Don't Leave It Until The Last Minute
A case study on rushing to get through an assessment
As I, again, have found myself rushing against the clock some fun details have been missed from this assessment. You'll find out later.
Our brief was to create two variations of "Picnic Bench's" with two variations of materials. An "Old" and "New." Part of the challenge of this was creating, and using procedural modelling techniques. We have been shown these over the past few weeks by our lecturer. Now, while I didn't start the assessment until a day before due date I DID do the homework. I wasn't completely lost.
I have decided to attempt the multiple plank bench (variation 2) first, as I believe this is the more difficult of the two and it will be easier to make the final model more simple than making a simple model complex. We shall see!
First thing I did after adding my human scale ref. was to find out the measurements of a park bench to make sure it was a relative size.
First thing I did after adding my human scale ref. was to find out the measurements of a park bench to make sure it was a relative size.
Next, get started...
I’m very proud of myself to have gotten this far. I’ve maintained a gap between the planks as the reference sketch indicates.
At this point I’ve mainly copied and pasted the table top to the chairs. I was charging ahead before I took a moment to glance back at the scope sheet and realised the beam holding the seat needs to span from one side to the other. I’m not 100% sure how I’m going to do that while keeping it procedural. Thinking I will use the Copy node and times the base width by 5, then bind the line to the MatchSize Node bounding box.
At this point I’ve mainly copied and pasted the table top to the chairs. I was charging ahead before I took a moment to glance back at the scope sheet and realised the beam holding the seat needs to span from one side to the other. I’m not 100% sure how I’m going to do that while keeping it procedural. Thinking I will use the Copy node and times the base width by 5, then bind the line to the MatchSize Node bounding box.
I was using Line node to create the seat beams, like we did for the barrel. However, when extruding there is a side missing, so I will need to use a box.
However, I also realised that, unless you’re playing a game like Grounded, you’ll never see the underside of these. I have decided to rotate the holes so they are facing the bottom.
For the table top I really wanted to show individual planks with gaps between them. To keep things
procedural, I wasn’t sure how to do that with the seats while also including the beams spanning from left to right. Because of this I used a bevel node to try and replicate the gap.
The most complicated issue I had so far was connecting things so the table would be fully procedural. I managed to do it after a lot of fiddling and I’m very proud.
Next I created a High Poly’ version where I upped the poly bevel. Maintaining the sharp corners of the planks. After this I began removing faces I believe wouldn’t be seen.
Deleting never seen faces
Taking off the wireframing I realised that due to the bevel where I made it look like two separate planks for the seat had created a weird fold on the edges, also.
Much better.
Now for the UV's
UV seam time. I find this part very dull, frustrating, and time consuming. Do people enjoy this? I have LoFi beats playing in the background so I can try and make it a calming experience.
Next we're into Substance Painter and on to texturing. It is also about the time that Past Sav stopped recording anything with words. Present Sav will try to fill in the gaps.
As the brief asked for wood, and I had wood as materials planned for this variation that is what I ran with. I enjoy making new shiny wood look old and weathered. Another lecturer was once amazed at how much I'd managed to age something he'd deliberately modelled to look new.
I knew I wanted the aged variation to look like paint was peeling off it. So we went with a nice forest green colour for paint.
I used the cherry wood, and a rubber-like material on top for the 'paint.'
I couldn't find any peeling paint in the starter assets. These are the ones I downloaded from the Library. I ended up mixing and matching the two.
This happened when I was playing with the Tiling parameter of a material. I thought it looked like a cartoon. It reminded me of the shading in Arcane. Thought it was cool, so I took a screenshot.
I played with moss covered wood, but the moss wasn’t falling the way I wanted it to. However, this texture did make it easier to see the carvings I’d put into the table. I really wanted this table to tell the story of being the hangout for high school kids. Y’know, those late night parties where you snuck out with your mates. Someone smuggled a permanently borrowed bottle of spirits.
SC is me and LN are for my cats, of course.
In the end I used 3 different textures. One for the height map, the other two for the chipped and worn away paint.
I like how the light hits the Height map here.
And here you can see more of the graffiti.
While admiring my work I realised an error that I'd forgotten to fix before I'd exported from Houdini. I did not have the time to go and fix it.
Added some dirt for final touches and then went to Unreal to see how the textures loaded.
Both of these variations worked perfectly (from what I could see) within Unreal. No issues. Which leads me on to variation 2.
It wasn't until a few hours of texturing, exporting, fiddling with the second variation for a few hours that I returned to the brief to see it states "OLD WOOD" and "NEW WOOD" for both variations. At this point it was too late to go back and fix as I'd decided on weathering some plastic.
With that being said, here is the 'New' variation.
And the 'Old.'
That is the end of the notes I have recorded for this. I exported them all into Unreal. For a reason I didn't have enough time to explore, the 'plastic' benches texture didn't work.
Here are the final renders. Ignoring the fact that I missed or ignored the section where it says to make them both wood, I'm proud of what I've done. What's the biggest take away for the future? Start sooner, stop procrastinating.
