Dev Notes, October 3

Posted by Citan on Tuesday, October 4, 2016

Well, this week flew by! I'm still working on the update. Aside from a one-day detour, the whole week has been me changing spreadsheets, playing with the changes for an hour or two, trying again, rinse and repeat. All day every day. It's not boring exactly... but it's also not interesting to talk about. So instead, let me talk about some questions and concerns players have had, both in the forum thread and via other channels. (And I'll talk about that one-day detour, too.)

Changing Gear on the fly: one concern players brought up is that it will be harder to switch to different skill layouts on the fly. Because gear is always focused on two skills, you can't just change skills -- you'll have to change gear, too. Those little buttons next to the ability bars are going to be impractical to use because you'll need different equipment when you switch skills. This is a good point, but I have a solution that will help.

By the way, those little buttons next to the ability bars were an experiment for newbies. Most new players seemed to stick with Sword+Unarmed (the default skills) for a very long time -- many hours -- even after they'd unlocked several more skills. As an experiment, I added those little buttons to see if newbies tried out more skills. And those little buttons definitely helped, so they were left in. (We'll try something a little more nuanced and useful in the next GUI, I think.)

Those little buttons on the side let you switch skills, but don't make it easy to switch gear or abilities.

Those little buttons are convenient but they were never the "main" way I envisioned people switching skills on the fly. Instead, there's a new feature called load-outs. Load-outs save the configuration of all three of your ability bars, and also save what equipment you're wearing.

Load-outs unlock when you get your first skill to level 25.

Load-outs work like this: once you get your first combat skill to level 25, you unlock two load-out slots. Thereafter, each combat skill you get to level 50 unlocks another slot, up to a maximum of 9. You can switch load-outs whenever you're not in combat. You can use the Load buttons in the GUI pictured above, or you can configure hotkeys to switch load-outs with a keypress.

The equipment used by load-outs is marked with a special icon in your inventory so that you can easily tell which ones are in use. And these items don't show up in the vendor GUI -- they are effectively "locked" from being sold.

I've talked about adding load-outs before, but it was postponed until the new GUI arrived. But funny story there: the other day, I decided to add them as text-commands so that advanced players could use something like "/layout 1" to switch layouts. That meant no GUI was involved -- just server-side work. I hoped that it would alleviate some player concerns. I gave myself one work day to finish those chat commands, but they ended up only taking a few hours. So then, since I was already working on the system, I added more features, and then the little GUI window you see above, and now it's a fully functional load-out system. All in one day. Ugly, yes, but fully functional.

I'm still tweaking and improving load-outs a little bit here and there as I play with them, but they are insanely helpful for my testing, and I'm sure you'll find many interesting uses for them. I really should have added them a long time ago. I just figured it would take a lot more work than it did!

Base Damage: we've talked about this in earlier blogs, and I'm still on the fence about the base-damage mods. For the moment, I've just made the drop-rate on base-damage mods very high, so if you're using Sword+Mentalism (for example), pretty much every sword will have +Sword Base Damage, every helmet will have both Sword and Mentalism base damage, and chests will have Mentalism base damage. This solves the problem of "I need these mods and I can't get them easily enough", but the opposite problem is also a thing: "I use psychology for healing and I don't want +Psychology Base Damage if it means I have less healing!" So ... it's a work in progress. I've experimented with a few replacement ideas but nothing has clicked here yet.

Combos: combos have always been a balance problem because they are either too annoying to use or too powerful. In other words, they're too much trouble to keep track of, so unless the effect is REALLY good, players don't use them. I've been iterating on this problem each time I have a chance to revise treasure, and I'll continue to do so. There was a suggestion to broaden more combos to use "wildcard" steps, and I think that's worth trying, so I'll change a bunch of the combos to have more generic steps in between, such as "Fireball + * + * + * + Fire Breath = combo", where * could be any kind of ability. We'll see how it feels, and iterate more from there.

Combat Refresh: several players wanted to avoid having to use basic attacks to trigger combat refreshes, instead adding the combat-refresh power to their favored ability. But that goes against the goal of making you use more abilities in combat! I want you to use a lot of abilities, and my metrics suggest that very few players -- like, almost no players -- use all twelve abilities on a regular basis. So you have slots you aren't using already. I'm adding basic attacks to each skill, so you'll always be able to put the basic attack on either bar -- wherever it fits. Since you weren't using all your ability slots anyway (statistically speaking), this shouldn't cause too much hassle.

And the big upside of pushing players to use a Basic Attack is that it adds some much-needed consistency between skills. And that lets me add more interesting treasure effects. I'll show you what I mean in the next blog post, if some of these ideas pan out, but the point is that when everybody has something consistent, like a basic attack, I can add interesting things that tie into it and let your two skills interact.

