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
That's literally the only thing you argue
I think this is a bit of an exaggeration - Dygz has plenty of reasonable points, and I think it's really important to remember that arguments are not soldiers. You don't have to support all of the arguments made by folks on the same "side" as you, and you don't have to oppose all of the arguments made by folks on the opposite "side". Rather, its best to evaluate each individually on its own merit.
For instance, I think Dygz's points about how the names of classes are more or less just names is pretty strong - Intrepid has to pick a name for a huge collection of possible playstyles, but what's really important is the primary/secondary combination.
Even so, that point has a pretty big flaw in the abstract, if you drop the requirement that every name match perfectly.
While Cleric/Fighter might not always be what people think of as a Templar, it is also possible to use that as a strong indicator of their design concept for 'What combination you should choose if you intended to play what they consider a Templar'.
I don't think that's too much to expect. If someone were to say 'I want to play a Samurai' (some have) and there was a class in the game called "Samurai" (it's fairly obvious why there is not), then it would be reasonable to expect that the standard 'expectation' (whatever that is) of 'Samurai' would be achievable by that class combination. Not that 'the class Combination can only be used as that', but that it works as a valid shorthand to guide players toward their ideas, especially in cases where the combination doesn't really tell you anything about the design innately.
And since they don't actually 'have' to pick names, we are left to question if they intend it to be 'a gentle guide' or just 'wanted to catch attention and make things sound cool'. Which is, as far as I've always understood it, the literal definition of semantics.
For what it's worth, I think we're totally on the same page. It's also a "problem" (if you can call it that) in other games. If you check the pathfinder 2e subreddit, you'll find a huge number of posts that are basically of the form "I want to play so-and-so from fiction, what's the best way to build that in pf2e?". Then people will be like "Well, you can start from the fighter class and then at level 2 take staff acrobat instead of your class feat" etc.
If they give us the label, then maybe it creates hype but also creates confusion when the expectations don't match up. If they don't give us a label, then it becomes hard for the community to communicate about your class identity.
The system is broad enough that you probably can build your class to be like how you'd expect, but it won't necessarily be viable. The more combinations there are, the more bad combinations and noob traps there are. Tough spot.
As I've noted, I don't believe there is any explicit situation in which bad combinations are a given. 8x8 is something I've been working on since 2013, and I didn't go for 9x9 because after months of working on it, 8x8 turned out to be the limit.
So from my perspective, it's theoretically possible to create 8x8 with probably zero objectively bad combinations and definitely zero noob traps, in Ashes. If the enemy diversity/design is good enough, I'd fully expect up to 150 viable (by geographic location) builds.
I don't find their chosen labels to be poor most of the time. On the contrary, I find them to be refreshingly clear, within the scope of their design plans, and only 'fear' them somehow not living up to the basics that the names imply.
I'm willing to believe that it's theoretically possible for all 64 combinations to be relatively balanced (all of them have a spot in PVP meta, the pve meta, and the pvx mta). I haven't seen it happen in games with just 3 classes. Even starcraft 2 has balance problems with just 3 races.
What I'm saying are "noob traps" are the ways to allocate your skill points within those subclasses. You're allowed to put your skill points into complete donkey builds. You can put a bunch of points into buffing bleeding effects and then not equip anything that causes bleeds, for instance.
To use a WoW example, you could spend your talent points into improved rend to increase the bleed damage of your rend ability by 45% after 3 points. Even after investing 3 talent points, rend still was a button not worth pressing for damage, so it was a total noob trap. Unless you did the math to figure out that it was bad, you might invest.
Oh, those. Yes, those are terrible. And easy to fall into.
I personally try to just design so that those abilities don't exist at all, hence spending so much time just crunching numbers the same way players would. Thematics be damned, if there's no way to get this ability to work most of the time or make it obvious that only X class should use it, then it doesn't generally get added.
