Selatria - A retrospective/post-mortem Part 1 - This is a new blog series I've started over holiday break, and will continue into 2025, at least as long as I have topics to talk about.
The development of Selatria was long and tumultuous. It became a long running joke from circles I was in on how long it was going to take to complete the game. It took 13 years, 6 months, and 29 days to complete the game, or 4959 days in total.
Selatria changed my life, and opened up a lot of doors and networking opportunities into software and game development careers. But it was far from perfect, and I wanted to write this not only for others, who may have a dream game they're thinking of doing but not realize how difficult it may be for them, but for myself to see how far I've come.
For veterans of the game industry, some of these points might seem like they're obvious, but these pointers and tips may be useful for someone who is thinking about making their dream game from the get-go.
You can read Part 0 here.
(Note: This was meant to be posted at the end December 2024. But then I got extremely sick with pneumonia, then the wildfires/power outages happened, and then my computer broke. Not a good time. We're back at it now and hope to be continuing this post-mortem where I can.)
---
Profit/Percentage-Sharing vs Commissions
We started Selatria in 2011 when most of the team was in college or working very part time. When we started the project, we didn't have any spare money to work with so we started the project as a profit/percentage share for the early folks. By the time me and the other two founders got stable career jobs, we began bootstrapping our projects. However, we kept this percentage share in place for those who stuck it out from us in the beginning. (These earlier team members also were paid on commission) Later team members were only paid on commission.
The total for Selatria's development cost as of this post came out to $37,297.82. As of now, the project is still in the red, and hasn't turned a profit. Unfortunately, this is the case of the majority of indie games that are released.
(As an aside, please don't take the above number as a "Wow, this money all went to waste." - While the game itself wasn't profitable, it did create opportunities for myself and others that would have never happened outside of the game. This project gave job and career experience to many, including myself, in other ways, and it paid off there. )
Selatria is the only project in Whim Indie that we've developed that used this model, and we went with just contracting and commissioning the team for their work on the project on an extremely part-time basis for every project thereafter.
Unless the game is developed in a game jam, where everyone knows that there's a small beginning and small end to the game, I do not recommend anyone work on a percentage-share. It becomes difficult to gauge exactly who is doing the work, and there will always be this bias on what work/percentage is more. You can't have a good RPG without good art, and you can't have a good RPG without good game design and programming, and you can't have a good RPG without good sounds and music. The time commitments on these are all different, yet are all equally important, and in hindsight, putting a percentage value on it now caused a lot of resentment.
---
Game Production
Disclaimer: Please note I have no formal project management training and I don't have any project management certifications. A lot of this I had to learn the hard way through trial and error through my process of finishing this game up.
My barebones definition of what I think a Producer and a Director really comes down to this:
Director - Determines or makes the final call on what gets put into the game.
Producer - Determines or makes the final call on how the game is going to be completed.
Different studios may treat these roles different, and it's not uncommon for the Producer to give creative input on the game or the Director giving suggestions to improve the overarching process.
While having a Producer and Director be the same person may speed up getting approvals done allowing the team to work faster, it's a lot of stress doing both roles. Even my role-model Naoki Yoshida "Yoshi-P" has several assistant producers and directors working under him! When we were working on Selatria in person (before we went remote before 2020) I would be spending most of my time being distracted trying to answer people's questions on assigned tasks as we would be mostly working on a Saturday in the studio.
I found that working remotely and asynchronously and letting the team just work on tasks how they see fit and going one at a time in a standup call once a week worked a LOT better to help concentrate my focus. It also gives a chance for each person to show off their work in front of others and creates this positive feedback. Tasks are getting done, each individual team member gets to be heard, and the consistent time of the meetings means that everyone knows the rough due date for their work.
And people say working remotely isn't productive...
That being said, being Producer and Director of an indie game may look cool for title, but it's overwhelming to have someone be both. We did not continue this practice for our other projects in Whim Indie, and we keep the titles on separate people. This also helps accountability.
---
Scope Creep
It became a running joke in friend circles about how long Selatria was in development. A series of broken promises and bad estimates with no thought put into them, caused the game to shift ever further into the future and for a good four years it really felt like there was no end to sight for this game.
- Selatria continuously suffered from scope creep
We had dungeons and battles planned but without a solid plan for where they were going to go. Because these dungeons and battles were already built, we had to make the story go that far to see it.
A lot of Chapter 6's dungeons which require going around the world in the airship to a lot of different places to be able to get certain items to advance the story.
