[Interview] 'First Descendant' at a Crossroads: What Comes After Season 1?
By Jung Jae-hoon ([email protected])
On July 2nd, 'First Descendant' kicked off its pre-season service, and the report card it received was pretty confusing. The developers couldn't help but feel a blow to their pride from the scores, but ironically, the game's popularity was inversely proportional. It was a weird situation where you didn't know whether to be happy or not, whether to call it a success or not. Eventually, the critics' opinions converged: "Let's wait and see until Season 1."
During the roughly two months leading up to Season 1, Nexon Games' response, while not perfect, was pretty darn quick. Even from an outsider's perspective, you could tell these guys were really hustling. Sometimes there were updates that made you think, "Why'd they do that?" and other times you'd be amazed at how fast they responded. Watching this rollercoaster of improvements, it was clear they weren't going to give up on this game easily.
Two days before the Season 1 update, I got to meet with PD Lee Beom-jun, who's leading the charge for First Descendant's development team, and Director Joo Min-seok. Having played quite a bit of the pre-season myself, the interview started off feeling a bit like a congressional hearing.
Q. Let's start with recent news. The update schedule was originally announced as three parts, but then changed to two, and the timeline was significantly moved up. Considering the development schedule, that couldn't have been easy. Are you guys okay?
- A. We got a lot of feedback saying the third update was too far off, so we knew we had to do something about it. The problem was that early October has quite a few holidays, plus there's something like a company anniversary, so we weren't sure if we could make the schedule work. In the end, we just decided everyone would come in to work. So far, no employees have objected (laughs). They seem to be hitting the 'like' button out of habit, but I'm not sure if they really like it. Taking all that into account, and ensuring the bare minimum survival of our staff, this is the fastest schedule we could come up with.
Q. It's been about two months since you launched the live service in early July. I imagine there have been a lot of internal changes during the pre-season. Have there been any shifts in your development direction or operational policies, like how you handle OP characters or content structure design?
- A. Let's talk about OP builds first. The strength of the OP builds that players are currently grinding out is within our expectations. However, the speed at which they're doing it is much faster than we anticipated.
Before the service launched, we thought we'd only need to worry about how the content would handle these powerful builds once we hit Season 1. But we've had to think about this since pre-season. We've been constantly brainstorming what kind of content in Season 1 can keep up with this.
Also, the direction of builds was somewhat different from what we expected. We thought a wider variety of builds would be used, but ultimately, builds that drastically increase damage output to skip patterns and reduce time became the norm. You could say the average user specs have skyrocketed towards offense rather than defense or utility.
While we didn't predict this, looking at the situation, it was inevitable. Players always seek the most efficient methods, so builds that allow for quick, repeated attempts naturally became efficient. This led to builds focused on extreme damage increase.
As a result, our direction has shifted from various gimmicks and diversity to faster combat, more monsters, and more loot. Overall, we're moving towards eliminating "gimmicks that forcibly take up time." For example, we've completely removed the mechanic of escorting while depleting shields from high-level infiltration operations, and we've changed all the time-consuming capture zones to pure extermination.
Actually, in the early stages of development, we only had 'extermination.' During internal testing, we added various types of gimmicks because it seemed like it would be more fun to have more mechanics. But now it feels like we've come full circle.
This direction of change applies directly to the 'invasion' mode introduced in Season 1. While 'invasion' has puzzle gimmicks, it's designed so that if gamers figure it out quickly, they can trigger an overload and progress without necessarily having to wait.
Q. Don't you think the time limits on each content and the uncertainty of ammo supply have created this meta of maximizing damage?
- A. We don't think the time for Void Boss battles is insufficient. During the growth phase, players often fail to clear due to lack of time, but now we're seeing 2-second and 5-second clear verifications popping up. The time limit for Void Bosses is more like a minimum spec cut-off concept.
The ammo supply is clearly a potential issue. The fusion reactor is particularly problematic. Since there are often no enemies around to drop ammo, players frequently die on purpose to refill their ammo. We're constantly adjusting the supply amount through data monitoring to prevent major problems.
One reason we think the damage maximization meta emerged is due to the systemic incompleteness of the pre-season build. Initially, we envisioned First Descendant gameplay as switching between different Descendants and builds depending on the situation. This was our idea of a "combat meta". You can see traces of this in the Void Intercept Battle.