Makes the game somewhat flat and simplistic in terms of what exactly builds can do, I think, pushing it back toward character skill and awareness except for specialized ability paths. I personally don't like games where 'the reward for being the savvy one with the time to test or understand' is 'being better at the game outright'. In our 'new world' it doesn't help either, YouTube will just negate it.
I feel like the days of 'feeling good that you figured out something new and cool in an MMO' are gone for all but a select few, and I'd prefer if we moved on to 'hey, do this thing you like as skilfully as possible'. But that isn't what all MMO fans want.
Actually, I'll clarify something since it might make my position more clear. My group got into the Ashes Alpha so early because by definition of what we find fun, being able to play during the development phases will almost 100% be more fun than playing after the game actually launches. The psychology of hope works out really well here, like Pascal's Wager on a gaming scale.
As long as we are playing before the game has any obvious flaws or imbalances (even if it isn't complete), we have fun, and have hope. Every time we find something that isn't fun, we can 'give feedback and hope they change it'. Another positive. Since most of a certain... type of people... won't invest anything into anything before Beta-1, the ratio of those people in the game is reduced, granting yet another positive.
If the development is delayed but Alpha-2 server stays up, then... for us, it's just 'getting to play longer while in this state'. I wouldn't recommend this for anyone whose joy is in 'progression' or 'endgame', but there's something to be said for 'being there on the ground getting the chance to poke at all the little bits while they are still fun', before balance is actually supposed to be in, etc.
If everything turns out great, then one can move on to playing after launch with no regrets. You lost nothing. Probably got your 'money's worth' easily. If things don't go as you prefer, you still got to play (without any true progression) for what could now be upwards of 2 years.
tl;dr, Alpha-2 Access means you get to play Ashes without having to be irritated at the little things, the noob traps, the weird balance, the YouTube-build armies... It's very much its own experience.
Instead, they just have a way they want their class to be played, and if players aren't playing the class that way it's a bug and they patch it out. They're able to accurately simulate how much damage players will be doing with those kits and design encounters around it. It completely removes the "character building" aspect of table-top gaming from the game, but you still have all of the other elements of a MMO, like economy, shared challenges, guilds, etc.
I don't think I would advocate for the full FFXIV approach, but I do find it interesting (and elegant) from a design perspective.
Back in the 'Ranger Fantasy XI' days of FFXI, people were able to actually do a pretty good 'calculation' of 'how much 'fun costs'.
5-7% performance.
When the gap between 'The best class for party' and 'the next best' is around that point, people don't wait around looking for that perfect class (in those days, Ranger). Unless they are going to be playing for upwards of 4 hours straight, they just go 'let's just take the Dark Knight, it's about the same, maybe something interesting will happen if they are good'. And ofc, a good DRK is equal or better than FOTM Rangers.
Past that though, it got interesting.
FFXIV's design always seemed like it couldn't get there, so I could never get into it. Just 'having not only my entire kit, but my entire playstyle' predefined for me, always made me feel like I might as well just watch someone else play it.
So I did. I still watch people play that game occasionally.
But I still feel like they overdid it, they got stuck because they tried to give each class 'too much identity', which doesn't combine well. So I was moreso saying that when designing, I prefer to do 'your abilities are simple, so that they can't fail, how you combine them is your style', and then leave it up to the enemy design to remove the 'meta'. FFXIV doesn't feel like that, to me, yet their abilities still feel simple. The Realm was Reborn, but the scars don't fade.
You think of what you want to play and try to find a way to play that.
The problem is if it's not possible to gain the abilities you want - not that the label you like isn't in the game.
Out of 64 classes, it seems quite natural each person has at least a couple of names they have an issue with.
Viability is determined by the Primary Archetype, so... shouldn't be many noob traps.
Normally I let your weirdness slide, but if this is your take, you definitely don't know what a noob trap is or you are doing that conflation thing where you believe that because it 'is possible to play it and not utterly fail at everything, then it's viable and shouldn't be considered a trap'.
