28/02/13 - Base file version 4

I had it working. It stopped at the right time, it did everything i asked of it. So of course i carried on with the animation, hoping to get it finished this evening. So finally, finally, i would be able to get to actually animating.

I hadn't saved. I should have saved.

I made a mistake and something inside the walk cycle died. Weighting the keys was non reversible and i couldn't take it back. Luckily i'd made a back up save just before attempting to make him stop and deleting a load of keyframes.

Unfortunately every time i try to replicate what i did this happens.

"kill me..."

Lesson learned (again). Save often.

Here's today's version. Other than trying to get him to stop, I've refined the players interaction following some feedback from Kayleigh on my last post (thank you  ^_^) added a camera and somehow acquired a twitch in the players left arm which i'll need to fix.

Feedback always welcome and appreciated.

I feel broken.

27/02/12 - Dissertation progress, life drawing

I had an odd problem when opening my files today. The textures are there.. and loaded but they aren't showing up on the model making them silhouettes... this makes animating difficult.

So instead i've been using this as a dissertation day and again it's been pretty productive. I feel like i'll have enough to show Robin on Friday for our second progress submission which is nice.

Life drawing this evening so i will try my best to upload the spoils later on.

A few hours later...

It's now later on! Here's the drawings! They're a little odd in colour, I've tried to get them to show up the lines to the best of my ability but its hard.

I've done some annotations on the actual work of things that didn't go so well. I swear there are some positive things I notice too though!

26/2/13 - Base File

I want this thing done. It has taken me too long but it shows that i need the practice i guess... here is the current version of the base file, feedback welcome. Very welcome.

Still to do on this:

• Create player camera
• Figure out to stop the walk cycle so I can animate the next part of the interaction...

At the moment i have a couple of ideas and have been searching the forums of the 11 second club. But it seems like i might have to duplicate the walk cycle until i want to stop it which would mean changing it in the future tricky.

Here's the forum link

23/02/13 - Feet Practice, Wildcard practice

So last night I went home and did some casual drawing of feet. I'm slowly developing a phobia as some of the references out there are not pretty at all...

I think awkward angles of the feet are easier to draw. It's the side on photos of feet that really trip me up. I figure as with hands if I keep this up a bit I'll eventually understand it better although the fruits of just a small amount of reference practice are that the 14th and 15th one here is from memory. They aren't perfect but at least they have a bit more shape and structure than usual.

Wildcard Time!

I've been thinking as Saturday isn't technically a work day anyway, I might use it for practice that isn't necessarily to do with my project and so begins wildcard Saturdays! This should hopefully stop me from getting distracted during the week by doing things that aren't relevant.

Today i've been working on my modelling skills, i happen to have sat in many of the lower years modelling lessons this week and it's making me want to improve my modelling skills. I don't really rate my modelling skills very highly at all at the moment, i can usually get the model to look like what i want but it takes a long time and i'm pretty sure the topology could do with a lot of improving.

I picked a character from the internet, even though it does ask people not to reproduce it I'm only using it for practice so I hope they won't mind.

Here's the image i've been using as it says on the image it's by Zara Alfonso.

I haven't finished.. I've modelled most of the day sporadically and this is what i have so far. I'll continue it next Saturday if I still have the modelling bug.

Things i'm proud of
There's actually interesting topology on the coat!
I only used 2 triangles in the whole model, the rest are quads.

Things that aren't so great
How long it's taken me to actually get this far

Trying to get the coat to match the concept is impossible and I got confused.. i swear there was more shape to it at one point.

22/02/13 - Walk cycles, progress, walk feedback, lines of action

  

These walk cycles are awesome. Filled with personality and flow naturally, exactly what i hope to achieve with a bit lot more practice. Not really any reason i wanted to put these in other than that. Inspiration and stuff.

Progress
Today's iteration of the base file.

As you can see there's more happening with the "NPC" now.. he almost falls over properly and for now i'm fairly happy with his core motion. Arms need a bit of work especially as in places it's a bit jerky but i think it might be fixed once i get the full flow of the movement. The walk cycle is modified but i'll talk about that in a second.
Problems i'm currently facing is figuring out how to get the "player" to stop walking. Animation layers practice aside, it's hard and the keyframes for weighting the layers look the same as regular keyframes and then you can t delete or move them once they're placed. =/


Walk feedback
I sent a clip of the walk cycle to Lynn for some feedback i knew there were some things off with it but i needed a second opinion. She pointed out the shoulders didn't have the right feel to them and needed to follow more of a figure 8 pattern to fit the build of the character, i've started to implement this but at the moment it's a little exaggerated and just makes him look sassy. She also pointed out some issues with easing on the hands/arms and obviously that the hands look a little robotic. I might want to look a little more at reference of hands in walk cycles to help fix this.