For example, 'Executioner' is a boss whose defense gradually increases, but this effect can be nullified by Esiemo's skills, unique weapons, or modules. While we secretly hoped for this, users just responded by overwhelming it with raw damage. It's probably because upgrading modules, weapons, or building Esiemo just to take down Executioner is too inefficient. Even if you do build it, changing builds isn't easy.
The reason users found 'Gluttony' difficult when it first appeared is because it's a boss that's hard to deal with using these damage maximization builds - in other words, it's tough to overwhelm with pure damage - but changing builds wasn't easy.
We designed 'Spiral Tidal Wave' to induce this "combat meta", but since brute-forcing patterns with high damage became more popular than utilizing the combat meta we intended, it inevitably felt difficult. The pre-adjustment 'Frost Walker' probably felt similar.
We've learned a lot from this. We realized it's better not to force a meta, but to create multiple clear routes. We've been thinking about how to allow more flexible build changes for a single Descendant without necessarily switching Descendants, and how to diversify content when a meta has already solidified.
Q. Is the socket type change being introduced in Season 1 a result of these considerations?
- A. Yes, that's right. As I mentioned earlier, we hoped players would rotate through various Descendants, but realistically, there were far more attempts to concentrate growth on one Descendant and make them an all-rounder. However, currently, once you expand a socket, you can't change its type, so it's difficult to change the direction of your build even using the preset function.
It's also hard to attempt build changes because even if you use additional catalysts to change socket types, you still have to level up again. So we're changing it so that once you apply a catalyst, you can freely change socket types. This allows for more diverse attempts, and you can save these to presets so you can switch to any build you need at any time.
Q. Isn't the fundamental reason that it takes too long to max out a single Descendant? If Descendant growth was faster, wouldn't we see more diverse use of Descendants as you originally envisioned?
- A. We didn't run any in-game events during the pre-season. Starting from Season 1, we plan to boost growth externally through in-game events, and we're also planning an overall update to experience point efficiency. I think growth speed will probably pick up compared to the pre-season.
Q. I want to ask about the structural holes you noticed during the pre-season. During pre-season, there was no way to use up parts from Descendants that were already fully crafted. Will there be a way to do this?
- A. We're working on a solution for the pile-up of part blueprints. I can't give you exact percentages, but it'll probably involve breaking down blueprints and converting them into other materials.
Q. Another thing I want to ask about is that there was very little motivation to play non-Ultimate Descendants. This was because Ultimate Descendants were perfect upgrades to regular Descendants. How do you plan to motivate players to use regular Descendants?
- A. This is an issue we're considering pretty important and in need of an update. We intended for players to experience and use regular Descendants first before moving on to Ultimates, but since regular Descendants pale in comparison to Ultimates in the endgame, most people just didn't bother with them and waited for Ultimates. We haven't decided on the exact method yet, but we're thinking of a system where growth in regular Descendants can be retroactively applied or transferred to Ultimate Descendants.
Q. I imagine you got a lot of feedback about reactor management. The difficulty of managing reactors was one reason why it was hard to switch between Descendants. Are there any planned improvements?
- A. There aren't any reactor-related changes in this update, but we're having pretty in-depth discussions about changing the reactor system. It's an issue I notice when I play the game too - just finding the reactor I was using becomes a problem every time I switch Descendants.
For now, we plan to let players categorize preferred items into 8 categories, making it easier to quickly find needed reactors. Later on, we're planning to allow players to expand the weapon types that meet conditions through reactor enhancement. So instead of constantly farming reactors for each weapon you use, you'll be able to meet conditions through enhancements as long as the additional options match.
We're also aware that farming reactors is a very grueling process. It's not just the repetitive content, but probably the most tedious part is examining and categorizing the options on each farmed reactor. So, we're planning to introduce an additional random option filter to the junk filter. This means equipment without pre-specified options will automatically become junk with just one button press.
Q. I also want to ask about ultimate unique weapons. Except for a few weapons, their usefulness is very low. Besides new weapons being added, are there plans to improve existing low-utility weapons?
- A. To be honest, I didn't think you'd believe me if I just said it, so I was planning to show results one by one later. But to tell you in advance, we're preparing reworks and balancing for all of them. All the weapons and Descendants currently deemed "useless" will have their day in the sun someday. For Season 1, major changes are planned for Ajax, Blair, and Jayber.
