Use Based systems

Last week I made a few posts to the Dev plus forums for feedback on the use based systems. As we get closer to release 21 I wanted to share these with everyone. Also, these are my usual super wordy TL;DR walls of text.

I will be posting the two super long posts right after this post in unedited form straight from the dev plus forums. Enjoy and you can look forward to the short, concise version later today if you can't get through my long winded rambling version.

Here is the more concise post on use based for those that don't like to read:



First I’ll go with the stuff for people who don’t care about the math and just want to play:

• You only have to train a skill once now to unlock it which will give you a 1 in the skill
• Skills can be set to train or not train on the upper left corner of the skill in the tree
• A small green bar on the bottom of each skill shows current progression on the current skill point
• Using an active skill set to train will slowly increase the skill
• Passive skills that are set to train will increase slowly as related active skills are used
• Players will be able to artificially train skills by using them in pointless ways but these gains will be rapidly diminishing until the player goes and uses the skills in real world settings
• In game terms we refer to skill level 40 as competent, 80 as master, and 100+ as grandmaster
• Skills will very slowly decay over the period of a month or more real time so you should not stop training a skill until you have a bit of a buffer on the level you want to keep
• Most skills require a 10 skill level in the parent skill before they can be trained
• Those who leveled more in R20 will find that they more quickly learn skills to a higher level than those who did not level up in R20
• For those who play with deck mode, glyphs are gained at 1,10,20,40, and 80

Now for those that aren’t afraid of the details:

• Underneath the covers, experience is still earned from adventuring and goes into a pool. A tiny percentage is applied to skills when they are used. Without earning exps, this pool will empty out over time
• For R21 only, that experience pool is set to the amount of experience players earned in R20
• Passives divide experience gained among all related passives set to train so they will level slower if all are set to train
• Most skills increase in effectiveness based on square root curve. So a 100 vs a 50 is not 100% more effective but rather 41% more effective.
• Parent skills that decay below the minimum level will auto train back up to the min level
• Decay is currently set to lose 1 full point every 40 days of accumulated loss
• Decay is only applied on death or skill advancement
• Decay only accumulates up to a maximum of 2 days worth of loss
• Behind the scenes there are still player levels but they are only used as a small modifier on hp and focus pools and for modifying attack chances on NPCs that are significantly higher level
• The curve for experience is exponential so every 10 points more than doubles the exps needed. So going from 90 -100 is more effort than 1-90
• No matter how powerful a target is or how many defensive skills they have, players will always have a chance equal to their critical hit chance to land a regular hit

This was obviously a HUGE switch for us to make so not everything is quite in place for this release. The following are some future features we plan on adding but are not in R21:

• In future releases there will be ways to stop decay (likely through the Oracle)
• Through the use of special crafted books, grandmasters will be able to create books that will let others use them and learn a skill up to a certain point
• Achievements and titles will be added for those who reach Grandmaster status
• Grouping with a player with grandmaster skills will allow party members to more quickly advance in those skills
• We are considering exposing more of the numbers behind the scenes
• Trainers will allow players to get a lesson in a skill that will give the player a buff that increases training speed in a specific skill for some amount of time
• Not all skills will be available at all trainers and prices will be adjusted to be more varied
• Some skills will not yet scale with skill level in R21

 

For those who haven’t been around for a while, for the first 6 or 7 releases I would write up a massive wall of text style post each release where I just broke down one or two topics in detail. A few clever people identified that the departure of the “Chris’ wall O’ text” posts coincided with the arrival of my daughter. Sadly they were dead on target with that guess and I really just had to stop because I wasn’t able to work 60-70 hours a week and something had to give. This has made Smack very sad…

Well, for R20 I’m making time to write up a few big posts because I’ve got some exciting stuff to talk about! Possibly the biggest change since we started the project. Drum roll please….

Let’s talk about the impending switch to a use base system!

So I’m going to break this brain dump into 2 parts. This first part represents how it appears to the players. I’ll let dev plus people give feedback on it and then assuming feedback is generally positive, I’ll post the second half next week where I explain the specifics of how the system accomplishes its goals.

