3D Modeling, what I have been doing.

So Low Poly.....

So low poly was a choice I made after practice within Semester 1 & 2.  I initially thought of setting up an environment with high poly gear but after initially working within the idea of hi quality models... The idea after discussing with friends was to convert the idea into a low poly design, as both a challenge as well as a practical test that'll allow me to specifically try to see what there are not that many low poly horror games.

Colour Palette

The Colour pallete was originally going to be designed to be a complicated texture field to allow for indepth texture design to compliment the environment.  Since my favourite films were like Alien, Star Trek and Star Wars (Original Trilogy) then idea of clean brisk colours to emphasise design and specific parts.

 Looking at the images, I love the crisp aesthetic design of the colours against each other, no chance of bleeding colours and everything still has a strong definition even if the colours are all set to a specific colour range.

With that, I looked into the colour that worked well with each other so i went for the following selection:

• White
• Red
• Orange
• Black
• Dark Grey
• Grey
• and a metal Material.

My design asthetics will be a little different that the colour range will be changed but I want to keep the style but with my own idea and flair.  I found a few different variations of texture design for low poly design textures but each one used larps and meld that diesn't really work well so i've gone for plain old textures.

L.O.D Design

I'm not going to be designing them with the usein my prototype.  Due to the low level complexity of the parts, I don't think it'll be necessary as the polys themselve will have around 200 per asset so the idea of now using many polys is the best process i feel is necessary and more helpful than optimising different LODs for camera use.

UV Unwrap

In regards to the UV unwrapping, I specifically didn't slice it down like i would do for a higher quality object.  Also as well, due to the lack of detail that would be put onto the asset, I have specifically designed the UVs to stay in the same size if possible except if a screen is being used then the screen will be as big as possible to allow for the best quality image possible.

Light Mapping

In regards to lightmapping, I was taught to specific versions of light mapping.

How I was taught

That in the lightmapping level that all uv shells should be small and a big distance away from each other to stop the bleed effect on the edge of the the shells.  I initially thought this was a weird process when being taught this as the amount of distance needed to be as much as possible which meant that the shell had to be small, but it did the job but I feel now we're having the stronger capability with shadows it maybe the idea may have to be looked at in the future.

What Unreal has taught me

To teach a lesson in light mapping to my students last year, I had to do some research into lightmapping just to get the rust of the old skills.  I found a video lesson from Epic devs in regards to lighting.  What was taught was in regards to different textures and materials with different material settings, but what was interesting was when the dev brought up information in regards to the light map on an asset. 

Instead, what I learnt from the Dev was that lightmapping should be different as an engine doesn't calculate lighting in the same way.  The way the dev explained it, it was to make the UV shells practically hug each other with no space at all in the UV Shell.  They gave an example showing the new assets within an environment with different lighting, the quality was almost the same with less pixelation on lower shadow mapping.  From testing it against the previous and looking at the CPU processing.. I can't see a significant difference.

IT. TAKES. SO. LONG. TO. MAKE. A. MODEL...

 

So I made a list of objects to make for the environment... It took ages to make the listing and trying to specially make the assets correlate around the objects.  After doing some work, I had issues with the collision detection as well as the lightmapping.  One thing I learnt though, is that modelling take soo long to make.  It actually takes ages, longer than expected.

That said, I didn't expect it to be so long as it took to model, I am though way behind schedule.  I must plod on and continue as best as I can but at the same time develop my prototype and work skills as best I can.