Which is technically the same thing, but at least it's 'disagreeing on what noob traps are' and not 'being unable to see the possibility of them'.
I don't suppose it matters if I point out to you that your way of responding is not helping even Intrepid? When was the last time other than something related to RNG, that you saw your response actually seem to make anyone more comfortable, confident, or relieved?
Do you even care about that?
While this works for FFXIV, I feel it's the opposite of what Ashes is going for. I'm hoping they continue to lean more into the Pathfinder'esque mix and match what you want to come up with what you like approach.
Is x/y going to be as good as y/x, maybe not. But if built right and played well I hope it can perform well enough to complete the content.
And the balance of the game is determined by the Primary Archetypes; not the augments.
(Also, yes, you can respec.)
Facts are facts.
Should be easy enough to review the dev quotes and say, "I think the dev quotes mean something else".
Summoner will be a great litmus test. Let's see if a "more Tank-oriented" Summon rivals a Primary Archetype Tank as main tank.
QUESTION: If i use secondary cleric can i heal others?
STEVEN: indirectly there are some augments when applied to certain skills that can do this. But it would not replace a cleric archetype.
https://www.youtube.com/watch?v=u8ne-4Xl3Rw&t=983s
You will need a tank more or less depending on what's happening in the environment. Certainly in dungeons you will need a tank. I don't picture us creating any encounters that wouldn't need that sort of control. Now we're not just focused on aggro in terms of control and managing mobs... It's not just about mitigating damage, although that's a part of the sort of calculus there, but it's also about making sure that the mobs are doing what you want them to do.
---Jeffrey
https://youtu.be/ndtjwBxhwtw?t=2009
It's a big, huge, gigantic thing to work on. But, the way we're approaching it philosophically is not 1v1 combat. That would be an impossible thing for us to do and it's not really a direction we want to go with the game. We're really focused on mass PvP, not one 1v1 encounters; not really even party v party encounters, but these big, big fights around Nodes...these fights around Castles, these big fights around Caravans...
So, a lot of it is trying to figure out balancing ways for classes to work together synergistically. So that, your focus as a team is building out the comp that synergizes best for what you've got. And we balance it that way.
So, we balance party on party, group on group and different configurations of those things. We look at it more as a macro scale balance problem rather than a micro scale balance problem.
So, there certainly will be classes depending on your build, because we're a build focused game, that just blow other comps out of the water. And that's OK for us. But, what we do want is for parties to be relatively well-balanced as long as they've been built with the encounter in mind.
---Jeffrey
The Active Skills stem from your base Archetype. So, from a design standpoint, even though augments do radically change the way your Active Skills provide you abilities, there's still a primary focus on the base archetype; not the 64 whole classes.
----Steven
That's a really good point. We're not really talking about 64 true classes, we're talking about eight classes with 64 variants... There isn't as much variance between the 64 classes as you might expect. It's not like there are, you know, 64 different versions of... radically different classes.
---Jeffrey
https://www.youtube.com/watch?v=Xg2l6DJgHV0&t=3661s
Certain archetypes are capable of moving the gap between their counterpart per-se. If I am a Tank archetype and a Mage is my counter, I can take a Mage secondary and kind of bridge the divide slightly; and then move my identity that direction ever so slightly.
---Steven
That doesn't directly talk about "the balance of the game is determined by the primary archetypes". For instance, it could be the case by optimizing a fighter/mage, you wind up with a build capable of being around 35% more effective than the best build you can put together with fighter/ranger, which no real appreciable loss of utility. This sort of thing has happened before.
This is, more or less, what the original post of the thread was about. The WoW devs reasoned that the hybrid classes, those that could both heal and do DPS should be worse than those that could only do one or the other. As a result, when players started figuring out the game, it turned out that parties composed of "pure" classes like rogues, warriors, priests, and mages were WAY more efficient than those that had hybrids in them, because people were able to specialize in their role and not have to pay the "hybrid tax".
