 |
Use-based systemsJuly 18th, 2006 |
Recently, there’s been some chatter on the [ultima] Yahoogroup about macroing. Lots of talk about how the fixes for detecting macroing software are coming years too late. But we waged a silent war against macroing from very early on — not because of the gold farmers, as that was much less of a concern back then, but because of the basic mechanics of a use-based system.
These days, use-based systems are most familiar to everyone because of Oblivion, probably. And sure enough, there’s plenty of tales of how amusing it is that the use-based system there encourages strange behaviors that do not fit the expected behavior in the game: people jumping everywhere in order to improve the stats and skills related to dodging, that sort of thing.
The basic definition of a use-based skill system is one where you have a chance of improvement every time you use the skill. So everyone has access to every skill from the beginning (or, in a variant, access may be unlocked by other means, such as being taught), and as you exercise the skill, your chance of success goes up.
The attraction is that it leads to a more freeform game experience. Instead of tracking arbitrary “experience points” that can be applied to anything, leading to oddities like getting better at crafting because you have slain a lot of orcs, you can get better at crafting by, well, crafting. The downside, of course, is that it lets you sit there and do something repetitively in order to get better at it.
In the real world, this is called “practice,” and we don’t regard it as a bad thing. We force kids to sit and do scales on their piano or violin; we tell them to do a billion math problems; and we send them to sports camp so they can keep honing skills. “Practice” isn’t a dirty word at all, and generations have bemoaned the repetitive nature of it. However, the logic seems to be that in game, repetitive action like that has no place.
(from A Grammar of Gameplay)
Do you have to prepare for the challenge?
…where prep includes prior moves?
…and you can prep in multiple ways?
Does the topology of the space matter?
…does the topology change?
Is there a core verb for the challenge?
…can it be modified by content?
Can you use different abilities on it?
…will you have to to succeed?
Is there skill to using the ability?
…or is this a basic UI action?
Are there multiple success states?
…with no bottomfeeding?
…and a cost to failure?
That’s balderdash. The issue isn’t the repetitiveness. Many games allow repetitiveness within their framework even if they don’t require it. You can practice headshots in an FPS, or jumps in a platform game. In fact, you pretty much have to. So it’s not practicing that is the issue — it’s that, in classic “theory of fun” style, practice that isn’t challenging your skills and helping you learn is boring. In the headshot example, you probably start out not able to hit headshots reliably — as you practice, you take on tougher and tougher shots, and keep challenging yourself. In the musical instrument example, new pieces and new techniques open up, much like new levels in a game. (There’s a reason we say you “play” a piece of music!)
But in use-based systems, at least how they are structured today, this isn’t the case. There’s no challenge in jumping off a hill in Oblivion, and therefore, the grind rears its ugly head.
Now, the biggest barrier here isn’t mechanical, it’s psychological. Players are bottomfeeding the system (fun checklist item again!). There’s no cost to failure. The topolgy of the space is irrelevant. There are no differing abilities. No skill required. All of these can be avoided if players simply quit doing unfun stuff. But that particular psychological barrier isn’t one that is going to fall; humans seek advantage, and just because it may be more challenging to forage for one’s own food in the winter, it’s smarter to instead hit the grocery store even though it’s distinctly less challenging.
So the systems must adapt to the psychology. Some players will “play it straight,” but even many of those will succumb to temptation once they see other players gaining advantage over them by using the bottomfeeding method.
UO used a straight-up use-based system. Each time you used a skill, it had a chance of going up. In fact, originally, it had a chance of going up if you observed someone else using it. There was no difficulty rating on the task, so there was no way to say “you won’t learn from this because it’s beneath your skill threshold. What’s more, skills that you didn’t use originally decayed away. A player was allowed only 700 skill points total — when you hit that threshold, and gained a point, a point was automatically lost from somewhere else — not even the least used skill, as I recall, but any of them, on the grounds that ify ou used something occasionally, it’d counteract a bit of the loss.
The first big problem that we ran into with use-based systems was that every skill was used at a different rate. Swords were swung quickly. Talking to the dead was done rarely. If every skill had the same chance of advancing and the same amount by which it advanced, then swords would be mastered incredibly quickly, and skills that could not physically be performed as often because of the lack of opportunity would advance more slowly.
The solution to this was to dynamically track the frequency of use of all skills. That way we could look at swords, see that skill checks for it happened ten times more often than our baseline, and that spirit speak usage happened ten times less than our baseline. We could then make the chance of swords advancing when used to be 1/10th the chance of the baseline, and the chance of spirit speak advancing to be ten times the baseline. If skills became more popular (lets say, a whole bunch of spirit speakers started to play the game) then spirit speak would be checked more frequently, and then it would move closer to the baseline chance of advancing. It would be a self-balancing mechanism for when payers rushed to skills because they were “easier to advance.”
It didn’t work. For one, I don’t think there was a baseline in place; instead, everything started getting slower and slower to advance. As things got slower, players responded by macroing the skills, chasing after the one time in a thousand that they would go up. This meant that now the chance became one in ten thousand. The thing players had an infinite amount of was time — it didn’t matter to them whether they macroed for an hour or overnight. What broke was the feedback mechanism — the reward of advancement came so incredibly infrequently that playing “the right way” no longer had any incentives.
A few thing could have fixed this, in theory. Only advancing for taking on actual challenges, rather than doing something you already know how to do, such as swinging at a practice dummy. Soon people started trapping powerful monsters in their houses to macro on instead. Ooops. Next solution?
One such solution was 8×8. I’ve mentioned before how UO internally divided up the map into “chunks” of 64 tiles, 8×8 square, and stored the invisible resources on a “chunk egg,” an invisible object that represented the amount of wood, grass, and os on to be found there. This also meant that the chunk egg was a unique object with an id.
