Project Intersection (my last school project)

Lexeus · Dec 2014

Intro
School is coming to a close and its time for my last school based portfolio piece. I wasnt sure where to start but it hit me to do a piece based on a character doodle I did at work.
I have a 10 week limit so lets get started.

Goals

· Quick pre-production
· COMPLETE environment as envisioned
· Creative design and framing
· Prove that I can do environment art too
· Create an excellent portfolio piece

Planning

· Thumbnails and concepts
· Basic blockout ( to place architecture and props )
· Refined blockout ( to refine prop placement )
· Concept final composition
· Lighting ( start rough, finish refined )
· Utilize following workflow:

[FONT=&quot]o [/FONT]Model each asset to completion
[FONT=&quot]o [/FONT]Unwrap UVs
[FONT=&quot]o [/FONT]Texture
[FONT=&quot]o [/FONT]Place
[FONT=&quot]o [/FONT]Repeat

· Model environment
· Model architecture
· Model props
· Correct and refine
· Final render
· Post mortem doc

Setting
The year is 2086 in Xian China, at an urban street. This particular year is meant to be the intersection of modern day China and the China of the future. We view the scene from across a street, looking over at an intersection at dusk.

Assets
Itd be a good idea to start big to complete the main scene and go small to fill it.

Main assets

Car
Van
Scooter
Railing
Street light
Background buildings

Filler props

Standing signs
Food stands
Cheap tables and chairs
Parking meter
Vending machines
Cones

Color
The scene will feature everything draped in the color of dusk, but will still have the harsh colors of vivid store lights.

Refs

References for china of today

Near future

Near future vehicle

Lexeus · Dec 2014

It's been a few days too long but the scene concept is done

Lexeus · Dec 2014

I'll try and do the last vehicles by tomoro's end

Lexeus · Dec 2014

I meant to concept the van and scooter today, but I fell asleep when I got back from work.

Modeling starts tomorow

tahakitan · Dec 2014

I think you should redo the scene concept using the rule of thirds. is a good sketch for a concept but the angle just feels wrong. The point of interests aren't as dominant as they should be.

Lexeus · Dec 2014

I know for certain that my concept sketch is barely perspectively accurate. I just needed something right away to get started

Lexeus · Dec 2014

first pass blockout is done. I will definitely tweak the view angle after I get some actual modeling going

Lexeus · Jan 2015

It has been way to long since I've made any real progress, but I will put more effort and time into this project.

Lexeus · Jan 2015

Background modular pieces are 95%done. they need some texture filling. Foreground buildings on the right are close to finished. There are some missing textures and some leftover modeling on the grocery. I'll place these things in engine when they're fully textured.

Lexeus · Jan 2015

This is where I'm at now. I will bang out most props by this time next week. When I'm done with those, I will get it in Pro and light it properly.

Lexeus · Feb 2015

long overdue update, but what can you do when jury duty calls
current prop progress all will be fully textured by the end of today

Matroskin · Feb 2015

nice pre-prod! Good approach with blockout and prop planning. However at this point i'd suggest to move on with the architecture instead of finalizing the set of props. The structure itself usually plays the biggest role in the environment, so you might want to start refining it a bit more and maybe some general textures

Especially small props, like folding ad board, road sign and similar things can be totally ignored at this point.

Lexeus · Feb 2015

I know what I said. Dont judge me.

Lexeus · Feb 2015

Also, Matroskin is right.

Lexeus · Feb 2015

scene update. The next step I guess is to update textures

Deathstick · Feb 2015

Make sure you have the scale of the textures right, currently the gravel in the road is far too large compared to the rest of the scene. (some of those pebbles look bigger than the bricks on the buildings)

The painted walk ways on the street should also be either a decal or a single plane with an opacity-masked texture that goes on top of the road (essentially a decal), rather than having each strip being a model. I'd also check the scale of your roads and consider widening them a tad, it currently feels as if they may be too narrow for cars to drive through each lane and not smash into one another, although that may have to do with the feeling I'm getting from the scale of the gravel/road texture.

That said, I like where this is going and think it'll look awesome after enough time is put into it. Keep it up!

Lexeus · Feb 2015

here we are as of right now. This the last week of school and of the project(that doesnt mean that I will abandon this though).

To help myself, here's a list of things to do to finish: (preferably in order)
-finish road decals
-finish traffic lights
-re-texture road
-XCM neon
-build street lights
-light in UnityPro
-model grey box items
-scooter -car

Lexeus · Feb 2015

Here is end of the day progress

That just leaves:
-XCM neon
-build street lights
-adjust crack decals
-light in UnityPro
-model grey box items
-scooter -car

Lexeus · Feb 2015

SURPRISE!!! The project was due today!!! It was mostly done, but I had to put a rush job on the last bits I could squeeze in(scooter 30min, lighting 5min) in order to turn in on time. This is what I ended up with, but I'm not done this project is not done yet.

I learned an important lesson: keep checking on the student portal page to ensure the due date and time.

Lexeus · Feb 2015

looking at it, it still needs:

the XCM neon
the car
to be re-lit
touch up road decals
retexture scooter
new camera angle

Limewax · Feb 2015

You need to stop bumping your posts for like 5-10 minutes worth of changes. You should also stop commenting random self-critiques on your posts to try and bump your thread to the top of the forum. Instead, you should edit your prior post and wait until you have a major update. I know this may make me seem like an asshole here, but no one is critiquing it because it doesn't seem like all too much has changed.

Your lighting is nearly entirely undefined. The texture resolution to texel density ratio is all over the place. Every model is basically the first pass model. In all, the scope of this project is FAR too large for where your current skill level is at. You could benefit more from doing a handful of things in no particular order:

- Step away from the computer and go outside.

- Buy a camera, or use a smart phone, go outside and take pictures, but not just any pictures... take "well composed" shots. Understand composition. Look at films like Drive, TV shows like Better Call Saul. Take the time to understand what drives composition. Look at some classical paintings, look at artwork from the renaissance.

- Draw from observation. Set up a still-life, draw the still life exactly how you see it.

- Paint from observation using actual paint. Start with acrylics, they are cheap and get the job done.

- Get a sketchbook while you're at it. Then fill it up.

- Once all of the aforementioned has happened; then you should head back in to your modeling program of choice and try to start modeling simple props from observation. Headphones, pencil sharpener, grenade, keyboard... not a CAR, not a TREE, not a FORKLIFT. Don't "observe" from a picture, observe by having the object right in front of you. If you cant hold it in your hand, you probably shouldn't be modeling it. When you feel confident with modeling, then learn how to properly make renders in any of the various 3D softwares that are out there. After you can render in software, then you can come back in to the game engines.

ALSO, forward this feedback on to your friend DanielR... I'm pretty sure he's sitting right next to you reading this because you guys posted from the same location and you have the same forum mannerisms and level of quality. Rather than comment on his thread and give him a bump, have him read your feedback, because it applies to him as well.

Project Intersection (my last school project)

Replies