In Chapter 3 (you can also see this if you played Advent of the Dakk'rian Empire) we put a LOT of work into three different paths the player can use to sneak into the main Capitol Fortress. That's two significant paths the player can never really fully experience. We were going to have a similar branching path in Chapter 5, and we got so far as to have the voice actors record all the lines for the branching paths, but we axed it fairly late in development to save on time to finish the game. We will likely not do branching paths where the player can miss a significant amount of content again in the future.
The whole final continent at the end of the game was meant to be a surprise addition to where the player can do a lot of endgame content and optional boss battles, including a randomized tower dungeon. This was also scrapped before the game released and the size of the continent was scaled down considerably to just include the parts that were needed to finish the story. Had the game did well, we would have likely put more development time into this abandoned randomized tower concept, but we may bring that back in a future game.
The game was originally 12 chapters, and we could not come up with enough plot points and dialogue to fit within our smaller game budget, so we merged things together into 9 chapters, and once more before we settled on the current 6 chapters. We were also running out of space for abilities and maps. The RPG Maker VX engine we used was "only" limited to 999 slots. We also did not manage the database well and made it difficult to be able to have enough content that felt new and fresh every time.
From a writing standpoint, we should have started from the end and worked our way backwards, this would have helped us figure out how to write a more cohesive story and allow for better setup of plot-twists.
- Saewo was a pain to design for
 |
| Saewo - Also because this blog entry needed a picture |
The development of Saewo was troublesome from the time we came up with a character to the end of the design. We wanted a fistfighting character that gets really Street Fighter Akuma levels of angry, who transforms into a dragon like Breath of Fire, with attacks as devastating as a Final Fantasy summon.
We developed the dungeon this character was introduced in years before we actually began developing the character, and the dungeon was a hit in conventions we showed it off at. The creator of the Aveyond series even left a comment when we showed it asking how we were able to pull off the design of Mount Oreya with its shifting tiles in an engine like RPG Maker.
Coming up with Saewo took a lot of back and forth between Myself and Jon. Jon was able to come up with a twist on the AP/Mana resource mechanic we were using and come up with something unique for Saewo with his RAGE meter. It helped solve a writing problem we had, since Dakk'rians lost the ability to use magic, it's the dragon and his rage that allow him to use his abilities.
- The importance of Task Management, Proper Communication Channels, and Version Control
We spent almost four years developing the game with no proper task management system. We were coming into the studio and working on whatever felt like fun. It was the most creative time of development, but there were no constraints or deadlines, and because of that, there was no light at the end of the tunnel. We used a Facebook group to communicate with one another on development of the game.
In 2014 when me and some of the other team members got day software jobs we began to learn about how useful these task management systems were and I wanted to start using one that worked for the studio. At first because we were all working from the studio and we didn't want to pay anything, we used post-it-notes, a whiteboard, and the free version of Slack.
I don't remember exactly when I learned about it, but I learned about HackNPlan while going through the gamedev subreddit, and it seemed like a cheaper/cost friendly alternative to Trello, another task management "kanban" system used in the industry. We started using that tool and it really saved us during the pandemic and development of our other games when we stopped working from a central physical location. We continue to use HackNPlan to this day, and even got ourselves featured in one of the dev diaries by the developer in 2017. Thanks, HackNPlan!
We used "version control" from the beginning with Dropbox and Google Drive, but we used professional version control REALLY late in development. I believe we should have tools that are user friendly for the team to be able to work. And I don't think even the professional game companies have got it figured out. Sure you can have an artist or musician try and learn command line tools and Git to be able to pull/push changes to a project, but it makes things time consuming if they need help or get frustrated, or a merge doesn't happen as intended. I've been working in software professionally for 11 years and I still have a guide next to me with git commands. That being said, we never used it for Whim Indie, and we continue not to use Git. Instead we use Unity Version Control (Plastic SCM) for programmers and Gluon for artists. It has command line tools for people who want to use it, but the pros for this is for the artists in our team where they can quickly and easily check in their work into the repo without having to teach them command line tools or Git. It works for us, but I think we're grandfathered into pre-acquisition prices. So your mileage may vary. Perforce may be a good alternative to this.
As I mentioned before a lot of this was an expensive lesson that I had to learn on my own. Perhaps out of my stubbornness.
Selatria Part 2 - About Difficulty, Spite-Based Game Development, Game Engine Selection, and why "Everyone Needs a Paul" will be posted in late-February.
---
You can support Whim Indie by wishlisting or purchasing Selatria on Steam (Leaving a review would be extremely helpful. We're trying to reach 10!)
Wishlist/Purchase Selatria on Steam
Wishlist/Purchase Spellbearers on Steam
Purchase Spellbearers on Nintendo eShop
And/or you can support by joining the Whim Indie Discord!
No comments:
Post a Comment