We're saying that these sorts of builds and the options to choose them are "noob traps", generally, and that it takes really careful design to make sure that players don't accidentally shoot themselves in the foot.
"The Active Skills stem from your base Archetype. So, from a design standpoint, even though augments do radically change the way your Active Skills provide you abilities, there's still a primary focus on the base archetype; not the 64 whole classes."
---Steven
"That's a really good point. We're not really talking about 64 true classes, we're talking about eight classes with 64 variants... There isn't as much variance between the 64 classes as you might expect. It's not like there are, you know, 64 different versions of... radically different classes."
---Jeffrey
The question they are answering in that vid is how they will deal with balancing the 64 classes. So I dunno how you could thing the answers are not about balancing around the Primary Archetypes rather than the augments.
Any Primary Archetype/x is viable in their primary role. So, "more effective" is irrelevant.
There is no competition between a Fighter/Mage and a Fighter/Ranger. It's just different playstyles.
Fighter/Ranger just needs to be viable enough to win the encounter - doesn't matter if the Fighter/Mage is more effective as long as the encounter is defeated.
Also, you can't properly determine 35% more effectiveness because it depends on how the individual characters synergize with the rest of the group. The focus is on the larger group and synergy; not an individual character - like trying to focus on an "objective meta rotation" for a Fighter/x.
I don't agree. The WoW devs didn't reason anything beyond caving to player whining about OP classes so they could make more money.
Just because you say something, that doesn't make it true.
You have 25 levels to learn your Primary Archetype and you use the same Active Skills for the next 25 levels.
The Primary Archetypes/x are designed to be viable in any dungeon or raid.
The devs are not balancing to ensure that characters can commonly find success when trying to make their secondary roles their primary roles. If you try that, you will probably have a very challenging time.
So...the noob move would be trying to swap out your primary role and make your secondary role your primary role.
Notice how he says tank and not tank archtype or tank primary.
Lol
If anything this saying it's not just about aggro and mitigation means I don't need to have all the tank archtype mitigation activated abilities, that I could use something else as long as I can make the mobs dance to my tune.
#bardtank
Also
I'm sorry
I read this and the only take away I get is "your choice is irrelevant." This has been determined for you already by your primary archtype.
If I can complete everything with whatever I choose then the choices don't matter... So shouldn't I be able to at least have fun with it and make weird shit choices.
A Bard/Tank could use something else to make the mobs dance to their tune - that doesn't mean they could out main tank the Tank/x in an 8-person group.
You could try it and find out.
- option 1 best for the individual and the best for the group
- option 2 best for the individual but not the best for the group
- option 3 not the best for the individual but the best for the group
- option 4 not the best for the individual and not the best for the group
More effective isn't irrelevant. It determines how much weight you can pull. If you're advertising in chat that you're looking for a group or a guild, folks will want to know what class you're playing. If you're choosing to play a fighter/ranger when it's 35% less effective than a fighter/mage, then you've identified as an option-2 player, and option-3 players will self-segregate and just ignore your advertisements and not invite you. (Because it's relatively easy to switch from fighter/ranger to fighter/mage)Can retribution paladins and shadow priests defeat the easy raiding content in classic WoW? Of course they can. They're just contributing far less than the fury warrior, rogues, and mages. Option-2 type players don't care about that and are more than willing to bring along players that want to play retribution paladins and shadow priests, and so they form guilds of mostly option-2 type players. Option-3 type players wonder "why wouldn't you switch to fighter/mage? why do you not respect everyone else's time?". This lead to thousands of retribution paladin and shadow priest players having an extremely hard time finding raiding guilds. This stuff actually happens.
Synergy can be (and is) simulated to calculate overall effectiveness for comparison.
I'm referring to a period at the very beginning of the game's life - back when their class design was in line with their own product vision.
