dim3d or upcoming Unity for this idea?

DavidJJ
Unregistered
 
Post: #1
I'm imagining a simple two player game, with two marbles. The goal is to manouver your marble using the arrow keys or the WASD keys and knock your opponents marble of the playing surface (a flat square floating in space). Think "marbles meets sumo wrestling).

I'll need to take movement, velocity, mass and collisions into account.

Would dim3d handle this? Should I wait for Unity (demo, to see if I can grasp it)? I've got 2D versions comped up in both TNT Basic and Runtime Revolution but I'd like to add a bit of a relastic 3D experience here.

Thoughts? Thanks.

EDIT- I've actually answered my own question. I finally got around to firing up version 2.3.4 of Blender. OK, certainly an odd interface and somewhat confusing key commands, but within two hours I'd created a two level, scrolling marble game with all of the functionality I'd needed. Including pannable camera and full 3D.

I'm hoping Unity provides this much power and a more Mac-like interface. The big thing for me within Blender are the logic building blocks ... the easiest way I've ever seem to program behaviour for objects.
Quote this message in a reply
Member
Posts: 70
Joined: 2002.07
Post: #2
DavidJJ Wrote:I'm imagining a simple two player game, with two marbles. The goal is to manouver your marble using the arrow keys or the WASD keys and knock your opponents marble of the playing surface (a flat square floating in space). Think "marbles meets sumo wrestling).

I'll need to take movement, velocity, mass and collisions into account.

Would dim3d handle this? Should I wait for Unity (demo, to see if I can grasp it)? I've got 2D versions comped up in both TNT Basic and Runtime Revolution but I'd like to add a bit of a relastic 3D experience here.

Thoughts? Thanks.

I'd be interested in you giving this a try in dim3 Smile I want as many different types of games as I can, and, if you've followed this message board, I'm very willing to help out if people are making non-standard 3D games. It's a good test for the engine.

[>] Brian
Quote this message in a reply
Member
Posts: 47
Joined: 2004.07
Post: #3
DavidJJ Wrote:I'm hoping Unity provides this much power and a more Mac-like interface. The big thing for me within Blender are the logic building blocks ... the easiest way I've ever seem to program behaviour for objects.

Unity certainly provides the power. It also has a very mac-based interface (lots of drag and drop). It does NOT have logic bricks - cool as they are they broke down completely when I tried to do a simple car AI with them.

We are more focused on providing a power tool, and would rather spend the time doing cool tech than a logic system that doesn't scale to serious games.

Nicholas Francis
http://www.otee.dk
Quote this message in a reply
Member
Posts: 70
Joined: 2002.07
Post: #4
NicholasFrancis Wrote:Unity certainly provides the power. It also has a very mac-based interface (lots of drag and drop). It does NOT have logic bricks - cool as they are they broke down completely when I tried to do a simple car AI with them.

We are more focused on providing a power tool, and would rather spend the time doing cool tech than a logic system that doesn't scale to serious games.

I think it can be done, and dim3 is proof of that. You can do something simple and cliche in dim3 without any trouble, or you can dig a little deeper and use the scripting to make something unique.

I think the solution is to give them a logic system that is a base -- for instance, in dim3, it's the 1st peson shooter -- but make everything changable. In that way, you can alter the base in the ways you want without getting deeply involved in every part of it.

For instance, you can just change the camera and now it's top down, or 3rd person. You can change the physics, so it's a space shooter, etc. Make everything touchable; don't hard-code. In this way, the base is really "default" values.

Also, I tried to make everything as separated as possible from one another, so an object script can be interchanged with another script in another game; therefore, you can use scripts as building blocks. A project goal is that you can grab a script from one project and drop it into another project and have it work.

That said, dim3 is out now and it grows by what users request, so, if the marble thing is still in mind, take a shot because I'd love to hear where I need to make changes to help create such a game.

[>] Brian
Quote this message in a reply
Member
Posts: 47
Joined: 2004.07
Post: #5
I can certainly understand your philosphy - it is a very sensible one... However, our goal has never been for non-programmers to create games. It has been to fascilate the creation of publishable games without requiring 3 years of engine work.
That being said, we will ship at first some scripts with Unity, then more as we get them made.

My personal opinion is that making 'publishable' quality games requires a team (or an individual) with a large skillset - hence I don't see any reason why GFX guys and programmers can't work together.

The other thing to realize is that we're working with finite resources: It is not 'would we like to have a visual programming system?'. It is more 'Would we rather add a visual programming language or features that will allow dedicated teams to create greater stuff?' I'd go for the second one any day. Maybe one day we will have (if we can see that it makes business sense), but for now that is a definate no-go for us.

Nicholas Francis
http://www.otee.dk
Quote this message in a reply
Member
Posts: 70
Joined: 2002.07
Post: #6
This is something I never wanted to do because I see much infighting between programs and content creators; if you have a movable engine, then core changes completely destroy previous content. This is one thing that has held back many a project.

On the other hand, if you abstract this by removing core changes, then you don't upset content creation.

This is one of the main driving goals behind dim3; to find a way to get everybody involved, regardless of skill set, without effecting the work of others, and speed up content creation.

