So after playing around with Mudbox, I couldn't help but think how much easier a tree would be in ZBrush - considering it can be built from scratch in it using ZSpheres.
After finding a trial version that should last long enough for this project I decided to go ahead and use it.
Here is my first attempt at a rough tree and then a further attempt at the main tree using ZSpheres:
ZBrush Start
I was happy with this progress until I started to sculpt. I felt my mesh was getting really untidy and have began to worry about the UV Mapping stage and using a low res piece with high res texture maps as I am not too familiar with this.
After searching around I found a few good tutorials and info that might help get through this:
UV Master
A useful approach to UV Map your model
http://pixologic.com/zbrush/features/UV-Master/
Using DynaMesh
An effective method of sculpting a rough layout before final sculpt
The Pipeline!
This digital tutors tutorial shows an effective pipeline. The only thing is it uses a software called Topogun which I have never used. Never-the-less, I can still follow this and retopolgize in ZBrush (even if it is a longer approach)
Tomorrow I will be trying to find a suitable method to complete my tree. Whether that involves; ZSpheres, DynaMesh or anything else. Once I have figured out the best way to do this I will begin on my tree.
Sadly I am now a bit behind where I wanted to be but hopefully it will be for the best!
So just a quick update relating to my Mudbox practice...
So I have had a half decent go at modelling a tree much like I sketched out:
Practice Tree
So I have now a good idea of how to sculpt and finish a good tree. I have also found a good looking tutorial on Youtube regarding workflow between Maya and Mudbox:
https://www.youtube.com/watch?v=7SR-p1qQ_tY
From watching this I have gathered the following order of workflow:
Model in Maya
Take into Mudbox
Increase Resolution
Sculpt
Paint
Extract Normal Map using High Res Model
Transfer all new maps to Maya (Maybe a higher res model too)
Personal Notes
So my rough plan is the following:
Sculpt a final tree in two weeks (Maya Week 1, Mudbox Week 2)
3rd week will be to set up scene and the basics (hill - maybe pond)
4th week start on surrounding foliage
Again, this is a rough guide but it would be good to stick to it in order to finish it by Christmas!
So I have began by learning the basics of Mudbox, as this project is clearly going to be completed with it. I have used ZBrush before so I am familiar with the basics and am confident that I will pick it up relatively fast.
I found a quick simple tutorial online that actually involves a tree stump (guaranteed to be in my final piece) so I decided to start with that:
https://www.youtube.com/watch?v=koCOjBv7PTg
This had two tutorials, showing how to create a realistic looking tree stump using the demo one provided. I was happy with my final result however, when using the low resolution model in Maya, it did not look the best:
Mudbox Practice
The low res stump is level 0 (in Mudbox) and the high res is level 5. Comparing each of these with the rest of the levels, level 2 comes out best with regards to look and polycount (1824). This is something I will need to sort for my final assets.
Planning...
So I have decided to attempt to merge all of the original concept ideas together - Pocahontas, Peter Pan & Alice In Wonderland. Here is a quick rough idea of what I want to try achieve:
Planning
The idea is to build the tree first - a mix between Grandmother Willow and the Lost Boys Hideout. The Grandmother Willow Tree part will lead to the Alice In Wonderland scene (Tugley Woods). The main set up of the scene will be from Peter Pan.
Personal Notes
I have began to make a quick rough tree in Maya (next step - try and sculpt in Mudbox!) This is the first time I have done this so I am hoping for a half decent result that I can improve on for my final asset.
So, after the completion of my last project (which I am very pleased with) it is time to begin the next one. The previous project was meant to be a relatively familiar type of project as my previous projects, to settle me back into constant production of 3D environments. This project took nearly bang on 1 month.
Using this last project to judge how much time and how big my next project should be, I have decided to give myself approximately 2 months (Up to Christmas) - with the idea of a simpler project over the holiday season. The reason for an extended time frame is I plan on doing a more organic scene in the forest - something I have never tackled before! This will require lots of practice and learning before the final model.
So What To Do?
The original plan was to create some sort of Cambodian style jungle setting - taken from images found online however I was struggling to find myself engaging with it. I felt like as these are personal projects, they should somehow incorporate my interests or at least, something that drives me and will of course be fun!
