Illuvium Dev Blog — November 2021
There’s only been a couple of weeks since the last blog entry, so please don’t expect too much new information this time. I’ll try to be as detailed as we can. It’s funny sitting down to make these blogs, knowing the community will see them. It reminds me of one of my first tasks on the projects. You won’t understand why, but eventually, you will. Let’s just say that I feel a strong sense of emotion and connection with the community when I write these. I’m so glad the community suggested it. Ok, enough sentimentality; on to the updates.
Business and Partnerships
Raise the Roof
The major news is that we are initiating some type of raise. Some already know, but we are looking to bring forward the release of mobile, as this is a significant driver of the player base and revenues. But due to how we build the game, getting it onto mobile will be an enormous task, so we need to raise some capital if we want it done faster. Our initial thoughts on this raise have changed, with Kieran being a firm believer in the community. And we think the final choice will be something that the community at large can get behind. We will always let you know the details before we launch anything.
Some people have wondered about transparency, feeling that we might do things without people’s approval. That will never happen. But keep in mind that sometimes we need to sign NDAs that don’t let us communicate everything until it is ready. But rest assured that the final say will always go to the Council.
New Council Elections
As I write this, Council nominations are underway. We are super excited to see who steps up. We offer guidance where possible and have a solid relationship with the Council. We hope for robust conversations next quarter as the number of IIPs and ICCPs start to flow through. Voting commences 5 December 12 am UTC until 7 December 11:59 pm UTC on Snapshot. Remember to cast your vote.
We are starting to partner with several large guilds to understand what we need to implement to make it successful. We have had some good successes here already, but keep in mind that even when we go public, we will still be in beta and do not expect to see a large number of guild features. That said, we see this as a huge opportunity and hope to build as quickly as we can to support them in as many intelligent ways as possible.
Tip of the Iceberg
The marketing team is mainly working in the background. The team are still doing all the things they have always done, but at the same time adding way more finesse to what we do, such as partnerships and advertising. They are working on laying the groundwork that leads to effective marketing: doing market research and gathering audience insights, putting in place robust processes and tools, selecting and engaging the right partners/agencies/vendors and platforms to work with and more.
They are currently focusing on building the go-to-market strategies and creating the marketing assets we need to launch the Land Sale, Private Beta 1, Illuvitars, and everything else. Some additions you will see that show our commitment to the customer are detailed ‘How To’ videos for the land sale and staking v2, an Illuvium Podcast, and many more guild partnerships. The cofounders will no longer be the ones writing and approving the copy for the website, which I can assure you is an excellent thing.
We learned a lot of lessons from the marketing team regarding the trailer. While it was technically strong and most enjoyed it, we can do better to make our trailer even more compelling and engaging. With their help, they will point us in the right direction from the start. By letting us know the key areas to focus on, we can create a script that will be punchier, reach further, and go viral.
Game — Overworld
We are nearly done with the basic structure of Overworld functionality and will have a video for you soon. This video will showcase a bit more of the world (unfortunately, no new regions) and give people a better understanding of how the Overworld will function. We’ve had a lot of talks internally about the nature of the Overworld and hope that we can strike the perfect balance of fun movement systems that take some skill to execute and more ‘grindy’ collection mechanics. People need to understand the Overworld is not meant to be a crazy action-packed adventure. It is a solo collection mission where players can absorb parts of an ‘environmental story.’ There are no quests such as ‘kill five rats,’ and there are no puzzles like the shrines in Zelda. This is mainly because the time it takes will not be worth the effort for a game designed chiefly for collection. We’ve always said the Overworld is like a pack of cards, but instead of opening them, you travel to them. What you collect will involve choices and some skills, but you won’t have to be a speedrunner to be effective. In the end, you are paying for this journey, and it shouldn’t be punishing.
The safety systems in our game are all in the backend. This is where we make sure people don’t cheat and ensure that everything is fair. What spawns in the world is not determined by the client. The backend is called to make sure everything is as it should be. This takes some time to set up securely. It also takes time to integrate it into the blockchain nicely. There are plenty of games out there that do it poorly. We’re writing many of these interactions ourselves to make sure that you find it seamless.
