Warhammer Dark Omen Forum

Check out Wilbefast's early thoughts on developing a Dark Omen/Myth/Total War game engine ->

http://www.ogre3d.org/forums/viewtopic.php?f=5&t=71409 (http://www.ogre3d.org/forums/viewtopic.php?f=5&t=71409)

http://www.youtube.com/watch?v=mUqik2ZgGd0 (http://www.youtube.com/watch?v=mUqik2ZgGd0)


I have contacted him and asked him to post his ideas/thoughts for us all to enjoy and assist.

"Thanks for the message: I'm certainly looking to make something that will be inspired by Dark Omen, but I don't plan to create a direct clone/port I'm afraid. I'm also experimenting with various pieces of Open-Source technology before I pick which one I'll end up using: I'll be making a little demo similar to the one I did for Ogre for jMonkey Engine and perhaps also Panda3D.

Since you're a fan of DO, would you mind taking a look at the brief analysis I wrote of its gameplay (what applies to Myth applies to DO)?

http://wilbefast.com/2011/06/04/dissecting-real-time-strategy/ (http://wilbefast.com/2011/06/04/dissecting-real-time-strategy/)

It's obviously a simplistic view, but I do find that that game mostly is about luring enemies into a cross-fire. I'd like my gameplay to have situations where the enemy outguns you rather than the other way around to shake things up".

:)

Nice quick summary thou a few points.

Rock, Paper, Scissors is a really boring game, if you’re going to do anything interesting there will usually be a range of counters, combined arms, and/or incomparables.

Saying the game (myth 2, and dark omen) is mostly about luring enemies into a cross-fire is a little disingenuous while yes it is usual central tenant to the players strategy, I can’t think of a single level that doesn’t force you out of it.  Dark Omen in particular while “glass cannon” is paramount in order to reduce casualties most kills will not be achieved by cannon/mortar crews in most levels.  

The outnumber/wave structure in both games comes down to the level designer sitting down  and creating a puzzle for the player to figure out.  This has 2 effects, one good a.i. is really hard to produce, with this method you can almost skip it, two, It makes the player feel powerful.  Slaughtering wave after wave of troops does that, thou it needs to be done right not to feel dumb/cheap.

Also of interest is that the multiplayer component of these games is completely different, okay they use the same world rules but they are both totally different game experiences.  

Thinking about it you could futher split RTT games into the puzzle & wave types (myth, dark omen, nexus), and the I dunno “world simulation??” or games with actual a.i. types like total war, close combat, Gettysburg, etc.

That's actually what I mean by "Rock, Paper, Scissors": in RTS/RTT games a lot of gameplay hinges on correctly countering the opponent.


Technical it's more softening the enemy up with ranged weapons then finishing them off with combat troops. The fact remains that since the enemy never has more or better missile-troops there's never a reason to charge at them. Granted this might not be able fun as raining death upon them and watching them burn. Indeed, Dark Omen added the "to the death" mini-game for the simple reason that watching combat troops fight isn't very interesting: you need something to do while you wait.

Reverse engineering the AI script is a big part of the single-player campaign. Specific locations, if entered by your troops, will trigger the next wave of enemies. A single fast unit or wizard can be used to "set off" those enemies waiting in reserve:
http://www.youtube.com/watch?v=ERCiuPA8Rfc#t=3m45s (http://www.youtube.com/watch?v=ERCiuPA8Rfc#t=3m45s)


You've hit the nail on the head there. Game AI tends to be pretty bad even today! You look at something like "Doom 3": when you get used to it the gameplay boils down to recognising when you've stepped on a trigger, stopping to kill any enemies that have suddenly appeared and then moving on, the idea being not to step on multiple triggers at once.
AI is my speciality at university, and while Game AI and "real" AI have little in common, I'm hoping to build something a little bit more intelligent.


I'd never thought about it that way, but it's true that the levels are rather like puzzles to be solved. Total War battles are never quite the same twice, though the AI isn't exactly a genius:
https://www.youtube.com/watch?v=fK-lBmWWg2k (https://www.youtube.com/watch?v=fK-lBmWWg2k)

Truth be told, I'm probably going to be aiming more for this sort of gameplay, as I want the multiplayer and singleplayer to work in similar ways. I do want to go for some sort of fantasy setting though, with monsters and magic.