Year 1 contents section

This post is a contents section linking back to previous key submission posts from my first year of the Nextgen course.

Room project

Inside the Box: Final renders submission post

ImAwesome Ident

Original ident: ImAwesome Ident project: Ident submission post

Improved ident: ImAwesome Ident project: Improved animation submission post

Break the Cycle – Walk cycle

Break the cycle: Walk cycle submission post

 

Advertisements

First blog post

Hi, this is my introductory post to this blog. I am Rachael Wingate, a NextGen student who is studying animation, 3D modelling, games design, programming, concept art and VFX. My main areas of interest are animation and 3D modelling, because it is my dream to become an animator, and my ultimate career goal would be to become a professional 3D animator. I expect that this course is really going to challenge me, I am completely new to all of the software and I know that the standards are really high and industry driven, however I am excited to take on this challenge and learn along the way. I love to be challenged, and I can’t wait to learn more about all of the above subjects.

Creative and technical model: Retopology

This morning was spent retopologising the base model, because there was too many tris on the model itself, and this would have caused trouble with animation in the future steps.

I did this by using the multi-cut tool to connect the vertices in messy areas better, deleting the unnecessary edges from the model, and merging any vertices that didn’t need to be there. I spent some most of this morning working out the wisest way to do it, and retopologising one side of the model. I then deleted the other half of the model, and finally worked out how to use the mirror tool so that I wouldn’t have to worry about connecting the model pieces and vertices.

I then began personalising the model to the character that I am making for my creative and technical model, Lilith. I will likely make Zach’s model around the same stages that I make the Lilith model, however the Lilith model comes first in this case.

I started by making her shirt. To do this, I extruded most of the torso area out of the model, and made a collar part by extruding faces out of the top of that previous extrusion around the neck. I then put an edge loop around the torso so that it would be possible to extrude out the corset part. This is all of the progress that I have made throughout the college day.

Lil shirt 1.PNG

So far, I am feeling confident about how this looks. Before the model was retopologised, I was having trouble thinking it was good, however now that the geometry has been fixed, it looks more like a professional model. My next steps are to make the shirt look more like a shirt, then make the rest of the low poly clothing and hair as soon as possible so that I can rig her up and get started on the animation while I work on the high poly version.

Imagined Worlds: Self assessment time!

So far in our imagined worlds project, I have been required to do some character re-designing, character 3D modelling, and in the next few months will be the lead animator of the project.

So far, I have managed my roles relatively well overall. The character re-designing worked out really well on my part. I remade some characters that we had designed previously into chibi like characters after talking through how we wanted them to look as a team. I have also been making the character models, which has been a very difficult task that for the most part has gone well, however there is still a way to go. The base model is made, it’s just that the actual individual character models haven’t been made yet.

I hope to finish these by the end of the week so that we can start animating as soon as possible. To complete this task, I just need enough time to actually make the assets. I’m probably going to extrude the clothing from the model itself, and then make separate models and parent them to the main model for the hair and the bow on Lilith. I’m not relying on anything else from my other team member for this, one of these models is my creative and technical model, and the other is important for the animation because it can’t be started without it.

Character re-vamping

Chibi Zach and Lilith.png

These characters were originally designed by both Ebonny and I as lifelike proportioned characters. After discussing styles and how to simplify outfits, I made this redesign of those characters, and this is what we have decided to make the characters based on. I feel like this turned out well because it would have been difficult to get wrong, but I’m happy with how they look nevertheless.

Model planning

This slideshow requires JavaScript.

This doesn’t look particularly neat, but it’s a plan for how I’m going to make my creative and technical model, Lilith. There is the plan for the base model, and the plan for what she will look like ideally at the end. I like how it looks, but I think some annotations would massively improve it.

Base model creation – Current progress

4th boy soft lines

This model has been a challenge so far, because it started off as literally a cube and a super low poly sphere. I made the head shape, then made the body, which for about 2 weeks looked like a lego figure. It now looks like a person and is smooth, so I’m very happy with where it is. It needs to be made into the two different characters next, and hopefully this will be done to a good quality by the end of the week!

Emerging technologies essay- Research collection 2

After trying to write the essay with the research that I had collected previously, I felt like there was still some holes in my research, so I decided to do some more research in order to support my essay. Here are some more of my findings!

GDC: The Future of Art Production in Games – https://youtu.be/7Rt0wOyCCAI

This talk was mostly talking about how methods of art production for games has evolved since the NES colour pallettes, and talked about how our 3D modelling techniques now will likely become outdated eventually. The talk then went onto photogrammetry, and talked about how using photogrammetry is phasing out the need to make 3D models by making photo-realistic models from pictures. They then talked about simulators and how potentially people could phase out thinking about polygons and vertices, because simulation generators make the need for that knowledge obsolete. This also applies to character models, because there is software out there that can be used to just generate characters. The talk then turns to algorithms, and how animators’ jobs could become obsolete because of deep learning neural nets with motion capture software that can capture an actor’s face, then store the information into a network of facial movements and gestures that an algorithm could then use to chose the necessary movement for the job, and have almost game ready animation. Artist’s intent was another big topic of this video, and how sometimes with all of this software that makes execution quicker and faster, sometimes you can miss the artist’s intent of the work because the assets were made too fast.

