Build an RTS for fun.

Hey guys. I've recently decided all RTS games aren't very good and none have done anything interesting for about 15 years.

I know the Planetary Annihilation is in the works, I'm a backer, but I simply cannot wait for that game so I'm going to build my own to pass the time.

Just wondering if anyone else has attempted an RTS and if you have any tips or tricks or things to warn me about.

wish me luck.

Cheers,
Antony.

P.S. This probably won't amount to anything as I'll probably lose interest when it gets hard as I'm doing it in my spare time and not for commercial purposes, but we'll see.

I'll keep this up to date on what i'm currently thinking about and exploring so you don't have to trall the thread to figure out what the current topic of conversation is.

Current ideas worth exploring:
* Line of sight - view cones and elevation.
* Sound generation and detection of units and buildings (kind of like radar in supreme commander i guess but possibly only applies to infantry detecting sound and vehicles making sound).
* Day/Night visibility - Units in the dark are harder to see and harder to shoot
* No grid, freedom to rotate things - more freeform base building.
* Active resource collection - Not sitting in your base camping on resource spots, you have to leave and go get retrieve the resources.
* Strong buildings - Make buildings so strong that destroying them is difficult and a more cost effective and faster way of winning if capturing or clearing a building out.

 

I have had lots of ideas for the best RTS ever, just need to wait until I'm rich to make it

 

I always wanted to see a RTS that takes from turn based strategies like Civ and AoW. I think Endless space tried something like that but it failed hard.

 

I just loved .... Populous this one is a favorite not quite sure why that is ?

 

I never played Populous, before my time. Great sound effects... Haha!

My favourites are C&C + C&C Red Alert (Resource collection), Supreme Commander (for the camera and some cool UI advances), Civ IV (So many different strategies!), Company of Heroes (Directional combat). I'll try mash in the best bits and eliminate the annoying bits...

...But first I need to build some basics.

Here's a quick scale test. Currently i'm liking this, it's sitting somewhere in between Company of Heroes and Red Alert.



I think i'm going to go for a reasonably small scale in terms of map size, unit counts and such as I want to keep scope small and don't want to have to deal with a lot of optimising for hundreds of units etc.

 

Yeah I feel that most RTS games either have too small a scope or too big. Getting that right balance can be illusive.

 

Hopefully Molyneux's remake will be good - Godus Kickstarter it certainly looked interesting with a nice visual style.

 

the two technically difficult things i have encountered are networking and enemy high level ai. only few documents can be found about these. also rendering hundreds of animated units can be a hard task if your game needs it.

good luck.

 

Paradox Interactive kinda makes games like this but they aren't really RTSs(RTSes, RTSi? Whats the plural?). You can pause and resume, and order armies around, but they are grand strategy games, and very complex. More complex then Civ even. Check out Crusader Kings 2, Europa Universalis, Victora etc.

 

Some of my favorite RTS games (though most are frightfully outdated)
7th Legion

Dune 2000

and, of course, Starcraft 2 was pretty impressive.

 

Definitely a lot of cool stuff to draw ideas from. I really need to flesh out the design before I start delving to far into the tech side of things. I really need to keep things basic as I am only 1 guy and I'm building this in my spare time.

I'm not going to focus on graphics at all. Units and buildings will stay largely grey boxed like in the screenshot above. The only colour or texture will be added to aid readability and communicate game state, like damage etc.

 

did you ever play the 3D version of Populous? one of my favorites... then there is Age of Empire my first RTS, Shogun Total War was pretty awesome to with Massive Armies!

 

As much as I want to keep this simple... I'm getting some cool ideas around line of sight, sound, and time of day mechanics.

Basically I don't like how infantry are quite useless in most RTS games and I think this is because a lot of their advantages aren't modled in the games I've played. They have less visibility or the same amount of vision, are slower, weaker, can often be crushed by vehicles, are just as visible to radar and don't get any terrain movement bonuses. I think these things can be things to play with to make infantry a real tactical unit. The heavy units also need weaknesses, like direction armour and decreased manoeuvrability in some situations.

I think company of heroes did a good job of touching on some of these aspects but I don't think they pushed it far enough. I also think their implementation got bogged down with micro management. I think I can simplify it a bit and still get all the gameplay bonuses that resulted from their system.

Edit:

Also some other advantages infantry have is they can be extremely quiet and stealthy. Tanks cannot. So, provided I implement line of sight and unit view cones I could also introduce a concept of sound (omg sound??!??!). Yeah so your units can gather intel based on if they can hear an enemy even if they can't see it. An icon of the enemy could appear at its approximate location. This means vehicles could be heard even if they are out of the line of sight of your units, where as infantry could remain undetected. Also the amount of sound generated could be based on if a unit is in motion. A moving unit would be harder to keep hidden than one waiting in ambush.