One area that we are behind on is the UI for the Overworld. We aren’t fully staffed yet. We have a great UI team but need at least one more UI programmer. I made a mistake a long time ago and passed on an excellent UI designer because I thought they were too expensive. Of all the people on the project, I’m probably the most financially conservative, and I want to make sure that everyone who joins the team is worth it, and the DAO gets their money’s worth. The downside of my conservatism is that sometimes I haven’t pulled the trigger on someone just because I felt their demands weren’t worth it. And now we are in a state where we still need one more UI programmer. We will find them, and we aren’t too far behind, but still, this is my B.
Game — Autobattle
The final primary piece of the Combat Engine was the move to a hex grid. That has been implemented but is in testing. A large part of our ability to run the first beta phase revolves around this. Hopefully, testing will be short. We’ve found some minor bugs but nothing major.
Additionally, we optimised a lot of the code base and segmented it in preparation for things to come, such as additional titles. Clean code takes time, and keeping everything clean has added to our build time. But long term, we feel this will be something that we look back on and thank our past selves.
A fantastic new addition (thanks to a new hire who is an expert coder we managed to pull away from Sony) is a caching system that effectively runs the whole simulation first on your local machine in less than a second and won’t be noticeable. But it will make the battle run smoother in real-time. It also opens up many possibilities (that won’t be in the private beta) in terms of cool visual features we can do. We can smooth out everything because we know the result beforehand. This extra polish pass also doesn’t affect the server runtimes and means that we don’t lose any speed with our reinforcement learning algorithms (since they only care about the simulation).
In the last update, we had the Cinematic Renderer working only in the editor, not packaged builds. That is now complete. We still need to clean it, but I’m relieved it is done. It will become a much bigger system that can manually move the camera eventually. I agree this is the eventual state, but don’t expect this even in the public beta.
One of the most labour-intensive parts of the game client was importing all the assets for the Illuvials. They are complicated creatures and require a lot of love to bring to life. We have worked tirelessly on this and now have a system to import them automatically. It has been a good exercise for us in categorising them and working out what makes each one unique. So now the only slow part of creating an Illuvial is the artwork. But we never intended to make procedural art. We’ll leave that to other crypto games. We’ll continue to do the high-quality custom stuff that we feel will make the characters iconic. Check back in 5 years to see which stood the test of time.
Eye of the Survivor
We are still scrambling to finish Survival mode. You will experience this first, so we want to give it some love. We decided to create a new battle arena for it, and we think you will enjoy it. It has a small story element with it as well. Not related to the main story, but something fun. It’s a simple mode, but it will get more complex over time. Most of the logic is there; we’re just finishing some UI elements.
Overall, it will be tight to get private beta 1 in by the end of the year. Many things can go wrong still, and we need everything to go right. One feature that we are excluding due to time is the Enhancement system. You fuse these things onto Illuvials to make them stronger for a battle. They are a part of the core design of all auto-battle modes, but it will still be fun without them. We pushed it to private beta 2. We might get them done before then, but almost certainly not before the start of Private Beta 1.
Game — Illuvium Zero
All the main features of the game are done. We will add more over time, but the original design (and then some) is in the game. Most of the user interface is done too. We’re just completing some of the authentication systems (which have been a real pain in the butt).
We’re behind on the artwork, but not by a significant margin. Overall, I think it is starting to look good. Some of the hand-painted tiles look amazing, and over time we will roll out more and more VFX and animations. It’s always meant to be a mini-game, and it still is, but it’s a nice one.
We’ve hired an extra person to take control of the backend of the game, and we will probably add one more so that others can focus on systems that need attention. Game development is always a juggling act, but we’re still in control.
Lately, the game design team has been focused on getting ready for closed beta 2, as a large part of the work for private beta 1 has been completed, which they finished at the start of the month.
Significant accomplishments include the data of the Synergy system. When we originally designed these, I was adamant that we could hit several features to make it all come together. The team met me with scepticism. However, they managed to pull it off. Undoubtedly, we will tweak Synergies as people start to play, but overall, I think we have the right level of simplicity of individual systems while having emergent complexity.
The abilities of basically every Illuvial have also been designed. A complete balance pass is needed, but overall we think they are fun and fill all the significant roles you expect in this type of game.
