Glorious Alpha Two Testers!
Phase I of Alpha Two testing will occur on weekends. Each weekend is scheduled to start on Fridays at 10 AM PT and end on Sundays at 10 PM PT. Find out more here.
Check out Alpha Two Announcements here to see the latest Alpha Two news and update notes.
Our quickest Alpha Two updates are in Discord. Testers with Alpha Two access can chat in Alpha Two channels by connecting your Discord and Intrepid accounts here.
Phase I of Alpha Two testing will occur on weekends. Each weekend is scheduled to start on Fridays at 10 AM PT and end on Sundays at 10 PM PT. Find out more here.
Check out Alpha Two Announcements here to see the latest Alpha Two news and update notes.
Our quickest Alpha Two updates are in Discord. Testers with Alpha Two access can chat in Alpha Two channels by connecting your Discord and Intrepid accounts here.
Comments
Yes, Meridian 59 never reached the success of EQ. Yes, it is pretty obscure and most gamers have never heard of it. It’s most notable for being the first 3D MMORPG but that makes it a piece of trivia more than anything else. EQ was king of the MMOs in its day, and it was synonymous with MMORPGs in the same way that WoW became synonymous with them (and amazingly still is; I’d wager that to this day more people in the western world know that game than any other MMORPG ever released).
I’m not objecting to any of that, but it was first, and it’s still around. Not making much money at any point but considering that it still lives, that is pretty remarkable. Sure it’s only around because it is cheap to maintain and it is free to play, but the same goes for EQ.
This has not happened with either EQ game.
The only reason Meridian still exists today is because the server code was essentially leaked, and the ex-developers took over.
It has never turned a profit, and likely never will.
I can claim this as a commercial failure quite easily.
I'd be interested in hearing how you could claim that a game with this history is not a commercial failure.
Edit; and I still debate that any RPG game with a server cap of less than 200 could be considered massive in any respect.
Would love to see a living breathing ecology of run by one.
Unless the tanks gained abilities to receive the damage dealt to teammates and had aoe buff to increase damage mitigating among other stuff but that wouldn't be a pure tank it would prob be a subclass of tank
An invisible boss is assigned to be the overseer of the encounter area; job description is to scale fights to the level of the participant(s) - Instead of scaling the players to the area the encounters are scaled to the players. The boss would learn and adapt so that the same player visiting the same area more than once would have varied encounters instead of seeing the same, boring loot pinatas commonly seen in MMOs.
The ecology could be run by a different "boss" that the encounter boss would feed off of using a resource-management type mechanic.
Is this a bit what you were driving at?
I.E. say you strike up a conversation with a blacksmith...you get your armor fixed and buy some items from him....but you keep coming back to him....over time he tells you things, gives you DEEP discounts, you form a bond and rapport with him...this is not honor or whatever you want to call it...reputation within the city, this is you and the blacksmith only. Also, he can give you tips on new weapons he has or has access to, I.e. the auction house.
Then that same blacksmith NPC starts noticing how much ore or ingots cost and begins adjusting his own basic goods price based on the price of ingots required to make his basic goods. Then he starts expecting a tip when he does other side options like identifying, upgrading, or repairing. To top it off he then tries to rip unsuspecting players off when they come to sell trash loot but that unlocks a bartering system!
Now one thing leads to the next and more of his shops are popping up around the city, he begins to take over other similar shops and next thing you know that damn blacksmith is running for mayor of the node only to win by offering larger PC guilds a discount to vote for him. The bastard wins then evolves into a villain and now the city is ran by hostile NPCs funded by the trash loot you were selling him. Then when you try to defeat him he pulls a Nagash and goes Lich form, raises all the dead NPCs, kills the players and now we have an undead army ran by an always learning AI blacksmith/mayor/Lich.
Do you join the undead army or wait for the server reset once he kills everyone and takes all the nodes?
In regards to an MMORPG, I think having the top raid bosses like this would be dope. Although just for dungeon type stuff. It'd have to be very minor for large scale pve content or else it'd just be unplayable organizing a zerg to do anything.
The first part is easy. You make a boss/monster with a bunch of attacks, abilities, and movement options just like any other enemy. Then you can use the output of a ML system to pick which actions to use, some or all of the time. The AI can have other components or restrictions on the actions (e.g. cooldowns) though.
The second part is basically all of the inner details of the ML configuration. There's a bajillion different vectors of data and various metrics that you could use as input to the system, but it's up to the devs to tinker away and find what works best for each particular AI. However, they are limited in how much they can use as input. Too much input makes the system overly-complicated, uncontrollable, and probably too expensive in terms of computation.
