Community Resources

Hi,

as I am just a coder with limited artistic talent, I was thinking about something like a page with community created (free) resources (Scripts, Artwork, Prefabs, Textures etc.), maybe even a marketplace with stuff like "will code for art"

I don't know but how do you think about that or something similar? (my game production experience is nil)

thx
Alex

p.s. thanks to the otee guys... incredible what you have done!

 

I wonder how to get it up and running... maybe we can start with a forum in here, and then see where it takes us.

These forums support file uploads, so at least technically is possible, if not great. Whoever knows of a better solution, speak up!

 

I suggested something like this a few weeks back too. I think it would be great. The thing is, I'm guessing that we would all like to see the OTEE guys focus their efforts on Unity. At least until Windows export, toon shading, the web-plugin, etc. are done. :wink:

So, I'm guessing that stuff like this is going to have to be more self-sustaining and community-run. There are so many great user-run resource sites for Flash, .Net, Director, etc-developers, so we have plenty of successful models to mooch off of.

While I can't set it up myself, I can promise to support such a site once someone gets it up and running. I think that the Unity community is poised to become the program's best, killer feature - especially when you look at how many of us there already are and how much we seem to believe in this awesome tool already.

So, anyone out there ready to set up the first Unity community website?

 

I (also) think, it would be best to first start out here in the forum until we have some critical mass and then go to a self-sustaining, community-run site - just to make sure you guys have enough time to improve Unity

I also can't get such a thing up an running all by my self, but like marty I would promise to support it, so maybe we could pool ideas, volunteers etc. for this site in this thread?

Stuff like what content should be found there?
Voting on contents quality?
License of the content (I guess important for Indies trying to sell finished games including some of this content)
Is there any interest in a "market place" like feature?
Constant "new features for Unity" poll?

So that by the time, we are "critical" the concept of this site stands (and we have enough volunteers).

Right now having something like "perpetual cool [scripts|prefabs|models|etc]" threads should be enough... I guess we will know when this becomes too complex.

You know I am really glad... Unity could mean really innovative, original "games", not just "everybody has to like it Blockbusters".

 

As a senior web designer I've got the skills (including some buddies who are database gurus) and motivation to get a Unity-related community site for files etc up and running.

It's the time I lack. I have a family and fulltime job and am still wanting to create a further tutorial for Unity.

But the idea is sound and a place with Unity assets (be they scripts, materials, objects, models etc) would be a good thing, as long as it doesn't take the community away from here.

I will contribute my skills when the time is right, though.

 

I agree with all of the above, the community will certainly help put unity on the map. I'm a partner in a small graphic design shop that also does web design and programming, unfortunately like everyone else here my time is already spread thin and I want to actually use Unity too! I'll be happy to contribute any time I have available as well though help make it happen.

 

Thats a good idea. I don't actualy have Unity and wouldn't be able to do anything with it any way because I'm a graphic artist not a programer. So if you need any graphic desing for the page I'm happy to help I have designed the sites below:
http://www.agesofilathid.com
http://nakurij.bluegillweb.com
http://games.bluegillweb.com

so yah I'm not that great but I'm willing to help, and support original mac game development!! Jeff

 

Very cool. I'm definatly going to get an inde license to mess around with. Just a question though douse the license only alow me one game built in it? Because if your learning it your obviously going to be making a lot of little things to learn it first. Well as I said before I'm all for helping with the cuminity site. Jeff

 

Ok I'll order my coppy soon then just got to dig up the money. And since its a new product and the forums are good and friendly I know I can get help from you guys. Now just need to get on learning Maya like I have been trying to do for a while. Jeff

 

You know, if some of us came up will all the scripts and stuff needed to creat RPG's then I could bring the entire ColdStone Comunity over to your engine probably. I know most of the comunity and we have given up hope on an update to get it working in OS X corectly so its an idea. Jeff

 

Ok sounds like a good idea. I have already talked with some of the ColdStone comunity (Coldstone = CS) and some of the CS guys are interested in looking at this engine. A few are not interested in going with a 3D option. So I'll see what I can put together some time in the not to distant future. Jeff

 

I am also one of these rpg freaks and currently thinking up a "system" on how to implement a rpg with unity (like what kind of prefabs, scripts, models etc. would be needed).

In some areas I am into some kind of pre-pseudo-code stage (exhaustion, exposure, injury recovery, disease) - some warning having studied sports science and bioinformatics i am into hardcore realism and do not care about play balance - more like a "life simulation" than D&D style roleplaying.

I think on monday i will have my notes compiled and will post them in the forum (in the hope of critics and tips on how to implement some of that stuff with unity).

Alex

 

If you have any interface requests for Unity, don't forget to post those as well (e.g. placing tile-like assets in adventure game would be alot easier if we had snapping ;-)

Not promising anything, other than they might make their way into my 4-page long TODO-sticky.

 

Yes snapping would be great, but I just want to wait a few days for some feedback for my stuff in the RPG Project Thread before making requests. Right now it is pretty hard for me to set priorities, having no experience in game design. But snapping will be for sure on my list. [/url]

 

Just take your time... I'm not running out of tasks anytime soon...