GDC: Believable Humans (In Video Game Animation) – https://youtu.be/Aj-zNjff7wY

This talk mostly started with the presenters explaining their past with animation, and they then went on to talk about their contribution to The Last of Us. These animators created the cutscene “Joel’s loss”, and after watching the cutscene, they talk about the emotional impact that even to this day they get from the cutscene. They then showed a recreation of the cutscene that they made in an 8bit style, and talked about how this version didn’t have as big an emotional impact as the first one, and how this links to the execution of how they made the original cutscene. They then started to talk about how photo-realistic animation often dips into the uncanny valley, and talked about how the animated film Beowulf has a significant place in animated history, for being one of the first films to use a photo-realistic animated character in a photo-realistic world, and then explained that the movie struggled to get an audience because of it’s uncanny valley. They then said that when it comes to photo-realistic digital animations of real life actors, “you know something’s wrong if it’s not exactly right”, and pondered whether if all of the characters were completely made up characters, if may have been easier to watch because we don’t know a real life version of that interpretation of Beowulf. They then talked about Uncharted 2, because the animator joined Naughty Dog just in time to start working on this game, and they talked about how they got a much better reception for this game because it is hyper-realistic. They talked about how they were brought into the production of an unreleased PS3 game, and how they were able to completely overhaul it just by using different techniques to make 3D models and blend shapes for animation with facial capture. They then talked about the photogrammetric scanning of actors’ faces to create a whole library of blend shapes that can be used to create emotions and expressions with only a small amount of tweaking from an animator.

In the second half of this talk, they managed to pull together a rig of a young baby girl and set her up with a simulated neural network so that when the presenter interacted with her, she was happy, and when he left her alone, her stress nerves kicked in and she got upset. They were even able to condition her to respond with happy emotions when they rang a bell, like the Pavlov’s dog experiment.

Creative and technical model: Feedback

Today’s lesson was focused mainly on feedback for our model’s progress so far. We were split out of our groups and partnered with people from other groups to give and receive feedback on our models. The feedback for my model’s progress so far can be viewed here

I then got working on some final steps for making the hard lines on the model smoother. I asked Matt for advice on what caused this and what to do about it, and his response was that too many edges were close together, and this made the lines look sharp even when smoothed. Matt then advised using the target weld tool to merge vertices of some of the lines so that they wouldn’t be there to cause trouble. This worked out wonderfully, and now those parts of the model look smooth. I then shifted some edges around and used the same technique to make the chest less pointy. Lastly, I moved the ears on the model down so that they looked more anatomically correct and appealing.

4th boy soft lines

At this moment in time, I am very happy with this model. I am currently very happy to share this model with Ebonny so that we can get started on making this model into Zach and Lilith, which is my next step.

Creative and technical model: Model smoothing update

Since Tuesday, I’ve spent some time making the model’s body smoother and more rounder. This was mostly done with edgeloops and edge smoothing, so no new techniques were used on this part. I then gave the model a nose and some ears so that it resembles a person more. This was done by extruding those parts and then moving around edges and vertices. Then, I attached the model’s head to the body.

3rd boy re-capitated

I’m quite happy with the way the model looks now. I’m happier now that the head has been attached, because it feels more like a character and less like a lego boy. However, I’m not overjoyed about the hard edges on the model’s head and chest, because they look bad from the wrong angle. This is something I plan to fix in my next steps. My next steps after that will be to make this into the characters that it needs to be.

Creative and Technical model: Model smoothing, and learning about texel density

Today I worked on making the base model less blocky and more life like. We also learned about texel density.

Texels are basically to textures what pixels are to pictures. A texel is a small piece of texture that forms together with other texels to make a texture. Texel density is the amount of texels that make up a texture, so high texel density is a higher resolution, and low texel density is a low resolution. This is important to keep track of, because if the texel density isn’t equal for everything on a model, then textures look weird and badly scaled, which overall makes the texture worthless. It is also important to keep track of texel density so that your scene doesn’t become too memory intensive for small assets that could easily just have a lower resolution texture.

This morning was mostly spent making the base model more rounded. I got some edge loops into the model, and then moved edges and verts around to make it look rounder.

Base model low poly rounding 1

I’m now working on trying to get the arms to look more arm like, and trying to overall make the body look smooth. I used edge smoothing and lots of appending to polygons to try and get the model as smooth as possible. I’m actually super happy with the outcome so far, considering that this morning he looked like a lego boy.

My next steps after making the model smooth are to make facial features such as ears and a nose, and to make any extruded clothing parts that are necessary for both models.

Storyboarding: Week 4 – Game storyboarding continued

I spent some time today making an effort to complete the storyboard that I started on Monday. It took a few hours but I really had fun making it!

Mail delivery storyboard.png

I took the time to remember how I used to play the minigame, and thought about some fun perspectives for the images for each panel. The sketches themselves may not look like the best quality, but I decide to just go with my messy emotive cartoon style for this one. I don’t have much else to say for this post because it’s really just a continuation of Monday’s post!