The third part is the most important and the only reason I'm making this comment: The ML system SHOULD NOT be designed to kill players and win fights. That is a terrible goal to set. Rather, it should be set to optimize for certain gameplay states that are correlated with player enjoyment.
For example, the ML system might aim for 80% success rate, because that's indicative of a decent challenge level. Or maybe it tries to to nearly-kill players (and then give them a moment to breathe), because that provides intense moments in gameplay. The AI might want to force players to use a greater variety of abilities or movement. Maybe it includes an optimal ratio of healing-to-damage that it tries to force out of the healers (because healing all the time, or not having to heal at all, are kinda lame experiences). You could even have the players give ratings to each boss, as an additional point of feedback to the ML.
The point is to let the AI design it's own behavior using Machine Learning, so that it can evolve over time, and so that the devs don't have to spend as much time making and tuning each fight. So the goal of that ML should be the same as the devs original goal: to make an enjoyable fight, not to kill everyone.
Any game with rotations doesn't have a top end raid game, so if learning rotations is all you think raiding is about, you have clearly never raided in a game with a real top end.
Several good games have set rotations (or close to it) and still have a lot of challenging end-game content, including raids. They just don't focus on making you adjust to your own random class mechanic. They instead test your ability to plan, execute, and adjust the plan when things go wrong. And executing a static rotation can be it's own challenge and fun for lots of players (assuming it's fast-paced and complicated enough), since you know, we're humans and not robots. We take time to learn rotations, and still make mistakes once we've learned them. I've watched plenty of top-tier raider curse themselves for messing up basic rotations, and then have to adapt to get things re-aligned somewhat. All parts of that experience (learning, practicing, mastery, adjusting for mistakes) are fun for me and a lot of other raiders.
Plus, once a player learns their rotations, they can focus more on specific fight mechanics (and thus those fight mechanics can be more complicated and fun). Also, fight mechanics tend to interfere with rotations, so that players have to develop a variation on their rotation for every fight.
So please, just chill and let people enjoy their static game designs if they want. And I'll let you enjoy your priority-based combat (I have enjoyed them too in the past), as long as you're not so elitist about it.
Rotations are the dumbing down of any game that is developed with them in mind.
As far as I am concerned, designing a class around a rotation is the in-combat version of designing a game around LFG - both allow players to be lazy and get away with actual incompetence, just in different ways.
Edit; I should point out that my hope in Ashes is that there are some classes that work on rotation, some that work on priority. Yes, some people may want to be lazy, and I'm ok with that. I would like there to be the opportunity for those playing classes without basic rotations to excel a bit, but that is a discussion for another day.
What I am not ok with is people playing rotation based games thinking that is all there is out there, and then making the assumption that all PvE must be the same. If they are playing the lesser version of a genre, they should be informed as such.
---
That said, I completely agree with your post about AI, it is a better written description than I could hope to do myself of what I consider to be the main drawback of attempting to use AI in encounters - that the AI will be programmed to win, and before long that is all it will do.
To add to that, I don't personally think AI in general is at the point yet where setting an AI up to specifically not win in a combat situation, but still make that encounter challenging and fun would be easier than setting up encounters with specific scripts to perform the same function.
Maybe in a few years that could be a reality, and if the AI is designed as you describe I'd be all for it, I just don't think we are quite there yet.
Most classes nowadays are actually not designed around rotations. Rotations were something developed by the game community, not the developers.
WoW Classic didnt have rotations originally for example. People will always try to optimize the hell out of everything, because it is our instinct to save time and energy (the opposite can happen if the person doesnt like what it is doing). Thats why we learned to use tools after all!
So, what you are saying is that rotations are a product of bad game design.
Assuming the rotations people use are near optimal, either they intended them to be how the class was played, they had no real design intention of how the class was to be played when developing it, or the way the class was designed to be played was not the most effective way to play it.
You've excluded the first scenario, which in terms of WoW I won't argue (though I know some games have designed around rotations). The second example is that of poor game design, and the third is of poor game development.
I can't think of a fourth scenario that could lead to rotations being near optimal, though if you can, I'll be here.
No i am not saying that rotations are a product of bad game design. I am saying that rotations are a product of the community in their approach of optimization for a PvE scenario.
There are no rotations in PvP for example, because there are too many variables to consider. The best spell/ability in a certain situation changes depending on the enemy and their build, their equipment, and their approach for example.
PvE though is not about a huge number of variables, but rather the perfect execution of a task under time pressure in a given environment. I myself dont like the approach of EQ2 in that regard tbh. They overcomplicated it immensely, but that is my opinion and you may like it that way.