A feature of how random numbers work on computers is that they aren’t actually random. They are seeded by a value. If you give a random number generator a given seed number, say 1234, then it will always spit out the same result. With a use-based system, we were rolling dice to see when people were going to advance. The inputs were things like what skill they were using, what skill level they were currently at, and a random seed.
I had already moved off the team by this point, but Sunsword came to ask me for advice. I suggested that instead of seeding the random number generator with the system clock (so that it would always spit out different numbers) we instead should seed it with the id of the chunk egg. This meant that you’d get a predictable result for a given person, skill, and skill level. If a guy who was standing at a forge started forging, and failed, he’d be guaranteed to fail there until the heat death of the universe. The only way to get it to succeed would be to change the random seed number. This could be accomplished by putting different inputs into the seed: a different skill level (by, say, going elsewhere and managing to advance) or a different location (by just going elsewhere).
The intent was to get people to move about — two or three moves would carry you a whole screen away from where you were previously.
Players still managed to figure out, via pure scientific experimentation, as far I know, that if they regularly moved 8 tiles away, they would have an improved chance of advancemrnt. And thus was born the phenomenon called “8×8 macroing,” which involved moving 8 tiles via an automated macro, after a given failed attempt. I regard the player discovery of this as a great example of one of the Laws:
No matter what you do, players will decode every formula, statistic, and algorithm in your world via experimentation.
Runesabre never liked the skill table, and once UO was under his domain, he removed it in favor of a hand-coded table. A hand-coded advancement rate table will always need to be updated as players discover exploits that drive them towards one skill or another, so I strongly recommend that anyone dealing with a use-based system maintain metrics of rate of use anyway; whether they then manually copy those values into their advancement table, or do it automatically is to my mind largely irrelevant. The challenge is fixing the feedback loop.
Once the table was removed, a series of increasingly outrĂ© systems such as “guaranteed gain” and “power hour” started becoming required, as rates were adjusted to try to bring some semblance of balance to things. I was gone by then, so I don’t know what the logic behind those were.
How to make use-based systems work? Well, one way is to move to “results-based” logic. This is the philosophy that it isn’t using the system that should grant you advancement, but instead, getting a concrete result. This is what leads to experience points; in SWG, we said that you could have results classified by what sort of result, thus giving us multiple types of experience.
The sorts of results we wanted, however, weren’t solely the act of building something (because that could still be done in a closed room making crap nobody wants) but instead results that were somehow socially valid. From this wan born the notion of crafting XP being granted when someone else actually used something you crafted: a validation that what you made actually worth something to someone. Alas, this system was marginalized in favor of themore direct feedback of getting XP when you made something; to really work, it would have needed to account for all of the XP, or else bottomfeeding would once again ensue (as it did).
In the long run, we ended up tracking the rate of usage during beta, and setting static XP requirements per skill box based on the rate of usage of skills, just like that table in UO. We also added the dfifficulty rating to each challenge that users were overcoming (be it a monster or something they were crafting) so that we could adjust the XP gain accordingly.
A fun little story for those who think SWG was still too grindy: three days before the game launched, the designer in charge of advancement rates went through them and made them all ten times slower. His concern was that players would max out the game in a matter of weeks, and then quit. This actually hit the live beta servers, and I reverted it back out as soon as I saw how excruciatingly slow it was: making literally thousands of blaster bolts to advance one skill box. I felt that I’d rather people exit the game having had fun, than stay and not have fun, or quit because it was too grindy.
Of course, it was still too grindy.
So, how to make use-based systems work? Well, I think you’d have to run down the checklist, and try to fix each of those problems. The result might not look much like a use-based system once you were done.
One system that we were planning with the cancelled Privateer Online was to drop this advancement metaphor altogether. Instead, you were certified to do things. You earned certification by doing a minimal amount of training, defeating a challenge of some sort, and then most critically, paying your dues. Dues would be currency paid to the NPC guilds in the game. If you couldn’t pay the dues, then you’d lose the certification.
The instant reaction to this system was, “but people could buy their way up to the top!” And my reply was, “so what?” It’s not it’s regular skills — instead, think of it as a mechanism for people to play the way they want to. If you want to set up a merchant empire, find a sponsor, and try to get going. And if you can’t cut it, you’ll lose all your money on expensive dues, and slip back down to the level of gameplay you can actually handle. For a game about commerce, this sounded fine to me.
At the heart of all this, though, is the problem that skills, levels, abilities — these are properly regarded as enablers, and not as the point. A skill is supposed to be a tool you use to defeat a challenge. If the game becomes getting the tool, then you have a nice little rat race going that is inevitably going to be a grind. You need to make the skills something you pick up in order to do something else. At that point, whether you make people earn them or just let people pick up any they want becomes much less important, because you’ve placed the burden for fun where it should be: on the game, and not on the advancement.
You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.
[...] Comments [...]
[...] So Raph has a great post up at his blog about use-based skill systems, and how they had designed the one for UO and how it changed over time. He also discusses SWG and its skill system. I don’t want to spoil the whole thing by posting the conclusion, but it’s the part that made me really think, so I’m going to post it here. [...]
[...] I always read his blog. Even if you agree or not with him, this man is doing great from the educational point of view, I feel I’m learning a lot from his reading. There are not so many developers writing theirs ideas so openly. Some nice thing digged from last article:http://www.raphkoster.com/2006/07/18/use-based-systems/”A fun little story for those who think SWG was still too grindy: three days before the game launched, the designer in charge of advancement rates went through them and made them all ten times slower. His concern was that players would max out the game in a matter of weeks, and then quit. This actually hit the live beta servers, and I reverted it back out as soon as I saw how excruciatingly slow it was: making literally thousands of blaster bolts to advance one skill box. I felt that I’d rather people exit the game having had fun, than stay and not have fun, or quit because it was too grindy.” [...]
