XNA Creators Club Online
Page 1 of 1 (10 items)
Sort Posts: Previous Next

2D or not 2D, that is the question...

Last post 4/30/2007 1:25 AM by Lynch82. 9 replies.
  • 3/15/2007 6:32 PM

    2D or not 2D, that is the question...

    I am more familiar with 2D than 3D but am keen to finally conquer my fear of 3D.

    The game that I want to cut my teeth on with XNA was originally a 2D top-down stealth game devised by someone I know. A techdemo was done but it never got the audience I think it deserves. That aside, the concept is simple and I think addictive, but not in a trigger happy sense. The general theme is quite dark, hence the need for a flash light effect and rain effects. Depending on difficulty, the enemies are stationary on some levels while patrol paths on others. I can use the original 2D graphics to get this off the ground and can probably use the XNA Tiled Sprites Sample as some base code at least to to do a basic proof of concept, before refactoring it all into classes.

    So after that ramble my question is, should I use the port to XNA to familiarise myself with XNA's way of doing things in 2D first or should I jump straight into the 3D pool and see how I fare? The 3D version would essentially be similar to the 2D version except that you would be able to switch between first, third and top-down cameras and player and enemies would have height and variable terrain to contend with.

     

     Any thoughts welcome.
     

    There are a lot of people who are dead while they are still alive. I want to be alive until the day I die.
    -= Paulo Coelho =-
  • 3/16/2007 2:27 AM In reply to

    Re: 2D or not 2D, that is the question...

    Thats a tough call. How much work do you really want to do. Its good to do something like porting the existing techdemo to 2D XNA, but will you then just be creating the game twice? Once in 2D then again in 3D?  I'm working on a game that shares the same camera perspectives you're talking about: top down, 3rd person and 1st person. Originally it was all conceptualized as a top down only game, but since I was building everything in 3D and the gameplay made sense to allow for the different cameras I'm going to allow it.

     How long do you think it would take you to port the game? If you know C# already and you're familiar with 2D, then I'd say jump right in to 3D. But again its all a matter of your comfort level, the time you want to commit and end goal.

  • 3/16/2007 9:15 AM In reply to

    Re: 2D or not 2D, that is the question...

    I think this is more a question about time and resources. Every time I make anything in 3D I then realize that I now have to make models and the textures to go with them and these both can be very time consuming. I haven't even started talking about animation that can take forever to get right sometimes. If you can get help with all the actual 3D things than go for it or if you have a lot of time to make this then go for it. But if your like me and don't have a large amount of time then try to stay in the 2D world, then you'll actually finish your game.

    McGrimm
  • 3/16/2007 12:57 PM In reply to

    Re: 2D or not 2D, that is the question...

    My last game was in 3D, and the sheer amount of mess involved really grabbed me and pulled me back to 2D.  It wasn't a bad game either--clicky for screens.
    James Silva
    Lead Dishwasher
    Ska Studios
    -
    Ska Blog FTW!
  • 4/20/2007 6:11 AM In reply to

    Re: 2D or not 2D, that is the question...

    Somersault:

    I am more familiar with 2D than 3D but am keen to finally conquer my fear of 3D.

    If you are new to XNA, and are comfortable with 2D development, then I would personally recommend that you stick with what you are comfortable with. It will provide you more time to learn XNA without having to learn, or stack the 3d math, and cameras and some of the more daunting 3d tasks. Get to know the framework, try some tutorials, even work with the Spacewar starter kit that comes with XNA which is a good 3d example, and has a lot of good techniques to be learned.

    Now with that said, if you are keen with the math related to 3d game programming and feel comfortable with that then by all means do it in 3D, there are many resources on the net that will go a long way in helping your learning curve.

    I do still recommend that you tack XNA from a perspective that you are most comfortable with, as I think that you will gain more, and be more interested/motivated to keep learning.


    Are you Inspired?
    --------------------------
    Matthew Randall
    Casual Games Architect
    Exhale Game Studio
  • 4/20/2007 10:40 AM In reply to

    Re: 2D or not 2D, that is the question...

    Somersault:
    I am more familiar with 2D than 3D but am keen to finally conquer my fear of 3D.

    Why? Is it the "Everyone's doing 3D, I should be too" reason? Do you think your game would work better in 3D than 2D? If the former, that's not sufficient reason IMO. If the latter I say go for it. :)

    Jim Perry - Microsoft XNA MVP
    If people spent a minute searching the forums and reading the FAQs before posting I'd be out of a job.
      Got some XNA Game Studio/XNA Framework development info to share with the community? Put it on the XNA Wiki.
        Please mark posts as Answers or Good Feedback when appropriate.
  • 4/20/2007 12:17 PM In reply to

    Re: 2D or not 2D, that is the question...