Multi-Skill Gear Effects: one suggestion that came up a few times was to do multi-skill treasure effects. And I definitely want to do a LOT more of these! (See the discussion of basic attacks above.) The most obvious way to implement it, though, has already been tried and doesn't work. For instance, a treasure effect that says "<ability from skill X> and <ability from skill Y> deal +50% damage" is only satisfying for people who happen to use both of those abilities. Players know that the ability's potency is a little watered-down because it can apply to both abilities, and they hate the feeling of "wasting" their potential. Even if the damage isn't watered down at all (which is only possible in certain circumstances), players don't seem to like them anyway, presumably because it still seems like it would probably be weaker.

And anyway, treasure effects that apply to only one other ability aren't broad enough, given how few treasure effects there are. I want treasure effects that can be used by a large number of builds. That's the reason I recently added the ability descriptors like "Epic Attack" and "Signature Debuff": to add treasure effects that apply to all the abilities in that category. And I'll also try some other ideas, like treasure that says "<ability from skill X> grants <bonus Base Damage to skill Y> for 5 seconds". It'll only be useful to people who happen to use both of those skills, but at least it doesn't lock you into two specific abilities -- only one, like normal.

In short, yes, making treasure synergize between multiple skills is definitely a goal. But it's a long-term goal. It will take more iterations to find the most successful ways to do it. In the next update I'll just change a few treasure effects to work this way so you can give feedback on how it feels and we can evolve it further.

Transmutation permanently locking in two skills: In an earlier blog post, I explained that the transmutation table will start to "enforce" the two-skill breakdown on gear, and it was asked if that was permanent or just transitional. In case it wasn't clear, yes: it's permanent. Otherwise, players would just use Transmutation to make single-skill gear again, undermining the whole point of this change. However, there will be a separate table that can help you cheaply adjust your existing Legacy gear, and THAT table is probably temporary.

On lowering the number of total treasure effects: I think the number is good where it is. For most skills there's really only about ten treasure effects per gear slot. That means that two-skill gear is only pulling from a pool of twenty or so effects. And if you're transmuting, you're pulling from a pool of only ten effects! (On average.) So the odds are already pretty great for transmuting a specific effect you want. Now, lots of the current treasure effects would never be desired by power-players, and I want to improve those. But throwing them out isn't improving them.

Iterating vs. Starting Over

And that brings me to a really important point about game design. It's always tempting to throw out stuff that doesn't work. It's boring to tweak and revise stuff, and it's so much more FUN to try all new designs! But no design works well the first time. Unless the design is being stolen straight from some other game, it's going to need iteration. *Lots* of iteration. Iteration is boring and time-consuming and it can be tedious, both for the designer and the play-testers. "Ugh, this system still doesn't work, why not try <brand new idea>?" And I do periodically throw away designs and start over... but I have to give ideas a fair shake first.

So that's my answer to a lot of questions: "Why not throw out Combat Refresh and do <X>", "Why not abandon the two-skill design and do <Y>", "How about if <Z> became the focus of Necromancy instead of the pets, since the pets suck?" Those ideas are sexy and exciting but they will need just as much iterative revising as what's in the game now. I'm iterating on these existing systems because I think I can make them work well with more iteration. If something just REALLY doesn't work, then of course I'll abandon it. But right now, this is the part of the game's development where I have to just grind out iterative improvements.

That's not to say I don't like reading new design suggestions! I can often take parts of those and apply them to existing designs, or use them when I make the next new skill or ability. (There are several more entire combat skills to add, along with a bunch of variant skills... but I'd like to get the existing ones working better before adding even more data to my gigantic spreadsheets.)

Playstyle choices: The idea behind this complaint seems to be that players liked focusing on just one skill, and by removing that option, I'm removing choices. But that's not true. The number of choices is exactly the same, actually. Right now, you could say "I want to focus all my gear on Archery", OR you can say "I'll focus on both Sword and Archery". That's two choices... but one is dramatically weaker than the other. After this update, both options are still open to you. If you want to focus only on Archery, go right ahead. Use only Archery and Generic treasure effects. You will probably be weaker than somebody who focuses on two skills and does a good job synergizing their gear... but so what? One of the two possibilities was already weaker. I just changed which is which. And I did that because I know that two-skill builds will hold players' attention a lot longer than single-skill builds.

