Forums » Crafting

Crafter's Roundtable: The concept of matrixed progression

    • 1528 posts
    November 4, 2019 11:56 AM PST

    Note:  Like all Crafter's Roundtables, this is also posted on Pantheon Crafters for discussion.  The goal of every Crafter's Roundtable is to generate community discussion that might help the developers at Visionary Realms as they move forward with implementing the Crafting sphere in the game.


    During the Crafting Developer Roundtable a few months ago, Ceythos mentioned that he was looking at the idea of a different type of progression for crafting.  Instead of having professions that you choose at creation and then level up inside, he mentioned that he was exploring the idea of a more free-form, skill-based type of progression and whether that might work out better for Pantheon.

    I call this concept Matrixed Progression, and today I wanted to post an example of how it *might* work.  My goal with this is to get people thinking about the concept, not to say that it should work in any specific way.  There are a few advantages to matrixed progression if it's done right.

    1) You don't have to treat each area of crafting equally.  In a profession-based system, you can't really make alchemy harder than cooking, because people need to be able to choose either one of those when they start out.  Likewise, you have to make sure you have enough things to make in each profession so that it feels like a real profession.  In a matrixed model, it's ok if you have 200 cooking recipes and only 20 alchemy recipes, and it's also ok if you make those 20 alchemy recipes much more challenging in terms of their requirements.  You can do what makes sense for each type of crafting and its role in the game.

    2) You can set up cross-area dependencies more easily.  In a profession-based system, you can't really make a high-level armorsmith recipe require knowledge of leatherworking, at least not without stretching things quite a bit.  However, in a matrixed model, you absolutely can do this, and set up dependencies in any way that makes sense.

    3) You don't have to tie everything into a strong vertical progression.  In a profession-based system, it's hard to justify meaningful choices like specializations in certain types of items.  For example, if you wanted your weaponsmiths to be able to further specialize in creating swords, or spears, or axes, you'd have to come up with similar specialization choices for every other profession you have.  In a matrixed model however you can have different specializations in different places.  You can set things up so that players can choose to go deep in particular areas or focus more broadly across an entire type of crafting.

    This isn't to say that matrixed progression doesn't have some potential downsides.  For starters, it's more complicated than a profession-based model, and that can be a deterrent to players, or at least require an innovative UI to help them understand their choices.  Likewise, depending on how you set the rules of crafting up, it's possible that matrixed progression might allow players to stack bonuses and make things easier than you, as the designer, had intended.  Like any other game system, it requires careful planning and thought during implementation.

    Anyway, because Ceythos mentioned this during his roundtable, I think it's something we should all think about.  Below is a very simple example of what matrixed progression *might* look like that I worked up.  It's just there to get people thinking.


    In this example, the crafter starts at Novice Artisan (at the bottom).  From there they pick a general crafting type (Blacksmith) and work their way up the tree, which branches out as they go.  Each skill box along the way has some requirements to unlock it and provides some benefits, whether those are additional recipes that can be crafted, new techniques that can be used during crafting, or bonuses to the efficiency of some of those techniques.  I've done my best to show this at the top with a focus on one of the boxes.  You could imagine this being a mockup for an in-game UI where crafters chart out their progress or figure out what they want to work on next.

    To make a system like this work long-term while still encouraging interdependence and socialization, there would probably need to be a limit on how many boxes an individual player could have unlocked at any given time.  This might be a hard limit (for example, you can only ever unlock 35 boxes, and you have to give one up if you want a different one after that), or it can be a soft limit, where the more boxes you unlock, the bigger the requirements are for each additional box after that.  Either way though, the system allows you to pick up the boxes that you care the most about, and ignore the ones that you don't.

    Just to be totally crystal clear, this diagram is just a very basic example to help kickstart the conversation.  If I was going to try to design a progression matrix for Pantheon right now, with all the different types of crafting that could be in the game, in as much detail as I think they deserve, the resulting diagram would be FAR too big for the forums.  I hope everyone will take time to talk about the idea of matrixed progression and how it can work in Pantheon.

    • 368 posts
    November 4, 2019 12:35 PM PST

    I love your ideas neph, especially this one. Being an avid crafter Im really looking forward to an elaborate system that the player base will actually depend upon in PVE content. Im really curious what the others will add, but I just wanted to thank you for this one.

    • 43 posts
    November 5, 2019 2:16 AM PST

    This is not a simple matrix, it is already a matrix cube. I like it.

    If you limit the number of boxes, for example: an expert swordsmaker cannot become an expert in alloys. But to make expert swords he needs expert alloys. Stimulating people to get or sell materials. And in this example both are blacksmiths needing each other. Or he creates another toon that can be an expert in alloys.

    But to be honest, i Love the idea of needing each other, but I hate to be limited in a single specialisation. So I rather see the soft limit.

    Now the bad stuff: You want to create this to increase socialization. But...lets say you are a expert in Alloys and you want to sell your alloys. In old EQ times you would yell in the area. You can consider this a social way of selling or buying stuff. Other games have a bank/shop where you can put your wares for sale or auction. Smart systems, but hardly increases socialization. Even your matrix won't really increase socialisation if you have a bank/shop system I just mentioned.



