3D Hard Surface Model

Hard Surface Model based on Character 

For this model, I wanted something that visualised, “To take a piece of home with you”, a heartfelt memento for my travelling Elf to carry with him.

Maintaining narrative continuity is something I realise is an essential driver of my creativity and ability to apply much-needed constraints to the brainstorming process.

Very rough concept sketch notes

Concept sketches 

The ideation process went from musical instruments, to travel collectibles, potion bottles, to stopwatches, until I landed on the idea of a portable windmill scene fixed in a glass jar. I thought that an Elf travelling for the first time would indeed carry a memento that reminded them of home, so I played with the idea of a ship in a glass jar and applied it to the logic of the Elf’s context.

I established the context, design genre and type of object I wanted: Stylised, Medieval, portable memento. I decided to finalise the design and plan the components.

Reference

To help inform my concept art I compiled references in Pureref of Cotswold Cottages and concepts related to the period and stylisation that would match the Elf, nodding to the period I wanted to model for, and see how I could stylise the model.

Blocking out

I realised I needed to acquaint myself with the necessary components of a windmill house and understand the physics, materials and textures of this scene. I referred to multiple materials online and practised many iterations of what the windmill could look like.

Rough Block-out based on concept art

wireframe of glass jar 

Considering scale and composition

To model more efficiently, I created mesh objects in groups which were UV unwrapped and then duplicated/altered to fit the scene. I modelled these wooden beams and adjusted their shapes to add variety to the scene without overwhelming myself with texture mapping, UV tiles and adjusting.

Wood planks

Windmill_Fans

Roof

MASH

MASH Editor parameter

MASH can be very finicky, it can randomise, distribute and assign the mesh to a plane and can flood the plane with the mesh selected to make roof tiles, however, problems arise because it isn't as intuitive as snapping. The distribute/connect mesh will sometimes result in scattered objects, far from the plane you wanted the mesh to flood. After some troubleshooting and manual scaling I achieved the roof tiles for the cottage and home.

Final Scene meshes: Windmill scene

Here you can see that I modelled the ground from a subdivided plane which I extruded and with soft select adjusted to fit the curves and floor of the scene. 

Lighting

For the lighting, I used the same HDRI as the Character model for visual continuity and ambience. 

I used an HDRI plugged into the sky dome. I then reinforced ambient lighting with area lights which were in blue and purple hues to indicate dusk behind the scene and to show a 3/4 lighting effect.

I then rendered the glass but was not expecting the results.

Back view rendered in Maya Arnold

3/4 rendered in Maya Arnold

Dilemma

The windmill and cottage were finished as was the glass jar however what I did not anticipate was the deliberation it took to settle on a good render for the glass jar. I could not achieve the glass jar in time and therefore had to move on without it. The glass also hid the detail and textures of the windmill and cottage models which was also not what I had anticipated in the concept phase. 

Modelling the windmill and house has been an experience. It takes a lot of reconfiguring and understanding of what works. I have found that MASH is a very powerful tool to use for roof tiling. I have made 4 variations of a roof tile, and using MASH I have been able to create a custom randomised pattern. 

Final Render

Final Thoughts

The past couple of weeks, even months have whizzed by incredibly fast, and It's partially due to the many phases in the 3D creation pipeline that demand a level of care and attention right from the beginning. I came into the assignment to understand every step and purpose of the workflow, integrating it into my modelling practices. I admit, that learning can be slow and stubborn and has cost me the integrity of my concept designs for eg the glass jar due to time constraints and time lost on troubleshooting various mistakes but I have come out of this challenge with the ability to produce two 3D assets, one an organic character and one a hard surface model. I have been able to identify my weaknesses in modelling which I will be building on over the next couple of months.

Going forward

I want to model more complex hard surface modelling and refine my application of edge flow. I aim to build/refine my technique in achieving the handpainted textures.