[...] [Quote by: Ralph Koster] These days, use-based systems are most familiar to everyone because of Oblivion, probably. And sure enough, there’s plenty of tales of how amusing it is that the use-based system there encourages strange behaviors that do not fit the expected behavior in the game: people jumping everywhere in order to improve the stats and skills related to dodging, that sort of thing. The basic definition of a use-based skill system is one where you have a chance of improvement every time you use the skill. So everyone has access to every skill from the beginning (or, in a variant, access may be unlocked by other means, such as being taught), and as you exercise the skill, your chance of success goes up. The attraction is that it leads to a more freeform game experience. Instead of tracking arbitrary “experience points” that can be applied to anything, leading to oddities like getting better at crafting because you have slain a lot of orcs, you can get better at crafting by, well, crafting. The downside, of course, is that it lets you sit there and do something repetitively in order to get better at it. http://www.raphkoster.com/2006/07/18/use-based-systems/ [...]
[...] At the heart of all this, though, is the problem that skills, levels, abilities — these are properly regarded as enablers, and not as the point. A skill is supposed to be a tool you use to defeat a challenge. If the game becomes getting the tool, then you have a nice little rat race going that is inevitably going to be a grind. You need to make the skills something you pick up in order to do something else. At that point, whether you make people earn them or just let people pick up any they want becomes much less important, because you’ve placed the burden for fun where it should be: on the game, and not on the advancement. http://www.raphkoster.com/2006/07/18/use-based-systems/_________________Trace Silverhawk - Weaponsmith, Smuggler Alts: Darice Starshadow - Master Shipwright Varina Llaspe (Apprentice Weaponsmith) Jhaime the Wanderer Curiousity may have killed the cat…. ….but the cat died happy. [...]
[...] Raph Koster wrote an article talking about use-based skill systems in MMO’s and about how they worked out in Ultima Online and StarWars Galaxies. Just another example of synchronicity since I planned on following up A Distinct Lack of Class with a post about skill systems. [...]
[...] Use-based systems on Raph Koster Use-based systems on Raph Koster Quote: [...]
[...] That targetting stuff sounds cool. They should have had that pre-CU. Could have called it, oh I dunno, /ui action targetAtCursor or something.And imagine the useful tricks Rangers and Wielders could have had if they differentiated between their combat and look-at targets. My oh my the old game would have been a lot better with these new innovations!That must be why the old game was massively unpopular and NGE remains the number one selling game of the day.Light reading. [...]
[...] Posted: Sunday, 23 July 2006 10:41PM If anyone is interested, there is a long (and recent) article about the advantages of use-based systems in RPGs over experience-based systems, on the site of Raph Koster (one of the original designers of Ultima Online and Star Wars Galaxies):Click Here [...]
[...] I made a mildly controversial comment over on Raph’s blog that use-based systems kinda suck for achievers. Levels make it much easier to compare e-peens and see who has more time to spend. [...]
[...] http://www.raphkoster.com/2006/07/18/use-based-systems/ It seems to me, as I’ve been discussing with the man Jesus, that a lot of things about MMOs, from the features down to the very mechanics, are only included in recent games because people expect them to be, so I think game designers have a tendency to design around what people expect rather than designing the game the way they think it should be designed. Use-based skills are a very good example of that. [...]
[...] WoW BlueTracker: Min-maxing the Arena system WoW BlueTracker Home | Recent | CS Posts | RSS | Search | Archive | News | Contact Poster: Igniferroque at 8/16/2006 5:26:18 PM PDTSubject: Min-maxing the Arena system I suggest the developers share the exact details of the Arena system scoring system now or during the beta. Experience has shown that players will a) reverse engineer any system you design and b) min-max that system once reverse engineered. So tell us how the system works now, we’ll tell you how we’ll min-max it, then you can decide if that’s the behavior you’d like to encourage. I was there for the birth of the honor cap on Bleeding Hollow. A few weeks after you began posting the honor scores and rankings, the system was reverse engineered and posted publically. We were in an AB group back in the day when you actually had to be in Arathi Highlands. A friend, an aeronautical engineer, was debating whether or not to push to GM that week or the week following. We disbanded the group for the hour while he "did the math." When we reformed, he had calculated that he couldn’t make GM that week, told a peer to push for it instead and that he’d get it the next week. So it began. I have no idea how the WoW designers feel about Raph Koster, but you’ll find another example of min-maxing here: http://www.raphkoster.com/2006/07/18/use-based-systems/ It shows how, by experimentation, people reverse-engineered UO’s use-based skill system. [...]
[...] The problems of a Use-Based System Saw this linked on the WoW general forums. It is a pretty interesting read. It looks like it’s excrutiatingly long, but the vast majority of this page is just comments on the main story http://www.raphkoster.com/2006/07/18/use-based-systems/ __________________ RIP - Pmoney - Level 60 Undead Warlock [...]
