Aloha Thrivers.
Last week, when we announced that we’re shifting our release date to January 18 2024, we also mentioned that we were going to shed some light on why and to bring as much clarity and transparency as possible to what led to that decision.
The Road Ahead
We had a meeting, going through a series of requirements that we didn’t want to end up compromising with. Here’s what we’ve set internally:
- No server crashes or items corrupting which leads to players losing their inventory.
- Minimum 30fps everywhere on PCs matching our minimum requirements.
- Servers can hold a high volume of players loading & playing the game and have been tested with over 150 bots and players actively playing as normal. Response time is kept under 30-40ms and players can queue for a server with no issues.
- Players are correctly spread across servers without major issues.
- Server Travel is reliable and bug-free.
Our Queue System ended up needing more work before it was ready, and we didn’t want to test things last minute before releasing the game in Early Access. Doing so would have likely uncovered some major issues that would severely impact our player’s experience. On top of that, we believe the game’s performance still needed some more improvement before Early Access.
What are we doing Server Side?
The keyword to understanding what’s going on with our server is “scaling”.
Getting a bit more technical, we are implementing Kubernetes and writing scripts for on-the-fly deployment of servers/databases/service apps. Some of our existing systems are also getting reworked to facilitate scaling. For those who understand what we are talking about, this means stuff like moving from static IP addresses to domains, changing info that used to be passed on the command line to be read from the database, using automated load-balancing, and so on.
We’ll also take advantage of the extra time to work on some features that were lower on the priority list but that will have a positive impact on the game. This is stuff such as improving AI (enemies/wildlife) performance when a lot of players are spread across the map, improving our caching system to reduce overall Database load, implementing various 3rd party services/frameworks for monitoring/reporting activity, or just adding more server functionality that we didn’t previously have.
So what about performance?
A lot is going on here. We feel we have reached a point where we believe most modern gaming PCs won’t have issues running the game, however, work on performance is just getting started.
First and foremost, we have built an internal system so that we can gather performance statistics and compare various PC Hardware to pinpoint issues with specific setups, locations or situations and then apply targeted performance optimization wherever necessary based on our findings. Since the Unreal Engine 5 upgrade was such a huge lift, we couldn’t dedicate resources to develop better testing systems, and this tool will play a pivotal role in helping us increase performance all around.
We are also spending some time optimizing foliage. Since our engine upgrade, we lost some features we had before, and one of these was the ability to de-render foliage as the player walked away from it. Trees, plants, flowers and shrubs will soon start disappearing at medium and large distances from the player, significantly improving performance.
We are aware that Unreal Engine 5.3 brings significant benefits under this topic, however, we are not planning on doing another engine upgrade until after we have our game released in Early Access.
Some other areas we are working on:
- Since allowing players to play with their friends is a priority for us, we also need to rework the menus and the UI alongside the logic for server and veil travel.
- We have moved our dungeons to their own levels, and that requires more testing, especially traveling back and forth to Lahaina.
- We are improving the visual quality of the game at Medium and Low graphics settings, without sacrificing performance.
- We are re-introducing some visual stuff that we lost with the engine upgrade, such as rain puddles.
- We are adjusting the lighting on Dungeons.
- Lastly, we are doing many quality-of-life improvements, bug fixes and filling the world with more stuff!
We hope this brings more clarity as to why we ended up delaying the game. We thank everyone who recently shared their support with us on our discord server and social media. We are doing this for you and we wouldn’t be here without you!
Questions? Send them via this form.
Below, our Delay Announcement:
