Kenshi - A free roaming squad based RPG (Trailer!)

[Captain_Deathbeard]

You are right that it will take a lot of careful thought to balance it well. As for limping to town being boring, there will be a fast-travel option similar to in fallout, plus healthy teammates can carry him. Vehicles would have been good too, but again, the tech level.

And your big question is so big, it already has an article:
http://www.lofigames.com/devblog/?p=22
The basic idea is that the player should never feel the need to re-load. There will be no permanent, irreversible wounds (thats why I'm still undecided against severing limbs) and everything that seems bad also has a benefit. Characters actually get slightly more XP from losing fights than by winning (learn from your mistakes). Characters can still go on fighting for a long time while wounded.
And the main point of this whole medical system is that it blurs the line between life and death. If a character goes down, you can still save them, and the most common cause of re-loading games is characters dying.

Jagged Alliance 2 had a similar system of slow-recovery, and that was an awesome game. Like JA2 it won't be a standard game where you just charge in and kill everything without thinking about it, and it also won't be the kind where you just wander the map non-stop slaughtering a nice steady stream of enemies. Battles are fewer but bigger things, so hopefully I can balance it well enough so that it doesn't become an endless repeated cycle of "fight, walk all the way back to town, recover, walk all the way back, fight...) and during your recovery there will be lots of other things to do (professions, team management etc).

So yeah, takes a lot of thought. Please feel free to challenge my ideas and help me refine them.

[Captain_Deathbeard]

Brainshack: I had to mod Ogre's animation system slightly. So I have a bunch of simple animation poses (hand to head, hand to stomach, hand to arm etc) that just animate the arm and hunch the posture, and i designate the arm bones as 'overrides' so they don't blend with the walk animation etc.

[CABAListic]

Well, there is no doubt that if a character dies I would reload the game (unless, maybe, we're talking about D&D games, the fight was really hard and I have a scroll of resurrection handy, but that's besides the point). But the main reason for me to reload is if something happens that I believe will hamper my progress in the game. A serious injury that will make a character useless until I found a doctor and paid good money for the recovery would be a likely candidate for a reload. For that matter, I *do* reload in JA2 if any of my soldiers receives a serious injury
Problem is that such things are ultimately luck-based. Of course you can go a long way to prevent them in the beginning (stealth kills in JA2 etc.), but sooner or later I'll get unlucky. Now, if I can retry my luck with little effort (i. e. savegame near to the event), I probably will. That's why, in my opinion, things like injuries, character deaths etc. work a lot better if they are story-driven, i. e. a matter of choice, not chance.

But that's just me. I'm certainly not representative of all gamers out there, and I'm certainly not a hardcore gamer. I just wanted to warn you that if you spend so much time on a detailled medical system, make sure it's actually experienced (in a positive way) and not just reloaded away

[Captain_Deathbeard]

I don't like it when things are hard-coded into a story, I like my games to be purely games, rather than falling back on hollywood storytelling and non-interactive elements, I feel like its cheating.

Anyway, you can treat injuries yourself if you have good medics, so they just need recovery time, and they can still fight when wounded, especially a strong character.

And yeah, actually I reload in JA2 when guys are seriously injured too. But ultimately you rarely actually expect injury, so its hard to make them predictable without making it feel sterile and artificial.

Thanks for your advice, I have much thinking to do.

[SaiCo]

So many JA2 fans ! truly great game.
I think there are at least 2 situations possible: 1. Team member is severly injured or killed - that usualy make you reload game in some point that would enable differet tactical approach. 2.not so severe injuries (fleshwounds , hits below 50% of HP) - you can roll with it.
Point is - I feel that in tactical game first scenario is result of tactical flaw (like, using last actionpoints to scale on that roof .. suprise you climbed right in front of sniper). On the other hand even fleshwounds can be serious - they can be infected, gangrene can kick in, small wound on outside can result in internal bleeding. Thats especialy important in a world which is medicaly undeveloped. (also - thats a point for using any armour too - even if it cannot withstand lethal blows, it can save from small wound that can go bad) ...

Wonder if you plan to add some armour system or say, bows ?

[Captain_Deathbeard]

Gangrene and stuff brings the effects of the medical system outside of combat, it no longer effects things and would likely just be a chore. I want it to effect the feel of the combat, rather than just adding to the micro-management.

How about this, medics can jury-rig injuries. That is temporarily patch-up/splint it with some painkiller so the character can go on fighting at near-full ability (and pay for it later, when back home and its not a problem other than looking cool). That way it affects the current battle until you get a break or a chance to treat the character.