 

CloneHelix that makes two of us not experienced in game design, but I think this is the best place to learn! I have contacted my two best budies (I'm out of country) and we have decided to put together a game studio. Our first project is going to be an RPG so all this stuff will be great. jeff

 

Is anything happening on this front? An Art support for Code support (and vise versa) exchange board is a great idea. Perhaps it could be section on this board to start? Many game dev boards seem to have this kind of thing...

 

Unless it's on one of those roller-type pianos, right?

 

For now I just created a new forum for teaming up or looking for help.

http://otee.dk/forum/viewforum.php?f=11

When it grows we might expand on the idea...

d.

 

Hi Scoo, well I am still with it. I started with the IMO toughest nut, the AI system. For that I am still in the research/learning (i.e. reading) phase and maybe a bit overly ambitious with that

Basically it will be similar to "The Sims", i.e. each AI object (= NPC) has a personality which will modify how important certain "needs" are for it "happiness". Needs are fulfilled by using other "gameobjects" (e.g. hungry -> "use" apple, "lonely" -> interact with other AI). The AI object will always try to satisfy its "most pressing" need. The information on how the use of a gameobject will affect the Needs, is stored in the gameobject. Needs are on a -100 (e.g. starving for hunger) to +100 (e.g. stuffed for hunger) scale and will decrease with time and by use/presence of "bad" gameobjects.

The difference is that each AI instance does not have knowledge on how other gameobjects influence its needs (i.e. it can't access a global list and decide upon that which visible object will improve the most lacking Need the most). Instead each AI will have to learn that (not sure yet if the learning is handled by a decision tree or neuronal network, probably the former).

Pathfinding will be part of the AI (A* as it is well documented and somebody else already had success in implementing it with Unity) and "volatile memory" (AI object remembers where it found gameobjects and will wander there if its decision tree tells it, that this object is the best way to satisfy it current need).... oh the schedule of an AI object will also be modeled as Need - meaning that the AI object might not stick to it, if it gets hungry AND conversations will be handled as "need fulfilling gameobjects".

 

I've been looking into this for a game I originally planned to create using Torque. In some ways game bots are state systems; flipping between different modes depending on what happens to them. For example, a group of enemies might have a built in state that makes them move together if they're close enough. If they then detect the player close by they could flip to 'attack' mode. If the players starts injuring them each bot might flip to 'injured' mode and try to head for cover, moving as a group until they get split far enough apart not to detect one another.

It's complicated stuff...

 

Yes, "state machines" are one way to build an AI. The more sophisticated ones (like Black White) seem to combine multiple methods, which "state machines" often being one of them.

The "problem" is that bots relying "exclusively" on them are not able to learn. Under the same circumstances, they will always do the same.

The idea I have (well and a couple of thousand people before me )is like that (simplified):

The Bot checks which need it has is the most "pressing" (e.g. hunger). Having identified the need the Bot checks it surroundings (Question: "Are raycast the best way to do that?") and picks the item it has learned which will satisfy the need most and "use" it (the used item will "decide" what using it means, it is not part of the AI... it simply "uses the apple"... the apple decides what using it will mean (in this case, telling the "user" object to activate the eat animation, and to destroy itself - it's eaten). Then the "apple" will modify the Needs of the AI using it (e.g. reducing it hunger). Each Object is categorized (e.g Apple: round, shiny, food, whatever) and these categories (deciding upon these categories will mean some hard work) are used by the decision tree to learn that "using" a "round, shiny, food" will reduce hunger by 10 points.

In case the AI does not know anything that will satisfy hunger it will try unknown objects until it has found one. In case it does not see any object it knows that will satisfy the need it will go to a location it memorized that contains objects which will satisfy the need.

So one feature, which might be cool, but well let's say a bit "politically difficult" in a real world setting, would be to give each AI "a starting table of objects (and its keywords) and how its use will effect it needs. This starting table could be created by say "General Knowledge your Mom and Pap told you" (Apples are healthy) which could depend upon say social standing of the NPC's family and "Religion" (thou shall not lie). First these tables will prevent totally F***ed up gameworlds, because the creation of a stable dynamic system is hard and secondly (which is IMHO the cool(TM) thing) the tables to not have to be true! So you could have "sad" npc's not being able to satisfy their needs properly because they have the "wrong" personality for their religion.

(The real implementation will be somewhat different... i imagine "social interaction" which a preacher of that faith, will count as a "learning experience" towards the religions values ... which could be wrong, and which does not satisfy any need - Piety being the exception)... so if you as the game designer find out that all npcs are after a while permanently drunk, add some preachers reminding them that alcohol is the root of all evil

So basically that "Needs System" is a state machine (hunger is the most pressing need -> get into "hungry mode" -> "use food" -> "learn" -> evaluate next need...)

 

If is looking at some game specific AI stuff and want a book... I'd recommend O'Reilly's AI For Game Developers. It starts at the beginning and gives some pretty nice ideas for real applied stuff, along with source code to help.

If you are reading this and are not looking for a book, carry on.

-Jon