Week 12

Week something or other, but academically, this is the last post before presentations next week! Now, this doesn't mean I'm going to stop working on this, I'm having too much fun with it for that. But for now, I have to present what I have so far and I'm confident that I have a pretty good set up going. Now, onto what happened so far.

 Now, here is what took up most of my time at the last minute before posting this. A bunch of "centipede1:"s in there. I had to go through my entire outliner and restore the proper naming conventions to everything. This took me (no joke here because I had no distractions and I was keeping track) 1.5 hours. Sweet cheeze wiz crust (I'm trying to stop saying Jesus Christ so I'm saying Cheeze wiz crust now). Anyways, I know that when you import a project, everything contained in that import get's their project name placed in front of it. The problem is, I never imported this project at any time except for the head last week when I was struggling to fix the antennae. The only thing that comes to mind is when using a different machine that also has maya, something occurs to cause this.


















I went back and looked at several videos of centipede rigs (not a lot of them are as intricate as mine) but the one video that has been my inspiration through this whole project gave me numerous ideas for this rig. That video is the only one I've seen where there are pole vectors for each leg, and so, I did just that.

I also went back and did a little regrouping of certain controls and re-parenting. One area was the rear legs. If you lift the back section, the legs should come up since these legs aren't really used for walking.

I also went and recreated the pincer controls since the antennae issue has finally been fixed (i had some more problems over the week but I solved it through proper parenting.

I even went in and included a flare (I can't remember if I made that already, but I did it anyways and perfected it.) I also put rotation limits on the pincers so that they wouldn't disappear under the head.





Voila! Everything rotates properly, no stretching, no errors, no controls going whacky, no nothing!

Next week's post is going to be a two part, one for my presentation on Monday and another on Friday as usual. Monday will basically be a summarize of everything that has been done so far, so no changes will be made, but if you want, give it a read!

Week 11

So this week was a big humungous mess of troubleshooting and just trying to fix issues. With only a week left to work on this before presentations, it looks like I won't have this done before Summer. Good, because I'm having too much fun with this thing to just call it quits when I turn it in.

Last week I ran into some issues with my antennae. The first problem was when I would rotate the head, the antennae would over shoot the rotation and thus exaggerate the movement. This was a clear sign of a double transform. I spent several days trying to figure out why this was occurring and it wasn't until I sat down with a peer that we figured out the issue was due to a poor pivot point. Returning that to a better spot fixed the issue. Sadly, that did not remove the largest issue I had. I would move the head up or down, but the antennae would seem to want to stay in one area despite moving with the head. To clarify, I would move the head up, and the antennae would seem to begin to sink as it moved up with the head, causing a weird stretch. This took me days to figure out and it wasn't until I sat down with my professor and spent a good amount of time until he was able to figure out my problem and create a fix.

In the picture above, you can see two control rings, but their control pivots seem to be in the same place. It is because there are actually two joint nodes, one on top of the other, but still connected to a hierarchy. I never thought to use this little trick despite just recently learning about it. So I did a parent constraint with the joint leading directly to the pincers and antennae to the second controller and then moved the antennae controllers underneath (parenting them to the head controller.) Before I did this, though, I had to reconstruct the head's skeleton, ik splines, and controls due to some serious alignment issues that could not be ignored. This project has been proven to be a pain in the butt, but has surprisingly been really fun to work with. Everyone tells me rigging is a pain and they try to avoid it, but it's a necessary thing, and if I can get past how frustrating it can be, it's actually really cool!


I don't even mind weight mapping. I may not be the best at it, but I just have to learn and get better. weight mapping an inset is one thing, because it's mainly just segments, not as fluid as a human being, so I have to ensure that a single joint controls a large portion of the geo to make it appear stiff and mechanical.

As you can see, that line of thinking is paying off, and now I have a row of mechanical legs ready to tread some dirt.










Sadly, with only one week left, I won't be able to have this thing completely done, but this only means I get a nice little side project to work on over the summer. But, for now, I have to recreate the controls I originally had before this incident, and see if I can get back to texturing, and hopefully, animating!

Week 10

Alright so I was doing alright, taking my time painting the skin weights for the centipede when I encountered a problem and I don't even know how I caused it.

The weight map will no longer show up on my models. I click the geometry and then go to paint skin weights, but no map shows up. I select the joints to the left, as you can see, but in the view port they aren't selected. This is a very troublesome problem and I may even have to take some drastic measures to fix it, which I don't wanna have to do. Unfortunately I've been trying to Google my problem all day and I have grown exhausted from searching.

I was able to at least paint enough of the models so I can just copy the weight map of one segment to the next, but I'm not sure how that will affect the map and which joint is affecting what area, so we will see.












I had a problem for awhile where the weight map of one pincer would affect the tip of the opposing one. Even painting the weights was not helping, so I had to do something a bit more precise. I went into the component editor while selecting the affected vertices and readjusted the values of which joints affected which vertices. Now they work beautifully without moving the tip of the other side.

I added a little flaring attribute to the pincer controls which extends the pincer out even further, but only for half of it's length since there is a segment in that region. This will enable the pincers to have a more grasping appearance when animated and can even look like it's "chewing" on prey.








I have been doing more research into how I will be able to rig the leg movements when the creature walks and I have an idea as to how to do it finally. I will have to create an expression that, after keying one leg, will affect all the others in the same row a certain way. I still have to get clarification on certain areas, but one of the codes has something to do with timing and placement. Coming down to the wire here and now it is officially announced that there are 3 weeks left in the project. Guess I had more time than I thought. Next week, I'll fix this weight map problem and do some more texturing along with starting the coding process.

Week 9

Like I said last week, I've lost count as to how I am marking the weeks, but I do know I have at least 3 weeks left until I am presenting this project as a final.

Last week I said I would do more texturing for my next post. Well I lied and decided to move forward with rigging the centipede further since rigging is my main goal here. If I can get that done, then texturing is secondary. Here is what I have so far.

 

So I had a little problem with placing the IK (inverse kinematic) handles on the legs at first. On one side they would work, but the other side, they would wig out. Then I found out that when I had made the joints for the legs, I had used symetry to create them, which meant that when I selected the side I created, the side that was copied was selected as well. I deleted one side and mirrored over the remaining with the mirror joint option. That fixed my solution and now both columns of IKs work splendidly. I then went in and created a control curve and had the handles parented to each control respectfully.

I created an IK spline controller that would allow the body to move like a tentacle or snake, which is technically what a centipede looks like when it moves forward.

Antennae are essentially the same thing.

I created a rotation control for the head. It's not perfected yet, but the basic function works. I had a little trouble because I had a "double transformer" (I hope that's what it is called) that would cause the skeleton of the antennae to over exaggerate on the rotation. I figured out that the spline curves were grouped under the head rotation. Once I removed them, everything rotated normally.

I also added rotation controls for the pincers. These are also not completely done, because I'm gonna add an attribute that allows them to flare out a bit more.

 

 

The back legs on the tail end I decided to have a sideways bend. It's an optional thing, some centipedes have sideways rear legs, others have legs at an angle, and a few even have some that curl downward.

After all is said and done, I eventually would have had to go in and rename everything. I've gone through everything twice, if not three times. The first time renaming things, and the second or third time changing the name to something that would be easier to understand

Next week I'm going to be perfecting the rigging controls and create some attributes for certain areas. Here's to another week. I need a beer...