So I started to think of films etc that had 'iconic' jungle scenes but even this proved a little difficult, until I stumbled onto Disney's iconic creations...
Tugley Woods - Alice In Wonderland (1951)
The Lost Boys Hideout - Peter Pan (1953)
Grandmother Willow - Pocahontas (1995)
I decided that I would do one of these scenes, but in a realistic game style. I love each of these films so I am already excited to start this project (I want to do all three!) - that might be possible...
Personal Notes
Tomorrow I will begin carrying out a little more research on organic modelling and look into using Mudbox for the majority of this project - yet another learning curve. Really excited to use it. Having used ZBrush before and with the interface going to be much like Maya, I am positive I will easily pick this package up.
So I have now completed my project. Overall I am really pleased with the finished result. This took me about a month to complete, which I feel fantastic about. Here is a reminder of what I was replicating:
Reference Photos
And here is my final project:
Final 3D Model
So whats next?....
I am away on holiday for a week (mini-break!) so I am going to spend that time deciding what to do next. At the moment, I am considering a much more organic scene. With this being the main thought, I am considering to a scene from Cambodia as my friend recently visited there and came back with lots of really interesting and nice pictures.
Anyway, for anyone who has been following (not expecting much, if anyone at all...) but thanks for takign the time to read my posts. I will be keeping this up to date and maybe even blog more times than I did for this project. I will finish with a link to my portfolio as this is where all the final pieces end up in the aticipation for a job!
This is the final week I am giving on this project - which takes it nicely to a months worth of work!
At the moment, I have just finished the texturing (and kind of the lighting) but here is what I plan to do for the rest of the week:
Mon - Final Texturing
Tues - Final Lighting/lighting maps
Wednesday - Final Touch ups
Thursday - Render final images - set up portfolio
Friday - Finalize portfolio - Metal Gear Solid.... sorry, did I mention I had this game and haven't had much time to play it?! (Yes the newest one!)
So thats my week plan and I am very confident of completing this. I look forward just to having another piece of work on my portfolio site. Here is a few images of the current stage of the project:
Final Texture
Personal Notes
I am extremely pleased with this so far. Minor things I wish to tweak are:
As I have been texturing the environment I have stumbled across some interesting and somewhat useful items on the world wide web! Again, this blog is purely for my own purposes and can almost serve as a resource of mine for future endeavors. So this quick update is just to highlight some interesting and useful sites (that I found):
Glass Texture
So a few days ago I did spend a fair bit of time trying to figure out how to create dirty looking glass windows and (ideally) include a checkerboard effect to them. This proved very difficult however towards the end of my searching I discovered this...
Basically this guy has managed to create something close to what I wish to do. With careful manipulation I might be able to figure out a way to get what I desire! I am leaving this for next week when I can spend a good day attempting it. This web link provides the material set-up 'this guy' used for his glass material.
Wood Textures
In the end I ended up using cgtextures.com but I thought I'd take note of this page as it looks very useful and has lots of wood textures!
Of course, when using textures from online, I end up editing them heavily as they usually serve as a background layer whilst I go over and over it, making it as original as possible.
Texture Sizes
When comparing my work to big AAA titles I always wonder how I can't get my textures to look as good (as detailed). I thought it might be because I am using texture files of 1024x1024. After carrying out a little research on what the next gen games use it was quite fuzzy with mixed answers. I think it pretty much depends on the teams budget etc but I did find this forum useful and interesting:
Creating Next Generation Assets
It might be a little late to be looking at this but I can certainly look to apply some of the techniques to my next project - whatever that may be...
This individual actually goes through the pipeline of creating a next gen asset which is very helpful. One thing I have always wanted to learn (which I will learn once I get a job!) is what the actual pipeline is. I know it will differ from company to company depending on a range of factors such as size etc, but I do look forward to starting a new job and immediately carrying out work in their ideal pipeline.
So this is a relatively long blog (compared to the others) but it is basically just a quick update some things I found interesting. I am currently looking to finish the texturing of the environment by the end of this week. That means next week will be used for:
Lighting
Glass Texture (hopefully)
Final Touch up
Apply and sort portfolio
I am really excited to finish this. I can already see an improvement to my last project (Honours Project) in terms of texturing which is what I wanted to achieve.