[...] WoW BlueTracker: Min-maxing the Arena system WoW BlueTracker Home | Recent | CS Posts | RSS | Search | Archive | News | Contact Poster: Igniferroque at 8/16/2006 5:26:18 PM PDTSubject: Min-maxing the Arena system I suggest the developers share the exact details of the Arena system scoring system now or during the beta. Experience has shown that players will a) reverse engineer any system you design and b) min-max that system once reverse engineered. So tell us how the system works now, we’ll tell you how we’ll min-max it, then you can decide if that’s the behavior you’d like to encourage. I was there for the birth of the honor cap on Bleeding Hollow. A few weeks after you began posting the honor scores and rankings, the system was reverse engineered and posted publically. We were in an AB group back in the day when you actually had to be in Arathi Highlands. A friend, an aeronautical engineer, was debating whether or not to push to GM that week or the week following. We disbanded the group for the hour while he "did the math." When we reformed, he had calculated that he couldn’t make GM that week, told a peer to push for it instead and that he’d get it the next week. So it began. I have no idea how the WoW designers feel about Raph Koster, but you’ll find another example of min-maxing here: http://www.raphkoster.com/2006/07/18/use-based-systems/ It shows how, by experimentation, people reverse-engineered UO’s use-based skill system. We have precious little information on how the Arena system will work. It will be like a chess rating system and there will be a minimum of 10 games required, as revealed by Jeff Kaplan during the Joystiq interview. To demonstrate our affinity for min-maxing, I’ll walk you through my ideas based on these two facts. My first thought is to get a group together and load them for bear: Flasks of Wisdom, Supreme Power, Titans, even Petrification for those times you need a three minute iceblock equivalent. Then, play the minimum number of games in this state. Don’t even have to play them all at once. Pick and choose our times when we’re likely to encounter the least resistance. During prime time raiding hours. In the middle of the night. Mid-morning. Whenever you’re likely to meet up with people who are PvPing honestly, for fun instead of simply to win. Am I on target? Probably not, but I might be somewhere in the neighborhood. Regardless, you see how we’ll think it through. Share with us the details of the system and we’ll tell you how we’ll going to game it. And then you judge for yourselves if that’s the game you want us to be playing. As an idea to tune the system as I understand it to be - I didn’t put too much thought into it because I’m just speculating here - give us an incentive to play more games. The current "grind or get demoted" disincentive lends itself to boring gameplay and you’re absolutely right to move away from it. To continue along the lines of your "honor point" currency, let us earn points to buy Arena loot. Instead of a winner take all system at the end of the three months, have a floor. If the system has a max rating of 10, everyone who has an 9 or more can ‘buy’ gear from the epic selection. Everyone who has a rating between 6 and 9 can ‘buy’ gear from the rare selection. Maybe allow us to carry those epic or rare points over from season to season, thus not penalizing the people who fail to play 10 hours a day but still rewarding those who do. These are conjectures based on an incomplete set of facts. So give us all the information of the Arena system before beta. By giving us the information now, we can give you feedback before its implemented and before it becomes so late you have patch in any changes. Let me finish by saying, I’m all for the changes you’re implementing in the Burning Crusade. Bravo. I’m very interested in getting this iteration of the PvP system right and many of us are more than willing and more than able to help.[ post edited by Igniferroque ] The enemy’s gate is down. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9552122 Poster: Drysc at 8/17/2006 10:01:17 AM PDTSubject: Re: Min-maxing the Arena system We’ve kept a lot in mind while creating the arena system to help make it as centered on player skill as possible, and not time invested. We’re looking to help create a place where players can track their achievements and improvements in a fun PvP environment. We’re no strangers to the tactics and extent at which players will take a game to help ensure they reach their goal sooner or easier. It can however be minimized by simplifying a system so that the possible benefits available to a player are specific and tangible to everyone, and that the potential for exploitation is reduced as much as possible. With a complex system there will always be that something extra that can give you an edge over someone else, it’s the size and availability of the edge that makes the difference though. We think we’re making some great progress in creating place where players can come and test their PvP skill against others, have fun, and get some pretty cool rewards out of it too. Just because you asked, here’s some more information on the arena system. Hopefully this quells some of your concerns, and squashes a few assumptions. Please keep in mind that this information is subject to change, but these are our current plans for the PvP Arena system in The Burning Crusade expansion. When a player enters an arena battle, all buffs and conjured items are purged/deleted. This is done to ensure no outside buffs or items from members outside of your team are able to be used in an arena battle. No consumables other than bandages and conjured items can be used while in the arena. There will be a waiting area similar to the current battlegrounds where players on the team will be able to conjure items and cast buffs. Abilities/spells/items with cooldowns longer than 15 minutes cannot be used while in the arena. Each arena team will be given a rating, and will be matched up in the arena queue against teams of a similar rating (matching within that specific cross-realm battlegroup). As time progresses the matching system will broaden its search up to a limit if a closely rated team cannot be found. Ratings adjustments are made in the same way that ELO ratings adjustments are made. This system works by adjusting each arena team’s rating from a specific match based on the rating of the team they are up against. If a team wins against a team of a higher rating, the increase of their arena rating would be much higher than if they had won against a team of a lower rating. This system will help promote players improving their skill and besting teams which may have previously held the top spots. At the end of each week, characters receive arena points based on their team’s rating. A team must have fought a minimum number of battles for the week, which we have announced as currently set to 10 games. A player must have played in at least 30% of the team’s battles that week in order to be eligible to receive arena points. Each eligible player will then receive a fraction of the teams total awarded points to spend on arena rewards. Players may spend their points on arena rewards, or save their points until they have enough to get the rewards they desire. There is a limit to the number of points a player can stockpile, which is currently planned to be as high as the most expensive item. At the end of a season, players on a team are given a title based on their team’s relative position on the ladder which will last through the next arena season. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9559694Poster: Drysc at 8/17/2006 10:11:37 AM PDTSubject: Re: Min-maxing the Arena system Q u o t e: So, Drysc, if you’re still reading this, that means potions are out? Not just health/mana, but things like Moongoose, Troll’s Blood, etc. Correct. Q u o t e:So rogues can’t use poisons, huh? Nothing posted would imply that you would be unable to use temporary item buffs. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9559861Poster: Drysc at 8/17/2006 10:11:37 AM PDTSubject: Re: Min-maxing the Arena system *edited post* Q u o t e: So, Drysc, if you’re still reading this, that means potions are out? Not just health/mana, but things like Moongoose, Troll’s Blood, etc. Correct. Q u o t e:So rogues can’t use poisons, huh? Nothing posted would imply that you would be unable to use temporary item buffs. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9559861Poster: Kalgan at 8/17/2006 10:23:07 AM PDTSubject: Re: Min-maxing the Arena system Q u o t e: Nothing posted would imply that you would be unable to use temporary item buffs. In addition, you can use spells/abilities that require reagents if you so choose (as long as they still meet the other criteria). So, for example, Blind and Prayer of Fortitude can be used, while Reincarnation cannot (due to its cooldown). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560041Poster: Kalgan at 8/17/2006 10:36:35 AM PDTSubject: Re: Min-maxing the Arena system Q u o t e: I wonder if it would be possible for a team to feed *all* of the awarded points to a single player? A rotation, like warrior + 3 paladins… the warrior stays in, the paladins rotate. At the end of each week, each character receives an equal number of the points. For characters that did not play in at least 30% of the battles, the points are lost. So no, there shouldn’t be any way to inflate the number of points any single player on the team receives. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560259Poster: Kalgan at 8/17/2006 10:39:51 AM PDTSubject: Re: Min-maxing the Arena system Q u o t e: It was not a good system, and it was too easy to exploit. All you had to do was throw a couple of games after a win streak to lower your ELL and then get easier matches. Sure your win/loss ratio would go down, but it was the easiest way to climb the ladder. I remember the 2v2 teams having mediocre win/loss ratios, but would always be at the top of the ladder. edit: spelling I think it’s worth noting that War 3 doesn’t use an ELO system… it has some elements that are based on ELO, but it’s significantly hybridized. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560304Poster: Kalgan at 8/17/2006 10:23:07 AM PDTSubject: Re: Min-maxing the Arena system *edited post* Q u o t e: Nothing posted would imply that you would be unable to use temporary item buffs. In addition, you can use spells/abilities that require reagents if you so choose (as long as they still meet the other criteria). So, for example, Blind and Prayer of Fortitude can be used, while Reincarnation cannot (due to its cooldown). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560041Poster: Kalgan at 8/17/2006 10:36:35 AM PDTSubject: Re: Min-maxing the Arena system *edited post* Q u o t e: I wonder if it would be possible for a team to feed *all* of the awarded points to a single player? A rotation, like warrior + 3 paladins… the warrior stays in, the paladins rotate. At the end of each week, each character receives an equal number of the points. For characters that did not play in at least 30% of the battles, the points are lost. So no, there shouldn’t be any way to inflate the number of points any single player on the team receives. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560259Poster: Kalgan at 8/17/2006 10:48:49 AM PDTSubject: Re: Min-maxing the Arena system Q u o t e: Will we see the class balance of the team we’re up against before we go into combat? No, you will not see the composition of the enemy team until the battle has begun. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560450 [...]
[...] I suggest the developers share the exact details of the Arena system scoring system now or during the beta. Experience has shown that players will a) reverse engineer any system you design and b) min-max that system once reverse engineered. So tell us how the system works now, we’ll tell you how we’ll min-max it, then you can decide if that’s the behavior you’d like to encourage. I was there for the birth of the honor cap on Bleeding Hollow. A few weeks after you began posting the honor scores and rankings, the system was reverse engineered and posted publically. We were in an AB group back in the day when you actually had to be in Arathi Highlands. A friend, an aeronautical engineer, was debating whether or not to push to GM that week or the week following. We disbanded the group for the hour while he "did the math." When we reformed, he had calculated that he couldn’t make GM that week, told a peer to push for it instead and that he’d get it the next week. So it began. I have no idea how the WoW designers feel about Raph Koster, but you’ll find another example of min-maxing here: http://www.raphkoster.com/2006/07/18/use-based-systems/ It shows how, by experimentation, people reverse-engineered UO’s use-based skill system. We have precious little information on how the Arena system will work. It will be like a chess rating system and there will be a minimum of 10 games required, as revealed by Jeff Kaplan during the Joystiq interview. To demonstrate our affinity for min-maxing, I’ll walk you through my ideas based on these two facts. My first thought is to get a group together and load them for bear: Flasks of Wisdom, Supreme Power, Titans, even Petrification for those times you need a three minute iceblock equivalent. Then, play the minimum number of games in this state. Don’t even have to play them all at once. Pick and choose our times when we’re likely to encounter the least resistance. During prime time raiding hours. In the middle of the night. Mid-morning. Whenever you’re likely to meet up with people who are PvPing honestly, for fun instead of simply to win. Am I on target? Probably not, but I might be somewhere in the neighborhood. Regardless, you see how we’ll think it through. Share with us the details of the system and we’ll tell you how we’ll going to game it. And then you judge for yourselves if that’s the game you want us to be playing. As an idea to tune the system as I understand it to be - I didn’t put too much thought into it because I’m just speculating here - give us an incentive to play more games. The current "grind or get demoted" disincentive lends itself to boring gameplay and you’re absolutely right to move away from it. To continue along the lines of your "honor point" currency, let us earn points to buy Arena loot. Instead of a winner take all system at the end of the three months, have a floor. If the system has a max rating of 10, everyone who has an 9 or more can ‘buy’ gear from the epic selection. Everyone who has a rating between 6 and 9 can ‘buy’ gear from the rare selection. Maybe allow us to carry those epic or rare points over from season to season, thus not penalizing the people who fail to play 10 hours a day but still rewarding those who do. These are conjectures based on an incomplete set of facts. So give us all the information of the Arena system before beta. By giving us the information now, we can give you feedback before its implemented and before it becomes so late you have patch in any changes. Let me finish by saying, I’m all for the changes you’re implementing in the Burning Crusade. Bravo. I’m very interested in getting this iteration of the PvP system right and many of us are more than willing and more than able to help.[ post edited by Igniferroque ] The enemy’s gate is down. [...]
