Reply
Reply
 
Thread Tools Display Modes
jmchar's Avatar
Old (#1)
Hi, just a quick question here i've been wondering about and that is when you create an environment in modular pieces, should these be welded together again in max and then imported into unreal as a whole mesh. Or is it better to keep them as seperate modules import them seperately into unreal and then rebuild the building again in the editor by lining them all up using the grid?
Offline , null, 3 Posts, Join Date Feb 2009, Location United Kingdom  
   Reply With Quote

Tumerboy's Avatar
Old (#2)
you shouldn't have to weld them together at all, but assembly is up to you. Whichever is easier I guess.

Never done anything in UE, but in our engine we can instance the same object over and over again, drawing it more cheaply than if each were unique objects. So we export pieces and do a large amount of assembly in game.
-Nick
-Portfolio-
Offline , veteran polycounter, 2,695 Posts, Join Date Jan 2007, Location San Jose, CA Send a message via AIM to Tumerboy  
   Reply With Quote

renderhjs's Avatar
Old (#3)
if the verts are on the exact positions of the other verts it shouldn`t really matter if they are welded. It might differ though from engine to engine.
It might be thus a wise decission to snap all values to internal values (e.g 8 unit grid) where they snap or should connect to other building elements.

I have a script collection that places the selected verts in 3dsmax to a internal grid snap and not the metric one - I could put it into a macroscript this weekend if you need it - I use it for a project where I wrote the engine and we work with internal units and no unit conversion (just like any other game engine).

Last edited by renderhjs; 02-19-2009 at 09:07 AM..
Offline , veteran polycounter, 3,037 Posts, Join Date Mar 2008, Location Sydney Australia Send a message via AIM to renderhjs Send a message via MSN to renderhjs  
   Reply With Quote

jmchar's Avatar
Old (#4)
Hey Thanks for the replies I am using the grid snap in max to snap pieces together and so far when testing it in unreal they piece together correctly, I just wasn't sure about the assembly part, whether to assemble it in max and export or assemble it in unreal, but I think its better to assemble it in unreal as it allows me to reuse the various parts of the buildings to create new buildings in the editor.
Offline , null, 3 Posts, Join Date Feb 2009, Location United Kingdom  
   Reply With Quote

Mark Dygert's Avatar
Old (#5)
It kind of depends on if you want to move the pieces around after you import. If you don't see a need to have a bazillion little pieces, might as well stick them all together. The point of modular design is to have a library of pieces that can be mixed and matched easily to make something look new and unique, but really is just rearranged while minimizing the number of assets.

So for example you're making a big brick building and you find that you've used the same brick tile 50 times without altering it. It would probably be a waste to weld it all together and import it as one giant unique object. Instead import that one piece and copy and arrange it in the editor 50 times.

Depending on the engine, there are costs and benefits to instancing geometry. Often there is a point that it becomes easier on the engine to import larger chunks. The general info that I remember is that it mostly saves on texture memory space first, and actual 3D data second but with some kind of weird cut off point. But that's a discussion for programmers and game designers, I just roll my face around on the keyboard until something pretty pops up...


Small library of parts makes a huge city.
Large library of parts makes a tiny but very unique city block.
Or a mix of the two...
It all depends on what you're trying to do, the scope of the game, and the engine. Even in the same game you'll often find a mish-mash of techniques, so there isn't one right way to do it, just a best possible scenario for area being created at the time.

[soapbox]
So many questions need to be answered before you sit down and decide, I'm going to do things this way. I really think the key to success is thoughtful, methodical planning. I once read that if you have 5 days to do a project, you should spend 3 days thinking about what you're going to do, and spend 2 days getting it done, and possibly work the weekend because there actually is 4 days worth of work to be done, ha! That doesn't mean jack around and surf for porn that means lock yourself in a room with nothing but your thoughts of the project on your mind. This isn't waiting for inspiration its creating a battle plan for the 2 day war you're going to wage. [/soapbox]

Good luck! Post progress as you go!

Last edited by Mark Dygert; 02-19-2009 at 09:22 AM..
(AKA Vig) Portfolio | Lab | Brawl | Decker |
Offline , Polycount.com Editor, 13,917 Posts, Join Date Oct 2004, Location Seattle, Wa Send a message via MSN to Mark Dygert  
   Reply With Quote

Peris's Avatar
Old (#6)
If you made them modular, you can can make them instanced, which is great for performance, so it's best to keep them as separate reusable pieces! It's also easier to work with, if you want to make adjustments you only have to adjust your pieces, and not an entire mesh.
Offline , polycounter, 963 Posts, Join Date May 2006, Location Stockholm  
   Reply With Quote

jmchar's Avatar
Old (#7)
Cheers for the replies guys, yeah so far I have kept them as seperate reusable pieces I just wanted a bit of clarity whether it was the right way to go about it.
Offline , null, 3 Posts, Join Date Feb 2009, Location United Kingdom  
   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Copyright 1998-2012 A. Risch