Work on the Xy Gaverrah completes.

While work on the UD Cathedralship begins.

I have begun work on what is likely to be amongst the last and most difficult Anahn models I ever make.

The Emperor-class.

Simply wow, dude.....

Glad you guys like 'em. Here's some detailing and layout updates for the Emperor.

well, you need to remember that model are for game and not for render in 3dMax or any pro-software... keyword is optimization at any level...

I will not comment about the shape of the model because taste is different for each people...

Based on your screenshot, i can only advice on poly reduction/optimization... it is not about changing the shape but remove non needed poly with the shape remaining like now...

Let's take the fortress... on the top of the arms who go in each direction, you have plenty of small quad when this can be reduced to only a few tri... same with the top front part of the Battleship...

If you really wish constructive critic, best is to post a .obj from your model... it is the only way to check if everything is good or not ( double edges/vertex, non planar poly, hole, bad normal, flipped faces, etc )...

A quick XSI Poly. Mesh->triangulate fixes this up before you export. 

THIS!

Though I'd restate it slightly as it's ok to make models without knowing how to texture IF you aren't planning to do anything else with them. But for the love of god don't start a modding project if you aren't willing/capable to do all the roles. So many mods appear to die on the vine (look at moddb) because somebody who could code or model (or worse yet, just had a cool idea) started up and then quit because no one would pick up their slack and join the mod.

Three new Undead ships have been created.

This capital ship had an unusual production life. I pulled the head off of Antharas from Lineage 2 and began my symmetry mangle spree until I had a base mesh, which I then proceeded to butcher in a thousand different ways until I finally ended up with this.

Simple UD Heavy destroyer.

An Undead Xy`Kranashian ship, formulated from Xy mental waves. Extremely rare, but fourth generation material.

IskatuMesk, love your Xy`Kranasha corvette.  I have drawn similar designs for a race I was working on, but since I have to graphic design background, made it hard for me to put my drawings into code, haha.  Well done!

Nice job. Would be fun to play a strategic level Homeworld (although I admit more of a weakness for the first one). There's a tabletop game that was in development, but it stopped just beyond functional. Tabletops aren't my thing anyway.

So nice Battlecruiser. Now I might have to create a massive Homeworld map. Shining Hinterlands, here I come!

It is where non planar surface ( non manifold surface ) can lead to problem... for example, a quad where two of the opposite vertex are somehow highter that the two other... with a auto triangulate, you have no way to know how the final result will be ( there is two way for triangulate a quad )... and both result give a different shape...

Well, more easy to explain with a picture that with text... look at the pic below...

The center piece is the original model... with the top surface of the cube being non planar ( non manifold )... left and right show the two possible way to triangulate the top surface... leading to two different shape...

Before make any auto-triangulate, best is to triangulate manually all the non planar surface manually... it is the only way to be sure about the final shape of you model...

In wings3d, "select > by > non planar faces" show the surfaces who are best to triangulate manually... in Meshlab ( my prefered free software for check/repair a model ), need to select/show the "non-manifold" surface... pretty sure that XSI or 3Dmax have a similar function...

I don't agree... a modeler need to know how to model-segment-UV-texture... same if he don't texture himself, he need to know how his modeling job will greatly impact the job of the texturer... by example, a good UV optimization give more pixel for painting job... and if you don't paint yourself, a dialogue with the texturer is needed from the beginning... by example if two identical pieces are planned to have text of number paint, don't optimize the pieces else on one piece, text and number will look like when you see them in a mirror when render...

As for doing all the roles, in my own case, i cannot... XSI hate me and i use somebody other for make the hardpointing and create the .xsi... it is not a big problem since i make models for other people and they own me...

Of course it is best to be able to make everything yourself and i will be able to make so in maybe one year, by removing completely XSI from the process... have already found a way for create hardpoint in Wings3d... with time, i will have a convert_obj-mesh script able to convert .obj to .mesh ( or reverse ) with everything ( hardpoint, tangent, bounding box, etc )... the trick will be to use wings3d and a special method of modeling ( hardpoint are a separate model that you merge with the main object without combine it... orientation is define by the line who go from the top of the little pyramid to the top of the big one, coordinate is now the center of the geometry but later, it will be the top of the little pyramid... wil only be for Linux version since i have no idea how to make a program for windows...

Point is that you don't need to know everything, simply adapt to the situation... you can collaborate with other mod ( i make your ability but you make my entity ) or create your own tools ( like the myfist hardpointing plug-in for XSI )...

Maybe it's semantics, but I don't see where you actually disagreed. 

I simply stated the following. It's ok to make models and never learn how to texture if you aren't planning to do anything with them. You clearly stated a scenario where the original modeller intended to make the model for a project which implies them being created for a specific purpose other than the fun of making the model. In this case it would be beneficial they know how to uv-map and texture. However, given the hobbyist nature of modelling I've found quite a few people that make models don't know how to uv-map them or texture them.

The other statement I made was "willing/capable" in regards to starting a modding project. I didn't say that the mod leader would need to fill every possible role in making the mod, but they should seriously consider their willingness and capability to learn it if they can't fill that role they fill is "important" or potentially tweaking whats left after a team member leaves. It's not a difficult concept and one that a potential mod leader should really ask himself and evaluate before committing to a long term effort especially before asking others to commit as well.

I could site some example mod projects that fit the problem areas described but it's fairly obvious if you poke around on moddb. 

 Yes please.

I am sure I can figure out what the hell iwo format is  .

As for the triangulate thing. I often do not use tools that automate because it you are a cook you dont get your spaghetti from a can, you should know how to create it. Same with the convert tools. I do all mine manually or with a script I created so I know exactly what is being done.

Thanks for that explanation Thoumsin. Can I post that info on Getting Rid of Non-Tris and Non Quads ?

I was responding to myfist0's statement of "Must learn to texture if your going to model..."   

Also this thread never predicated the model will ever make it in game.