Q. There were also many criticisms about the UX/UI. First, the menu was split in two, there were too many required inputs, and the lack of a ping system was also an issue. Are changes planned for these as well?
- A. We've been constantly updating UX/UI issues through hotfixes. The split menu was primarily designed with gamepad controls in mind. Currently, over half of First Descendant's users are console players, so in designing an interface suitable for gamepads, switching between the full map and inventory requires quite a few inputs.
Because of this, we're developing a feature to switch directly from inventory to map. The ping and chat functions are also quite lacking due to being designed for consoles, but we're working hard to bring them up to a level that's not inconvenient through various methods. Starting with the party finder interface, we're preparing a pretty substantial update.
Q. It was also pointed out that the 'shield' factor in defense has very low utility. The current meta focuses on health and defense, and in extreme cases, even defense is ignored. Are there plans to increase the utility of shields? Isn't the shield important for Descendants like 'Kyle'?
- A. I don't think the current meta will change just by improving modding. What's crucial is that the content guides the player's patterns. There need to be content where shields are necessary for shield-related research to naturally occur, and for shield-based modding to be created. I think the health meta has become dominant because we're not ready for that yet. As more content is added in the future, I think research on shields will naturally follow.
Q. I also want to ask about the balance between 'gunplay' and 'power fantasy'. Although all Descendants have unique abilities, their utility varies greatly. For some Descendants, it's better to just use gunplay without their abilities. What do you think about this?
- A. I think this issue will also be mitigated as we add more content. For example, 'Lepic' and 'Bunny' are currently showing powerful performances. Lepic is strong at dealing massive damage, while Bunny excels at handling wide areas. On the other hand, it's true that Descendants like Freyna, Blair, and Esiemo don't have specialized uses in the current content structure. Once we provide content where these Descendants are highly useful, research will naturally progress. I think a time will come when players will fulfill their power fantasy by utilizing Descendant skills in addition to simple gunplay.
Q. The composition of 'outposts' was also an issue during the pre-season. There were many situations where players had to forcibly play Sharen. Are there plans to improve outposts as well?
- A. We're aware that outpost play tends to be repetitive because some blueprints drop at high rates from certain non-standard enemies you can encounter there. We also know that even the reduced waiting time of 1 minute, down from the original 5 minutes, still feels tedious.
We currently believe that outposts need a rework. We're envisioning a change where combat can continue without downtime, and players receive loot proportional to this combat time.
Q. Let's talk about Season 1 content. Can you explain the 'Inversion' system in detail?
- A. The Inversion system is a kind of season-limited buff system. You level up the Inversion Reinforcement device by earning specific experience points while playing invasion dungeons, and unlock buffs one by one. You can choose up to 8 buffs from different categories.
This Inversion system mostly consists of buffs that make it easier to overcome the gimmicks in invasion dungeons. It doesn't take too long to fully unlock. We designed it so that even users with limited daily playtime can easily unlock it, and it meshes well with existing builds.
Q. When you say season-limited, does that mean these enhancement effects will disappear after this season?
- A. No, that's not the case. In the next season, it will probably return under a different name. The season-limited buff system is designed to increase the convenience of that season's content. Since invasion is the main content for Season 1, it's designed around invasion operations. But in the next season, it will likely take a different form.
Q. How difficult is it to obtain the new Descendant 'Hailey'?
- A. The method of obtaining Hailey is a bit different from before. We've tried to minimize random variables. When you clear an invasion operation, you'll get a trophy grade based on your results (clear time and score). You'll then receive a fixed amount of blueprint fragments according to your grade. For example, if you get a gold trophy, you'll receive 3 blueprint fragments. You collect these to craft her.
However, invasion operations are limited to twice a day, so you can also farm fragments from other infiltration operations. But these drops are probabilistic, not fixed amounts. So, even gamers who can't play for long can definitely craft her, while those who invest more time can craft her faster.
Q. One more question about Hailey - there was controversy because the module update schedule was separated. What's the reason for this?
- A. First, let me be clear - we didn't do this out of any malicious intent to torment users. For Descendants prepared after the initial batch, starting with Hailey, we had planned to update the transcendence modules after a bit of time. We expected to show that the Descendant is highly useful even without transcendence modules, and then let players design their final builds with the modules later. But thinking about it now, that was probably a misjudgment on our part. If people dislike it this much, we can't do it again next time, can we?
