This is a follow up/progress report from my previous roadmap post that you can see here: https://steamcommunity.com/games/744900/announcements/detail/1585754447302643497
Open World
I must admit that this Open World update has been a lot more challenging than I initially expected. I figured that I could work on it 3-4 days per week, leaving time to make smaller updates to the game, and still have it all finished in a reasonable time. However it quickly became apparent that in order to make decent progress I needed to spend a lot more time on it which is why things have been so quiet recently.
The general idea was to take the system I had used to procedurally generate the existing streets, and simply expand it to handle more road layouts. From there I would have it generate around 100 "tiles" which could then be loaded in on the fly. So as you reach the edge of one tile, the next one would be loaded in and one of the older tiles removed giving the effect of either a very large, or even unlimited game world. I would try to keep view distances fairly short in order to keep the game running around the same as it does currently.
However the first problem I faced was that as each new tile was loaded in, Unity would cause a long stutter of up to 2 seconds each time. This made travelling through the city very annoying. It took me a long time to debug what was causing this issue and optimise it. I've now managed to get the loading to happen without any noticeable stutter on my test machines (one high end, and one low end laptop) but I am aware there may still be more work to be done to have this work for ALL systems.
The next issue I faced was that the existing random generation system simply wasn't fit for purpose to generate the new tiles. It was all based on a simple single street layout and it simply didn't work for anything more complex (crossroads, tjunctions, corners etc). In the end I actually completely rewrote this system in order to give me more of the needed flexibility.
At this point I was quite happy with the world, but I was aware that it simply wasn't interesting enough. Yes it was "open world" but what we had was still pretty much similar looking streets repeating over and over. One of the big complaints people have is the lack of visual variety for the outdoor areas so I knew I needed to expand things further. I decided to work on creating more interesting area types, such as woods, fields, construction sites, grave yards, parks etc. In order to do that I of course needed to add some new environment objects to game.
Unfortunately I had forgotten just how fiddly and time consuming it can be to setup new objects to be included in the random generation. Each one needs a lot of work in terms of sourcing the artwork, adjusting materials and shaders, creating accurate optimzed colliders, setting up LODs and then defining the rules for how the system should place that object. In total I've added about 140 of these objects and I still have a few more to do. This again took me far longer than I imagined.
As it stands, I now have an Open World which I am pretty happy with. There is a fair amount of visual variety and a lot more to explore. The world feels large and realistic. Just running around and hitting zombies in it feels a lot more entertaining than before. But I still have the following things to do:
- Make it work well with zombie spawning. As it stands the way zombies are spawned/despawned is causing some more stutters so I need get this system optimised further (this should also improve OA performance)
- Modify the server so that it understands the new world layout and correctly connects up all zones and buildings
- Make everything work with multiplayer
- Modify mission and event/OA systems to work with the Open World
- General testing and polishing
This is a lot of work but to be honest I much prefer working with code/systems than I do fiddling around with things visually, so I believe this phase will be a bit quicker than the last one.
I am sorry that this has taken so long, but I'm hoping that you guys can see from these screenshots that I am making progress and that the project will be finished in the near future. I suspect that it won't be done before the new year as I still have some prep work to do for both DF1 and DF2's Christmas events and there is always a lot of personal/family stuff going on at this time of year. But I do hope to have things finished in January, maybe February at the absolute latest.
Level 30-50
As I'm sure you can imagine I haven't been able to make much progress towards this as I'm busy with the Open World and I am waiting on new artwork. However I can report that Milen has finished all of our new concept art for the new creatures/bosses and Dobrin is now working on the actual 3D models, texturing and rigging. I expect that this new enemy art will be complete by February which should line up perfectly will the release of the Open World update and so should mean I can jump right onto this next. I think it would be silly of me to give an ETA for the Level 50 update at this stage, but it will almost certainly be less work than the Open World was.
Thanks again for reading and for your continued patience. I know that waiting around this these improvements must be annoying, but I believe I can safely say that by this time next year DF2 will be like a totally new game, and something with enough depth to stand the test of time much the same was as it's predecessor has done :)
I've leave you with these concepts of the new enemies below. We are planning to add all of these into the game which represents more than a 50% increase in the number of enemy types. Hope you like them as much as I do!