So outside of a handful of not so constructive people, I think most people acknowledge that we listen to players and change things when needed, almost to a fault. (PVP people, I swear you’ll get one of these style posts soon! I’m looking at you Tek/Abydos/Isiah) So this was an idea I started fishing around the office about 6 or 8 months ago. Our office policy for when to take PTO is “There is never a good time to take PTO so any time is a good time!” I think we’ve finally realized the same is true of major project design changes… “There is never a good time to make a major design change so (if it really needs to happen) any time is a good time!” Well, now is not a good time to make this change so that must mean it is a good time!

Based on a number of reasons we finally decided this needed to happen. The largest reason for it was definitely player feedback. We were also struggling with how to make crafting compelling and immersive in the long run and too many times the answer started with “Well, if we were using a use based system…” Also, while we all still feel we’re not a game that requires extensive grinding, switching to to a use based system will make it even less grindy for players.

The reason we hadn’t switched yet was because no one had a design that we felt was both implementable without massive rework of systems, which was fun, and that solved some of the more serious flaws in the use based systems of UO and SWG. After several more months of thinking and planning, I believe we have something that meets all of those criteria!

So there were a number of problems we were trying to solve with the system but the biggest problem can pretty much be summed up as: “Stack of Pennies”! For those not understanding, in UO/SWG, since everything was purely use based, the most efficient way to advance your skills was to take a stack of pennies (or a wrench if you’re Burningtoad!), put it on the A key and then go to bed for the night. Today that stack of pennies would be replaced with a keyboard macro or robot program. While it is fun to hear the stories about people who became Jedis in SWG by dancing for a day and then tumbling and using medic on themselves for a day completely through macros, this is not the game we really want to make!


So let’s get to talking about how it will appear to the player:

Skill trees will generally look the same as they do now. The only real visible changes is there will be a small toggle icon for train/do not train, a small bar going across the bottom of each skill icon, and the number above each skill go from 0-100 instead of 0-5. There will also be a few new skills at the tops of some of the trees that are effectively tier 0 skills that represent things like polearm base attack skill, bow base attack skill, etc. Mousing over skills will show when the next glyph is unlocked for those using random decks. The effectiveness of the skill will now be tied to the 0-100 number though and not the number of glyphs unlocked.

Skills will start at zero and training takes them to a 1. That unlocks the skill and allows the player to use the skill. For active skills, when a player uses the skill successfully, the small bar on the skill will go up slightly assuming they did something useful. If they failed or missed, it will still go up some smaller amount. For passive skills, they will simply progress slowly while the player is actively playing. Skills will default to “Train” but can be toggled to not train which will allow other skills to progress slightly more quickly.

Repeatedly using a skill in a pointless manner (attacking a training dummy, casting fireballs at cows) will actually improve the skill some but will have rapidly diminishing returns until you go do real stuff. Adventuring, PVP-ing, crafting, will all show much faster advancement in skills.

All skills go from 0 to 100. Skills that gain glyphs generally gain an extra glyph at 1,10,20,40, and 80. Skills that have more levels than 5, gain at evenly distributed increments. For example, for the Strength passive, you gain a point of strength for every 2 skill point increases in the strength skill because 100/50 = 2.

Passives will advance pretty slow but you will be able to go to a trainer and pay for a lesson with most skills. The lesson will give you a buff that helps you increase skills faster for some amount of time though that will come at a cost to the speed at which other skills advance.

All skills will VERY slowly decay over time. Current target is that if you set a skill to not train, it would take about 40 days to lose a full point. That decay will only be applied at death or when gaining another skill point. There will also be a cap to how much decay can be accrued so players who don’t die or gain any skills for a month, aren’t suddenly going to get a bunch skill decreases on their next death. Also worth noting that death does not accelerate that decay process, it only applies it.