It’s important to note that there are so many limitations due to being a blockchain game, not just little things such as transaction times and technical challenges but also HOW people see the space. We are trying to take a very physical approach to design. Travel is equivalent to opening a pack of cards, for example. We are doing this so that people can understand the purpose of systems. We have also tried to design the game to make the language easy to digest. Tier ALWAYS means ‘how good and rare something is compared to other types of things.’ Stage ALWAYS implies a progression. This will help people learn the game more quickly, which is essential.
Last month we added two more members to the team and continued work on the land sale contracts. We have decided to make the land sale in a very decentralised way, which has proven challenging. But most of that work has been done. The overall land sale application (which will become the IlluviDEX) is epic. It takes time to polish things, but we would rather make sure everyone loves something when it is ready than hate it when it is not.
Staking v2 has passed the internal audit and is now with Quantstamp. The IIP will formally be submitted soon, and we expect it to pass. We are excited to show you v2.
The dev team continues to work on the land sale and IlluviDex implementation. There’s still work to be done, and this has caused some of the delays with the land sale, but it’s not precisely the fault of the Web team but more an issue of us working on our communications between teams and starting to focus more on projects rather than groups.
This is one of our first huge integrations that span many departments, and we need to get better about coordinating teams. It’s easy to make excuses about how difficult it is to coordinate, but the fact is that we need to get this done, and excuses don’t produce results. We’ll learn our lessons here and get better.
One of the ways we will solve this is by bringing people with experience working for large and decentralised teams. We have some candidates in mind from some massive game studios.
We are adding an extra UI designer to the web team so that we can start to become a bit more targeted in our focus. Our lead UI designer is working full time on the game.
The backend team has integrated the auto battler simulation into our execution flows. The Overworld will use this too, but more excitingly, it could be used in the upcoming private beta to verify players’ battles.
They have also supported the other teams in reaching their authentication goals, and this vast body of work will also be deployed as part of the first closed beta.
Our current focus is now making sure we hit all the targets for a successful launch of the beta and the land sale. I’ll be honest, we’re a bit light here, and this is an area where we are behind on. While this is not ideal, we are looking at creative ways to fix this because we only want to work with experts, which means many of our applicants are rejected for not having the expertise we need. We will do some internal restructuring to help meet our goals and hire people who might not have ALL the skills we need but have some of the skills and work in intensely communicative teams.
But if anyone knows any C# backend engineers with IQs of 150+, please let us know. It turns out they are rare!
The infrastructure team is growing, with two new members joining the team this month, including a Security Lead and an extra DevOps engineer.
We completed an initial SaaS product audit with another scheduled within the next six months. We’ll have a new SaaS product onboarding procedure available soon to ensure we do our due diligence when selecting vendors.
We completed Phase 1 of our proof of concept for a decoupled CI/CD pipeline that is highly fault-tolerant and scalable for future requirements. We’ve begun Phase 2, focusing on artifact promotion and deployments from dev to production.
This might seem overkill, but we need to put on our big boy pants if we expect to have a seat at the big kid’s table.
The team has rewritten the reinforcement learning environment for ranked to match the newest vision by the game design team. We think you guys are going to find this tech super cool. We extended the reward function to learn not only on wins/losses but also on remaining team members & health. This is due to the way that our ranked game mode will flow. This allows us to balance all the attributes in the game (not just values, but the modules that make up abilities, it’s pretty crazy) and create bots that act as agents in the matchmaking queues. You’ll be able to play against a computer if you feel self-conscious playing another human, and if the queue is low on human players, we can add them in to ensure that you can always find a match quickly.
Finally, we generated grid coordinates for the land sale map and the formula to create sites on the actual plot. It’s not as easy as it looks, but that is now done.
I’ve hinted at this up higher in the blog, but this is an area where we need to be better. I wear a lot of hats on the project, but this is something for which the responsibility falls purely on me. I’m getting better. We’re getting better. And we’re adding the people we need to get the engines working. So far, I’ve been carried by excellent leads that don’t need much production and outstanding contributors that don’t need much guidance. But as time goes on, we need to form teams that produce massively complex products spanning many groups. It doesn’t matter how good the team is; at that point, you need perfect production and product management. I’ll build that out, and we will get better.
Eventually, we will be the model that everyone else will use for decentralised teams in terms of output and efficiency.
The story is now complete from a “beats” perspective. I have to be a bit cagey about what this means because nobody knows exactly how we will deliver the story. But put simply, we are working on a long-form story that could eventually turn into a movie or series. This will NOT be what goes into the game. It just provides the structure for the world.
