Anim8or Community

Regarding UV mappers I use LithUnwrap which has a free downloadable version, but there are others.  Google is your friend in that respect.  Having said that, Anim8or's built-in UV control tool is actually really useful, for example the glowing panels and switches on the console in the second image above were all added to the object and aligned with it before the object was exported for rendering (mainly for consistency I render most things outside Anim8or which is a shame really, since Anim8or renders always seem to look richer somehow).

Re your comment "Since we cannot actually reflect the background, how does specifying a texture in the "Env. Map" help me?", I should point out that ART materials are absolutely capable of reflecting the background using the ART renderer.  ART materials are complex but seriously powerful, the manual is very detailed on these and is definitely compulsory reading before tackling them!

Regarding what happens when placing different textures in different channels, I would recommend creating a single scene with a single object and light source and experiment away, rendering the same scene with materials in each channel and swapping them about for each render.  It really is the best way to get to grips with how the different material channels work.

If your maquette (I think that is the word) is essentially made out of a collection of parts, like a wooden figure with moveable joints or a toy robot, you can skip the skinning stage.  Skinning is useful for objects that are bendy, like a maquette made from plasticine or something similar.

To create a non-skinned, jointed figure I would generally have each body part in a separate object, and another object that contains the full assembled figure.  The skeleton can be constructed using the full figure as a guide which is then removed, and each body part is then added to its corresponding bone.  It's important to have the orientation and centre point of each object correct in object mode or the objects may not align with their bone as expected, but this can be corrected in object mode easily enough if required.

Once the object is posed in Sequence or Scene mode, it can be exported as a posed mesh using Sequence > Export... and choosing the 3DS option (the Anim8or option exports the sequence as a complete Anim8or file rather than a posed mesh).  I should point out that the 3DS export function may have problems exporting subdivision objects, though this might be something that has been resolved in later updates (I'm using V1.0).  If this is the case it may be necessary to convert any subdivision objects to meshes first.

I'm not sure if there is any way to directly export a posable figure from Anim8or, this isn't something I have had need to do but others may be able to assist.  I assume you would need to know the format your buddy requires (ie. the software he is using) at the very least.

Sure you didn't move anything or re-name any folders?  Anim8or files store texture locations as absolute paths, if you rename or move the folder they are stored in it will throw this error.  You can open the file and fix any materials with missing textures.

Brilliant!  If a picture tells a thousand words a 3D picture tells 10,000.  With not much more than a cursory glance it's clear how your batteries are set up, would have been quite wordy and easy to misinterpret to cover this with text alone.

From a basic concepts point of view, the workflow in Lithunwrap (V1.3) is as follows:

1) Load a 3D model (File > Model > Open)
2) Choose an UV mapping method that best suits the model (some experimentation required here, I most commonly choose Box but it depends on the model and what it will be used for)
3) If appropriate for the chosen method, tweak the spacing and positioning of the UV map
4) Export the UV map (File > UV map > Save)
5) Export the mapped object (File > Model > Save)

There are a bunch of available settings relating to the size, colours and other parameters of the exported UV map, the size will again be driven by what the model will be used for.  At this point, open the exported UV map in your 2D editor of choice and use it to construct your object's texture.  How you do this will depend on the 2D editor you're using and the complexity of the final texture.

In Anim8or (or any 3D editor):

1) Create a material with the finished UV map in the Ambient and Diffuse channels
2) Import the 3D model you exported from Lithunwrap
3) Apply the material to the imported model created by Lithunwrap.  The original model you imported into Lithunwrap will not have updated UV coordinates so the UV map won't match up.

I hope this helps!

Double-click the object with UV tool activated to access the Texture Coordinate Parameters dialog box which includes a 'Reset to default' option.

Achieving 'realism' in a CGI image is not easy, but it all boils down to attention to detail.  We all have pretty good built-in bulls&%t detectors and can tell when something doesn't look quite right, but in some cases it can be difficult to know exactly why.  There are a bunch of elements that all need to be considered when attempting to create a realistic-looking render:

1) The model: Using Rudy's excellent examples above, the 'Knobs' image in particular, notice the tiny gap between the knobs and the font panel, the rounding of the edges of the front panel as they meet the case, the gap in the case suggesting it's made of two halves.  In the MorseKey image, notice how the shiny metal armature is a bit convex on top, just as would happen when polishing a piece of metal like that.  Notice how its edges are not perfectly sharp, which is easy to achieve in the digital world but much harder (and more dangerous!) in reality.  This sort of attention to the reality of engineering is important and when it is overlooked the image can look too 'perfect', a sure sign that it's not real.

2) Materials: While Anim8or's materials are relatively simple compared to other 3D packages they are still complex with a massive range of possibilities.  Getting to grips with the ART parameters is critical, it has its own set of instructions and permits things like complex reflections, glossy surfaces, dielectrics, refractions etc.  What makes glass look like glass, considering it's effectively transparent?  Again, using Rudy's Vase6 render above, it's the complex interplay between transparency, dielectics, refraction and reflection that makes it look so real.  The best advice I can give is to learn what the material settings do, reconstruct the samples in the literature, and do a bunch of test renders to see what's going on.  There's really no trick to this, it just takes time.

3) Lighting: As has been said before, lighting is one of the most critical things to get right.  For example, consider a render of a comms satellite in Earth orbit.  Technically there will only be a single light source (the Sun) but it will also be illuminating the Earth below, which will be acting as a second, diffuse, bluish light source from 'below'. Parts of the satellite hidden from both the Sun and Earth will be totally black with nothing to illuminate them (other than starlight I guess).  However, a scene in a room lit by sunlight streaming through an open window will be very different, as the sunlight will be bouncing around the room banishing most shadows and casting different shades of light on the objects.  This 'secondary' illumination can be simulated in a rudimentary way with the ambient light setting and some more complex systems can simulate secondary (tertiary etc.) lighting, but well-considered positioning of fill lights in an Anim8or scene can go a long way towards improving the realism of a scene.  Having an understanding of how light behaves in different situations is critical in the accurate positioning of these fill lights.

4) Filth: As I said above, perfection is a sure sign that something's off.  Looking at Rudy's MorseKey image, imagine if there were fingerprints on the shiny metal part and the Bakelite knob or a bit of rust pitting; if this was the case I'm not sure I would have recognised it as anything other than a photograph of a real object.  Imperfection is everywhere, from the rough and damaged edges of a desktop to rain spots on a window pane or machining marks on a polished metal surface.  Simulating imperfection can go a long way towards achieving realism, mostly it will be done in the materials but thinking about it at the modelling stage is also worthwhile.

There are other things that can destroy the perception of reality, such as movement in an animation that doesn't respect physics, but I would say these are the main ones.  I don't believe there are 'tricks' to achieving realism, every realistic CGI image I have ever seen (here or elsewhere) has been created by someone with lots of experience and folders full of earlier renders, many of which they would be embarrassed to share now.  It's a learning curve, every render you do will be better than the one before it.  The best advice I can give is build, render, post in forums like this one and take on board what others say.