 

Definitely want to do some cool stuff with lights now! Who needs textures when lighting can add so much depth and such a dynamic look (when it's in motion).

Day:


Night:


Unity rules!

 

Looking good Anthony !
I'm also designing a rts game (although it's more tactical than strategy) , and the focus on my game is the infantry, as you said infantry has been made as an easily disposable unit without much value.
Something that I believe hasn't been used much, is interactivity of the units with the scenario (opening doors, accessing consoles, disabling turrets, etc), making the gameplay a bit different than just build build, attack attack.
Keep posting

 

Yeah, the more I think about it the more I'm heading towards Z Soldiers. In that game infantry were everything because vehicles had to have infantry to drive them. Infantry captured flag points and the zones you captured had factories rather than you having to build them. think Battlefield 1942 but in RTS form.... and 15 years ago.

 

Sorry don't wish to offend the OP with Hijacking the thread,

I skipped over the 3D version but at some date afterwards played Populous II, Which it seemed to have lost the Innocent beginning's, certainly did not favor the greek classical period.

 

Starting my RTS interface last night I discovered that I have no code to handling right click very well... Developing iOS games has taken its toll.

 

I would like to see some asymmetrical warfare for example maybe there was a race that wouldnt actually build anything but would instead steal everything from the enemy. So instead of controlling a whole army you control a group of heroes.

 

Dawn of War tried this a little bit, not as extreme as you're suggesting though. Yeah it definitely would be cool to see how far you could go with this. Supreme Commander 2 also has a similar Hero theme as a valid but controversial tactic of Commander rushing works pretty well... but it does create a boring match though.

 

After thinking a lot today about things I've decided that RTS games have real trouble trying to make collecting resources an interesting part of the game.

**** If you think this post is long then skip this bit ****

Westwood style, Dune and early c&c had a reasonable mechanic where you had to send a vulnerable unit out to collect ore/tiberium/spice and return it to the refinery for conversion into money. As the resources near your first base dwindled you were forced to send your harvesters further into dangerous territory to gather resources or spend your hard earned cash on setting up a second base to make harvesting safer/faster.

Chris Taylor changed things a bit by making resource gathering a choice. In Total Annihilation and Supreme Commander 1 and 2 the metal and mass deposits on the map were simply a kick start for you at the beginning, once you had enough energy you could simply make metal/mass in your base and turtle for eternity until finally your super weapons finished off the opponent (provided you survive the dreaded peewee and ACU rush).

Relic seem to have settled on capturing command points that allocate you resources for as long as you keep control of them (Dawn of War, Company of Heroes). This was interesting to me at first but my multiplayer experiences of it have been plagued with exploits and annoying gameplay. The game is often won in the first five minutes by player who manages to rush, grab and keep the most command points while avoiding combat. Generally this resulted in getting the upper hand pushing home and winning with a rush soon after.

Age of Empires was interesting as it required different resource to build different things, some of those resources were not always easily accessible and there was also a limit to those resources making every battle count towards the end of the match. In my mind it sounds like a good idea but where it works extremely well in a game like Civilization I can't help but feel like it's to much to think about in an RTS.

To be honest I think the Westwood's approach is probably the best implementation of resource gathering to date. The reasons why are: a) It fits the fiction well and, b) It provides you and your opponent opportunity to strike the opponent's resource gathering without having to strike their base and all their forces head on, c) It's simple, easy to communicate, understand and d) Resource quantity is clear and plain to see as part of the terrain itself.

That said, I don't think I'm going to implement any of these types of resource gathering. The reason is that in all cases it tends to fall into the background noise of the game. It's a chore to have to keep gathering resources to keep your production up with the other guy. If you don't then your blob of units won't be as large as the other guys blob of units and the game is over.

**** and continue here. ***

I'm going to try make resource gathering an important event or series of events. Resources will be collected in large bundles rather than a consistent stream. Those bundles will literally be an item or a small number of items on the map that must be retrieved (think capture the flag). Once retrieved they need to be converted into usable materials. This process can take time to prevent one really successful resource gathering event from instantly ending the game.

My goal is to make an RTS that plays out more like the board game GO (if you have not played it then you must learn it's super awesome). In Go a player can lose a battle early but can easily recover to even footing if they win the next. The game is decided by the player that slowly accumulates the most small victories in what appear as isolated battles but end up uniting into one epic conflict!

Hope to many didn't 'to long did not read' this post. Please post your thoughts and suggestions around the topic. I think this is one of the most overlooked parts of RTS games and is probably the area we can push the genre the furthest.

 

Im going to combine harvesting with node management.

