Feedback

14 Jul

Modelling

What went well: Some realistic models in terms of shape and texture (i.e the shape cushions and toy car).

Even better if: More detail on some of the models such the black box which really confused the viewers as to what it is. Same goes for the puppets at the side.

My comments: I couldn’t agree more. Whilst the car and cushions were the highlights due to lack of last minute previewing the render ended up making my objects seem obscenely dark. The box has an UV texture which again had a broken link so it didn’t render out (something I noticed after rendering for a long time already). It was all due to silly mistakes really, that costed the final results.

Environment

What went well: Colouring, spaciousness and realistic looking ceiling.

Even better if: The room was too empty so more objects would have added realism; better lighting and shadows.

My comments: I was trying to go for a minimalistic room but I understand that it clashed with my realistic approach. The lighting and shadows really was my weakest part. I spent a lot of time on it but failed to adjust it properly in the end and it cost me a tone of rendering time and clarity. The lighting is my biggest regret. This would be the department I need to work on the most in future.

Animation

What went well: Smooth camera / object movements; well rigged models.

Even better if: Jerky car movements; unnatural ending; a couple hiccups with camera movement (need to work on keyframes).

My comments: I love how this one gave everyone conflicting feeling and that’s understandable because of several reason. 1. Regarding the camera whilst parts of it were naturalistic (I spent a while keyframing imperfections) there were some overly shaky parts (but alas I went overbaord at times with creating dirtness to the camera movement). 2. Focusing on my car, whilst it was well rigged (thanks to expressions) the animation lacked fluidity,  mainly due to the fact that I’m an absolute amatuer in the world of animation so naturally I would need to work on that (by reading up theories and methods whilst using it in practise of course).

 

Advertisements

Opening statement: the journey

5 Jul

It’s a bit odd to end my 3D blog with an opening statement but it makes perfect sense to end it by detailing my journey from researching, planning to my final product. I set out in Maya to model 6 objects before advancing to modelling and lighting an environment (the playroom/shelf) so that I can produce an animation with these assets. That’s the project in a nutshell and without further ado let me sum up my final products:

Original models and final results

Models screenshot

models side by side

Original sketch and final environment

environment

environment side by side

Final animation (will be updated once everything’s rendered):

animation

https://docs.google.com/file/d/0BxExkArU7_gOWlVTMVhBaktSWmc/edit?usp=sharing

So what did you think?

19 Jun

Now that you’ve watched the 3D animation (or a quarter of it, depending on when you watched it) it’s time to give feedback! Complete the survey before and remember all opinions are welcome (remember to be polite though I want to keep school work SFW).

My 3D Animation- Playroom

18 Jun

Due to (greatly) underestimating rendering times the below is only the beginning of my full 2700 frames (yeah I don’t know what I was thinking either) animation. I’ll update it at a later date when it’s finally finished rendering/spiking the school’s electricity bills.

https://docs.google.com/file/d/0BxExkArU7_gOWlVTMVhBaktSWmc/edit?usp=sharing

In Conclusion

15 Jun

It’s been a really insightful and chunky course/project. Maya is a software project with almost bottomless depth and every google and tweaking with the options here and there brings new knowledge. I’ve absorbed an incredible amount of keyboard shortcuts, modelling tips, mini physics lessons and how 3D animation requires a lot more procedures than I thought.

The world of 3D contains an incredible amount of jargon and mathematical qualities (starting up the words, rendering, UV texturing, bump mapping, mesh, particle collision, locator and the likes were as familiar as ancient glyphs) and that made it the most intimidating software for me to date. Once I knew how to navigate around inside the software and started; maneuvering geometry; creating my own custom shelf and even marvelling at the physics simulations everything started giving me a sense of pride. There’s nothing like looking at a finished model, all rigged up and textured with mental ray.With the knowledge that you, yourself have created that after 1000 google searches, forum digging, note taking and tutorial watching (I have a whole playlist on Youtube dedicated to Maya).

Mind you all these new knowledge has made me extremely analytical of any 3D graphics. It’s hard for me now to play a game and not comment on the animation of the characters or ponder at whether or not the ground is an UV texture. It’s a bit scary but I’ve always wanted to understand how things were made (ignorance is not in my dictionary) so in a way I don’t mind it. Is just that I can barely focus on the story when during a cutscene the character’s hair is penetrating with their clothe (where are the colliders!) or if their animations baffle my mind (how many rigs must person A have)!