Most players will gravitate to gameplay styles that are easiest. (That's just human nature: smart people find the path of least resistance to getting what they want, whether it be in a game or in real life.) In fact, in an MMO, this is often a very big problem because even players that want to try weird ideas often feel "obligated" to follow the easiest path due to peer pressure. If your friends are all using a build that's vastly better than your build, you'll feel pressure to change toward their builds, even if you were having fun with your slower build. And there's no way to avoid that pressure -- I can't make all play styles equally easy and equally powerful. Instead, my job as a game designer is to make sure that the "easy" path is the one I think is most fun, in the long term. Speaking of which...

On making the game "less fun": This concern is sort of similar to the last one, but it's a little meatier. The idea here is that it's fun to be able to kill monsters with just a few button presses, so my push for longer combats with more combat-actions makes the game less fun. And that is ... true. For a little while. But not for thousands of hours of gameplay.

This is a problem that ARPGs run into all the time. It's their Achilles' heel, actually... they start out incredibly fun, but they get boring too quickly. Hmm, how can I talk about this without badmouthing a fellow MMO team? : Ah, I know. Did you ever play Kingdoms of Amalur? It's a really fun single-player game, but if they'd managed to turn that combat into an MMO, it would have been a flop. I was bored after "just" 70 or 80 hours. I'd seen all the iterations they could come up with, and there wasn't a lot more they could do with the combat except make the numbers get higher. There was nothing left to think about after I'd mastered the action parts, so I got bored.

I didn't feel ripped off... eighty hours of gameplay is a lot of fun for a single-player game. But MMO play-time is often measured in four-digit numbers... or five-digit numbers for long-running games! The core gameplay has to stand the test of time. It's my job as a designer to get this right, or my game simply will not survive for years to come.

So yes, for a short game, short combats are more fun. But that gets less true the longer the game is supposed to last.

This isn't the first time I've had to reduce short-term fun for long-term viability. For instance, there used to be no global cooldown timer and animations could be interrupted, so you could spam all your abilities instantly, interrupting the combat animations to just blast out all your attacks within a few seconds. That was actually really fun at first, but it couldn't stand the test of time. Since monsters only lived for a few seconds, they had to hit VERY hard in order to pose any danger. That meant combat was very lethal. A surprise addition was very hard to deal with. Combat was either very easy or very frustrating, and I couldn't get a middle ground. After iterating a while, I declared it a dead-end and added a traditional global cooldown timer, like most MMOs have. (It's 1.3 seconds if you're wondering. That's a little faster than the average MMO, but in the same ballpark.)

Another example of "removing fun" is that I'm eventually going to remove the ability to run really fast during combat. It's a lot of fun to outmaneuver monsters, racing past them, shooting them in the ass as they desperately try to regroup, then doing the same thing again and again. But it's not sustainable for thousands of hours of gameplay. I can add trick monsters, like ones that stun at long range, but a trick monster is a special case. Special cases can't be the core of gameplay.

Play-tester burnout: In the past, a few play-testers have quit each time I make big changes like this, because they're nerfs. That's always a danger when you nerf something: players feel personally slighted, like you just shat on their ice cream cone. They are now having less fun, and they don't have to take that kinda shit from a game, so they go play something else. Nerfing something after a game launches is a very hard call. Sometimes you really need to do it, but you agonize over it, because who wants to lose players?

But this is alpha. You are not players. (No offense!) You are play-testers. I can't afford to lock down my alpha choices in order to keep the tiny audience of alpha gamers happy. I'm iterating on all kinds of design elements, and sometimes a fun thing isn't sustainable. I have to do what's right for the final game. Alpha-testing is a lot of fun if you approach it from the right viewpoint. You guys get to experience a lot of things that won't be feasible in the final game, when there's 1000 people logged in. If you've ever gotten your Sprint Speed up to 20, that's a run-speed of 45 mph! It's hilarious and fun... when there's a few dozen people online. Not as viable when there's a thousand people blurring around everywhere, causing the monster path-finding service to explode trying to keep up. The fact is that some kinds of fun can't last... but that doesn’t mean you shouldn't enjoy them while they're here. Just understand that change has to happen eventually. Stay fluid... that's the key to avoiding alpha burnout.

Anyway, based on some of the feedback I've received this week, I can't say I'll be surprised if some players quit over this change. I hope you find fun elsewhere, and can come back and try out the final game. But I do hope that most of you will stick with it to get past any disappointing first impression of "nerf", and can help me make this design work!

Lots More to Do

I'm still working very hard on the next update. By my best guess, I have something like 50 more hours of data-crunching to do to convert all the skills and treasure (and make sure it's vaguely playable), so I should get that wrapped up this week... but then I need a week to polish everything. So we're at least two weeks out. The scope has broadened in various places, but it remains focused on the same core mechanics issues -- there are just a lot of facets involved. This is a big update, and it improves how the game feels in a lot of ways.

Please have patience, and please keep communicating!