#anm220

Oh, real time.  I have such a love-hate relationship with you.  One one hand, you are responsible for every game I love.  But on the other...  

You can suck to work with.

The large team based project last trimester was based in a real time platform - Sketchfab.  In addition to this, we also had to provide assets for the games students (and will again).  So, I now have a bit of experience with how my own workflow needs to change in order to work with real time engines. 

First: poly count.  Learnt this one the hard way.  While 3ds Max on a half decent system can handle a model with thousands upon thousands of polygons, most real time engines are not too thrilled to be faced with that.  Last trimester, the model I made apparently crashed their game engine.  gg Unity.  

I’ll likely never make that mistake again.  Which is a good thing, since optimisation can never hurt.

Rigging remains rather the same.  I doubt I’ll ever purposely want to cut corners on a rig.  However, any issues with the rig and skinning can sometimes be hidden in an offline render with clever camera placement.  There’s less of an option for that in a real time environment especially if it is a game and the camera can be moved freely by the player.

That camera freedom brings me to the next point - animation.  Sometimes in what will end up as an offline render, poses or cyclical animation may only need to look good from one angle - as that will be the only angle the audience will see it.  This is not really an option in games, as you are likely able to view an animation from every angle possible.  So in theory, game animation should be as polished as possible.

That being said, it doesn’t always happen...

Real time animation can also get away with errors that would not fly in a rendered work.  One particular example is when a character runs and the solid armour they wear stretches around with the rig.  I know that I can overlook that in games.  Though if an error like that happened in a film, it likely wouldn’t go unnoticed.  (See: that weird ponytail thing from Frozen.)

Anyway, this ended up as a large wall of text.  Oh well.

---

A particular specilisation of animation I’ve decided to research is character animation.  It is a specific part of animation that appeals to me the most as most of my own work is very character driven.

Naturally, character animation has existed since animation itself was “invented”.  In order to tell even the most simplest of stories, we need a character - human or otherwise.

Early ways of depicting movement with devices such as the Phenakistoscope and the Zoetrope can be considered very early character animation, as they often focused on a human or animal performing a particular looped action.

In the early 20th century, animators began to form their own studios (eh. Bray Studios).  Their works were starting to be shown in theaters.  

Famous characters from early cartoons include Woody Woodpecker, Betty Boop Felix the Cat, Bertie the Dinosaur, and Mickey Mouse.

One of the most notable successes of early character animation was in 1928 with the release of Steamboat Willie.

In the following “Golden Age” of American animation, the industry was obviously 2D focused.  During this time, many animation techniques were developed that we often still use in 2D animation today.  For example, exaggeration, especially of expressions and actions to make a character’s intentions very obvious.

Many features of classic Disney animations are linked with the basic principles of animation that we are taught today.

Over the weeks, I will be looking into character animation more, especially since it ties in well with my studio project.

---

In the first week of this trimester, I’ve been working on my documentation for my project.  I’m finding that, while I hate doing it, it is very helpful to have.  

I also found that when planning out my Gantt chart, that I came to realise how little time there is so get everything done.  That’s why I am aiming to keep my art style and character designs more simple compared to my usual.  That way, I don’t have to sacrifice too much aesthetic value for the sake of animation.

Visual effects started out as in-camera effects like jump-cuts, miniatures or matte paintings. Optical effects were created when artists experimented or accidentally discovered a new way to create an effect using the physical film, lenses, chemical processes etc Many optical effects are now created digitally like fades, blue/green screen and dissolves.

Timeline

1856 Oscar Rejlander – Trick Photograph

1895 Alfred Clark – Stop motion of actors

1896-1914 Georges Melies – Stop motion of film, multiple exposures, time-lapse, dissolves, hand painted colour film, incorporating animation with live action, miniatures and matte painting

1910-20s Norman Dave – developed matte painting compositing effects that were the beginnings of green screen

1920-30s Effects were refined, some taken from theatres eg. Schuefftan process. Beginnings of prosthetics with casts of faces and make-up effects. 2D and 3D (plasticine stop motion) animation were included with masking eg. King Kong 1933. Optical printers, invented for creating copies for distribution was used by Linwood Dunn for illusions and effects.

1940s – Colour films were being created more and more which meant effects had to develop to keep up. Some effect processes had to be adapted such as Norman Dave's matte masking. Ray Harryhausen created and refined more techniques, and is most famous for Jason and the Argonauts skeleton fight sequence.

1950-80s – Effects continued being developed and refined. Of note is Pavel Klushantsev, Stanley Kubrick's Space Odyssey, George Lucus' ILM with effects supervisor John Dkystra, and Steven Spielberg's CLose Encounters of the Third Kind.

1990s – CGI was being developed, becoming cheaper and less resource intensive by the end of the millennium. Prime examples are Jurassic Park (which only had a few digital effects, mostly using practical and optical effects) and Toy Story.

Now

Footage is filmed. The raw files can be put into After FX or Premiere, though sometimes this didn't work because Adobe. When this is the case, me especially, we used Media Encoder to convert* the MXF files to whatever we wanted. In this instance it was MP4's. We then look through the takes and choose one that will work best for us.

Once we've fiddled with the take (cutting around it for our purposes), we start with adding Keylight to get rid of the green screen in the background, and masking out objects out of the green screen. This is a standard industry practise to use this plug-in, though there are other programs to do it as well, but you have to buy those.

Since my scenes has a couple of issues (Azam had plenty too), I switched a few settings in the Keylight to make an inverted matte. This was so I could use the Inverted Luma Matte setting on the layer below.

Once that was all set up we created the effects. I used several tutorials to help me with some, and I extrapolated some aspects by myself.

Rendered out the sequence as MP4s.

Imported them into Premiere Pro to add the sequences together and add sound. Then render out the finished product as an MP4 at 720p.

This is the process I went through to get the GUI for the game done. I had done sketches from what I recall in one of my books, and some on my laptop. This gave me broader idea pf what to do. Drafting is just the stage of making iterative saves and trying things out. I also take a lot of feedback from peers to what makes sense and what works well at this stage. For example: I wasn't understanding how to go about the wood textures for the GUI and then I had a chat to azamatqantm and l00s about the best methods of making textures sit well. Finalising and Checking with the group as to which design is the most suitable and desirable. This was wholesomely dependent on the communication I had with my team members. And Hand-Off is explanatory. I handed the final product or rendering to the Games team to handle the rest.