[...] This was my original post:Igniferroque wrote:I suggest the developers share the exact details of the Arena system scoring system now or during the beta.Experience has shown that players will a) reverse engineer any system you design and b) min-max that system once reverse engineered. So tell us how the system works now, we’ll tell you how we’ll min-max it, then you can decide if that’s the behavior you’d like to encourage.I was there for the birth of the honor cap on Bleeding Hollow. A few weeks after you began posting the honor scores and rankings, the system was reverse engineered and posted publically. We were in an AB group back in the day when you actually had to be in Arathi Highlands. A friend, an aeronautical engineer, was debating whether or not to push to GM that week or the week following. We disbanded the group for the hour while he "did the math." When we reformed, he had calculated that he couldn’t make GM that week, told a peer to push for it instead and that he’d get it the next week. So it began.I have no idea how the WoW designers feel about Raph Koster, but you’ll find another example of min-maxing here:http://www.raphkoster.com/2006/07/18/use-based-systems/It shows how, by experimentation, people reverse-engineered UO’s use-based skill system.We have precious little information on how the Arena system will work. It will be like a chess rating system and there will be a minimum of 10 games required, as revealed by Jeff Kaplan during the Joystiq interview.To demonstrate our affinity for min-maxing, I’ll walk you through my ideas based on these two facts. My first thought is to get a group together and load them for bear: Flasks of Wisdom, Supreme Power, Titans, even Petrification for those times you need a three minute iceblock equivalent. Then, play the minimum number of games in this state.Don’t even have to play them all at once. Pick and choose our times when we’re likely to encounter the least resistance. During prime time raiding hours. In the middle of the night. Mid-morning. Whenever you’re likely to meet up with people who are PvPing honestly, for fun instead of simply to win.Am I on target? Probably not, but I might be somewhere in the neighborhood. Regardless, you see how we’ll think it through.Share with us the details of the system and we’ll tell you how we’ll going to game it. And then you judge for yourselves if that’s the game you want us to be playing.As an idea to tune the system as I understand it to be - I didn’t put too much thought into it because I’m just speculating here - give us an incentive to play more games. The current "grind or get demoted" disincentive lends itself to boring gameplay and you’re absolutely right to move away from it.To continue along the lines of your "honor point" currency, let us earn points to buy Arena loot. Instead of a winner take all system at the end of the three months, have a floor. If the system has a max rating of 10, everyone who has an 9 or more can ‘buy’ gear from the epic selection. Everyone who has a rating between 6 and 9 can ‘buy’ gear from the rare selection.Maybe allow us to carry those epic or rare points over from season to season, thus not penalizing the people who fail to play 10 hours a day but still rewarding those who do.These are conjectures based on an incomplete set of facts. So give us all the information of the Arena system before beta. By giving us the information now, we can give you feedback before its implemented and before it becomes so late you have patch in any changes.Let me finish by saying, I’m all for the changes you’re implementing in the Burning Crusade. Bravo. I’m very interested in getting this iteration of the PvP system right and many of us are more than willing and more than able to help. Online Index » General Discussion » PvP gear will equal PvE gear; access to PvP gear will be limited [ 1 ] [...]
[...] Poster: Igniferroque at 8/16/2006 5:26:18 PM PDT Subject: Min-maxing the Arena system I suggest the developers share the exact details of the Arena system scoring system now or during the beta. Experience has shown that players will a) reverse engineer any system you design and b) min-max that system once reverse engineered. So tell us how the system works now, we’ll tell you how we’ll min-max it, then you can decide if that’s the behavior you’d like to encourage. I was there for the birth of the honor cap on Bleeding Hollow. A few weeks after you began posting the honor scores and rankings, the system was reverse engineered and posted publically. We were in an AB group back in the day when you actually had to be in Arathi Highlands. A friend, an aeronautical engineer, was debating whether or not to push to GM that week or the week following. We disbanded the group for the hour while he “did the math.” When we reformed, he had calculated that he couldn’t make GM that week, told a peer to push for it instead and that he’d get it the next week. So it began. I have no idea how the WoW designers feel about Raph Koster, but you’ll find another example of min-maxing here: http://www.raphkoster.com/2006/07/18/use-based-systems/ It shows how, by experimentation, people reverse-engineered UO’s use-based skill system. We have precious little information on how the Arena system will work. It will be like a chess rating system and there will be a minimum of 10 games required, as revealed by Jeff Kaplan during the Joystiq interview. To demonstrate our affinity for min-maxing, I’ll walk you through my ideas based on these two facts. My first thought is to get a group together and load them for bear: Flasks of Wisdom, Supreme Power, Titans, even Petrification for those times you need a three minute iceblock equivalent. Then, play the minimum number of games in this state. Don’t even have to play them all at once. Pick and choose our times when we’re likely to encounter the least resistance. During prime time raiding hours. In the middle of the night. Mid-morning. Whenever you’re likely to meet up with people who are PvPing honestly, for fun instead of simply to win. Am I on target? Probably not, but I might be somewhere in the neighborhood. Regardless, you see how we’ll think it through. Share with us the details of the system and we’ll tell you how we’ll going to game it. And then you judge for yourselves if that’s the game you want us to be playing. As an idea to tune the system as I understand it to be - I didn’t put too much thought into it because I’m just speculating here - give us an incentive to play more games. The current “grind or get demoted” disincentive lends itself to boring gameplay and you’re absolutely right to move away from it. To continue along the lines of your “honor point” currency, let us earn points to buy Arena loot. Instead of a winner take all system at the end of the three months, have a floor. If the system has a max rating of 10, everyone who has an 9 or more can ‘buy’ gear from the epic selection. Everyone who has a rating between 6 and 9 can ‘buy’ gear from the rare selection. Maybe allow us to carry those epic or rare points over from season to season, thus not penalizing the people who fail to play 10 hours a day but still rewarding those who do. These are conjectures based on an incomplete set of facts. So give us all the information of the Arena system before beta. By giving us the information now, we can give you feedback before its implemented and before it becomes so late you have patch in any changes. Let me finish by saying, I’m all for the changes you’re implementing in the Burning Crusade. Bravo. I’m very interested in getting this iteration of the PvP system right and many of us are more than willing and more than able to help. [ post edited by Igniferroque ] The enemy’s gate is down. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9552122 Poster: Drysc at 8/17/2006 10:01:17 AM PDT Subject: Re: Min-maxing the Arena system We’ve kept a lot in mind while creating the arena system to help make it as centered on player skill as possible, and not time invested. We’re looking to help create a place where players can track their achievements and improvements in a fun PvP environment. We’re no strangers to the tactics and extent at which players will take a game to help ensure they reach their goal sooner or easier. It can however be minimized by simplifying a system so that the possible benefits available to a player are specific and tangible to everyone, and that the potential for exploitation is reduced as much as possible. With a complex system there will always be that something extra that can give you an edge over someone else, it’s the size and availability of the edge that makes the difference though. We think we’re making some great progress in creating place where players can come and test their PvP skill against others, have fun, and get some pretty cool rewards out of it too. Just because you asked, here’s some more information on the arena system. Hopefully this quells some of your concerns, and squashes a few assumptions. Please keep in mind that this information is subject to change, but these are our current plans for the PvP Arena system in The Burning Crusade expansion. # When a player enters an arena battle, all buffs and conjured items are purged/deleted. This is done to ensure no outside buffs or items from members outside of your team are able to be used in an arena battle. # No consumables other than bandages and conjured items can be used while in the arena. There will be a waiting area similar to the current battlegrounds where players on the team will be able to conjure items and cast buffs. # Abilities/spells/items with cooldowns longer than 15 minutes cannot be used while in the arena. # Each arena team will be given a rating, and will be matched up in the arena queue against teams of a similar rating (matching within that specific cross-realm battlegroup). As time progresses the matching system will broaden its search up to a limit if a closely rated team cannot be found. # Ratings adjustments are made in the same way that ELO ratings adjustments are made. This system works by adjusting each arena team’s rating from a specific match based on the rating of the team they are up against. If a team wins against a team of a higher rating, the increase of their arena rating would be much higher than if they had won against a team of a lower rating. This system will help promote players improving their skill and besting teams which may have previously held the top spots. # At the end of each week, characters receive arena points based on their team’s rating. A team must have fought a minimum number of battles for the week, which we have announced as currently set to 10 games. A player must have played in at least 30% of the team’s battles that week in order to be eligible to receive arena points. Each eligible player will then receive a fraction of the teams total awarded points to spend on arena rewards. # Players may spend their points on arena rewards, or save their points until they have enough to get the rewards they desire. # There is a limit to the number of points a player can stockpile, which is currently planned to be as high as the most expensive item. # At the end of a season, players on a team are given a title based on their team’s relative position on the ladder which will last through the next arena season. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9559694 Poster: Drysc at 8/17/2006 10:11:37 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: So, Drysc, if you’re still reading this, that means potions are out? Not just health/mana, but things like Moongoose, Troll’s Blood, etc. Correct. Q u o t e: So rogues can’t use poisons, huh? Nothing posted would imply that you would be unable to use temporary item buffs. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9559861 Poster: Drysc at 8/17/2006 10:11:37 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: So, Drysc, if you’re still reading this, that means potions are out? Not just health/mana, but things like Moongoose, Troll’s Blood, etc. Correct. Q u o t e: So rogues can’t use poisons, huh? Nothing posted would imply that you would be unable to use temporary item buffs. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9559861 Poster: Kalgan at 8/17/2006 10:23:07 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: Nothing posted would imply that you would be unable to use temporary item buffs. In addition, you can use spells/abilities that require reagents if you so choose (as long as they still meet the other criteria). So, for example, Blind and Prayer of Fortitude can be used, while Reincarnation cannot (due to its cooldown). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560041 Poster: Kalgan at 8/17/2006 10:36:35 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: I wonder if it would be possible for a team to feed *all* of the awarded points to a single player? A rotation, like warrior + 3 paladins… the warrior stays in, the paladins rotate. At the end of each week, each character receives an equal number of the points. For characters that did not play in at least 30% of the battles, the points are lost. So no, there shouldn’t be any way to inflate the number of points any single player on the team receives. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560259 Poster: Kalgan at 8/17/2006 10:39:51 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: It was not a good system, and it was too easy to exploit. All you had to do was throw a couple of games after a win streak to lower your ELL and then get easier matches. Sure your win/loss ratio would go down, but it was the easiest way to climb the ladder. I remember the 2v2 teams having mediocre win/loss ratios, but would always be at the top of the ladder. edit: spelling I think it’s worth noting that War 3 doesn’t use an ELO system… it has some elements that are based on ELO, but it’s significantly hybridized. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560304 Poster: Kalgan at 8/17/2006 10:23:07 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: Nothing posted would imply that you would be unable to use temporary item buffs. In addition, you can use spells/abilities that require reagents if you so choose (as long as they still meet the other criteria). So, for example, Blind and Prayer of Fortitude can be used, while Reincarnation cannot (due to its cooldown). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560041 Poster: Kalgan at 8/17/2006 10:36:35 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: I wonder if it would be possible for a team to feed *all* of the awarded points to a single player? A rotation, like warrior + 3 paladins… the warrior stays in, the paladins rotate. At the end of each week, each character receives an equal number of the points. For characters that did not play in at least 30% of the battles, the points are lost. So no, there shouldn’t be any way to inflate the number of points any single player on the team receives. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560259 Poster: Kalgan at 8/17/2006 10:48:49 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: Will we see the class balance of the team we’re up against before we go into combat? No, you will not see the composition of the enemy team until the battle has begun. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560450 Poster: Kalgan at 8/17/2006 11:12:41 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: Are bombs and grenades considered consumables, and therefore unusable in the arena? If they are considered consumables, I have no problem with that restriction. I’d just like to know, so I can adjust my tactics. Correct, bombs and grenades are considered consumables. Q u o t e: So, since the rocket helm has a cool down of 15 minutes, it can still be used in the arena, correct? Hmm, the goblin rocket helmet I’m looking at has a 20 min cooldown. Is yours different? http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560869 Poster: Kalgan at 8/17/2006 11:32:12 AM PDT Subject: Re: Min-maxing the Arena system Q u o t e: whoa, the points that you are capped to is = to the most expensive item? I hope that isn’t true because what if you are like 1 point away and then win a match just for 1 point? Make it about 15% over so you can be able to buy the item but still be able to keep the points for the battle you just won to put you over the top. It’s more accurate to say the point limit is approximately equal to the most expensive item (slightly higher). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9561222 Poster: Kalgan at 8/17/2006 11:12:41 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: Are bombs and grenades considered consumables, and therefore unusable in the arena? If they are considered consumables, I have no problem with that restriction. I’d just like to know, so I can adjust my tactics. Correct, bombs and grenades are considered consumables. Q u o t e: So, since the rocket helm has a cool down of 15 minutes, it can still be used in the arena, correct? Hmm, the goblin rocket helmet I’m looking at has a 20 min cooldown. Is yours different? http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560869 Poster: Kalgan at 8/17/2006 11:32:12 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: whoa, the points that you are capped to is = to the most expensive item? I hope that isn’t true because what if you are like 1 point away and then win a match just for 1 point? Make it about 15% over so you can be able to buy the item but still be able to keep the points for the battle you just won to put you over the top. It’s more accurate to say the point limit is approximately equal to the most expensive item (slightly higher). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9561222 Poster: Kalgan at 8/17/2006 10:23:07 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: Nothing posted would imply that you would be unable to use temporary item buffs. In addition, you can use spells/abilities that require reagents if you so choose (as long as they still meet the other criteria). So, for example, Blind and Prayer of Fortitude can be used, while Reincarnation cannot (due to its cooldown). http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560041 Poster: Kalgan at 8/17/2006 10:36:35 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: I wonder if it would be possible for a team to feed *all* of the awarded points to a single player? A rotation, like warrior + 3 paladins… the warrior stays in, the paladins rotate. At the end of each week, each character receives an equal number of the points. For characters that did not play in at least 30% of the battles, the points are lost. So no, there shouldn’t be any way to inflate the number of points any single player on the team receives. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560259 Poster: Kalgan at 8/17/2006 10:39:51 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: It was not a good system, and it was too easy to exploit. All you had to do was throw a couple of games after a win streak to lower your ELL and then get easier matches. Sure your win/loss ratio would go down, but it was the easiest way to climb the ladder. I remember the 2v2 teams having mediocre win/loss ratios, but would always be at the top of the ladder. edit: spelling I think it’s worth noting that War 3 doesn’t use an ELO system… it has some elements that are based on ELO, but it’s significantly hybridized. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560304 Poster: Kalgan at 8/17/2006 10:48:49 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: Will we see the class balance of the team we’re up against before we go into combat? No, you will not see the composition of the enemy team until the battle has begun. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560450 Poster: Kalgan at 8/17/2006 10:48:49 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: Will we see the class balance of the team we’re up against before we go into combat? No, you will not see the composition of the enemy team until the battle has begun. http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560450 Poster: Kalgan at 8/17/2006 11:12:41 AM PDT Subject: Re: Min-maxing the Arena system *edited post* Q u o t e: Are bombs and grenades considered consumables, and therefore unusable in the arena? If they are considered consumables, I have no problem with that restriction. I’d just like to know, so I can adjust my tactics. Correct, bombs and grenades are considered consumables. Q u o t e: So, since the rocket helm has a cool down of 15 minutes, it can still be used in the arena, correct? Hmm, the goblin rocket helmet I’m looking at has a 20 min cooldown. Is yours different? http://forums.worldofwarcraft.com/thread.aspx?fn=wow-general&t=9552122&p=#post9560869 _________________ [...]
[...] See “Use-based systems” for further details. [...]
[...] For those that enjoy computer RPGs, this might be of interest. Raph Koster (lead designer behind Ultima Online, Star Wars Galaxies and author of A Theory of Fun for Game Design) has a great article on his experience with Use-Based Systems. In an RPG environment, the most free form realistic character development will occur when players get compensated for doing actions, rather than killing things (similar to real life). You wanna be a good basket weaver? Then practice it. You wanna be a swordsman? Then hack things up. You wanna stop being a swordsman and learn to cast spells? Then do it. This system should be clear to everyone that has played Elder Scrolls: Oblivion. But unlike real life, it also encourages aberrant behavior (such as people jumping everywhere to get more jumping skills). How do we simulate a realistic environment that maximizes fun, yet encourages open activity? I don’t really know. Maybe Raph does.(Post a new comment) [...]
[...] Raph’s Website Use-based systems The basic definition of a use-based skill system is one where you have a chance of improvement every time you use the skill. So everyone has access to every skill from the beginning (or, in a variant, access may be unlocked by other means, such as being taught), and as you exercise the skill, your chance of success goes up. [...]