    For myself as a lone-wolf indie, I find that 3D actually is a much better fit for me than 2D, even though as a 2D past master, I could lay down stuff really really fast in 2D.

    3D is and was a hard slog to learn all the math involved - hell, to my embarassment I had to learn a piece of grade 9 trig the other day, mixed in with the linear algebra, calculus and analytic/differential geometry. :) But the point is - it's all learnable.

    And why I believe 3D is a better fit for me than 2D is simple - even though some of my game concepts WOULD work in 2D, they'd be better and easier to develop in 3D. For instance, I could make that turn based tactical party-based combat portion of the one game I want to make in 2d/Isometric, but it'd look much better and be hella easier both in assets and programming to make it in simple 3d.

    Hope that helps.
  • 4/20/2007 12:37 PM In reply to

    Re: 2D or not 2D, that is the question...

    The best steps to take are to do some 3d inside the 2d world you are comfortable in. Both SpaceWar and Marblets do this so takea  look or start with one of them. It will get you used to some of the concetps without worrying about bone animation or other stuff.
    Play Kissy Poo - a game for 4 year olds on Xbox and windows
    The ZBuffer
    News and information for XNA
      Follow The Zman on twitter, Email me
        Please read the forum FAQs - Bug/Feature reporting
          Don't forget to mark good answers and good playtest feedback when you see it!!!
  • 4/20/2007 5:50 PM In reply to

    Re: 2D or not 2D, that is the question...

    Are you comfortable with creating art assets for 3D?  If so then there really isn't a reason to utilize 3D.  3D offers many benefits in terms of flexibility.  It is a bit hard to pick up at first but once you learn the basics its really not that hard.
    --
    Gamer Outfit - XNA News and Reviews
    Maker of Exisled - Helicopter 2d/3d Shooter available now!
  • 4/30/2007 1:25 AM In reply to

    Re: 2D or not 2D, that is the question...

    Hi Somersault,

    The maths of 3D is not that hard... XNA (and Managed DirectX) abstract so much of it. That is not to say it is not daunting; but you can over come it quite fast.

    The problem is 3D assets... it is much easier to make 2D graphics than it is 3D.

    I myself have made an MD2 model loader. MD2 models are a dime a dozen, and while they are not the fanciest 3D graphics around they do look good.

    There are plenty of MD2 models for the fantasy genre and the modern 'shot-em-up' genre. The other great thing is that MD2 models can be customised by essentially 2D editing methods (its amazing how much changing a model's skin changes its look and feel).

    I am currently in the process of writing a library to help people like yourself make the transition to 3D by abstracting all the 'complex' 3D work; but at the same time not making it invisible (it is open-source, and my encourage method of installing it is for people to just add the classes to their projects).

    The current release of my 'framework' allows you to load an MD2 model, position it by defining a vector3 (X,Y,Z) and rotate it by degrees (you can currently only rotate on 1 axis by this method in the current release).

    The model can be animated by simply defining the start frame and end frame.

    So the only difference between 2D and 3D is that you have to specify a Z position.

    Currently you have to define the matrix of the camera which sounds scary, but the tutorial clearly demonstrates it... however the current version I am working on, and intend to release by wednesday includes:

    - A camera object that sets it self up, you simply attach it to a model (Camera.AttachTo(Model)) and specify how far from the model you want it to be on each axis (Camera.OffsetX = 10, etc)

    You don't need to worry about perspective, or anything. (You can ofcouse manualy set the perspective, view, etc matrices if you are so inclined).

    - Model loading is much simpler... was simply already, but now it is very simple. I have also abstracted everything out in to abstract classes and interfaces so when you are ready to delve in to 3D you can do so in steps.

    There are three things my framework is missing (or for the public, will be missing after wednesday):

    - Maps... by the end of this week I intend to include atleast a basic map maker / loader.

    - Lighting... going to squeeze this in sometime soon, mabey even by wednesday.

    - Physics... specifically gravity... need the map load first for obvious reasons.

    I don't intend the mapping in my 'framework' to be great... it will basically be textures blocks at first, and then textured cut off blocks (a block with its corner triangled).

    This is not top of the end 3D graphics, but that is not my intent ... my intent is to allow people to make 3D games simply, with the resources that are plentiful (2D textures and older 3D models).

    If you would like to have a look at my model loader: http://forums.xna.com/thread/7103.aspx

    If you have questions about 3D: lynch82@gmail.com

    If you want a bunch of MD2 model to play around with, contact me and I can point you in the right directions... at the moment I am trying to colate over 200 models in to a single library but I am waiting on permissions to republish from the various authors.

Page 1 of 1 (10 items) Previous Next
var gDomain='m.webtrends.com'; var gDcsId='dcschd84w10000w4lw9hcqmsz_8n3x'; var gTrackEvents=1; var gFpc='WT_FPC'; /*<\/scr"+"ipt>");} /*]]>*/
DCSIMG