Ergo, the game must be designed around rotations being optimal. If it wasn't, then they wouldn't be.
Unless you are telling me that the community is designing and balancing the class abilities, which I don't think you are, if spell rotation is the optimal way to play a class in any given game, then that is how the game was developed.
Now, you can argue that it isn't how the game is designed, but then the question becomes - if it isn't how it was designed, what happened to that design? Why couldn't the developer make that design work? It just goes back to poor design.
So, no matter which way you look at it, games where players use rotations are that way because the developers either designed it that way on purpose, or are bad developers. It isn't, and simply could not be, a result of players.
You just left out the whole reasoning behind that argument...
The problem is that you want to push you outdated agenda on other people and csnt seem to be able to hold a valid argument. You are repeating yourself again and again without actually giving solid backing except: "EQ2 was better, it was sooo complex and didnt depend on a rotation, but rather a spell priority."
EQ2 is a nearly dead game with around 180.000 players and a population of around 7.000 players daily.
But... the only way that can be optimal is if it was designed to be optimal.
Ergo, the game must be designed around rotations being optimal. If it wasn't, then they wouldn't be.
That is a ton of assumptions right there. Only because players create rotations doesnt mean that the game was designed for rotations. Have you EVER SEEN A GAME DEVELOPER TELL PLAYERS HOW TO USE THEIR CLASSES IN HIGH END PVE SCENARIOS? No you havent, because game devs dont think of that.
Unless you are telling me that the community is designing and balancing the class abilities, which I don't think you are, if spell rotation is the optimal way to play a class in any given game, then that is how the game was developed.
No, i am not saying that the community is designing spells, obviously, but you are justs aying that because players create rotations the game designers created all spells to be used in a rotation. You can not use a rotation and still fight enemies cant you? You can use your spells however you want. Do you achieve maximum efficiency? No. But can you fight enemies without a rotation? Yes.
Game devs can only create what they envision their game to end up as. The rest depends on the community. Tht doesnt mean that they failed, it means that the community took that idea and took it in another direction.
You seem to be stuck in a spiral of negatively reinforced ideals my friend, saying that something appears to be the fault of the devs doesnt mean that it actually is.
And look what i found: a rotation for the wizard in EQ2!
"- Pre-cast frost shield on mt
- Cast magic debuff hex doll
- Frigid gift
- Rending icicles
- Surging tempest
- Freehand sorcery
- Ice nova
- Forge of ro
- Surge of flames
- Protoferno
- Ball of lava
- Incapaciate
- Irradiate
- Firey convultions
- firestorm
- Fusion (if its up)
Then its just a matter of spaming ball of lava, irradiate, rending icicles, firestorm, icesheild, and firey convultions until the mob is dead.
Dps will vary from 1000-2400 depending on crits / spell quality / and the good ole random # generator"
So, the game designers of EQ2 ALSO DESIGNED THEIR GAME FOR ROTATIONS? Thats your argument isnt it?
That isn't a spell rotation that works. Sure, it's basically a list of spells that a wizard would cast in EQ2, but it is not only not optimal, it will see you kicked out of most raids. It may get you by in a pick up raid, but since the discussion here is about optimal output from the class not passable in a scenario where no one knows what to expect, I don't think this works.
The hex doll there made me laugh a bit, but only someone that was raiding EQ2 at the very specific point in time when they were a thing would understand why.
It also isn't taking in to account what buffs you have on you, as that would greatly alter every decision you make. Even though Freehand Sorcery is specific to one of five AA lines that was available at the time this was written (some time late KoS), it doesn't mention anything else about any other AA's, nor does it say what you should do differently if you went for the line that reduced cast time over the line that increased crit damage - as you could only take one other line on top of having taken the line that gives Freehand Sorcery, and of the five available, crit and casting speed were the only two that were worth taking fora a raiding wizard.
Even the very first spell in that list isn't a guarantee - if there is another wizard in the raid you would first want to see who has the more powerful version of that spell, as you could overwrite a more powerful version with a less powerful version (a sure sign you don't know how to raid in EQ2).
Developers absolutely DO think of that.
That is one of the key things they think of when designing a class. They think of the pace the player will need, the positioning they will want, the skills they will be using and how all of that interacts with everything else they may have at their disposal.
Not only do they think about how the class will play, they then monitor the game to see if players are playing the class as they expect it to be played.
If developers add the best ability to a class that needs three of a particular combat resource to use, and also put in a spamable ability that generates one of those resources, the developers expect players playing that class to spam that ability three times, then use the big ability. If they don't give you anything better to then use, the developer expectation is that the next thing you will do, is spam that ability three times to get three more resources again, and then use the big ability a second time.
