Powerpoint Submission of Renders
Here is the powerpoint file for submission which contains all renders for the project.
Finally the deadline is upon us! This post will be a comprehensive wrap up from start to finish. So rather than tread over old ground again, I’ll simply post links to old posts and discuss.
Having chosen to do the spaceship interior for the brief rather than modelling a real room (from my home or elsewhere), research and prep work was a little bit different to picking up a camera and gathering research materials. I ran through a few old video games and movies for aesthetic inspiration, then put pencil to paper and sketched out an environment.
I deviated slightly between the original design and the finished product. Some of this was down to not planning ahead with the geometry, limitations on the type of modelling I was doing (need to be very careful using planar rooms in the future, they do have their limits) and being honest some of it down to time again. However none of that means I’m disappointed with the end result, far from it! Having kept the textures from the concepting stage I was able to re-use some of them on the model, keeping it spot on to the original artwork.
Modelling the Room
Okay here is where the waters get a little muddy, this project ended up being partially paired up with my Unity project which has been mentioned before. So for my first WIP modelling post, I’m going to link to a Unity post made during half term. This is where some of the assets came from.
After this there was a rather large time jump before I posted anything else. This was mostly me texturing some of the previous assets and the creation of a few new ones. These would all be the basic items used to populate the scene.
Not long after I finished the room itself and began populating it.
Now truthfully there should have been another post after that, instead I chose to do the mad dash to the finish line method. So overall this post will be a little longer than it should have been.
There is a lot I can take away from the whole modelling process and thats before I get into texturing. Sometimes a bit of previous experience can lead to bad habits, I see now I started the course guilty of this. For far too long I was oddly reliant on vertex mode and only vertex mode, which limits you immediately if your initial sub-division geometry isn’t perfect from the start. Thanks to classes and online content my eyes were opened to the wonders of edge loops, after that modelling was FAR easier. Not to say I didn’t pick up on other useful tools along the way, I find myself using extrude and offset for new face creation a lot too. In the wise words of one tiny green old Jedi “You must unlearn what you have learned”. Too true Yoda, too true!
Now texturing was my hurdle all the way through this entire project. Previous posts have talked about this. Single small objects, absolutely fine. Unwrap as normal, export to Photoshop, done! However given my environment was a single mesh large planar mesh, exactly like how I’d built my Unity environment, the problems never stopped. This was eventually solved by the discovery of UVsets, allowing me to UVmap individual sections of a mesh, apply its own texture and setup the UV relationship in the relationship UV Centric window.
I saw this as the solution to all my problems, I could even go back and texture my entire Unity game. It would look great! Until I tried to import the whole thing into Unity. Turns out Unity doesn’t like multiple UV sets, it’ll only read whats on the default map (UV0), so unfortunately the mystery continues as to how on earth I texture large environments.
Here is the really cool, fun bit. If you go take a look at the renders of my individual items I’m most proud of, you may think wow that is a huge quality jump from the other renders! You’d be right. Two days before this submission I invested a chunk of cash into the indie bundle for Substance Painter. I’ve had barely any time to learn it but it was used to render out the single props, my Maya refuses to use Arnold however Painter doesn’t have a problem. The computer console was fully textured in Painter, I think it looks okay for someone with zero training in the program. The Xmas holiday will involve a crash course I’m sure.
Here are all the links for the three models I’m most proud of.
The computer model doesn’t show up in any of my previous posts purely because it was created in the last 48 hour mad dash to finish the project. As already mentioned it was quickly textured using zero knowledge in Substance Painter. For the speed I put it together I’m pretty happy with it and it really helped add something extra to the overall scene.
Originally a model from my Unity game, now textured. I know the model itself isn’t anything fancy or complex but I spent a lot of time adding metal textures, tinting them and blending layers together to get something I was content with. For that reason alone it earns a top spot for me. If I’d had more time I would have created a bump map for the crate to try and highlight the scratches in the texture.
My personal masterpiece! Again not a complex model but certainly the one I spent the most time texturing, before realising if I spent that amount of time on everything, I’d never complete the project. It has rust, it has grime and overall is the most real looking model because of it. They really pulled the scene together and made the crate storage area stand out as the best part of the scene.
Reflection from Start to Finish
I don’t wish to repeat some of the comments I’ve made in the past on other posts or even comments here so I’ll keep this as more of a summary.
I’ve learnt a lot from all of this, mostly from making mistakes. Maya is an unforgiving mistress sometimes and I’d say learning what not to do is a far bigger list than learning what to do. I’ve already mentioned this has all helped changed the way I model for the better, taking all these new skills forward I finally feel like I can progress onto bigger and better things. Especially since the speed of my modelling has also increased. Maybe one day I’ll dig out my college final major project from 14 years ago and compare the two.
The part I’m still most frustrated with is texturing large items. I found a workflow, it worked, the environment was completed and then I discover its a useless method for game engines. It was a low moment for me and I’ll be back to the drawing board for a new method, which may involve Substance Designer.
Overall though, time is my biggest hurdle. It’s been mentioned time after time in my posts but it really is the biggest obstacle in large creative projects such as this. Spend too much time modelling, your texturing quality is low and vice versa. I wish I could have added all the items that I had planned, and used a fully textured version of my glaive ship rather than the very basic first attempt ship that was used in the end. Thankfully, like anything in life, the more time I spend doing and learning the better time efficiency will get.
Bring on the next modelling project!