Total Noob Conceptual Question.

Fix

Ok so i am struggeling to get this question answered, and its really bugging me.

When I have broken some people models down into their separate components (i.e barrel, grip etc) I see that alot of these shapes are in no way join either with a polygon or by having the overlapping verticies welded. I was told u cannot do this, as then all u have is a collection of "free floating" polygons, and not a solid object.

If this is not he case, and I can make a model simply by overlapping boxes, cylinders etc..what then is the use or welding verticies together (other than lowering the number of verticies themselves), and how is this collection considered a "solid" object?

I know its a confusing question, but i simply have not found the answer to this question.

Thx

Psycho-Kinetic_Hyper-Geek

It's generally frowned upon because somtimes it makes the shading wonky and it also mean a loss of detail as you have to use space on your skin for the parts that will be covered up.

Wolf_Kahler

It's always an illusion. As long as you THINK it's a solid object when you're looking at it in-game, it doesn't really matter whether you welded or overlapped. One of the modelling tutorials I skimmed at one point highlighted that fact. It doesn't matter what it looks like when you render it in Brazil, it doesn't matter what it looks like in your modelling program, it doesn't matter what it looks like in Lithunwrap, it doesn't even matter what it looks like in HLMV. The ONLY thing that matters is the end result that's seen when you're playing. Take whatever shortcuts you want on your way to that goal. It's all personal opinion, anyway. All artists work differently. It's good to get the opinions of various artists while learning, but you take what fits, modify it to fit even better, and leave the rest without regret.

BrigadierWolf

Usually the reason why models which are taken apart seem disconnected is because when a model is decompiled, It is grouped according to the Mesh. meaning, every group of vertices that is assigned to a different part of the mesh is made into its own group, regardless of wether or not it was grouped otherwise before it was compiled. And usually, simply overlapping boxes instead of welding everything together is avoided because it makes for sloppy work, However I tend to agree with Kahler that the ends justify the means.

Owen1

<!--emo&???--><img src='http://www.unknownworlds.com/forums/html/emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif'><!--endemo-->

Wolf_Kahler

<!--QuoteBegin--|Owen|+Oct 23 2003, 08:03 AM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (|Owen| @ Oct 23 2003, 08:03 AM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin--> <!--emo&???--><img src='http://www.unknownworlds.com/forums/html/emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif'><!--endemo--> <!--QuoteEnd--> </td></tr></table><span class='postcolor'> <!--QuoteEEnd-->
??

Ollj

you can make models out of a heap of tettaeders (4 polygon pyramids) and intersections of really simple shapes but thats no good idea unless its super lowpoly.

the disadvantages of this are:
- An exhausted z-buffer determining whats visible and what nod. You will have long render times.
- no smooth shading on curved surfaces.
- most times a big lack in detail when you look closer

Its exactly that what made polygon engines > voxel engines.

blocky voxel models are just perfect for huge open areas and large amounts BUT they basically suck in any detail.
<a href='http://www.effectware.com/download/images/efx_voxel2.jpg' target='_blank'>http://www.effectware.com/download/images/.../efx_voxel2.jpg</a>