Lines of action

Lynn brought up a video by Kyle Balda which looks at lines of action within animation something that I've seen before in Force: Life drawing for animators which is an amazing read and has been very influential on my thought process when drawing and animating which can be seen below by the drawings i uploaded last semester (i think i'm gonna spend some more time on it). Here's the video link as it wont let me upload it. 

https://www.youtube.com/watch?v=YqyQEH30YEA

Above: Force front cover and some images from inside.

Below: My own drawing influenced by the theory the book provided.

21/02/13 - Dissertation, Animation, Life drawing

Today has been a good day so far. In regards to the dissertation, i've managed to reach and exceed my word quota for the day. Yes, it's not the best thing i've ever written but for the moment just getting the words out of my head is a good start, after all, it's more difficult to get feedback on nothing. I've got a resonable way into getting back to where i need to be with the animations and i'm hoping to be able to start animating the fight or flight tests properly again this weekend with my base file finished and ready to roll. Clip of it's current stage here... sorry for the poor quality video.. had to do it quickly.

Current critisisms.

The walk seems a little floaty, more up and down movement needed.

Hands need to be fixed.

Finally life drawing stuff for this week. It was one long pose, roughly an hour and a half,  which is nice practise at drawing tone and colour but not really what i'm after for pdp as im attempting to get better at weight and posing although practising with the face and hands and feet was nice.



Self criticism
I think the main things that i'd need to improve on from this session is that pesky mouth area. The lips and cheek definition just haven't worked as well as i'd have liked. The eyes are much better than normal and the hand worked out well enough. Choosing pencil crayon to work in was an interesting experience, dont feel the tone went too well in the skin sections, the forehead is too dark and the arm doesnt feel like it has much actual volume to it.

18/2/13 - Presentation Review, 4 week plan

PRESENTATION FEEDBACK

The main points of feedback from my presentation were:

• Possibly separate the factors so they all produce different scores.

This would make a lot of sense as some of the factors may not have an affect on specific interactions, possibly causing an inaccurate  response. So work will need to be done once a list of interactions is formed into which factors affect what interaction and why...add it to the to do list.

• Don't worry about breaking the model.

I kinda already knew that breaking the model wouldn't be a bad thing and that i could probably learn more from making a model that fails than having one that seemingly works by the end of it. I think i'd be suspicious of a model that worked because i'd probably have missed something vital from my research. Despite this, it was nice hearing it from the lecturers.

• Worry less about theory for the moment and more about animating.

This is kind of the point I've reached at the moment anyway. I feel like I've reached a point in my research that making the animations wont be premature.

• Self criticism for the actual animations

There was a suggestion of looking at acting DVDs to help analyse the effectiveness of my animations which i'll look into later this month. 

I was the last person to present this time round which meant the lecturers had been listening to presentations all day, they had collectively managed to run over by about an hour so understandably they'd started to become a little brain dead so i'm very grateful for the feedback i was given.

FUTURE PLAN

The thing i like most about presentations is that it forces you to take a step back from your project. Look at where you've come from and where you're going instead of focusing on the one tiny bit of it that you're actually working on. Its a nice little checkpoint to check you're still heading towards where you thought you were going at the beginning and flags up areas you may need to think about more. There's a small list of things i'm going to need to investigate at some point that will be crucial to my final showcase/presentation that this checkpoint has flagged up. So, so i don't forget them;

• List of interactions

START NOW

To know what i'm going to be animating i need to know what interactions are needed. To do this i'll need to spend a lot of time watching other people play games and making notes of things they do to NPCs. This will be pretty boring and time consuming but its vital. So do it when you have no energy to move.

• Work out a default animation for each of the interactions made

ANIMATE FIRST

For time sake, probably have to use general consensus of peers to determine what would be the default reaction to a situation.

• Test the score model as it is.

ANIMATE FIRST

Without splitting it up into separate parts, like the feedback suggests, use the factors to create score for existing characters from games. Ask around to get a general consensus of how people think the character would react to situations (and why?) and then evaluate it against what the score says they would do.

• Part two of testing score model

ANIMATE FIRST

Because it's going to go wrong somewhere... use the results from the previous test to investigate which factors shouldn't be included in which interactions and try and point out why this is.

I'll add to this is i remember anything else, it seems a weekend was a bit too long to wait...

BONUS 14/2/13 - Presentation 3

It's my 3rd progress presentation tomorrow and i'm glad it's 15 minutes not 10. Robin mentioned it should be focused on practical work but i need the theory behind my practical work up so it makes sense plus I've had troubles with Rigs and things so i'm behind where i'd like to be. I'm putting a run through of my presentation up because when i did it before it helped me find the right words to use while actually presenting so i thought i'd do it again. 

Slide 1

My project is currently going by the name Credible Performance of NPC's within Story-Driven Games.

Slide 2

At the end of the last presentation i was working on a test showing an NPC's reaction to first seeing the player based on their past encounters. My intention was to then try and incorporate other factors such as stereotype and the NPC's personality but i had no real ideas as to how at that point in time.

