Inspirational work .
Thanks for sharing, really impressed.
Hi , I’m really interested in you work.
I’m currently working on a large scale RTS and am almost at the point where I need to think about crowd management and pathfinding.
Do you have any updates on your system?
I would love to see this on the marketplace, or anywhere really,
as I’m sure many of us are!
I love seeing how other’s implement things for my own ideas.
keep up the amazing work!
Hi ,
Any update on your progress?
I would love to be able to see your implementation, as I know many others would too!
keep up the amazing work!
Really impressive stuff. Can’t believe I’ve missed this thread until now.
You really seem to cover pretty much everything there is to different RTS games. I’ll definitely keep an eye on this.
@Dragonstorm97 : sorry, somehow I missed your comments. of course, the project is in development, but recently I had to totally focus on my other project, which is the AI system of NPCs in a really cool WW2 FPS/RTS hybrid game called Our Ghosts of War, using UE4’s navmesh, crowd and behavior tree systems. I don’t plan to share the code until it is suitable for making a playable game without too much extra work, and I’m still hesitating to convert it into a plugin (and sell it on Marketplace), maybe I will publish it as a project template somewhere else… (I should do some market research)
@kreivi_krapula : thanks! basically I want to make my own ancient/medieval RTS game, so have a long list of todos 
in the last weeks I have implemented a damage system (ranged, melee, collision), which resulted in further works on dynamic map data handling and group formation updates, so in the next video I’m planning to present some fight and destruction 
, hi! I sent you a PM about your project. Pls check.
honestly if you put this on the market place id buy it. solid work always loved RTS games however id try to convert it to vr im addicted vr atm.
yeah a VR RTS could be cool, I have already found one attempt in the forum, it looks fine, but I get dizzy when watching VR gameplays recorded… anyway, a god game should definitely be fun, or playing a battle game as a warlord standing among the soldiers
btw I spent my time with detecting and fixing strange and rare bugs, experienced only in certain situations at complex crowded places (resulted in units moving into invalid tiles), requiring a lot of testing and causing headache… but finally solved by implementing functions realizing and handling illegal unit movements as early as possible, hiding the errors from the player and avoiding units to get stuck, so I can move on with the development to deal with destruction…
Recently I have packaged the project (Windows 64bit), so if anybody interested in testing, please send me your email in a PM, then I can share it with you via Google Drive.
It is a development built, so you have access to console commands and some debugging stuff. Features are limited currently to spawning/splitting/merging various groups of units (infantry, cavalry), pathfinding and movement in the test level you could see in the video.
If i wasn’t so busy at the moment I’d definitely love to test it out. I’ll keep this page bookmarked
Long-time lurker here but this is my first post. Saw this project and found it to be really cool. Are you still working on this, ? Or perhaps any plans to get back to it? It’s not far off what a colleague and I are working on. We are inspired by Age of Empires and Company of Heroes but have come at it from another angle. Would be good to chat to see if there’s anything we can share or even collaborate on…
The development has slowed down a bit in the last 2 months because of personal life difficulties, luckily it has led to many testing, identifying and fixing bugs instead of new features. Stability is great now. Moreover, time was spent also on establishing my game concept…
A little video about smoothed movement, dynamic group path update, and some debug visualization:
@ Good to see that it goes straight forward with better implementation of movement.
Some questions:
-
How works the formation, are able to use a greyscale bitmap or shapes and change while runtime?
-
Did you have tested it with the UE4 mannequin with its skeleton as performance reference in extreme situation?
Im really interested in this scenario.
yeah, movement is never good enough
- no but a very simple csv file (can be made with excel), any shape does not work, a line based approach is okay like line, wedge, square
- UE4 mannequin is quite high poly, and in an RTS less bones are enough, e.g. no need for so many finger bones, they just eat performance but nobody can really see it… anyway, soon I will test with some custom characters using the UE4 skeleton, made for our new game called Edge of Chaos
Another short video showing ballistic aiming:
Groups and units do weapon range and ballistic aiming calculation to check whether their targets can be hit or not. If more than 80% of the units of a group can hit the target, they stop approaching and start to throw their javelin (or shoot arrows).
To use a Excel generated csv for formation is really simple but very clever. And yes i have seen this vid last week on your YT channel, and must say this is better implementation as Company of Heroes has, there soldier’s shoot in building corners.
lol
Did you know this Plugin?
https://forums.unrealengine.com/community/released-projects/125045-voxel-plugin™
Are your phatfinding system adaptable to use this plugin out of the box with its terrain destructible features? This would be so heavy, The Ultimate RTS Kit.
An extreme test with your custom characters and Mannequin skeleton and a eventually comparision with UE4 standard pathfinding is what i will see and how its scaled. 
Yes I know this plugin, and have ever wanted to implement the support of it, but usually had more important features to complete first… My pathfinder supports no caves, only a heighmap like terrain (made of landscape and static meshes), thus I talked with its creator, he promised me once he is going to make a more simple deformable heightmap terrain plugin too… Only question of time and money Anyway, if the voxel terrain collision channel is set to my custom terrain channel it must work, what I need to implement is runtime area update (measuring and storing terrain surface height and slope data), which is not a huge problem.
@ have send you an pm.
Hi it me I messaged you about two week ago and I wanted to know if I could maybe have email to chat to or if you have private way of talking?