I don't so much like the idea of the player using semi-suicide missions to scout out the enemy and re-load. Even though it is a standard practice, I would like to avoid it as much as possible and create more of a game where mistakes can be undone in fun ways.


And yes there will be armour that has different body part coverages + strengths, and bows. I love bows. I'm gonna make the arrows stick in people like in oblivion.

[sisyphus]

Is this going to be open-source so others can learn from your code?

[lonewolff]

Sounds like a promising project

Any ETA on a release?

[Captain_Deathbeard]

No, not open-source, but I would be willing to write articles on certain elements if there was demand. The custom animation functions I used to do the injuries I have submitted to ogre as a patch, so they will be available for all soon
http://sourceforge.net/tracker/index.ph ... tid=302997

As for release I suppose 2-3 years, its still a way off, and I don't even have any real art assets yet.

[Jabberwocky]

How has your experience using AdVantage Terrain been so far? I think this is the only Ogre game I've seen screenshots of which use it.

Are you using the paging functionality? How big are your landscapes?

[Captain_Deathbeard]

I think its great. Current landscape is 10x10km (8000x pixel heightmap), but I intend to use 20x20km eventually. I have not yet started on anything like paging or loading/unloading world sectors. Performance is great. I still need to work on a better collision mesher, and better shader.

I intend to generate all the splatting layers procedurally (cus dont need artists for it), and page them in with advantage.

Advantage is the only decent large-terrain plugin for me that I could find. I tried PLSM2 but it was too slow and small, and all the others are too small too

[buckED]

hey captain deathbeard.

First of all: you are making great progress with your game (from what I have seen and read and what I can judge so far)
I like the medical system idea. It came to my mind the other day aswell.
Why hasn't anything like that ever been done in a commercial game yet?
(I know Jagged Alliance 2 has something similar, but without a location based system) Is it so resource intensive or is it just the developers fear not to sell enough copies if the game gets to realistic/difficult/complicated?
As for my part: it can not get realistic enough

But apart from that. When can we expect first videos showing all this nice stuff in moved images? Screens are cool and all, but seeing it in action is a completely different thing.
You keep stating that your art is not that great. In my opinion it's not that bad either. I especially like the shadows.

So everything you need to make a nice video is given. Why not just go ahead and do it (I do not intend to put any pressure on you. If you can't spare any time you should keep working on the game)

cheers,

Ed.

[Praetor]

Because a system like this is "niche." I claim to want realism but even I'm a bit iffy on this idea. But you know, that's exactly what an indie developer like this is supposed to do. Make a game that conventional wisdom says would be too realistic, or too hard, or too complicated. I'd love to see where this goes.

[Captain_Deathbeard]

Well I can always see how things turn out and make adjustments. It still has a good impact on your enemies, seeing the effects of your attacks. And if it turns out to be too much micro-management for the player, the AI is powerful enough that I can make it help out. And if things still just don't work well, I can remove or reduce the stat effects and then its just an awesome cosmetic effect.

BuckED: I think the reason commercial developers don't try things like this is, as you can see from peoples comments, its just too risky to try things that have not been done before. Especially as casual gamers are the biggest part of the market.

I will upload a video in the next few days.

[akiraoni]

I have an idea on how to approach the save/load situation. Regular auto save that automatically loads default file upon entry into the game (then has to be manually switched if desired, somewhat like the GTA3/VC/SA loader). Instead of punishing the player by taking away their saves, you're just creating a more real-time environment.

I did want to ask how you were going to do events. Are all of your events going to be scripted (a pain with a game as open as yours), or is the gameplay going to be open-ended with events happening randomly as factions vie for power?

I like the concept of the game. I was looking for a progress feed on your dev blog, yet one does not seem to be available. I did notice that you were looking for a concept artist. If you would pm me with an appropriate email address, I would like to discuss what that would entail and how I might be of assistance.

[sisyphus]

@Captain_Deathbeard: You said you wouldn't mind writing an article about how you did certain parts of the system, in particular, I would be interested in how you implemented the scarring system. i.e. how you went about making realistic slashes/scars appear where you struck the enemy...

If you don't mind of course, it would be really useful IMHO...

[Captain_Deathbeard]

I finally got round to making a video:

http://www.youtube.com/watch?v=wBW9zhkGdhs

sisyphus: I would be happy to when I can finally get round to it. And remember how I did it. As a summary, I think I pre-partitioned all the verts by bone association and facing, and then I choose the nearest bone, then look at its verts and choose the closest to imact. Then there was some way of looking at the mesh data to get the UV coord, then you copy a pre-made wound image into a custom texture layer on the material.

akiraoni: I dont have scripted events (except dialog). The plan is that events will be dynamic and basically handled by emergence.