Agreed, which is why I generally back up the things I say with a large amount of supporting evidence. I try to start with things that are agreed upon as true and then logically build to a conclusion.
Do you think I've advocated for this somewhere?
I understand that that is the elitist meta-obsessed perspective, but it is not objectively true.
It's just how some gamers like to play.
What's most important is what Primary Archetype you're playing because that's the way encounters are balanced. What class helps determine some tactics, sure. Who will try to synergize with whom.
There may be some obviously wise choices, but it doesn't have to be the most efficient. It just has to be good enough to win.
Your option-2 and option-3 concept, again, is irrelevant.
I dunno why you keep trying to include that concept.
Ashes is not WoW.
The class systems are really nothing alike.
There is no way for you to objectively determine that with all of the augment combinations and also somehow factor in individual player prowess.
LMAO
That is not what you actually do.
Did I say you advocated that?
No corner to reach.
Maybe try 'extrapolating things that must be true from other things said' until contradictions appear' instead for a step or two. The response 'should' be an attempt at obfuscation, with any luck you'll get one of those 'links to a video that doesn't mean what Dygz thinks it means' and therefore a closer range?
Who do you even play in Smash? I'd guess Marth, you strike me as a 'top tiers when serious' sort of person.
And also in context with all of his other quotes. I'm not going to do all of your research for you.
You don't have to have all the Tank Damage Mitigation Active Skills.
You could try having your Bard make mobs dance to your tune. That doesn't mean you would commonly be able to out main tank a Tank/x in an encounter designed for an 8-person group.
You can purposefully make shit choice if you want to. Sure.
You could try to make your secondary role your primary role, for instance.
Why is it irrelevant?
I think the class systems are close enough (you pick a class that allows you to make build choices that define your character's available options, class identity, role, synergy, and effectiveness) that the historic social dynamics are worth referencing.
Why do you feel at all confident in this assertion? We have artificial intelligences that can operate self-driving cars and machine learning algorithms that learned how to play starcraft 2 that only operate from the data on-screen. Are you sure this is beyond human engineering? I think you're out of your depth here.
Player prowess doesn't have to be considered at all, you can just simulate them as playing "with mistakes" or perfect - simulationcraft can be simmed for either.
No, just felt like you thought I did or something. Maybe it was just a weird tangent.
Fox and Falco, though mainly falco. Marth is actually the only top tier I've never dabbled with. And yeah, it's top tiers all the time in every game. Usually strictly whatever character is the absolute best no exceptions no secondaries
That's your biased interpretation of what I actually wrote.
I've already told you why your meta options are irrelevant. I don't know why you keep bringing it into the conversation.
Well, we disagree that they are close at all.
You often think that, but your counter-argument is patently absurd.
That is false. I mean you could do that...and the resulting data would be irrelevant.
You don't believe in self-driving cars?
Or quantum computers?
I forgot to respond to this!
In a lot of ways that's not even the cost of fun, that's the cost of waiting around for a few more seconds compared to the marginal benefit of efficiency. For instance, if a Ranger does 400 DPS (example), and the DK does 380 DPS, the ranger outperforms by 5%. But, in a 6 person group (had to google ffxi group size) that might be 1600 total dps vs 1580 total dps (the rest of the group does 1200 dps). Now you only have a 2.25% benefit. If you have to wait around for 2 more minutes to put the group together, and you're only actually doing dps like 1/2 of the time you're actually farming (waiting for respawns, etc), then now you're farming for ~180 (2 / 0.0225 * 2) minutes before you make up those lost 2 minutes.
I think people have an intuitive sense of "getting started sooner is better than getting started later", and "doing more dps is better than doing less dps", but the breakpoint for how long you should wait for how much extra dps is often surprising until you math it out a few times like above.
I keep looking for where you did this, and I haven't found it yet. Can you please repost it?
What makes it absurd?
What makes it false? What makes the resulting data irrelevant?