Hm let’s see: the suit is most likely a mental ray metallic texture, point/ spot lighting at the shoulders, rust effect is probably an UV texture applied on top and the glowing strips are… Wait a second, what am I doing?

There’s still much to learn though so if I ever start the project again I’d make some adjustments. For one I need to change the schedule so that I leave more time for rendering and texturing. To better prepare myself I’ll need to read up on some physics relating to the ones used in Maya as well as study some principles of animation. One thing I wanted to do at the start was giving my animation a distinct visual style but due to my high level of curiosity I ended up using a whole load of realistic textures and wasn’t really the graphic style I loved. Personally I’ve always wanted to try my hand at toon shading or low polygon modelling so for future projects I might just stick to those styles to give a cleaner look to my work (and reduce the rendering times as well by cutting down the number of photons and mental ray textures I’m using).

Another thing I’d love to improve on is the lighting. In my final render the scenes came out too dark and the rendering took too long. Next I need to better preview my work and also adjust the figures of global illumination and final with a little more caution. Another way is to avoid having too many light elements in a scene that uses raytracing when it’s not the main light source (I wanted a glow effect on lights but ended up ticking the raytracing option for shadows which dramatically increased rendering time). So all in all I’d need to give lighting a lot more focus and attention. No matter how well your models and animation are if the final render/lighting doesn’t illuminate the scene the end result will be ruined.

Original Objective

15 Jun
Screenshot of the overly large prezi.

My original plan.

For the most part I stuck strictly to my plans for the entire project. So close that even I was a little surprised that I managed to pull everything off (okay so I never added the wooden textures to my puppet models). Even the car ended up actually looking like a Mustang and not something out of an 80’s Disney movie.

For the animation part I added more bits to what I had planned in the storyboard- after all once production began it’s natural for me to get more interesting ideas which I couldn’t resist adding.  I even got to experiment around with the smoke effect even (by making the car breakdown of course, even though I know in real life toy cars can’t really breakdown).

That doesn’t mean everything I planned was implemented exactly though for one I wanted to animate the Rubik’s cube to rotate in numerous directions but ended up having to stick to basic rotation due to lack of expertise on how to rig it. I found an already rigged rubik’s cube on the ever generous interweb though I opted to use my own for the sake of originality.

Problems Encountered- Trials and Tribulations of Maya

12 Jun

As any followers of this blog would know I’ve encountered many problems during the making of this 3D animation. I’ll not repeat the whining and ranting about how silly my modelling antics were but there were many other problems you haven’t heard about in the other procedures! So here’s a whole post dedicated to explaining why I have less hair than before entering the world of Maya (which I love just to clarify, despite the stress it gives me).

Tip 1: Don’t move around, pan, zoom or anything that requires you holding alt key in the camera view. If you do it will change the view in every frame of your animation and as it’s a UI element to move around in a scene you can’t undo. So control the camera from other views not inside the camera view.

This view gave me plenty of nightmare.

This view gave me plenty of nightmare.

Bleached out renders: this is what happens when you turn the light intensity up too high.

Bleached out renders: this is what happens when you turn the light intensity up too high.

Maya is just about one of the most unstable software packages out there. It could quite literally crash when you do something it doesn’t like (i.e. using the interactive edge tool in the wrong places). So save frequently.

Where am I? It becomes difficult to track your scene when you viewport camera go wary. Locating them is the hard part...

Where am I? It becomes difficult to track your scene when you viewport camera go wary. Locating the cameras is the hard part…

I've moaned about troubles in modelling enough times but I can't stress enough how important it is to keep your mesh clean. Don't add too many edge loops and always delete edges and vertices with the 'delete edge/vetex' tool. Obsolete vertices are your worst enemy.

I’ve moaned about troubles in modelling enough times but I can’t stress enough how important it is to keep your mesh clean. Don’t add too many edge loops and always delete edges and vertices with the ‘delete edge/vertex’ tool. Obsolete vertices are your worst enemy. I knew that first hand.

There are times when rigging can go wrong (for instance when a geometry is distorted by the rig) hence the object would need to be re-rigged. That ladies and gentlemen is ruddy difficult if you don't have an original bind pose to go back to.

There are times when rigging can go wrong (for instance when a geometry is distorted by the rig) hence the object would need to be re-rigged. That ladies and gentlemen is ruddy difficult if you don’t have an original bind pose to go back to.

 

The biggest problem I’ve encountered though has to be the rendering time. When your render takes two days to produce a 45 frame render you know something wrong (very, very wrong). It could be because I added too much elements without thinking or I seriously needed to spare a whole month rendering.