This is from the Weekly Famitsu article that was published in the issue released on August 25. I know there are other translations elsewhere (Reddit, at least), just figured I'd give it a look for old time's sake. There's another Famitsu interview that was published in the last day or two as well from Gamescom so I'll take a look at that one as well. The article in the weekly magazine was part of a 3rd anniversary special with some other information from FFXIV developers, so I'll take a look through the rest as well to see if there's anything worth bringing up. Frankly this interview was pretty boring and I didn't really see much information that hasn't already been discussed in other interviews, but I guess they are really holding back for the Fan Festivals at this point.
The "chat with Yoshi-P" has become a staple event outside of the game. In these, players and others such as Level Five president Akihiro Hino or voice actor Yuichi Nakamura get to chat with Yoshi-P. Indeed, it's something we'd all like to have the chance to do. We started this project with a thought like that, so this time Famitsu will chat.
Before we get into the chat, first please tell us your feelings about celebrating the third anniversary.
Yoshida: Sorry, I don't really have any feeling at first (laughs). When we released the expansion pack last year, I had the feeling that two years had finally passed, but with regards to the third year being over, it's more like "I wonder if we'll get to another three years."
There might also be many players who are wondering if the game will last another three years.
Yoshida: With the volume of development for the major updates coming every three and a half months, it's also surprising how fast the time flies and that we're already thinking about release of the next expansion pack.
Looking back on the continuing updates every three months for the past three years, what did you find to be the successes and challenges of keeping that pace?
Yoshida: Speaking about successes, the pattern of organizing the specifications for design and content to neatly manage the production tasks, being sure to design the item hierarchy into the content, and performing fine tuning adjustments based on player feedback definitely benefited the team. As a result of that, we were able to deliver a steady pace of content to players. Rather than just the leaders working hard, the entire development team is now able to work at their own pace and come up with proposals. At the same time, we understand that the game also needs to change. There's a lot of focus on the content that players partake in once they reach level 60. However, if we focus only that, we'll be leading ourselves to a plateau that eventually takes us to the termination of the game. All MMORPGs will come to an end eventually, but I think it's one of the jobs of the producer and director to hold off the timing of these transitions for as long as possible. I'm not looking at a particular time window like two or three years out, rather I'm focused on providing experiences unique to FFXIV and to continue with development that allows more new players to start at any time.
With the 3.X series coming to a climax, will we get a hint about the new jobs, content, etc., that will be central to 4.0?
Yoshida: Well... I can't (laughs). In addition to the ones in the game, our activities outside the game are also important for the fun of the game, so to keep the expectations high for the Fan Festivals later this year and for patch 3.4 where foreshadowing for 4.0 will begin, we still aren't putting out any information. I'm sorry.
Well, now for the chat...
TOPIC #1 - Although "Hall of the Novice" has been implemented with training for each role, we still occasionally see people who aren't doing a rotation of combat skills. Are you preparing a system where these players can work on their performance?
Yoshida: Fundamentally it's a game, so there's no one correct answer and every player should be able to play as they like and shouldn't be blamed if they aren't doing a basic skill rotation. There's no sense in getting mad every time you see this; rather, I like to think of it as the additional attack power of a stronger player can help clear a little faster. If the person was on the same job, perhaps give a little advice. It might be a bit difficult because of communication issues in Japan, but since it's a party play game you should feel more at ease. That said, we do recognize the problem that it's getting more difficult to achieve the highest performance from each job which has caused the gap in player skill to become more apparent. Because of that, we're planning on significantly changing the difficulty of operating each job in the next expansion. In the 3.X series, the rotation for each job includes many quick decisions that must be made which created room for improvement, but looking back on that, it ended up creating too much of a disparity.
After the implementation of the actions through level 60, some jobs felt entirely different. Will there be significant changes like that in 4.0 as well?
Yoshida: We'll be going back to basics. In FFXIV battles right now, managing buffs is necessary and it's too much for some jobs, I think it's become too difficult. The battle team was working along the lines of gamers enjoying the opportunity to devise ways to improve in battle, and I think we achieved that. However, setting the bar higher also created a lot of angst. In 4.0, we're talking about reducing the amount of things that need to be managed and emphasizing the characteristics of each job. The top layer of players may feel that it's a bit too easy, but I think I want to reduce the gap between the top players and casual players a bit more. Since this is just a general image of our plan, please just think of it like "well, it will be easier than it is now." (laughs)
So you're saying that you might review the actions and make changes to the ones that already exist at level 60?
