Hi everyone,
I wanted to give you an update on where we stand on the production side of Endless Legend. We are knee deep in development of a great beta version, but you do not necessarily realize that. So here are a few clarifications.
Process and updates
Since the release of the alpha version by the end of April, we have delivered a number of patches for the most crucial bugs, then resumed the development of new features which were dearly missing in the game you played so far. The way we work is in sprints of 2 weeks, which result in a build review and the start of a new sprint. In parallel, we send game versions to our QA on a regular basis, which results usually (we are not perfect
) in bugs to fix. The difficulty for us is that fixing bugs prevents us to develop more features and improvements (including the ones you requested). So for us it is a matter of balancing the time spent on fixing bugs and developing new features. Also, there is a strong temptation to become feature complete to avoid new side effects and bugs.
The bottom line is that it is difficult right now to deliver a build with the quality standards you are used to. Still, we are considering releasing a “work in progress” build in a parallel branch to the main build. It would be your choice to download this build (however imperfect) and give us feedback on the current development. We will update you soon if we choose this solution.
In any case, we are really fully focused on the game, we restaffed the team a few months ago with additional developers, so we have what it takes to reach our objectives.
Current developments
Since the release of the alpha, we have worked on the following aspects on the game.
A new faction is almost complete, with graphical assets (units, cities, animations), design content (affinity, traits) as well as its specific gameplay. This faction was more complex to develop as its gameplay is really exotic and brings what some of you thought was missing in a Fantasy game (I think I revealed too much already
). We have also started working on the 6th faction of the game, which contains a very specific gameplay as well.
We have implemented new modes of transportation, especially with the ship navigation system, which now allows us to play on maps with several continents. Sadly we need to teach the AI how to use it, so we have to wait a bit more before we release this. We have also added a road system which can connect cities and speed up transportation on land. We are developing the trade routes aspect in addition to that, which will greatly encourage the use of the diplomatic system.
The diplomatic system was pretty well developed feature-wise in the alpha, but it was missing a proper AI behaviour. We already have developed a good deal of peace/truce/war management by the AI, and are progressing on the ability for the AI to setup contract terms and handle more complex options. The diplomatic AI has been developed with an Adaptive Multi Agent System (AMAS) which is already operational and shows good potential to emerge interesting AI decisions (we used it on the low level city management with good results). It is a huge step for us because the tools which come with this AMAS make it easier to monitor how the AI is reasoning, instead of having to understand a bunch of interconnected algorithms.
We are also developing other capabilities for the AI, such as acquiring resources on the terrain, designing units relevant to the owned resources, and assembling more consistent armies. We are working on an improvement for the military behaviour and a better interception/avoidance of enemy armies by the AI. A new city governor has been implemented and seems satisfactory. We are improving the deployment / targeting system in the battle. The deployment itself is almost done, and should take much better account of terrain elevation and defensive positions. It should greatly help the development of the new targeting system in battle, so that units choose more meaningful targets.
Speaking of the battle, we have already applied numerous improvements to its behaviour, for example the display of the units list in deployment. We still plan on adding a game option to select the number of resolution rounds per order phase. For players avid of complete tactical control, you will be able to give orders for each execution round. However we hope that the new battle AI will be greatly improved and will allow less tactical players to enjoy the battle just as well. We are spending a great deal of time improving the feedback and understanding of the battle, especially displaying exact stats in real time, which is missing right now. It is a complex task, because we are transforming a round-based simulation to an almost real time simulation, and I suspect that will not make things easy for multiplayer synchronisation…
There is a number of smaller features that I will enumerate rapidly and which are complete or all but complete: an immersive faction selection interface at game start (instead of a droplist), improved Vaulters holy resource gameplay, improvement of the Siege system, city militia to avoid losing your city to a single scout, several visual improvements and animations.
Multiplayer
This is the one big feature we want to release for our beta version. We have been working on it for months now, and even before then we built a new messaging architecture from the ground up. The architecture is solid and we have been able to play the game for tens of turns without desync. However desyncs do occur from time to time, as new added features do not always take multiplayer into account. It is a tedious task for our multiplayer developer to track all these mishaps. Our new architecture seems to be robust enough (much more than Endless Space) and we also have developed specific tools to track desyncs and fix them faster.
Again, the temptation is strong to complete as many game feature as we can before we deliver the multiplayer, to avoid regressions, but we cannot wait forever and are impatient to see you play the multiplayer. And besides, that is what the Games2Gether is all about: adding new features. On a very positive note, we have a much better management of the host dropping the game. When this happens, a new host is elected and all remaining players enter a new lobby, ready to restart the game. We have considered having an auto migration, but this could result in an unwanted loading time when the host leaves the game. Instead we prefer that players decide to continue or not, in the lobby. Of course we will listen to your feedback when you try it.
Mac version
As I stated in a previous post, it is difficult for us to deliver a Mac version right now, as it would slow down our QA and build delivery processes. Adding features and improvements is the priority right now. Still, our engineers have started looking into the Mac version. The biggest issue usually consists of making running on Mac the external components used by the game (DLLs, video and sound playback). The game is running on Mac now, although we have to fix some graphical shaders for OpenGL. Also, we had a small setback as a burglar (for real) stole our development MacBook last Friday… So a new Mac is coming, we are on it, but we need more time to avoid jeopardising the development of the full game.
Next update
As stated above, it is difficult for us to deliver a new version right now with sufficient quality. However, we will look into a “Test” branch for the bravest ones. For the rest of you, we will probably deliver a new build between 2 to 4 weeks, depending on the quality of what we have in 2 weeks.
Thanks for your attention!