    This post was edited by Qulash at November 5, 2019 2:45 AM PST
    • 1023 posts
    November 5, 2019 5:02 AM PST

    First I love the general idea.  It looks very SWG skill trees to me which had one of the most interesting character progression systems in MMO history.  Making a “build” based on a few goals and a total number of available matrix boxes was a great way to make each character different and open up the option to make either a laser focused character or a generalist but not both.  WTB this system for adventuring as well (never have been sold on hard character classes as it is an artificial limitation).

    Ideas on how to do it:

    1)      One mechanic to earn points to spend on skill matrix and another mechanic to unlock the ability to spend a point on a specific box.

    1. Points would be earned through generic crafting experience and quest rewards.  Mastering a group could give a few bonus points that can only be applied to adjacent cells.
    2. Each cell would have an amount of a specific type of EXP earned, required processing skills and recipes unlocked and a level of mastery in said tree.  Additional social or opportunity requirements could be added for specialties.

    2)     Have a “mastery” line of skills that provide a huge benefit but also consume a large portion of available points to unlock to create specializations within a free form system.

    3)     I would not cap the number of skill points that can be earned but I would introduce an exponential growth in difficulty to unlock more such that there is an effective cap but if someone wants to put in 1000% more time to unlock 10% more options than the average crafter nut then more power to them.

    4)     The requirements to open up specific matrix boxed could be effectively epic quests themselves.

    The issues I see:

    1)     Named Itemization vs Template Itemization.  Pantheon is being touted as having “iconic items” that are recognizable.  This massively limits the total number of crafting outcomes that the game can support as each must be recognizable.  Small incremental changes that would gain in a matrix system could not equate to changes in outcomes as variable item characteristics are not acceptable in the "iconic item" system.

    2)     Drop balanced and not crafting balanced item economy.  If it takes 10 hours to farm a great item and 100 amortized hours (between the crafting process and the amount of time required to grind crafting to a point where the item can be made divided by the number if items made at that skill level) to craft an item that is at best 90% as good as the dropped item then no one will want the crafted item.

    3)     Non-competing but complementary itemization between crafted and dropped items.  Crappy band aid to make crafting relevant that will be quickly violated as new areas are added with dropped competing items that once were only crafted.  The amount of time spent to gain the item between the raiding, crafting, grinding, camping should be what sets the reward level that should then set what is best.  A raid drop should not be more powerful than an item crafted from drops from the same mob as it requires that same raid encounter plus all the crafting and likely other valuable resources while still having a chance for failure.  Crafting could be value added by changing a drop of one item type into another item type of the same relative power level in order to decrease the amount of rotting raid drops.

    4)     Slot based inventories vs Mass+volume inventories.  The number of unique items a character can hold in a slot based inventory system is finite but in a mass+volume system it could be nearly infinite if the mass+volume ratio to inventory space is low enough.  A limited number of unique items also means a limited number of possible ingredients which limits the complexity of the crafting systems and its stages.  This can be worked around to some degree by having nested bags of different item sizes and appropriate item sizes but at that point it might as well be mass+volume for all the UI benefits.

    • 508 posts
    November 5, 2019 6:15 AM PST

    I enjoy and support everything that makes each character different from the others as well as giving us more options to customize. I would personally love some recipes that would incorporate many crafting "trees" - so lets say a regular plate armor [metalwork], could be laced [leatherwork] in order for it to be more comfortable and silent, then we could attach magical gems to it providing certain bonuses [jewelery] and so on. I would opt for a soft cap though.

    • 72 posts
    November 8, 2019 4:17 PM PST

    I do love the idea but would rather it be more like it is in "The Repopluation" where it is all skill based.

    What I mean by that is you have no restrictions and can choose to level none or all craftings,  as you choose.

     In order to craft a an item you would need several ingredients.  Those ingredients could be obtained by drops, quest rewards, Crafted or whatever else they want to offer.

    Should you wish to craft each of the ingredients yourself you would need to be proficient in many skills.

    Crafting is much more interlaced and dependant that way.

    An example of how it might work for a player to craft just one of the ingredients necessary to craft that sword...

    Long Sword Hilt....

    Hilt Crafting Skill (making the final product)

    Hilt Production skill (makes the base hilt)

    Hilt Template Production (makes the mold for the hilt)

    Ore refining (Makes the metal)

    Pommel Production (makes the pommel)

    Leather Stripping (making the leather)

    Adhesive Production (making the glue to bind the parts together)

    Ore Refining (makes the metal)



    And this could be as easy or as complex as they desired. 

    My personal preference is to make it kinda simple to begin with like say a wooden training sword might only require using a knife on a log but getting more and more complex as you get further into it

    so if you wanted that Blazing Vorpal Sword of Gnoll Slaying it would require say 10 sub components and each of those would require 8 sub components and each of those 6 and so on and so on.

    and each component would require higher and higher skill levels so that while technically feasible that one character could do it all,  it would be much faster to achieve by working with others.




    This post was edited by KatoKhan at November 8, 2019 4:24 PM PST
    • 422 posts
    November 26, 2019 7:08 AM PST

    As always I'll try and put the short version here.

    The time required to get past the early stages in each profession can determine how likely it is that you'll see a lot of players having the same skills. 

    This could have it's impact on early tier content and the experience of being meaningful as a starting player.

    The matrix can be so complex and daunting that it actually discourages players to tackle this mountain of skills. The paradox of choice is a useful thing here. And if there is a long term vision there, to prevent maxing out on all things over time, it could work well. The paradox needs to remain efficient or you risk implementing a drag of a design.

    Allowing fast progression in this matrix could put pressure on the devs to put out more content as people start to max out skillboxes. 

    If you implement the exponential factor, you'll have to consider how much fun it will be over time. As each expansion would require you to up the stakes ever higher and higher. In a way you're removing the value of earlier tiers using this method.

    Players should be able to explore the different styles up to a point, restricting their final class choice seems fitting into the overall design of VR. Where choices have their pro's and con's and one needs to learn how to cope with the consequences of their choices.

    1) You could allow multi class skill progression if you put a break on how fast you skill up at ANY skill based on your OVERALL SKILLPOINTS earned. Initially you'll experience faster advancement in your chosen skills and as you grow your choice in skill will weigh down more and more. It will become ever more important where you skill up as it will take you more and more time. Both stages have their own specific motivating incentive. At first you more easily experience the affect of your chosen skillpoints. As you mature in your skills, the weight of your earned skill point will become more valueable to the player itself. And towards the outside world, players with high amounts of skillpoints  / expertises will be something to look up to and distinguish. This meaningful decisionprocess allows multiclass skill progression to have some fashion of uniqueness to it. Mastering more than one profession or skill within a profession is not a easy feat. In short, this is mathematical alternative to the exponential approach.

    2) A more complex take on this last suggestion would be that you put communality of different classes into the equation. Some skills of a provisioner might be closer linked to an alchemist than to a blacksmith for example. This mutuallity could allow for faster progression in that skill if the player already has related skills in the alchemist skill lines. The level of a skill box could be a factor that has an impact on the pace of advancement. If you have 1 level 2 skillbox you'll progress at standard pace within that box. If you have another level 2 skillbox initiated already within that same tradeskill class, you might advance faster in a level 2 skill box of the same skill line or profession. If you have progressed in a level 2 skill box in a different profession, you might get a beneficial boost on advancement/progression but that might be lower than if it would be in earlier mentioned scenario. So in this equation, you have something putting a break on the pace of earning skillpoints (1), but at the same time you have benefactors that impact your progression in the other direction. This might seem complex, but it can make sense and it could make for a logical impact on the choices players make as they hone specific skills. This suggestion does not require a maximum of skillpoints per skillbox. The actual amount of skillpoints within a level X skillbox could be the actual number to put into the mathematical equation, thereby increasing the value and meaning of the time spend by the player within that skillbox.

    This post was edited by Barin999 at November 26, 2019 7:37 AM PST
    • 2626 posts
    December 12, 2019 8:45 AM PST

    If you were the crafting developer I would comment in great detail about the suggestions. Alas, you are not.

    Assuming the developers are considering basic options at this stage of development - and if not all of this is merely entertainment for us and will have no impact on the game - what I suspect is important is the big-picture question which your matrix was intended to help us - and them - focus on. And done very very well too.

    Profession oriented crafting is the most common choice in MMOs. It is easier for players to understand in any detail and makes sense to them faster because it is analagous to the profession-based system for adventuring. We are used to the idea of being either a cleric or a warrior but not both. If we have played typical MMOs we are used to the idea of being a weaponsmith or a cook but not both combined together. A matrix or skill-based system is almost surely harder to create well in a way that takes the approximate amount of time and resources to master and use that the developers intend. And that doesn't end up either too powerful or uselessly weak. A matrix or skill-based system will quite likely reduce the number of people that seriously try to craft and cause many who do try it to get significant things wrong as they learn the ropes.

    In other words I strongly favor this type of system over the traditional approach if VR feels that they are able to make it work.

    As with many things I go back to Vanguard as being a model and an inspiration - in approach if not always in how well things were implemented. Much more so than EQ in the area of crafting - which is fair it came far later and saw further because it stood on the shoulders of giants - one of which giants was EQ.

    The basic approach in Vanguard was that crafting and adventuring should be different spheres with neither one necessarily requiring the other. Maybe equal in importance though that didn't *quite* come close to working out. 

    In adventuring not too many of us would enjoy a simple class with an easy roadmap to maximum level and able to hit maximum level quite fast if the resources were available to work with. Well, neither do I wish such for crafting and a nice complex system appeals to me. Those that do not want to spend a lot of time developing their crafting as if it was as important as developing their adventuring can just stick to the latter or not reach the highest levels of their crafting.


    This post was edited by dorotea at December 12, 2019 8:46 AM PST