There are no hard caps on how many total skill points a player can have. Due to the slow decay there is a sort of a soft cap where at some point, no matter how frantically a player tries to advance everything, they will not be able to keep everything at max. Also, the higher the skill is, the tougher it will be to maintain BUT the same 40 days to lose a point still applies!

Player hit points and focus pools both slowly increase over time and are based on their stat passives. Just as with the existing system, players can expect this to not be a huge curve though.

A handful of real crafting skills will be coming online this release as well and they too will use this system. Also, the salvage system will make it for R21. This will let a player break down finished items into its base components. This conversion is a lossy process but how lossy will depend somewhat on player’s skill in salvaging.

Also, keep in mind that players will be able to salvage many of the items that they find on creatures. So now those useless rusty swords suddenly become not so useless as they can be taken by a blacksmith and turned into iron scrap which can then be crafted into an ingot. All the while earning the player skill up options with the use based system!

Also, for those worried about there being a wipe to drop this in, don’t fret. We will be wiping skills BUT the way the system has been engineered means that players who are highly experienced now will advance FAST in the new system. So don’t feel like you shouldn’t adventure in R20. Go adventure, PVP, and craft like the wind!

That is it in a nutshell! Next week I’ll post the second half of this brain dump which will include details of some of the mechanics that are going on behind the scenes. See you next week!

 

Ok, as promised, I present to you the second half of the use based wall of text! My First post was all about how it appeared to the player. This one is going to be for those who want to dig a bit deeper and understand what is happening under the covers. If you don’t care, or if you don’t like math, you should probably stop reading here.

So when I’ve referred to the system before I’ve called it a hybrid use based system. In a pure use based system, just using a skill would advance it and players could become masters at skills through repetitive, often macro-able methods. While we wanted a use based system we also wanted to limit that type of artificial behavior as much as possible.

Our system actually still has players earning experience through in game actions. Those experience points go into a pool of “to be spent” experience. Then using skills spends points out of that pool on the skill to level it up. How many points spent on each skill use depends on the current level of the skill, the size of the experience pool, whether the skill use was successful or not, and a few other secret items. The amount would be a certain % of the pool for the base value so having a large pool of experience built up will accelerate the process.

The amount applied per use will be capped so it requires a minimum number of uses to advance. The minimum number of uses starts around 5 for most skills and advances up to 100+ at the highest levels.

Each skill tracks its own experience and levels up similar to how players level up in the current system. Each skill requires far less (varies widely but around 1/100th) experience than the current leveling system. This means that players will see tons of incremental progress and if they choose to only work on a couple of skills, they will increase at a MUCH faster pace than leveling.

Skills will have a difficulty value associated with them that we likely won’t show. This is how many times more costly a skill is to level. Some things that have a very limited opportunity for use, like taming, might level more easily. Other things that are immensely powerful will level more slowly, like raising stats. Leveling str/dex/int will be in the order of 5-10X average skills.

Active skills get experience applied on use. Passive skills gain experience very slowly over time. Players will also be able to pay trainers for a lesson for a specific skill which will grant a buff that applies more exps to a skill per use.

Now for the part that players will hate BUT that is key for long term survival of the game and allows us to have really fast non-grindy progression, skill decay! So in addition to being more realistic, the purpose of decay is to allow us to not have hard skill caps and to make it so we can let people level up skills faster but still keep some control mechanism. I’ll get into the math of that in a bit but first also let me talk about why situation X isn’t a problem.

What is situation X? Well, hopefully that is your situation no matter how you play! The decay is only accumulated for around 2 days max loss. The accumulated decay is only applied on death or on skill gain. The decay is based on real time whether players are in the game or not. So players who want to go PVP and expect to die 20 times in afternoon are not going to suffer any more than someone who went adventuring and died two or three times. Players who are just logging in for some role playing and saying hi to friends won’t suffer any loss. Players who are in the military and get deployed for 2 months will only be coming back to around 2 days of decay.