What WILL go into the world is 30% complete from a story perspective. That might not sound like much, but this will get released over a long period of probably six months post-release. There is still much work to do, but we are on track to deliver something pretty cool. As I’ve said a million times, we don’t have a quest driven story system. We have a simple loop that allows you to collect things and battle. No story there. But in addition to that, you’ll be able to get story elements that hint at a larger vision we have for the world, what it is, and what it was. And if I haven’t told people before, the story will be collectible just like everything else. So that will be exciting for the few people who collect it.
The next cinematic is underway, and we have bullish news regarding our ability to create these faster than we have in the past, but I can’t say anything now. Stay tuned.
Another area we have gotten even better a pumping out great content. We are done with the Lynx line and finished with Tier 0. These are huge lines representing a large amount of work with a different production line. And even though it is Tier 0, it needs to be the same quality as everything else we do. Some people love the designs. They represent something different to normal Illuvials. But you’ll have to play the game to understand.
Texturing and Fur
We continue to get these done faster and faster and at a higher quality. We’ve discussed internally that we now need to go back over the first few lines to ensure they are up to the same standard, but this won’t happen until the public beta begins. We’ve designed 3 of the battle boards as well, and they look great. We think you’ll love the Survival mode board.
I issued a considerable challenge to the animators. A six attack ‘sequence’ that we think looks dope replaces the old system. Ten Illuvials are complete, with 15 more WIP. These were started three weeks ago. That’s fast. Other than this, they are working on bringing to life the plants. We don’t guarantee that every plant will have animations on launch. We just don’t have the time. But it is something that we will continue to work on until they all have some movement. I’m sure everyone would love to have a fully finished non-beta game in March, but most will appreciate that the things in the world that aren’t completed will be fixed fast. We could have made things simple and scrapped the Overworld entirely and just had a game where you buy packs of cards. We could have let people buy them now to fund something. But this isn’t in our DNA. We love fairness. We love how raw everything is. So you’ll just have to come on the journey with us as we improve the game over time. For anyone who has been a fan of a game since its inception, you’ll know that this is a satisfying experience to constantly see improvements and watch as your feedback gets made real.
Being late in the process means that VFX always are one of the last to finish with an asset. So there ARE still a couple of Illuvials needed to be finalised for the private beta, but we should get through them. The biggest challenge will be the Ranger weapons. The modelling and animation teams have not completed these yet, so it will come down to the wire if we get them in. As I said above, we’ve already decided Enhancements won’t go into Private Beta 1, but I feel strongly that all the Ranger weapons need to be there. It will be tight. Moving to the Attack Sequence system has added extra work, but we’ll be releasing some footage of Rhamphyre soon that will blow your mind. We are also working on the environmental VFX and sprite renders for Illuvium: Zero.
As I type this, I’m getting ready to meet with a new technical audio engineer who will finalise our transition into Wwise, which will help us rapidly develop and integrate sounds. We also brought on a contemporary composer who will specifically work on music production. We are delighted with the style of the music, and from the comments, we’ve seen you are too. But we need to add a little bit of extra quality. There are no significant blockers with the music team, but they are so creative and always looking for ways to increase their workload in exchange for providing you with even better sounds. I can’t wait to leak a few of the older songs that have been improved. I think it shows a maturity to our work that wasn’t there even six months ago.
This month has flown by, just like every other month. We continue to work as hard as we can to get everything done. There are certainly challenges. As outlined above, some areas of the DAO have more work to do than others. Private Beta 1 by the end of the year will be tight. We could probably release something next week, but it wouldn’t have everything we need, and it wouldn’t be polished. And while I do understand that private betas mean rough edges, we want those that are lucky enough to join to have a fun time playing this small slice of the game.
So I thank you all for your patience. You know that we can’t be more transparent than we already are. If you read this and look at the leaks in our discord, you’ll have a pretty good idea of what is going on. We don’t mention everything or leak everything, but we try to provide visual context for the journey. I will endeavour to get these out at the start of each month. It feels excellent to discuss everything openly. I’ll get better at writing them coherently and concisely, but for now, you’re just going to get some raw output from me regarding everything that is going on, good or bad.
See you in the next one, Rangers.