Hopefully, when dim3's map editor gets good enough, one person CAN make a good game. There are certainly people with the skill set, they just need a stable platform with which to build. I want to bring a end to the multi-year million dollar games, which is my other goal. Smile

[>] Brian
Quote this message in a reply
Member
Posts: 47
Joined: 2004.07
Post: #7
I'm not quite sure exactly what it is you never wanted to do, but I DO have some feedback.

Core changes happen a lot less often than most people think. We have ancient demos that still run fine...

Currently I'm playing with hierachical fuzzy state machines for AI. I would certainly not want any visual newbie editing to get in the way of my playing around. I would also not want to spend a lot of time implementing an FFSM editor - far too specialized to warrant the cost. Our approach has always been to expose the raw interface first, then find out what we _actually_ used & how we used it. Then we build easy-to-setp functionality on top of that.

As for ending multi-million-dollar game projects, have you ever seen budgets for those? The majority of those multi-million dollar projects is marketing. Then you have content creation. True, we can alliviate it, but the majority of time spent is done by graphic artists sitting in Max spitting out core assets. Not a lot we can do there. That doesn't mean we should just give up... There is also a market opening for smaller titles (e.g. GooBall ;-)), but I think your dream of ending large scale productions is a long way away. Tools can only do so much.

Something else: Why are you spending time making a good map editor? AFAICS, we cannot compete with the teams who do dedicated 3D modelling programs... We have taken the decision of saying that we only do object placement. No building of geometry. If you want a corridor, you have to build it in one of the (many, many, many) apps that specialize in modelling. You can then place said corridor in Unity, or you can build your complete scene in one of those programs.

On a side, I'm glad we're targetting different groups. You want to get the single persons & newbies, we aim for getting small & medium sized teams who want the power to experiment.

Nicholas Francis
http://www.otee.dk
Quote this message in a reply
Member
Posts: 70
Joined: 2002.07
Post: #8
NicholasFrancis Wrote:Core changes happen a lot less often than most people think. We have ancient demos that still run fine...

One of the big complaints, and reasons for projects being late, that you hear the most is core changes during content creation. Right now, you have your engine that you are just working with yourself; when people attempt to extend it themselves, that's when the fun begins.

This is something that has plagued the industry a lot.

NicholasFrancis Wrote:Currently I'm playing with hierachical fuzzy state machines for AI. I would certainly not want any visual newbie editing to get in the way of my playing around. I would also not want to spend a lot of time implementing an FFSM editor - far too specialized to warrant the cost. Our approach has always been to expose the raw interface first, then find out what we _actually_ used & how we used it. Then we build easy-to-setp functionality on top of that.

Note from my experience that you DO NOT know how people will use it, only how you will use it Smile Get ready to be surprised. Smile

NicholasFrancis Wrote:As for ending multi-million-dollar game projects, have you ever seen budgets for those? The majority of those multi-million dollar projects is marketing. Then you have content creation. True, we can alliviate it, but the majority of time spent is done by graphic artists sitting in Max spitting out core assets. Not a lot we can do there. That doesn't mean we should just give up... There is also a market opening for smaller titles (e.g. GooBall ;-)), but I think your dream of ending large scale productions is a long way away. Tools can only do so much.

It depends on the tools; more later.

NicholasFrancis Wrote:Something else: Why are you spending time making a good map editor? AFAICS, we cannot compete with the teams who do dedicated 3D modelling programs... We have taken the decision of saying that we only do object placement. No building of geometry. If you want a corridor, you have to build it in one of the (many, many, many) apps that specialize in modelling. You can then place said corridor in Unity, or you can build your complete scene in one of those programs.

Because you'll never be able to optimize your engine that way. Engines optimize by geometry reduction, and you need specialized things that regular 3D creators don't have. You can create trees and other optimizations above that, but to get to the true mass-reduction (like portals), you need to specialized tools. Now, in the future, I will hopefully allow importing from said sources, but specialized tools will ALWAYS beat out generic tools for speed ... and, in the future, content creation.

You're gonna pay a price here. You have to learn a new tool but gain high-end optimizations, or you use generic tools and have things slower.

You should create some enormous map, like the old doom or marathon sized maps, and see how well the engine holds up. Now is the time to test this, before you first release.

NicholasFrancis Wrote:On a side, I'm glad we're targetting different groups. You want to get the single persons & newbies, we aim for getting small & medium sized teams who want the power to experiment.

Not really Smile That's another reason I'll open-source the engine, to give small, medium, and large sized groups a shot at it. My first love is the mod makers; smaller developers, single people, newbies, but I'm shooting for the high end, too.

[>] Brian
Quote this message in a reply
Apprentice
Posts: 18
Joined: 2005.06
Post: #9
I see that the engineers are butting heads a little. Sometimes this is the best way to decide on a tool, having the designers discuss it. Obviously, they like their creation and won't say bad things about ( neither would I) but this does show possible turnofs for game creators shopping around. Both of these look very good right now, although the price point of dim 3 ($0.00) is nice Rasp
Quote this message in a reply
Post Reply