If that is the most effective way to play a class in any game, it is either because the developers developed the class so that it is the most effective way to play it, or it is because the developers didn't design their class at all. Not didn't design it well - didn't design it at all. Throwing a random bunch of abilities together isn't design, thinking about how those abilities work together is.
Now, if players want to play a class poorly by using a rotation that is not optimal, that is on them. But that isn't what we are talking about here - we are talking about when a rotation is the most effective way to play a class. A situation where a rotation is the optimal (or near optimal) way to play a class as outlined above can ONLY exist either by developer intent or incompetence. Which, to be fair, every single thing in an MMO can only exist because developers want it there, because developers don't know it's there, or because developers can't figure out how to take it out of the game.
That is, for the record, intent followed by two cases of incompetence.
There is no other reason something can be in a game - yet your argument seems to indicate that you think things can end up in an MMO due to the community. This is why I didn't see any reasoning in your post - you simply didn't provide any.
I would say in the next 2 to 3 years we will see it in most new games
That timeline seems a little short to me. I'd give it 5 years before it really catches on (probably when Unreal Engine implements it into a core feature that everyone can use). Even then, we won't really "see" it. It will just be used to optimize some small aspect of the game; something that designers of eld had to tune by hand. We're never going to see ML designing entire levels or bosses, at least not until the Singularity
And yeah, we won't see it, but I have no doubt we will hear all about it - developers of the first few games that implement it will no doubt want everyone to know.
Anyway, going back to the issue of rotations vs priorities and how it relates to machine learning in boss fights, the issue is one of balance. In order to have a true priority dps system (tanking and healing is much easier to produce a priority system for) you need to have a bunch of constantly changing variables. The problem is that if you try to add in too many variables you ruin the balance of the game, especially if a lot of those variables are RNG-based. On top of that adding too many variables becomes pointless because players won't be able to calculate and account for them in real time.
How does this relate to machine learning? Well adding in sophisticated AI like that is yet another set of variables that need to be balanced around. Unbalanced combat isn't fun for anyone and over-tuned bosses are a raider's worst nightmare.
I fear that even if the developers could design every single dps class around a priority system (which I doubt they can whilst keeping every class unique), adding machine learning on top of that would make balancing the fights impossible.
In other words, you might have to choose between better AI on the bosses OR more engaging class design.
Thing is, with how much people defend this type of play, I think they may be right.
What you say about players figuring things out though is absolutely true, and I can think of at least one example where an entire class thought they were sub-par by design, but it turned out to just be that the entire class didn't know how to play properly.
Thing is, developers still design the class. They look at all the abilities it has, and how they interact with each other. While they wouldn't tell players how to play a given class, they still design it with they expect it to be played - and that includes designing the expected output of the class.
So, if a rotation is the optimal way to play the class, then either players have figured out the optimal way to play the class (which, as it is a rotation, isn't hard), or they have come up with a way to play the class that goes above what the developers intended.
When a class performs better than developer intention, it gets nerfed - this is a thing I think we all know. If developers are nerfing a class, they will attempt to nerf it in a way where the original way to play the class is maintained.
Thus, if after many years, rotations are still the optimal way to play a game, then that is what the developers designed.
You are right that you need at least a few constantly changing variables, however, the only way it is unbalanced is if those variables are RNG based.
If the variables come in the form of teamwork (with Ashes having dedicated support classes, this fits in well), then those variables aren't unbalanced - or at least need not be.
I wouldn't want all 8 classes designed around a priority system. I'm all for as many players finding just the right class in Ashes as they can.
This means some classes should prefer action, others tab; some prefer fast paced combat, others prefer purposeful combat; some prefer positioning, others prefer standing back; some prefer rotation, others prefer priority. There is no reason at all these can't all be catered for.
The only issue here is balance between the different styles. By it's very nature, there is little difference between a good player and a poor player in terms of output in a rotation system, yet the difference is vast in a priority system (it is perfectly possible to find yourself with nothing up to cast for several seconds if you mess up). To me (though I'm sure others would disagree), the rotation system range should have a designed output right in the middle of the priority output range. This would mean that a poor player on a priority system will be worse off than a poor player on a rotation system, a good player on a rotation will be worse off than a good player on a priority, and an average person would be about the same on both.
as to your last point, while I disagree that it is a true statement, if it were, I would pick the more engaging combat system.
Players spend an average of less than 5% of their in combat time fighting top end boss fights - even top end raiders (most raid bosses are not top end fights). So really, this choice is "do you want to make 5% of combat more interesting, or 95% of combat more interesting?"
To me, that isn't even a question.