items that could be stored, a number which will continually increase over time. Granted, it's not actually that huge a number (the Eorzea DB lists just over 15,000 items at present), and I do hope they do something like this at some point, but my point is that it's unlikely the system already exists as you described it.

To go into a bit more detail, let's see if we can get a decent low-end guess of how much space inventory actually takes up.

The items themselves are listed in the DB by hex codes, with the ones I've seen having 11 digits, so that's a minimum of 44 bits. Dye color could be stored as 3 bytes (RGB), so 24 more bits. Durability and spiritbond are displayed as whole %'s but definitely don't increase that way. My guess is 16 bits each for those. I suspect materia is stored very inefficiently, but since there are less than 255 varieties of materia at present, let's be generous and say 8 bits per materia, times 5 slots. Character IDs are past 17 million now, so for safety let's say those are stored as 32 bits. In total, that's a maximum of 172 bits per item, or (rounding up) 22 bytes. The item ID itself is 6 bytes. The result is, if you store four times as many items this way as you have slots, you're still using more data. Not sure if that makes it a good or bad idea, but there it is.

Quote:

---

Originally Posted by Sokar

Realistic Solution to Inventory Space (Programming idea that would lower server data)

First and foremost, I want people to read the full idea before jumping to any conclusions. This is an idea meant to solve current glamour issues in game while at the same time reducing the amount of data that is stored on the server for every character out there. To explain this I want everyone to understand a few things about how Square Enix is looking at things and how they make their decisions:

1) Money spent on the game must be kept within current budget

---

Whoa, I'll stop you there. This is not a "data expense" problem. Nor is it a storage problem. It's a synchronization problem. I'll tell you straight up from a game I had a NDA with that there is nothing necessarily stopping the developers from having infinite space, but it comes at one extreme penalty.

More items you can have = larger packet sizes, too large of packet means fragmentation. If you know how and why MMO players tamper with the MTU settings, you'd know that the maximum packet size is just under 1500bytes. Those 1500 bytes have to be sent every 200ms (for 5fps of game play.) Within those 1500 bytes you have to store your character's coordinates, gear, physics, etc

So if you sit there and watch a packet monitor you'll note that opening, the inventory, picking up or dropping an item, or consuming an item, re-polls every item. Hence part of the reason FFXIV can do all these AOE dances in the first place is that the game doesn't constantly have to send updates to 100 inventory spaces every 200ms when the inventory is open.

So if you had say, 1000 spaces instead, now instead of 100 items fitting in one packet, it now takes 10, which means it takes 2 seconds to open or close the inventory, and everything else in the game grinds to a halt until it does. That makes playing dungeons impossible without "locking down" the inventory in some way so only part of it is available.

So here's where I'm going to suggest another idea.

Reverse how the Armory system works. Instead of having 25 items per slot, have one armory page for each class/job. While in a Duty no gear can be swapped except what is in the Armory for the class/job the Duty is for. The reason it works the way it does right now is to allow re-using gear between classes, which is all and good... if all your classes are the same level and you really have no desire to glamor anything. In V1.0 they removed a few item slots to reduce the complexity of the models, not necessarily the data sync requirements.

Then with the game's main inventory, have a "Field" and a "Dungeon/Duty" inventory. Field inventory consists of the current inventory we have. The Dungeon/Duty can only contain finished consumables (and item drops from the Duty.) When the Duty ends, the player can keep the consumables in the Duty inventory for next time, but the gear drops must be transferred to the field inventory, a retainer, or destroyed before another Duty can commence. The "field" inventory would not be available during Duty's.

Quote:

---

Originally Posted by KisaiTenshi

Then with the game's main inventory, have a "Field" and a "Dungeon/Duty" inventory. Field inventory consists of the current inventory we have. The Dungeon/Duty can only contain finished consumables (and item drops from the Duty.) When the Duty ends, the player can keep the consumables in the Duty inventory for next time, but the gear drops must be transferred to the field inventory, a retainer, or destroyed before another Duty can commence. The "field" inventory would not be available during Duty's.

---

This would piss of a lot of players that do duty one after another. It solves one problem but generates another.

I believe it was you that mentioned storage items, and that's what I thought of as well when I heard about the data transferred. Instead of giving players more inventory space, give them items that can store items, but that need a cast time and an animation and cannot be used in combat. Make the tabs within those easy to customize by means of filters (for example, "food only", "gear only", "materia and glamour only", but also between 5 and whatever amount of slots the player wants that the developers feel comfortable, 25 or 50 per tab would be a fine range I believe) and the ability to name them. Make one tab for current run which you can access only before leaving the dungeon (as in, while the run is ongoing, all the way to the moment you leave the duty through the exit) with there being an option to automatically relocating those items to other spots based on the filters. If there is not enough space in them?! Drop em in normal inventory. None there either?! THEN lock the players until they decide what to do with the extra items. If someone doesn't have space, normally they just couldn't even get it in the first place, so it's already a more forgiving system.

That way, people could go from duty to duty with no downtime for clearing inventory, all the while not getting ridiculous amount of clutter to bring with them all the time. I daresay that lowering the standard inventory to 25-50 items for the stuff that needs to be possessed "here and now", with some extra spaces for random stuff picked on the field, from quests, from merchants or just from not fitting into the other "bags" would be good, to give more of that "bag space" (because let's be honest here, Square Enix could but does not WANT to give unlimited, or even "more than enough" inventory space, just enough to "manage with").

While there is no actual confirmation how the armory system works, all the limitation of it highly suggest the items are stored just by the ID. Such as, dropping each value that could have variable on it like repair condition, dye and spiritbond. If we look on layered storage on our character at the moment, there is actually another concept it like to streamline.

Currently we have armoire, where we have had option to store selection of items, which to our best knowledge is I ly stored on owns/doesn't own basis. Then we have Salvager, which apart from couple other functions, sells us gear on unlocked basis. To some extent these two lists are overwlapping, which I as a bit of waste, since their practical use is near the same (for this function of salvager).


I'd like to see those two combined. The resulting thing would remember all items we are eligible for. When taking out an item, it would check if we already have it in our possession, if yes we couldn't get second one until it is discarded or passed on, I which case we can check out a new copy. This amalgamation of armoury and salvager would not be as lore apparent but wouldn't need multiple overwlapping lists.

Just use something like this, it would save on coding and remove some items from game.

https://biobreak.files.wordpress.com/2015/04/outfit.jpg

On the right hand side of the character sheet there are 1,2,3 etc and you purchase additional ones up to 10, this would allow 10 variations of fashion to appear there and you click on the one you want to load for you current mood.

To implement this:

1. New code to allow multiple fashion sets
2. New NPC to buy the additional slots [1st slot free, then you would pay 50,000 gill per new open set]
3. Items placed inside the new area would not be restricted by class
4. Items placed inside the new area would still be male/female restricted unless the item was unisex
5. This system would work with the current gear swap system as the glamour is not effecting your stats.
6. When you dye an item inside the glamour slots the colour change would be instant

I cant help but wonder how many of you are coders with a realistic idea of what the issue is...

That said Yoshi P has said a glamour wardrobe would be feasibly possible (IE. one off character that stored records of unlocked looks) but his UI team was fully booked out so if they did it, it would have to come later.