Dog research

I looked at multiple different videos about dogs and the following few links are the ones i find most helpful for at least.

The first link is to  a video of a man training dogs to walk on a treadmil. this video has shown me how dogs move so i could emulate that for my own animation when i have finished rigging the model hopefully to make it as realistic as i can.

 the secound is mostly about if dogs could sense emotions for their owners but it also has a little section and how they show their own emotions and i will be using this so i can add set driven keys to the ears so the dog can look excited or happy when they are moved up and sad when they are moved down.


https://www.youtube.com/watch?v=BqtqEbOilfs

https://www.youtube.com/watch?v=eK7wp2xWOo4

Dog rig

The dog was left until last because I was focusing too much on experimenting on the shark. But when I go the dog model and started rigging it. My thoughts changed and I started enjoying the whole rigging process again. While creating the skeletal system I used lots and lots of references compared to the other two models, which I feel that I only messed up on the front legs where I added an extra joint. But apart from that the joints just seemed to be falling into place after I looked at my references.

I added set driven keys to everything I thought should have one including ones that I would normally put on a human just not making their movement as drastic because of the minimal movement dogs paws and legs have compared to a humans. I really enjoyed doing all the different set driven keys. Which I took care in to not mess up and time to work out which is the driven or driver so I wouldn’t waste any time. Many people would find them difficult but I found them really easy when I got back into it. Im disappointed in the animations for the dog but I was trying too hard to make everything else prefect I forgot about the animation. So it’s a bit rushed.

Shark Rig

When modelling the shark I used quiet a few images that I had picked up in the primary research section of the project, which I thought, would be enough for me to be able to create a decent model that would be somewhat look life like but I was really wrong. My model looked like a squashed toilet roll with some triangles pulled out. I put a few hours into that model trying to make it look round and more life like but with not avail.

Later on when I got the model from online I could tell the differences between the two and where I had gone wrong. The main problem with my model was I wasn’t looking at it being circular more of a semi circle, which made it so I was losing a lot of the shark’s density and bulk about the model.

I had no major problems with rigging the model. Thought I did waste a lot of time trying out different skeletons when I had made probably the best one on my first try. I left some of the experiments in the project just in case I had some spare time at the end so I could experiment with them using deformers and maybe placing a few IK handles on it. Overall I would say the rig was good for the knowledge I knew at the time but now I know about IK splines, which would give my model a more curvy and wavy movement, and its main use is for that kind of thing. When weight painting the shark it didn’t need much doing it too it just making some of the dorsal fins stop influencing certain places. The animation was simply but easy until I tried to do an attack animation where everything seems to crash down on me. This is where I realised that I should have made the tail a separate thing and makes it more like a pair of hips than just have it all combined together.

Human Rig

The modelling I did for human was really bad in my eyes mainly because I don’t have much confidence when it comes to modelling so I would probably not like it even if it was prefect in everyway possible. I wasn’t using topology for any of it I was using references but they only get you so far when you’re trying to model something. Ideally using the references and the topology would of made my rig a lot better and be a bit more professional. Also the legs were a bit too thin and ropey for the body, which also makes me think I need to look at proportions some more to make a “prefect”.

I didn’t even attempt to UV map this model which I probably should of seeing as I was trying to increase my abilities within that field as well but I later changed from my own personal model to a model I got from online.

The model I got from online when I first got it seemed to be what I needed it showed muscles the body and topology seemed fine to be me but later on I found it wasn’t the best model I could of selected.

When rigging the human I virtually copied a real skeleton but added some extra joints to make it more animation friendly. For example a real skeleton doesn’t have a bone in the middle of the hand but I put one in the middle to make it easier when I wanted to do set driven keys and to make it so I have more influences in the hand when moving it. But after binding it the model I thought was really good turned really bad. It had too many vertices so I couldn’t affectively weight paint it to the standard I wanted it to be. I was able to animate it a small amount but cracks would appear all of it whenever I moved any limb on the body. I later found that this is most because the model was just to stand there and look pretty not to be animated so all the time I took to rig, weight paint and animated was lost because I thought I had the prefect model but it wasn’t.

On the bright side of things I think the animations are pretty good even though the model is cracking everywhere I feel that I did a good job on at least the animation part of this rig.

Rigging/joint systems

Rigging is adding a digital skeleton to a 3d mesh. It’s basically a real skeleton within the mesh with the rigger adding joints and bones, which can be used by animators to bend the character/ 3d mesh in anyway or pose they want. A rig can take many hours to get to create even the simplest ones where it’s only using a few joints.  But when finish a 3d model will be giving to an animator for them to work their magic on it. In theory the rigger has one of the hardest jobs by having to do multiple things to make the models move correctly and so it’s an easier job for the animator.

Joints chains/joint systems are the same as bones and joints within a real skeleton. When two joints are created one after the other they are joined and make the start of a chain. The first joint in the chain is the root joint which when select can move the entire chain. But if you select the joint below the root all other joints below the one you selected will move as well. The join hierarchy will eventually get bigger and when you add other chains/systems to each other where you have to remember which one is your root so you can easily move things about.

When a joint system is bound to a 3d mesh they have to do a thing called weight painting when the 3d mesh is freshly bound to the skeleton the weights can be all over the place . So weight painting is changing that so everything can move more affectively. You do this mainly because sometimes strange thing can happen where the shoulders can effect near the stomach area on a rig. It’s really simple how to weight paint or it is for me at least. You have to decide where each joint has influence on so no strange things happen. You don’t want to animate the legs and a few faces from the shoulder move up and make some spikes. I use the weight painting tool in black and white mode because I feel that is more helpful for me. White is the maximum of influence and black is no influence in the tap its 0-1 so having a 0.5 is a grey that gives an ok bit of influence to the selected joint.

The link below has helped me a bit to get back in the groove of things with rigging and to find/see if there are any new tips and tricks around I could use.

http://www.digitaltutors.com/tutorial/502-Beginners-Guide-to-Rigging-in-Maya

Curves/controllers

I will be using nurbs curves/ shapes to mostly make all my controllers adding a text letter here and there in case I want to have a certain think being controlled. For the feet and around the body I will be using circles because they are one of the easier shapes to select when animating and it seems to be the general thing to do when I was looking at other people’s rigs. For the shoulders I will either be using an upward arrow or a curve in a tunnel kind of shape to go over the model. This is most to show the animator what this is actually going to affect when they are moved. For the head and neck I will be using and capital H and N respectfully so there isn’t any confusion when animating.

Another thing I noticed when I was looking at other peoples rigs that their controllers where coloured. The controllers in the middle of the rig were yellow, controllers on the left were red and controllers on the right were blue. Which I will be adding to all my rigs because it seems to be common practice and if I wanted someone else to animate my rigs it might make it less confusing.

http://www.youtube.com/watch?v=o6Mfpdspo2c