Ok, next tidbit I’ll throw out is our first rough target for how much loss players will suffer. I won’t share exact numbers for how we’re doing the loss but I will give you an actual example of a real world situation. A player is awesome and levels up 20 of their skills to 80. Then they get a job and at the same time have a kid. They login every day for a year straight and die every day without earning any experience, you know, because they have a kid. This is pretty much the worst case, 365 days in a row of 1 day’s worth accumulated exp loss. So at the end of 1 year what do you think their skills will be at? 1? BUZZ! 20? BUZZ! 40? BUZZ! At the end of one year they will only have dropped to 72. If instead they only logged in once a week and died and earned no experience, they would be at 77 at the end of a year.

Ok, now I know after seeing that example, I imagine most of you are concerned that it isn’t enough decay to slow anyone down. Well, actually it is because of how it combines with the difficulty curve for raising skills at very high skill levels. With the decay used in the example, a player would hit a soft cap, due to reaching a steady state of losing experience at about the same rate they are earning experience once they hit around 400 times their daily earn rate.

So even casual players will be able to get a fair number of skills into the 80 range and keep a few dozen at 60+. The real challenge is going to be for those players that want to keep 1 or more skills in the 100+ range. Remember, that the difference between a 100 skill and 80 skill is around 10% but that last 10% costs 5 times as much as it took to get from 1 to 80! If players want to push beyond 100 they can but that curve gets steeper and steeper!

To give players a rough idea of the effort involved in leveling up a skill from from 1 to 80, which is what we consider to be master level, the effort required is about the same as it takes to get to level 28 right now. With the effort it currently takes to get to level 50, players could master 5 skills and have 30 additional skills to 40.

Things are MUCH more forgiving for those who are less concerned with becoming grandmasters. If a player just wants to get skills to 40 (which unlocks 3 out of 5 possible glyphs), then they should be be able to get 15 skills to 40 in the time it currently takes to go from level 5 to 20 under the current system. And just a reminder, 40 is approximately 70% of the effectiveness of a grandmaster.

So why spend the effort to be a grandmaster? Because to some players, that last few % is all the motivation they need to work for it. Additionally, becoming a grandmaster in a skill will let you unlock the grandmaster title that you can proudly display. What better way to attract new customers than to proudly display your “Grandmaster Weaponsmith” title. Also, if you’re trying to send a message to would be challengers, just fly your “Grandmaster Swordsman”. Those titles likely won’t make R21 but those, and probably some Steam achievements, will make a later release.

For R21 players, it will feel a bit like a player wipe release because all of your skills will be reset. So for a brief moment, all players will be on a pretty even playing field! Actually what we’re doing is moving all your current experience to that pool to be spent on skills. Also, recall that how much gets applied to skills on each use is based on a tiny percent of the size of that pool. So if PlayerX currently has a half million exps at the end of R20, they will start R21 with a half million exps in their pool to be applied to skills on use.

So the other thing that this means is that those players with lots of saved up experience will actually be able to advance quite a bit just from practicing. Over time though, they will burn through their pool and need to go adventure, PVP, or craft more challenging items to refill the pool. The other thing to expect is to see the hard core people behaving oddly as they try to be the first person to become a grandmaster a particular skill.

Ok, that is probably enough blathering to satisfy my wall of text for today. I look forward to the questions and answering concerns as I’m sure there will be many.

See you guys in game!

 

I think you're trying to focus on the implementation details too much. That difficulty was on us and is done.

For the end user the system as simple as use a skill and you get better at it. Not sure how it could be any simpler or more immersive.

 

Prereqs are set to 10 and it really only takes a few dozen uses to get from 1 to 10. We will continue to adjust and possibly remove some of these as we go forward. There will be no measurable hindrance to having lots of skills at low levels while you try to decide what to specialize in.

 

What sound bug?

As for the system, I don't expect players to sit there and think about experience pools. I don't want them thinking about them. I felt it necessary to tell people what was going on though so players wouldn't be endlessly trying to data mine only to come to false conclusions. What we hope the players do is:

Step 1: Go on adventures and/or craft stuff!
Step 2: Use the skills you want to get better at
Step 3: ...
Step 4: PROFIT