Q. Let's talk about 'Invasion Operations'. What's the dungeon structure like? You mentioned there are puzzle gimmicks too?
- A. It's created using the existing 16 dungeons. Every day, invasion events will occur in two dungeons, and you'll play these dungeons in reverse order or with altered configurations. Invasion dungeons are divided into three sections, where you repeatedly complete gimmicks and open doors. As I mentioned earlier, there are no sections that forcibly take up your time. Even with the puzzles, players can reduce the time taken based on their discretion.
Q. The update for 'Ultimate Freyna' was quite unexpected. Ultimate Valby was relatively easy to create, so how difficult is it to craft Ultimate Freyna?
- A. You can think of it as somewhere between the initial Ultimate Descendants and Ultimate Valby. It's not extremely difficult, but it's not as easy to obtain as Valby. We always struggle with this when designing these. If we make it too easy, we run out of content too quickly, but if it's too difficult, that becomes a barrier in itself.
Q. Before 'Ultimate Freyna', it seems like we need changes to the DoT (Damage over Time) dealing mechanism to increase Freyna's utility. Aren't the current DoT-based Descendants pretty much useless?
- A. The problem is that mobs die too quickly. DoT-based Descendants are designed to have higher damage potential than other Descendants, but situations where they can show this potential don't come up. However, if we greatly increase enemy health to make DoT dealing viable, that would cause its own balance issues.
Ultimately, we're planning an overall rework of DoT-based Descendants, and Freyna will be reworked soon. Blair can be somewhat effective just by laying down AoE effects, but Freyna's structure requires mobs to stay alive to be useful, which feels like too strict a condition. Here's an interesting fact: Blair's utility is quite high in invasion dungeons. Even with just 4 rebirths, you can earn gold trophies with Blair.
Q. Are there any new modules being added in Season 1?
- A. We don't have plans for new modules yet. There are so many builds that diverge based on modding choices, so adding even one module requires a lot of consideration. While we don't plan to add new modules in Season 1, we're constantly planning new ones.
Q. Can you tell us a bit about 'Deathstalker', the new intercept battle boss?
- A. It's hard to reveal all the patterns, but think of it as Gluttony-level difficulty. Currently, Gluttony is the hardest boss, and before that, it was Frost Walker. When Gluttony came out, we adjusted Frost Walker's difficulty. Similarly, when Deathstalker comes out, we plan to adjust Gluttony's difficulty as well.
Q. You must have had a lot to think about since launching the service. As you approach Season 1, can you share your current mindset?
- PD Lee Beom-jun: At first, I was lost in thought over the Metacritic score. I wasn't expecting a high score, but I wasn't expecting 56... no, 57 points either. Initially, I hoped for better ratings even if the metrics were a bit lacking, but it turned out completely opposite. The ratings were much lower than expected, but the metrics were better than expected.
From there, all our plans went awry, and we had to change our mindset. We swallowed our pride, accepted reality, and tried to understand the immediate trends. The problem was that users were progressing much faster than we anticipated.
So now, we're focusing on building our foundational strength. Until now, we've been pulling forward and continuously implementing future content we had planned and prepared, but we haven't had time to build up a reserve. That's why the message I recently gave to the development team was that it's most important to build our foundational strength to maintain an update speed that satisfies gamer needs while creating a reserve.
Two years ago, I attended a lecture by 'Bungie' at IGC. It turns out they were grappling with the same issues we are. I think the key is 'speed'. Rapid changes, organic responses to feedback, quick decision-making. That's what we're focusing on now. - Director Joo Min-seok: I think we're lacking in many areas. For me, everything we're experiencing after launch is a first. It's the first time a game has been this successful, the first time we've received this much feedback, and the first time we've had so many issues to ponder.
Since everything is a first, rather than drawing a big picture, I'm working with the mindset of tackling the immediate problems in front of us to the best of our ability. Sometimes I'm taken aback by harsh feedback, but most of the time, these are issues that I and the development team feel the same way about. We're also hardcore gamers of First Descendant, and sometimes we even receive feedback from employees that's not too different in tone from what gamers give us.
I want to create the foundation for a game that can be enjoyed for a longer time, a game that can be serviced for a long period. There's no abandonment.