Yoshida: There may be some job changes as well. We'll be revealing more information as we discuss the expansion pack in the future.
TOPIC #2: For ranged DPS, one of the benefits is that you can attack without approaching the enemy, but if you're targeted by an attack you may be forced to move. There's a risk when you move, but we feel that this is high for casters (especially black mage) - how do you feel about this?
Yoshida: The base concept behind black mage is that it has a high attack power but has a high risk associated with not casting, so that is the intent. However, since there are multiple people in charge of content design, there's occasionally content where there isn't an opportunity to update Enochian.... At first I complained about the phase transition on the extreme Nidhogg battle, but eventually I just changed my mindset to think of a way to deal with it. (laughs)
You've also increased the ranged attacks that can hit anyone in the party.
Yoshida: Everyone eventually ends up focusing on their own job's DPS, but when we work on the balance we are looking at an average "party power." Rather than talking about an individual's firepower, I think the focus should be on high party power, like for raid clears or a strong PvP party. Rather than wasting time focusing on who makes mistakes or individual DPS, the priority should be on raising the bar for the entire party. However, when looking at individual jobs, it's true that there are strong and weak points depending on the content. For example, when looking at extreme Nidhogg from a black mage's point of view, I know very well that there are certain moments when you should use full buffs. However, it's important to keep getting better to win as a team. Even for mechanics which hold certain party members captive, you should think of the one person like a decoy and the others as the attack team. If you think of things that way you'll have more of a fun mindset and I think it will help push towards a clear.
I see... that makes sense.
Yoshida: That's true for melee DPS as well. Not putting out auto-attacks also creates a difference between players, and we'd like to work on that. Regarding that, please hold on a bit before we can discuss it further.
TOPIC #3: The difficulty of the highest level content seems set in such a way that not everyone is intended to be able to clear it, but with the way the compensation is set, a lot of people want to try it out. What are your feelings about this?
Yoshida: I think there are different opinions about this, but there is a tendency among Japanese players to feel like they need to go because of the items there. However, not to say this is the case for everyone, but looking at the data it seems like there is a stronger tendency for people to be interested in going because there's content rather than going just for an item reward. I think one characteristic of Japanese gamers is that they have a much higher obsession with clearing any content they see than players in other regions. The clear rate of the high difficulty raids and the extreme primal battles is top here, it's as much as a threefold difference compared to other regions.
It's a Japanese characteristic?
Yoshida: Perhaps part of it is a sense that you see other people clearing so you feel that you also need to clear it yourself. When the latest raid is implemented, there's about a one month lag in item level between the raid players and casual players. There's really only a difference of five item levels on the weapon within the patch series. So rather than that being the motivation, it seems more that players want to conquer the mountain ahead. I do want to eliminate the feeling that players need to do the raid if we can, but I think the characteristic is unique to Japan.
But there are some people who feel like they aren't enjoying the game if they aren't trying the raid.
Yoshida: Part of it might be that this is the first experience for Japan's online game market with high difficulty raids. You can't just clear with strong equipment if you don't have skilled players. You can't just win by keeping at it, you need to come up with new ways to approach the fight and focus your efforts, but it can be really exciting when you finally clear. I think the result of playing like that leaves a much greater impact. So, in order to further widen our range, one of the challenges I am thinking about is horizontally expanding the content.
TOPIC #4: Frankly, I think the difficulty for savage Alexander is too high. Tell us about your future policy for raids.
Yoshida: Well, for Gordias the DPS check was too high, so we tightened that up for Midas. I think significantly more players would be able to clear Midas if the level cap was still 50 with the corresponding skill rotation. Taking a look through the data and logs, it seems like the abilities of players to understand mechanics, make judgments about avoiding them, and maintain the level 60 skill rotation led to the large disparity we see. There are a lot of things that need to be managed and maintained that can be interrupted by phase changes or enemy attacks. It does require a lot of split second decision-making on the part of the player. Looking back at Bahamut's Coils, I think there were some quite difficult mechanics, but the difficulty difference of that skill rotation was directly connected with the content difficulty. Since it will be difficult to change the rotation in 3.X, we're moving in the direction of reducing the content difficulty for the last tier of savage Alexander.
