During the morning class we were taught how to do a run cycle with Alec and he told us that it should take less frames to do compared to the standard 32 frame walk cycle we were taught. In Alec’s example run cycle file we had to fill in the missing up poses before adjusting the timing so the animation doesn’t go by too fast, and later in the class he showed us a comparison between the standard run cycle he made and a panicked run cycle made by rotating the back and having the arms in the air as even though the leg animation is the same the animation of the upper body is completely different and conveys a different emotion and meaning to the audience as a result from these changes. He advised us to not do a generic walk or run cycle for our submissions and to instead be creative and experiment with them. Teaching us how to do a run cycle was quite helpful, and it made me realise I still need to make some major adjustments to my fearful run cycle before the blocking for it is complete as the reference I used for the timing wasn’t as reliable as the timing Alec used in his animation so I’ll adjust mine accordingly.
The rest of the class was spent doing the in-betweens for my weight animation including animating the hair, and when Alec came round to see it he said the feet were moving oddly and didn’t look right as both feet shouldn’t be off the ground at the same time when moving, and advised I record myself doing the same pulling actions to see what my own feet would do in that situation and use the video as reference for the feet movements.
During the week I recorded a video reference for the feet movements to help with the in-betweens which is shown below:
Although it’s not entirely accurate to the animation since I didn’t bend down as much which affected how the legs would act, I was moreso focused on how the feet themselves would act in terms of what foot would be on the ground and how the other foot would move while the body is balancing on the first foot on the ground which the video reference does convey so it was still quite useful.
The playblast for the in-betweens is shown below:
Having a video reference for the feet movements helped a bunch when I initially thought it wasn’t necessary to have because I had still image references but ended up needing it to make the animation feel more authentic in movement. I’m not completely satisfied with the hair movements as I feel like it does these weird jagged and jittery movements in some parts of the animation, but I’ll see if I can improve it during the polishing stage so it’s not as noticeable. Overall I believe it’s looking really well and the animation is almost complete, with the only adjustments in the polishing stage being the aforementioned hair movements, making sure the mesh of the body doesn’t poke through the clothes, and potentially adding facial movements with the eyes, eyebrows and mouth as a final touch for added authenticity to the animation before sending it and my other polished animations to Alec for any last minute adjustments I need to make before submission on the 7th January.
During the afternoon class Sorcha showed us how we would go about creating 2D characters to be rigged in After Effects and showed us examples of the characters she already created. The characters would be in ¾ poses and she explained that it’s because the ¾ pose doubles as a front pose as by moving something like the eyes it looks like the character is facing you in a front pose. The size of the canvas you’re drawing the character in should be something like 4000 x 4000 pixels rather than 1920 x 1080 as it’ll prevent the character from getting blurry or pixelated during closeup shots. The characters would be rigged using an After Effects plugin known as DUIK which makes rigging in After Effects a lot faster.
In preparation for next week’s class, we had to watch a series of tutorial videos from Henry on blackboard about how we should go about lighting our environment scenes through either adding lights to the scene like point lights or directional lights or making an object in the scene emit light by using the material editor to create an emissive material, the latter of which I’ll likely use for my 3D environment as the primary light source of the scene is the map of the constellations on the roof shining down on the rest of the observatory. He also included how to bake the lights in the tutorial so they don’t have the “preview” watermark as the lights would look different when baked and optimising lightmap density by overwriting the light maps by powers of 2 (64, 128 etc) until the objects turn green which means that the objects are lit in a high enough resolution, as well as having the light bounce off walls and objects in the scene.
Another thing he mentioned was using post process volumes to increase and decrease the light intensity in real time if the lighting needed adjustments instead of manually adjusting the lights and tediously baking the lights each time to get the lighting right as the post process volumes make it a lot easier to adjust the already existing lighting. Finally he showed how to add a camera into the scene and having the camera focus on certain objects as the camera moves for each shot.
References Of The Week