It was at around this point when i started to write my proposal which killed me a little and took up pretty much the rest of the year to write and then recover.

Slide 3

In the end this turned out to be a good thing because while writing my proposal other factors from my research began to fall into place and an actual system began to form which hopefully i'm going to explain now. To start of with all NPC reactions are influences by 5 factors which are as follows...

Slide 4

The NPC's Individual Personality taking into account prejudices and phobias that the character might have.

Slide 5

Past experiences of times when the NPC may have encountered the player.

Slide 6

The pre-existing in game relationship between the NPC and the players character. 

Slide 7

The reputation the player has gained by performing good or bad deeds in the world.

Slide 8

And the environment the situation is taking place in.

Slide 9

These factors will then provide the NPC with a positive or negative score.

Slide 10

which will then affect the default reaction in either a negative or positive manner changing their animation. This would mean that all i'd need to do is make a list of possible interactions and create one set of animations for each. The library created could then be used by any character of the same build and gender. Of course the actual scores effect on the reaction would need to be figured out probably through trial an error by estimating an existing characters reaction to an event, working out their score and seeing if the reaction that would be generated matches what was expected.

Slide 11

With this in mind, my current mini project is influenced by the fight or flight reactions of the background NPCs in GTA 4.The fight or flight reaction is what happens when someone is put in an extremely stressful situation making it a good basis for major and minor physical interactions such as being pushed or being punched.

Slide 12

To start, i went about researching the existing reactions. Using GTA 4 as my primary source footage i interacted with NPCs and waited to see their response. Generally i found that the main problem was under reaction or over reaction.

Slide 13

This led to this particular interaction needing a 7 point likert scale instead ranging between an extreme fight reaction to an extreme flight reaction.

Slide 14

To actually plan the animations i used a combination of defensive and aggressive body language and live reference footage, which proved to be quite difficult to capture because apparently none of my friends can act naturally. 

Slide 15

I also looked at physical reactions caused by fight or flight that might affect the animation. For example, because it's an instinctive reaction this would cut down on the amount of time allowed for thought processes resulting in a change of timing.

Slide 16

I then created storyboards which also helped to form what would happen, for example the bottom storyboard was for the most mild fight reaction, initially i made it so that the NPC commented to someone near them but what if no-one was there? I tried a second variation which keeps the same gestures but the change in subtlety and where the character is looking changes the action to more of an inner gesture as if they are commenting to them self.

Slide 18

So at the moment i'm in the process of animating the major physical reactions. Things have been held up a little by problems with the rig. I started with the moom rig because i'm most comfortable with it but because i'm making the library for an average build male character the moom rig doesn't work. The second rig i tried had issues when scaling the animation that i couldn't solve so i'm currently using the abertay rig and trying to get to the point i was originally at... again.

Slide 19

Finally, for personal development life drawing has started up again, I've been looking at animation layers as they've recently showed up as a skill gap, I've been drawing hands to get a better understanding of their structure  and natural shapes which should help make my animations better and I've been drawing more faces to improve my knowledge of facial expressions and their meanings.


QUESTIONS!?

And if anyone... anyone reads this and does actually have a question, please ask it. I need all the feedback i can get.

14/2/13 - Animation layers, Rig problems

I haven't posted in a while so here's what i've been up to.

Animation Layers
Last time i posted i was animating the fight reaction for the minor physical section of the library while animating i ran into some issues. I'd never used animation layers before and they were causing me grief when trying to write over the walk cycle i'd prepared so i took some time to try and fill that skill gap. I read the help section in maya and watched a few tutorials on youtube and did some simple animating to try and get the hang of it so i could apply the knowledge in my tests. I came up with this

Each movement for each ball is on a separate layer, so for example the one that bounces has a layer which moves it from one side to the other, one that makes it bounce and one that makes it rotate.


Rig troubles.
I finally listened to my inner voice that kept screaming at me that the moom rig, despite being really easy to use and one that i'm used to, is not suitable for the tests i'm producing. As I've mentioned before, I've had to narrow down what character i'm animating for and settled on an average build male. the moom rigs proportions are not of an average male and even when producing the initial base file i had to compensate by making the brush past be into the arm not the shoulder.

So I mourned the animation I've already done and went in search of a new rig with better proportions and found this one.

 Everything was going fine with this rig, it's slightly larger than normal head wasn't an issue as it would display any facial animation better from a distance and it was fairly easy to use. But then i scaled the animation and this happened.

It went from being a perfectly normal walk cycle to this monstrosity. I tried to fix it by only selecting the Ctrl shapes i was using as well as other combinations but every time i tried it did this. So i had to find a new rig...again.

I remembered Robin mentioning that the university was putting together it's own rig for students to use and so i went and asked Lynn about it. She said it wasn't finished yet but it was usable and if i told her of any problems i had she'd try to fix them. And here he is...

So i'm currently trying to get back to the point of animation i was with the moom rig using this one instead.