So you'll be simplifying the content.
Yoshida: We want to maintain challenging aspects, but we're working on finding a good middle ground.
TOPIC #5: On the topic of job balance. Thanks to the ongoing adjustments, I think there's a great balance and a niche for each job. However, other jobs in the same roles feel like they are missing a little something compared to warrior and scholar. Do you think this is an issue?
Yoshida: It's not just a little, they have features which are hard to go without. We're not really looking at solo performance, but in terms of group performance, I think they are quite a bit ahead in terms of the things they can do.
By group, do you mean 8-man?
Yoshida: Not just for 8-man parties, I think that by far warriors and scholars are two jobs that have the most secure roles and can easily fit into a party configuration for just about any situation. Warrior and scholar also have a deeper skill set which allows for player skill to impact them moreso than other jobs, so it puts us in a tricky situation.
Do you have any plans for future adjustments related to this?
Yoshida: We have nothing planned during the 3.X series. Also, we won't be making adjustments that are focused on just those two jobs. With the next expansion pack, we'll be making adjustments on the whole that take the new actions into account. The basic idea is we want other jobs to get caught up, but not just by simply strengthening them. We'll have to plan carefully considering both the primary strength and the pet strengths.
TOPIC #6: I'd like to watch the raid race by veteran players from around the world (the competition for world's first). Can you set it up so that the clear situation can be checked on Lodestone?
Yoshida: I don't think the top teams would want to be monitored like that. There's a lot of factors that people aiming for world's first take into account - not just to clear it, but often also to clear it a certain way. For example, if a specific tactic is found out to involve a bug, you would be looking to see who can clear it in the intended way.
It might be difficult to officially determine that based on movements in the battle.
Yoshida: Also, it would look like we were endorsing players to go for fast clears of the raid if we had something like that, so I don't think we want to go much further than a post acknowledging world's first and a mention in the producer letter. Also, if we did it officially like that, it wouldn't be as exciting to see the exchange on the internet.
TOPIC #7: With regards to parameters like critical hit and determination, even if you increase the value by, say, 100, it's kind of hard to feel any difference. Also, with the values of the parameters continuing to increase, it can create situations where it's impossible to update your gear but also maintain the accuracy needed. Can you tell us your policy on this?
Yoshida: We already have the differences in player skills, so if we have to factor in parameters on top of that there would be too much of a difference, so we won't be adding elements which widen that gap. For parameters which are considered less useful, I think we'll be making adjustments with 4.0. Whether or not to keep accuracy is a hot topic right now. In the end, we want things to be as simple as possible so that you can concentrate on playing the content.
So there will be big changes in 4.0, including in this area.
Yoshida: That's right. We don't want to make it too complicated or make the gap with the top players too high, so we intend to make careful adjustments. Even before that, the essence is that we want people to want to enjoy it. Moving from level 50 to 60 felt like a totally different way of playing, so next time we don't want the change over the level range to be so extreme.
TOPIC #8: In content, there are mysterious pillars players have been calling "towers." What's the theme behind those and why do players need to step into the area around the tower?
Yoshida: I think it was created for turn 4 of the Final Coil of Bahamut. As far as the basic idea behind it, it's that you need to step on it to prevent the monster from powering up, it's kind of like energy pushing up from the ground, and the mechanic is that you're blocking that. We had the idea that it was probably something Allagan. At the same time, I was also nagging the team working on the extreme primals to try to unify the sloppy icons for mechanics. When we look across the content, the people developing it might be different, but it's the same group of players of FFXIV. For example, we were saying that the shared damage mechanic should have one shared marker. There were mechanics of the same nature with different looks and it becomes unfair. The tower marker is also used in other content later on, so the idea basically is, well, maybe the Allagans did it (laughs). So you step on the tower to prevent it from activating and powering up the enemy.
Yeah, I didn't even really notice when the shared damage markers were unified.
Yoshida: When you learn the shared damage mark after being targeted in a certain content and then you clear that content, it can be stressful if it uses a different mark every time in the future. The idea is that when we create new mechanics we will use new markers.
TOPIC #9: How many members on the team work to produce end-game content and how are mechanics developed?