[yapdakilla]

Hi,

Very nice vid. Especially the part where the injured character just fights with one arm. Will the characters know which specific limb to attack as well? As in the one armed character will know how to disable her opponent?

[buckED]

I finally got round to making a video:

You made my day captain deathbeard. Absolutely awesome. As I am defently the biggest Kenshi fan under the sun what I say must be true

Neat. Keep it going.

[Captain_Deathbeard]

I'm glad its made a good impression. I've been looking at it so much I just start thinking its rubbish, I'm interested in some critiques (aside from the many graphical glitches). Whats your first impression as gamers?

I'm thinking bodypart targeting should be random for now, otherwise everyone will just target the same vital area all the time. The exception being if you specifically order your characters to disable and not kill.

[Jabberwocky]

The thing which stands out to me is the blocking. It looks like you're on your way to implementing a fun, active, and realistic combat style. In the video, it also looks like you can block at different angles.

[akiraoni]

Why don't you try dictating area injury by the characters actions? For example, if the character blocks they would less likely find their head or torso injured (very vital points) and more likely find their limbs injured. However eventually the limbs become useless, and they begin taking impacts to those vital points. This is very similar to a concept I was working before for a mud, so I thought it might help.

[SaiCo]

i would just add $0.02 : when I played things resembleng WMA, my tutors emphasised that block is final solution that really gives you only time before next fatal strike from enemy. 1st- do not make enemy attack first/seize initiative in any way( finish him before he would even have weapon in hand, preferably), if you fail that - 2nd- avoidance , only then 3rd: parrying, always with flat of the blade, never staticaly- rather set aside than "hollywood static block", wind in, stab in face still blocking the opponent's blade (were using mid-XVc bastards) . Though its most tricky - better stab him while he's not looking There are also "Meisterhauwen" in western tradition that are used to seize initiative and end fight in one strike - (I know youre aiming in more eastern style, but i guess there are smilar techniques in kendo, or so i think). Also things easily forgotten are footwork, movement, and range.
Still never saw a game that even got close to "real" fight mechanics (just roll that d20 ) . So much speakin from my rather smallish experience

Edit: very nice vid

[Captain_Deathbeard]

I based the system on my own experience with kendo / bokken fighting (wooden katanas). So its about blocking or avoiding the attacks until someone slips up and gets cut. Not how a real swordfight would go because like you said seizing initiative and more complicated things go on, and real swords can't take much blade-to-blade impacts.
But it looks good and works nicely as a game mechanic because its quite simple, which is important.
Much of it actually, especially footwork, is left up to the animations. It just sort of turned out that way, the characters move about by themselves so I kept it.

To be honest though, I have been working on mainly the combat system for most of this year, so I am now trying to simplify and wrap things up as much as I can so I can get on with the rest.

[buckED]

As for my opinion on cambat:

Combat has not yet been realized in a beleivable way in any game.
There are different ways of simulating combat in computergames and of course they can only ever be an aproximation to reality. The quesion is how close are they to reality?

From what I've seen you are going with a more action based approach. And even if it were only NPCs fighting for now (at least I think so) it already looked pretty convincing. Especially the blocking. I think blocking is the most important factor whatsoever. In reality, who is able to slaughter his way through hundrets of opponents, take 20 blows to the head and still stand on his feet slaughtering on and on? It should rather be like most of the time you are blocking, keeping the distance between yourself and your opponent in order to strike as soon as he opens his defence in a way which gives you an advantage. So defeating him should not consist of hittimg him until his healthbar is empty but merely in getting through his defence. A real hit should either have great impact on how he is able to attack you or defend himself (i.e. you can't block as effectivly with a wounded arm, can't move as quickly with a cut leg. Your damage system holds a lot of potential for such an approach), or kill him altogether.
The same must of course be true for the player being attacked, which would make a good defense vital for surviving. One would be forced to plan his movements. Trying o avoid to give your enemy to much space for a fatal blow and at the same time putting preasure on him which will force him to let his guard down.

A location based hit-system is perfectly suited for such an approach and it might remove the need for a healthbar completely.
Think about it. Defend yourself, cripple your opponents defense and wait for the one perfect chance to stab your weapon into his vital organs.

I guess I have daydreamed enough? It is just an idea, something I would love to see in a computergame one day. Maybe we are not yet at a technical level to achieve it, maybe it would detract to many casual players due to it's complexity and it would of course be darn hard to balance. But I still think that representing everything with numbers(e.g healthbar) when the computer is perfectly suited to hide these numbers takes quite a lot from a game's atmospere and believability.