I initally didnt have nodes, but I found it resulted in all units just steaming towards the enemy base, and once the base had a few enemy close, you couldnt build anything without it being destroyed as soon as it rolled out the door. Sure this is going to happen anyway, but I didnt like that it could happen so easily.

So I put in nodes. Currently there are around 5 in the map. The idea will be that you have to hold all nodes before you can advance on the enemy base, other wise the base will have operation defence systems that will cut down anything that comes close. After reading about your resource ideas, it gave me another idea. Resources that are generated by the nodes. In my case, capturing a node, will stop generating money for you. Why? so that the team that is possibly on the back foot, will have extra resources sent there way to assist them in building vehicles to defend or capture nodes. The nodes wont generate a lot of resources, but they will help.

I figure this will offer some reprieve for the team being dominated. Chances are, if they cant get out of the base, neither will the harvester, so at least they will have something.

 

That sounds like it's open to a turtle exploit. I sit in my base and build units. I let my enemy capture all the points sending me more resources to build units. I when I have a large enough blob, I charge and win the game!

 

Except the plan is 5 v 5 multiplayer, with a maximum of 4-5 units each player (will be determined based on how much performance I can squeeze out), with limit restrictions on each vehicle (per person)

So at most you will have 25 vehicles per team.

Also, turtling would not be a winning strategy, because you need to go and cap 5 nodes before you can attack the enemy base.

A large force will have to be divided to ensure a fast jeep/gunship doesnt race around the back and retake a node.

 

Sweet, looks like you have your bases covered (pun intended).

 

hahaha call!

yeah hoping so. one day, when im getting closer to having a completed game, this may all go out the window after some play testing

 

Haha, yeah. That's the kicker. How do you develop something as complicated as an RTS when you're the only developer.

It's going to be difficult to design and build for a possibly heavily multiplayer game when you have no one to co develop and test with. I'll figure it out I'm sure, even if I have to create SkyNet along the way.

 

two words, supreme commander, try that game its amazing, imo based on graphics and just pure awesomeness, its one of the best out there. just watch a 1000 units vs 1000 units, its extremely epic, especially with the strategic zoom that really lets you get into the battle.

 

Yep, SupCom is one of my favourites to date. I feel the first one was superior to the sequel in many ways, but some of the simplifications were welcomed. I never played the forged alliance expansion pack though, maybe I should give that a go as I hear it's the best in the series.

SupCom 1 and 2's interface, with its strategic zoom camera is definitely the best of all RTS games and I will be stealing that.

 

Real Time Strategy games are not complicated to create. Of course that depends on what you define as real time strategy.

RTS = Sins of a Solar Empire
TBS = Galactic Civilizations 2

If you want to start out small, try creating a Unity3D game of Travian style first. Simple construction using primitives as you have done so far and go from there.

 

I was meaning less in terms of implementation and more in terms of testing/balancing and exploit discovering. I'm only one guy, and play RTS games in a specific way... I guess the only real way to do it is open alphas/betas.

 

True, but for that just look at what is out there and model your design and implementation after that.

Travian would be a good model to follow. http://www.travian.com

 

If you want, I can help with my latest package, Stagebuild. http://www.stagebuild.com

http://forum.unity3d.com/threads/178258-StageBuild-A-guide-to-level-design-in-Unity

 

Looks useful Curzedd. I don't know how i'm going to do terrain yet. My goal is to do something like the Planetary Annihilation guys where you fight on whole planet spheres but I might make that leap after I get a lot of other things working. For now I think i'm just going to use a height map or something basic like that, just to get things rolling.

Progress: I spent a few hours on the camera the other day and it's working nicely now, panning around with right click drag and I have a nice Supreme Commander style strategic zoom working also. I don't have the units lodding out to icons or anything like that yet, that's a job for way later when I know what units I'll have in the game. Right now I'm just focussing on the basics. When I have the cameras fully working I might post the code. it's a bit intertwined in my input system but that can be easily removed. Stay tuned.

 

Hi, just discovered this post. This is interesting, I'm looking for something like this, a good RTS, but I'm more focused on the military/strategy aspect. Any progress? I posted some of my thoughts here . Interested in your point of view and ideas. I'm not a programmer so I'm trying to get code from everywhere and figured it out how to do things.

Regards.

 

Hey. This has taken a back seat as you may have guessed. I saw what the planetary annihilation guys were doing and they were accomplishing many of the ideas I wanted to try. I though I might be able to wait and mod that instead of building all from scratch.

 

@Antony Blackett
I'm working on an RTS myself and started out with the boring stuff (physics, networking, etc.). If you're interested, check it out here: http://forum.unity3d.com/threads/dphysics-and-lockstep-framework-demonstration.318829/.

Maybe we can work together to make something great.