Yoshida: The production pattern is roughly the same for everyone. In the planning stage, the person in charge of creating the content comes up with a design document. In the case of an extreme primal, there's a template that includes the beginning part, as well as the second part after the ultimate ability, but it's basically up to the person in charge to decide what they want to do. Once the battle system is confirmed, there's my check and once I give the OK we move on to talk about specifications. The specifications include which mechanics happen in which part of the battle, how the phases are divided, and how many minutes from the battle starting until a wipe. If it's considered too difficult at that point, we'll pare back whichever parts were considered too much to fix it. Once the checks are completed, the monster and animation teams will receive an order to develop the model and animations for the boss in accordance with the content. The effects team creates effects like lightning or other skill-specific effects that the monsters emit for each skill. One programmer remains as the representative for the content until it is released. The planner and dedicated programmer team up in meetings to discuss implementation. First they'll place a dummy monster model and the mechanics, then work on the boss AI (artificial intelligence), and build up the phase piece by piece. The planners will play through this on a development version made by the programmers and test each technique separately. Once all of the implementation is just about complete, we'll temporarily move to the debug phase. It's still not over once most of the bugs are removed, though, as we then gather some developers who are skilled at playing to do test play for the final balancing. This is the last, and most important stage of the development process.
So it ends with the play test.
Yoshida: After the members participate in the test play, they'll receive an explanation of the mechanics from the person in charge and then enter to try the battle. To maintain the major patch cycle, the adjustment period is about 5 to 7 days for each piece of content. As an example, if you include the normal and savage modes of new raid content, there can be eight pieces of new raid content in one patch. Since there's also a primal battle in addition to this, it's roughly nine boss adjustments. To understand all of the behaviors and phases perfectly, they'll start by playing completely invincible. Once they have a good understanding of the mechanics, this allows them to test the solutions that the person in charge indicates without being knocked out. We use the invincibility as a way to shorten the process. It would be wasting check time if they had to keep wiping and restarting, so we begin with the invincible state. After that's complete, we remove the fully invincible state and apply a different invincibility where they will take damage, but the HP will refill if it hits zero. This allows them to check that the damage inflicted by abilities is appropriate, that the load on healers isn't too high, and that the DPS check isn't too extreme, etc. They'll take another look at all phases through the final phase in that state and as much as possible aim to clear the fight without dropping to zero HP. After they can clear it like that, there will be a final check with debug commands off where it's possible to wipe. This allows them to check the sense of tension in the fight and check how many members can be KO'ed before the fight becomes impossible.
One hot topic that came up in the past on the internet was whether play testing was being carried out in a non-invincible state.
Yoshida: Right. It seems like I was misunderstood, but the aim of the checks is to clear in the same environment as on the public servers. Once we clear and determine that it's appropriate, we look more finely at the DPS and healing checks for each phase and determine how groups focused on clearing the raids as fast as possible will fare. We continue with adjustments up until the last minute like increasing a DPS check by 5% or cutting 15 seconds off the enrage timer.
Do you decide on the script for enemy actions at the first stage of the specifications? Playing through content, it often feels like buff timing matches up perfectly.
Yoshida: Since this area requires more experienced developers, veteran members are in charge of the planning stages. There are many abilities with 60, 90, or 180 second recast timers and the people in charge of the development are aware of it. The developers are players as well, so they take that information as well as their own experience into consideration. The team members in charge of battle content all pay the monthly fee for their own private characters and aim to clear the content on their own. By doing that, they can get to see the content after it's actually released and understand the efforts and tactics that players use to clear the content, taking that into account for the next production cycle.
TOPIC #10: Enemy attacks come in three kinds: physical, magical, and non-attribute [[darkness]]. Will we be able to identify which attacks are which attribute by appearance?
Yoshida: That would be tough. In the planning of the specification design phase, the team developing the content gives general orders for the look of each ability to the effects team. At that stage, the abilities are made as much as possible to look like their skill attributes. However, as the content comes together and we begin balance testing, we sometimes discover that it's too easy to avoid particular actions for certain jobs. If we have concerns about one job becoming an "essential job," then we make adjustments to eliminate that feeling. However, if we went back to re-create effects and wait for the checks, we would miss our schedule. So changing the attribute of an attack is kind of a last resort for performing job balance, so please forgive me for that.