Interation Summary.

This is a summary of the artwork throughout my creative process. However, text explanations and reasoning and problem solving can be found further down this PDF/blog.

Last minute game fixes & completion.

Fixed the scrolling issue in S2 -- the scene wouldn't pan with a camera follow actor behaviour and the seedling would just stop dead in the middle of the fall. Fixed this by creating a gradient background in photoshop, loading it into Stencyl and setting it to vertically scroll upwards at a speed of 90. This allowed the whole scene to pan without looping at the same rate the seed fell. I also created a separate seed actor duplicate and removed the 'die in pit' behaviour, which appears to have stopped the problem where the seed would stop midway due to scene boundary limits.

I also added a start and end screen. For the end screen, Stina provided the artwork and I followed her example to create a 'Thanks for playing' actor that looped the player back round to the start of the game. For the start screen, I created the text, interactive actor 'Start Game' and used Stina's atwork to create an opening screen. I achieved this by layering the two colours of background over each other and using a soft eraser at Stina's advice to blend the two.

Game update.

Changes:

Managed to fix the jumping behaviour when an actor walks off of a platform (not jumps off). I downloaded the 'Fall Jumping' behaviour from Stencylforge, but as one commentor advised there were a few blips. For one, holding down the jump button caused the actor to continue flying off into space. Additionally, the animation was not switching correctly from fall to jump when fall jumping. I examined the 'Air Jumping' set of behaviour and used information from that to fix the issues with Stencylforge's Fall Jumping.

Stina implemented basic scene transitions. These work, and I set them up to point them at each subsequent scene. Reloading the behaviours seems to have also solved problems with single frame animation blips.

v1.3

Game revisions.

Changes to game:

Put in some of the tiles Stina made, and updated the collision boxes so that the seed doesn't fail to fall from a platform until you walk a few steps into mid air. Also corrected an alignment issue with the fall/jump animations not looking smooth.

I've created a thread about an issue I've been having with air jumping behaviours within Stencyl, to hopefully get some community help.

Help thread.

Latest version.

Triple jumping fixed, animations smoothed a bit. I was told the animations work well, especially when cancelling out of a dash.


v1.2

Changes.

Run, jump, walk, and fall animations all implemented. I can also toggle a double jump option on and off, and adjust jump values for height and such. Using the platformer behaviour set saved me a lot of grief there, as I don't think I could have done the 'coding' as well as creating and smoothing all of the animations. I discovered the in-engine editor, Pencyl, and it's wonderful. I was able to adjust the placement of each frame of animation from there, without having to continuously re-import my sprites.

Things to clean up if I get  the time: Movement into the fall animation looks a tiny bit jerky.
Additionally, my game tester tells me the seed is triple jumping. I'll fix that too.

v1.1

BA5 Changes, Etc

Dash implemented.

Things to fine tune if I have the time:

Slight forward jerk when pushing into the walk animation. It looks like the first frame is misaligned. It's only heavily noticible after the character drops from a run into a walk.

EDIT: Fixed the jerking walk. It was caused by a pixel difference on the walk right animation. However, the dash animation has a similar problem.

Helpful Troubleshooting Links.

Origin Points.

Collisions/Falling.

Animation Manager.

These were all very helpful in my initial troubleshooting stages.

BA5 - Changes

I managed to solve the aforementioned problem by importing my frames in as small sheets, instead of individual frames. Stencyl splits each frame up well enough, and seems happier with it when I do this. I currently have my character walking and idle, using the platforming behaviour set Stencyl provides. I encountered some issues getting this off of the ground, due to behaviour problems.

An explanation of behaviours, as far as I can tell from tutorials and help sites:

Camera follow player: Means that the game scrolls with my character, as opposed to my character making his way through static screens that change when you reach the edge. This will probably have to be modified when I place my character in Stina's environment, as she's been talking about having the character reach the edge and causing the scene to transition around it. For my own tests, it's suitable.

Walking: Walking behaviours, including the animations, speed, controls, and acceleration.

Always Simulate: Continues simulating the actor even when going off screen. Prevents character freezes when the player actor moves outside of the screen. I might need to experiment with this in Stina's environment, due to the above transition behaviour.

On Ground: Necessary for the walking/jumping/dashing animations to function -- determines if a player is on the ground and sets appropriate attributes. This was one of the reasons my character wouldn't walk originally, but not the main one.

Die in Pit: Respawns the actor and resets the level if they fall into a pit.

Animation Manager: Prioritizes and manages animation requests and changes. Not having this enabled was the reason my character was originally walking by gliding eerily across the terrain in his idle animation. The walking animation did not have priority over the idle animation without this behaviour, so couldn't take over.

A cap of the animation screen is above to show my functioning frames. I had to tweak the origin points a bit, as Stencyl was setting custom ones and causing my character to jump a little whenever I changed direction. The walk is simple, but I have added detail in the movement of the character's shoots and the realignment of his eyes.

Here's a link to the game in its current form. I'm using a background and tileset that came with the program, so as not to have to worry about creating those just to test my character.

v1

If the link doesn't work, manually right click and open in tab. Pop up blockers don't like it.

BA5 - Troubleshooting

Having now finished my animations, I've started importing. Immediately I ran into a snag in regards to the consistency of sprite placement, as Stencyl doesn't allow you to redefine character borders. As far as I can tell, having character borders below your character mean that it includes these borders within the character's animations, even if the excess area is transparent. While I could work around this easily if it was the only problem, Stencyl also disagrees with animations that are of a 'different size'. For example, my dashing seed has a couple of frames wherein it is wider than the rest of the images. If I were to make my image size bigger to accomodate that, it would cause issues with centering and collison detection, as noted here:

It is very irritating.

More Progression, BA5

Just creating the various frames to be used in Stencyl.

Art Style Experimentation!

Messed around with pixel art for an hour or so. I don't think it's viable to animate, unfortunately, as I'm shaky enough with pixel art as it is. Cute, though!

EDIT: Nevermind.

BA5 - Another redesign!

After I put together all of my poses, Stina volunteered to throw them together into a gif animation on photoshop. It took a long time and a lot of lacking motivation in order to throw together the poses for the animation, though it came out fairly well. However, I was intimidated by the notion of trying to animate a run or a jump for the same kind of anatomy, and was worried about the time it would take me to do so. I felt like animating a walk alone wouldn't be enough, as I wanted our game to be fairly explorable, in order to give Stina enough room to show off her own artwork. With a walk, I felt like we would be restricted to a single flat plane to explore.

Original animation, unfinished.

Having been doing a little more study into 2D platformers, I noticed that many of these platformers used a 3/4 view of sorts, or at least had both legs visible. I wanted to incorporate this, as my current profile animation felt very stiff and stunted. I began another redesign, basing it on one of the most simple shapes possible - a sphere. I returned to my seed studies, and found one in this:

I took some of the features of my design, and placed them upon this new base.

A circle was incredibly easy to animate in various poses, and I quickly came up with the idea of having the creature move in a roll to dash, in order to get into practice with the concept of animating something a little more.

And with Stina's handy photoshop skills to preview:

Animation!

While Stencyl takes each individual image to make into a gif on its own, I felt it was really good to be able to see what each animation will look like without the fuss of importing into the engine and fiddling with file sizes. With the body of my creature now so simple, I could focus on the little things with less worry, such as the physics of weight on the creature's feet and its sprouting shoots. Stina also refrained from doing any between frame tweening, so the animation above should be the same as it is within the engine. Satisfied, I began cleaning up my lineart and adding further details like wind motion:

BA5 - Animation Pains

Have been struggling with the ins and outs of animation. Thus far, I've managed to put together a few poses and determined a vague order, though I've yet to test it in game. Now I've started trying to get opses and motion into my character, I've begun to feel very intimidating by the GameSalad and Stencyl engines. I have no inclination towards either but Stina seems to feel more at home with Stencyl, so I took a look into a quick interactive tutorial. It's complex on the surface, to say the least.

Stencyl tutorial.

I also looked into some more animation tutorials, specifically for upright/humanoid movement. I wanted to keep whatever I did simplified from a human walk, though, as a realistic human walk has so many nuances in order to make it believable.

Animation Careers: How to.

From the above video I took to heart the concept of tracing over previous drawings in order to keep the artwork consistent, and also added some extra poses to my walk.

BA5 Project 1 - Further concept refining

After speaking with Nigel on Friday, I came away with some new considerations for my character design when it came to animation. Since the creation of my character, I had begun to realize that even a character like the 'hamster seed' that I designed would be very difficult to animate at my level of skill and drawing consistency. I had already recognized an issue I have with keeping something looking similar in multiple poses and positions. While I wanted to work on this a little, I knew it was unrealistic to attempt to do so and also successfully place my asset within a game engine in the time left to me.

Nigel spoke about creating a character that still was 'seedlike', but would be less fluid/animalistic in animation. Having viewed a few videos on animation for animals, I was inclined to agree.

Run Cycle.

Another.

Nigel suggested using a design that was more upright and stiff in nature, which would mean less animation would be required to create something smooth than if I had stuck with an animal design. I set out, at first, trying to keep my drawing looking consistent through opposite stages of the walk. While this design is very simple, I think it has a similar look to the previous design, except upright. Another thing I decided to implement into my design was the inclusion of more a more flowing interpretation of the leaves on the creature's head. While this makes them less clearly leaves, I feel it synergizes better with Stina's twirling, entwined background. I left the body of the creature blocky and simplified, so it would still stand out against what was behind it.

I also spoke to Lothar, and eased some of my confusions/worries in regards to the brief. While it was originally thought that we would have to design a fully realized level, it seems as though just implementing a functional asset within even just a screen's worth of space would be acceptable providing it was interactive and worked well. This is a relief, considering I thought I would have to create far more terrain than I now do.